User talk:Chris Day/Archive 6

From Citizendium
Jump to navigation Jump to search

Chris' Talk Page

I am an editor in the biology workgroup | 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, Current talk page (94,027)

Useful links on Citizendium

Z field experiment to define layers
Z field experiment to define layers
Z field experiment to define layers
Z field experiment to define layers
Z field experiment to define layers
Z field experiment to define layers
Z field experiment to define layers

.jpg thanks

Hey, thanks - my brain was in space (name-space space :-).

PS: You really need to archive some off the top of your talk page! J. Noel Chiappa 17:13, 7 April 2008 (CDT)

Images version

I think the table version eliminates a lot of needless work compared to an image-based solution. --Robert W King 14:21, 8 April 2008 (CDT)

What worries me is the complexity as browsers upgrade etc. The work is trivial as the number of elements is limited. i wonder whether the aesthetics are improved too as the coded table is much harder to manipulate. I'm very aware of all the coding that has occurred but is it really necessary? Certianly it makes it more versatile to change colours in the future or the elements in each class. I wonder, however, will there be a need to change this in the future? Chris Day 14:27, 8 April 2008 (CDT)
Since we're operating at the mediawiki level I suspect nothing will change much. --Robert W King 14:41, 8 April 2008 (CDT)
Well, if we don't use the thing, at least it was a learning experience. Lets face it, the jpgs look alot better.--David Yamakuchi 14:52, 8 April 2008 (CDT)
Right now :-)--David Yamakuchi 14:53, 8 April 2008 (CDT)
This isn't a fair comparison. We can always add the class of element to the table, and the colors can always be adjusted. The code version is dyanmic and superior, in my professional opinion. Not to mention it's less work, and takes up less space. Plus with style sheets we can make it look any way it has to. This is very web 1.0 vs web 2.0 thinking. --Robert W King 14:54, 8 April 2008 (CDT)
I do like the fact it is dynamic, big advantage, i just question if in this case we need it to be dynamic. If the style sheets can improve the aesthetics then great that will be one of my worries gone right there. If you consider all the images stored it might take up less space but if you consider what is seen on a single page it does not take up less space. See pre expand numbers below. Chris Day 15:02, 8 April 2008 (CDT)

Have you considered the size of the code? Using only mediawiki the this version of {{Elem_Infobox}} has a pre-expand size of 1,900 kilobytes compared to using an image, as in this version that has a pre-expand size of 10 kilobytes. I'm not sure if this will be really detrimental to performance or not but wikipedia has its max set at 2,000 kilobytes for a single page. Chris Day 14:58, 8 April 2008 (CDT)

But that doesn't take into account the subpages system either. WP doesn't use clusters. I'm suggesting that we are extremely limiting our capability if we choose to go with static representations. --Robert W King 15:01, 8 April 2008 (CDT)
I know, we are set up at 4,000 kilobytes. As soon as the limit is hit then citation templates stop working, for one. The subpages template is much smaller than it used to be and one goal was to bring the pre-expand limit back down. I don't remember the exact size of the subpages temlate, I'll check. Chris Day 15:05, 8 April 2008 (CDT)
Well also take into account that this is for roughly 100 articles, and the "extreme version" of the data is on a subpage anyway. Do you really want to take the time to change the color on 100 little colored boxes and upload all of the files? It's a question of time and effort. --Robert W King 15:06, 8 April 2008 (CDT)
I don't change any colours, i just move the black box to a new location. Everything else is in layers. Upload is the time consuming part, but that could be staggered, we don't have this template on many pages yet. Chris Day 15:11, 8 April 2008 (CDT)
Comparison of media wiki code vs jpg version.

From an aesthetic perspective here is the comparison from the two examples on the {{Elem_Infobox}} main page. I'd say it's up to you guys, there are definitely some advantages to having a dynamic version, but to the authors not to the readers. Chris Day 15:13, 8 April 2008 (CDT)

I have a feeling that the concern is just only over "the way it looks"; its a relatively moot point since it can be adjusted at any given time. Is that correct? --Robert W King 15:15, 8 April 2008 (CDT)
It can be adjusted. But obviously not as easily, as the mediawiki code. The real issue is how often do we anticipate adjustments in the future? If alot then I'd go with media wiki. If few I'd go with jpg, especially as it is smaller too. Chris Day 15:18, 8 April 2008 (CDT)
If we're going to overflow somebody's buffers then it's not just a question of asthetics, thats a functional problem. If ele were smaller would it be a more streamlined design pre-expand-wise?--David Yamakuchi 15:24, 8 April 2008 (CDT)
The smaller the better. Chris Day 15:28, 8 April 2008 (CDT)
How's the size now...BTW, how do I find out that number for myself?--David Yamakuchi 15:42, 8 April 2008 (CDT)
I didn't see much change (see table below). By the way these values are more accurate than ones I quoted above since in that example the template was called on twice. The numbers in the diff col. below relate to the template values alone without the subpages template included. In summary the mediawiki version is 42 times larger for pre-expand and 32 times larger for post expand size. Where is all that extra stuff coming from?
Different sizes of Scandium (units are kilobytes)
no template mediawiki version jpg version latest mediawiki version
diff diff
Pre-expand include size: 381 1058 677 397 16 1058
Post-expand include size: 54 382 328 64 10 382
Template argument size: 10 44 34 15 5 43
You can see the numbers yourself by going to any page and using the view source option in the view menu (on safari). in the html code you will find the values, just search for "expand". Chris Day 15:54, 8 April 2008 (CDT)

Catalog for elements

Can we have a seperate subpage for MSDS information, or do you think that should all go in the all-encompassing catalog subpage? --Robert W King 23:04, 8 April 2008 (CDT)

Good question. My sense is that it would be better to have specific subpages. Originally larry was mentioning there might be hundreds of subpages. We have isotopes for example, although experimental, I think MSDS makes perfect sense. The real issues is that for any one topic we want to limit the subpages used, only so many tabs will fit. But if we are below the limit then the more unique the subpage name the better, IMO. Chris Day 23:08, 8 April 2008 (CDT)

Muntzing the design

I have a story you might enjoy...

Back in the days when cars were black and radio still only came in the AM flavor, many people were trying to figure out how to make a television cheap. One early television company was owned by a fellow named Mr. Muntz. Now, word around the engineering campfire was that when one of Mr. Muntz's developers had a design they thought was ready for production, Mr. Muntz wold come to the their bench and take out his side cutters. He would then begin literaly cutting parts out of the TV set while it was running! (I suppose he had insulated He would stop cutting when the set stopped functioning with the instructions..."you can put that one back in". That was how a design was approved for manufacture.

I dunno why I just remembered that story, but I thought I'd share. :-)--David Yamakuchi 23:49, 8 April 2008 (CDT)

The difference being i keep getting electric shocks. I hope it does not collapse. :) Chris Day 23:51, 8 April 2008 (CDT)

cellpadding for Periodic

If we leave the cellpadding=0 in Periodic that gets rid of the "white borders" and I like that. What do you think?--David Yamakuchi 00:43, 9 April 2008 (CDT)

That's what i was thinking. I'd say just keep experimenting until we get the right look. Chris Day 01:11, 9 April 2008 (CDT)
I think we're getting close. I can still tweak some kB out of Periodic, and it looks like you had the right idea with the xxxxx on and xxxxx off templates...But I'm done here for tonite....zzzzzzzzzzz--David Yamakuchi 01:23, 9 April 2008 (CDT)


Nice polio there, right on your talk page. Do you ever get the feeling that we're laying an awful amount of groundwork that probably won't be fully appreciated whatsoever until we're both long gone and/or dead? I'm starting to feel that way. --Robert W King 13:04, 11 April 2008 (CDT)

I figured that I need to get to grips with the z field. That was one of my experiments. I don't feel it will go unappreciated. In fact, I think it is essential if there is any hope in attracting good wikipedians here. In addition it is in these early days that massive style and usability changes can be made with absolutely no protest. Enjoy it while it lasts. :) Chris Day 13:14, 11 April 2008 (CDT)

Template Recursion

Chris, I have a programming approach that can work well for programming simplicity in C, but I'm uncertain as to the possibility of using in wiki-markup with templates.

I would like to call a template, pass it a arbitrary number of strings, (names of links, which I think will then show up in the template as:{{{1|}}}, {{{2|}}}, ..., {{{n|}}} ), and then have the template do something to the first data, but then call itself, passing "in" the remaining data, n times until the data is done...Is there any way to do that? I was thinking something like this:

Consider a template named Template:Myself

{{ #if: {{{2|}}} 
      |{{DoSomething|{{{1|}}} }}{{ {{Myself|{{{2|}}} }} }}
      |{{DoSomething|{{{1|}}} }}

Basically, #if there is additional data, then do something with the first data, and call youself again, passing in the rest of the data.

The problem with the above example is that it does not pass in {{{3|}}}, {{{4|}}}, ... ,{{{n|}}}. Any ideas on how this might be done? Is what I'm looking for even doable? Is there a way to refer to 2,3,4,etc collectively maybe? Is there another approach that does the same thing? --David Yamakuchi 18:25, 13 April 2008 (CDT)

There's some stuff at Help:Advanced_templates which might be useful (it talks about composing argument names, etc...). I'm not sure if there's a builtin which tells you how many arguments have been passed to a template. J. Noel Chiappa 01:39, 14 April 2008 (CDT)

I'm not sure i'm the best one to help here, I normally just muddle through. I'm not sure of the exact context for what you are doing but the way I would do what i think you are trying to do is the following:

{{ #if: {{{2|}}} 
      |{{ #if: {{{3|}}} 
            |{{ #if: {{{4|}}} 
                  |{{ #if: {{{5|}}} 
                        |{{DoSomething|{{{1|}}} }}{{DoSomething|{{{2|}}} }}{{DoSomething|{{{1|}}} }}
                         {{DoSomething|{{{3|}}} }}{{DoSomething|{{{4|}}} }}{{DoSomething|{{{5|}}} }}
                        |{{DoSomething|{{{1|}}} }}{{DoSomething|{{{2|}}} }}{{DoSomething|{{{3|}}} }}
                         {{DoSomething|{{{4|}}} }} }}
                  |{{DoSomething|{{{1|}}} }}{{DoSomething|{{{2|}}} }}{{DoSomething|{{{3|}}} }} }}
            |{{DoSomething|{{{1|}}} }}{{DoSomething|{{{2|}}} }} }}
      |{{DoSomething|{{{1|}}} }} }}

So you're fine if you have a small number of variables but if n is large you're in trouble. BUT, i only about things i have used and that probably represents about 1% of the options out there. Chris Day 09:53, 14 April 2008 (CDT)

I had first posted this to Noel's Talk, but since you and Robert were in on the initial discussion, I thought you might perhas have some insight also...

Well, I saw that where you linked me to the MediaWiki docs it says we can't do this, but this[1] is basically what I was talking about _trying_ to do. So, dey do dat? (Category:Editorial_Council)

Now, the template that is "called" to produce this, {{Editorial Council}} "calls" another one named {{Community}}, and that one kinda hurts my brain...or at least I'm having trouble seeing how we end up with what we do. Thing is, it's not really what we want for this template, I don't really does look to me as if the author of {{Editorial Council}} didn't intend this to be the result. So now it's maybe really two things I'm asking...

  1. how dey do dat?
  2. how do we not do the recursion here, and so get the intended results?

Ain't computers fun!? :^) --David Yamakuchi 22:35, 18 April 2008 (CDT)

20/20 scientific hindsight

Anytime I hear something of the form "oh, that was actually easy", my response is always "if it was so easy, how come nobody else did it before X". The Web seems so obvious in retrospect - but a lot of really smart people were working in this area and didn't work it out. J. Noel Chiappa 01:35, 14 April 2008 (CDT)

Good point! Although there are those who claim that if he'd had access to Rosalind's data, he'd have worked it out too. Still, he's not exactly anyone - he did win a Nobel (I don't count the Peace one as a real one :-). J. Noel Chiappa 14:40, 14 April 2008 (CDT)
Good point about her input. It's too bad she died when she did; my understanding is that had she lived, she'd almost certainly have been given a share (Wilkins was I gather included somewhat in her place, although of course he'd done a lot of work in the area too) - which would of course have gone a long way to disposing of the inevitable complaints that she'd been robbed. Ironically, while everyone seems to complain about "Double Helix", it contains a lot of evidence (as best I recall its contents, without going to check) of how much she helped. J. Noel Chiappa 16:07, 14 April 2008 (CDT)
The double helix was just classic Watson (honest Jim was its original title, I think), even Crick complained about it. He was a showman and that rubs many scientists the wrong way. As we now know, an ignoramus too, with respect to race and sex. He did many good things but destroyed much of his legacy with poorly thought out comments throughout his career. The book of his life should be titled "The double foot in mouth". Chris Day 16:14, 14 April 2008 (CDT)
I didn't say the complaints weren't warranted! :-) Oh, and that rubs many scientists the wrong way - not just scientists! :-) J. Noel Chiappa 16:31, 14 April 2008 (CDT)

Metadata hack

You know how you have it so the metadata page, when displayed directly, says "this is the page for the foo cluster - how about adding a link to the main page of the cluster, as in Foo cluster? I'd do it, but I wanted to check with you first to make sure you're OK with it. J. Noel Chiappa 12:24, 15 April 2008 (CDT)

Wow, that took me a lot less time to find than I thought it would! :-) Thanks for adding all that doc on the {{Subpages}} page. (Speaking of which, I notice it's all still on {{Subpages}}, not {{Subpages/Doc}}?)
BTW, would it make {{Subpages}} any less complicated if the template called at the end of the /Metadata page was some other template? (I.e. {{Subpages}} wouldn't have to have code to deal with being transcluded from a /Metadata page.) I mean, I understand for the article page(s) you want them all to have the same name, to make the whole thing easier for users, but on a generated page like the template, it could be different, right? (And yes, I know it would take a bot to change them all now! :-) J. Noel Chiappa 12:38, 15 April 2008 (CDT)


Yeah, they're gonna be a pain to find; we don't have a category that includes all the bios, right? I saw someone catching a lot of them - if it was a bot that did them, I'll keep an eye out for them too.

BTW, how come you changed "status = 4" to "status =4" - does it not work with the space? That's going to be something else that's a pain to change, if it doesn't work with spaces there. J. Noel Chiappa 13:42, 15 April 2008 (CDT)

ToApprove cat

Hi Chris, I'm not sure if there was something that changed, but after approving Rottweiler this evening, the draft page ended up in the ToApprove Category. Hopefully it's just an easy fix! --D. Matt Innis 19:24, 15 April 2008 (CDT)

Might have been database lag - it seems to be gone now? J. Noel Chiappa 19:52, 15 April 2008 (CDT)
Dang, that was easy! :-D. Matt Innis 19:58, 15 April 2008 (CDT)

I do miss those nice orange dividers you made :( --D. Matt Innis 19:30, 15 April 2008 (CDT)

Checklist_ templates

You probably don't care one way or the other about this, but I'll ask anyway... :-) I've been updating {{checklist_blank}}, which is used on a number of CZ: pages to document the metadata, and created {{checklist_doc}} and {{checklist_basic}} to allow dropping various parts of the text around it. It strikes me that it would have been better to (re-)name them all metadata_<foo>, as what's being documented here is the metadata, now. Is it OK with you if I rename them all? (I'll leave Template:Checklist_blank as a redirect, so I don't have to edit all the obsolescent pages that refer to it.)

PS: Take your time with the obsolete templates; they aren't hurting anything, and I have the ones I've run across all corralled up for now (i.e. they're not wandering around totally lost). J. Noel Chiappa 17:01, 17 April 2008 (CDT)

The names have historical significance rather than functional significance. I agree we should change them all to something more logical. Chris Day 17:20, 17 April 2008 (CDT)
OK, done, thanks. J. Noel Chiappa 17:25, 17 April 2008 (CDT)

OK, what do we want to do with CZ:The Article Checklist? I have gone through it and fixed up all the obsolete stuff (e.g. references to the now-obsolete Category:Checklisted Articles). However, the (historical) name now seems a long way from accurate, since its content is now most about the metadata. Do we want to basically completely kill the term "Checklist"? Maybe this should be renamed CZ:Metadata contents or something like that? J. Noel Chiappa 18:41, 17 April 2008 (CDT)

PS: Is Template:Subpages3 in actual use, or is that just your test article? If it's used, you might want to rename it to something a little more descriptive... J. Noel Chiappa 18:57, 17 April 2008 (CDT)

I can't take credit for "automagically" (creating it, not using it). It's an old hacker (the originals, not the teenage vandals who misappropriated the title) word from MIT; it's in the Jargon file. (For the true defintion of "hacker", look in there.) J. Noel Chiappa 21:35, 17 April 2008 (CDT)
Hmmm. I see what you're getting at. There are definitely subsets of the metadata, of which the "approval metadata" is clearly one. I'm just wondering about the "checklist" part... Let me ponder that one overnight.
PS: Did you see that thread I started on the forums about ditching the "By" field, and using the Metadata history instead? J. Noel Chiappa 21:55, 17 April 2008 (CDT)
Nothing off the top of my head, but I'm pretty tired. Let me give it a think tomorrow when I'm fresh... (I see Aleta wants to get rid of "cleanup" :-). J. Noel Chiappa 22:38, 17 April 2008 (CDT)

wow, this is awesome Chris

Amazing work! Tom Kelly 11:46, 18 April 2008 (CDT)


Can you tell me if the tlsubevent section Template:Timeline/Sample looks all screwed up? Also, can you check in CZ:Howto if the {{tlsubevent}} section also looks screwed up? --Robert W King 15:52, 18 April 2008 (CDT)

Dead templates

Do we want to move User:J. Noel Chiappa/DeadTemplates‎ to, like, CZ:Obsolete Templates, and throw a little text into it? (PS: Still thinking about the metadata division.) J. Noel Chiappa 10:23, 19 April 2008 (CDT)

Oh, another one: It seems like we aren't using Template:ToApprove any more (although Template:Blank metadata still links to it linked to it at one point? I haven't poked around to figure out what template is producing the notice, but it seems from a wording difference that's not it. Of course, since the blank metadata links used to link to it, every metadata page in the world now links to it, so we can't delete it. J. Noel Chiappa 12:12, 19 April 2008 (CDT)

I went through the list at CZ:Templates, and using what-links-here, corralled a bunch of other test/obsolete ones. I notice that in a lot of cases <foo> is obsolete, and we're using <foo>2, or <foo>22, or something. It would be nice to put the templates we're actually using at plain old <foo>. Of course, if <foo> contains history it would be good/useful/interesting to save, we can merge the histories; it's not too hard (once you get the hang of it - of course, I was the main history-merge expert on Wikipedia for a while, so maybe it just seems easy to me now :-). J. Noel Chiappa 19:25, 20 April 2008 (CDT)

Minor subpage bug

See Talk:Church of Scientology/Archive 1; it seems to think it's the talk page for the subpage Church of Scientology/Archive 1. The code probably doesn't grok Talk: archives. This may not be easily fixable until we get the Strings: package; no big deal, it's pretty obscure, no need to fix it anytime soon. J. Noel Chiappa 21:04, 20 April 2008 (CDT)

New Periodic Table...table

Hi Chris,

I noticed a thing you were noodling around with and thought I might just go ahead and tweak some things. Check it out...
User:David_Yamakuchi/Sandbox2 (<--"top level")
{{Resizable Periodic Table of Elements}}
{{Resizable periodic cell}}
and, last but not least
{{Physical properties}}
Whew! I guess it got a little involved. :-)

Now, what I did was to take the Lead physical properties data and stick it in a template similar to the metadata one. This way we can call up the data onto other pages (like the periodic table for instance) thru the template..."#if" it exists...and when the data is eventually improved (like with better precision / more accurate measurements or whatever), the numbers get automatically updated thru the wiki. Please let me know what you think. I'm interested in your opinion on the direction I went with your table, but I'm particularly interested your opinion of the scheme with the physical properties and the templates. Thanks for taking a look when you get a chance...--David Yamakuchi 22:28, 20 April 2008 (CDT)

{{Resizable periodic cell color}}--David Yamakuchi 23:53, 22 April 2008 (CDT)

Jump in if you like :-)--David Yamakuchi 19:50, 23 April 2008 (CDT)

User plan mess up

Chris, I completely messed up my User Plan and rollup to work-group. Will you help me get on track. You could delete everything needing deletion, and I'll try again. Would appreciate your help. Thanks. --Anthony.Sebastian 16:45, 22 April 2008 (CDT)

Chris, thanks. Will try deleting and starting over. --Anthony.Sebastian 19:40, 22 April 2008 (CDT)

header footer

You mean, as in, 'there's a default header and footer they all use', or 'there's a default header and footer that get used if there isn't one defined for this subpage type'? I think it might be nice to allow customer headers and footers - but I agree that might be a little overkill.

Also, the documentation is a 'work in progress'; I fully expect it to get changed as we tweak things. Still, I wanted to get a start on it, so we're not facing a huge backlog of untackled documentation at any point. (It seems easy to tweak existing documentation, but hard to do new stuff from scratch.) J. Noel Chiappa 17:22, 22 April 2008 (CDT)

Font style

What font style settings are you using for the [?] I want to add this to a section of the periodic infobox to the upper left, and add a bit of hover code that tells people they can navigate it. I'm not sure that right away it is instantly recognizable that you can click on the periodic table within the infobox. --Robert W King 10:36, 24 April 2008 (CDT)

In {{periodic}}, where in the code does it display the text that displays the label of what group is currently navigated? On that cell, I want to insert <font size=1 style="align:left;">{{H:title|This is a navigatable Periodic Table of the Elements. You can hover over each unit and click on it to see its current article!|[?]}}</font> --Robert W King 11:11, 24 April 2008 (CDT)
When you say "group" do you mean element? If so, it is just the elements article name that is used for the wikilink. Chris Day 11:14, 24 April 2008 (CDT)
|[[Helium |{{ {{#ifeq:{{{elSym}}}|He
     |element on
     |{{#ifeq:{{{elClass}}}|Noble Gas
          |Noble on
               |Nonmetal on
               |Noble off}} }} }} |ele color={{{ele color}}} |noble color={{{noble color}}} |nonmetal color={{{nonmetal color}}} }} ]]

Above is the code for one cell. The part that give you the link is bolded here: "|[[Helium | style code ]]" I'm a little confused by your question so this answer may not make any sense. Chris Day 11:18, 24 April 2008 (CDT)

I'm looking for the part where the elClass is; I want to insert the hover on the upper left corner (it should be the left hand of that table cell). --Robert W King 11:21, 24 April 2008 (CDT)

Got you, you're looking at the wrong template. It's in {{Elem Infobox}}. See the code below:

{{!}}style="border-top:2px solid #bbb" width="100%"  align="right" cellpadding="3" cellspacing="0" bgcolor=#f2f2f2 
{{!}}<font size=1>''' [[{{{elClass|}}}]]:'''</font>

Chris Day 11:27, 24 April 2008 (CDT)

That doesn't seem like a great place for it; I'd rather it went in the very upper-left hand corner of the interface (across from the group name), but I don't have the motivation to fix it properly. I'm going to put it back next to the group name until I get around to putting it where it probably should be. --Robert W King 12:06, 24 April 2008 (CDT)
It looks right now, except it made the whole thing wider. --Robert W King 12:36, 24 April 2008 (CDT)
We can restrict the width to the correct dimensions by figuring out the minimum size of the question mark and fixing the cells (? and group) in pixels. Currently i just left the group one at 100%. Not good. Chris Day 12:39, 24 April 2008 (CDT)
I just fixed it. I removed the 100% width. --Robert W King 12:40, 24 April 2008 (CDT)
Does that fix it? If so, that is easier than figuring out the minimum size. Chris Day 12:41, 24 April 2008 (CDT)
Yeah, check it out. I think it's a little more intuitive now. I wish the WP folks were jealous of this widget we (You, Myself, and David) put together. I really am proud of it works and looks. We did a great job. It really makes WP's chem box look amateurish. --Robert W King 12:44, 24 April 2008 (CDT)

For the record, I think it looks great, but my only gripe is that the MSDS/catalog subpage is horribly organized and needs to be arranged in some kind of logical way. Right now it's just a hodgepodge of unformatted data. --Robert W King 15:52, 24 April 2008 (CDT)

I'm just worried about the concept now, not the format. Tell me if you see any problems I have missed. Chris Day 15:54, 24 April 2008 (CDT)

Per-article subpages

I saw the header/footer templates; at first I was confused, because I thought they were to be added to the subpage manually, but I looked at that /Isotope subpage and worked out that you must be calling them from {{subpages}}. I think that's a very good way to go.

My impression is that many users here prefer not to interact with media wiki when possible. That was my rationale but there is a problem that misspelled subpages such as Isotpes will slip through (see below for more on this theme).

Here's an extra tweak for you, if it's not too much extra code: if Template:<tab>_header exists, call it instead of the generic tab_header, and similarly for the footer? Yeah, I know, it's a little silly, and we're threatening to drown in complexity/features... but it might be useful. Whadja think?

I think this is a good idea since it gets around the problem of the generic header being too general to be useful, or grammatically incorrect, in some cases. Being able to write a specific header at a later date is very desiriable. As you point out below, MSDS tab could have Material Safety Data Sheet (MSDS) in the header.

You have a point with a new header/footer slowing down proliferation of new tabs, but... it's a pretty low speedbump, I think. In terms of checking for proper spelling of the tab name; why don't you, on the talk page, if {tabN} is defined, check for the existence of /{tabN}, and emit an error message if it doesn't? Does that take too much code?

But does this solve the issue? It all depends on how the tab is created and where it is misspelled. If the subpage name is initially incorrect and the tab is correct your suggestion would work (works the other way too). But what if the user starts by adding the tab and then clicks on the red link in the new tab to create the subpage (I suspect that is what happened to give us Isotpes)? Now we have the wrong name in both cases so the error check will not catch it. Likewise, both tab and subpage might be created with a subtle variation so that it is distinct, but acceptable, from other clusters with a similar subpage. For examples, an isotope subpage might be created as opposed to an isotopes subpage. At present the best bet for identifying an error is when the category is seen as a red link at the bottom (assuming all the relevant category pages have been created, which is not the case at present). Even that is not foolproof since what is to stop us having the category pages for isotopes AND isotope, so even a blue link at the bottom might not be a good indication of a mistake? As CZ grows I see this as a major pitfall with regard to having more versatility to create novel subpages names, yet it's probably not enough of a problem to nix the idea. For me the real thrust of this proposal is to increase the transparency (or descriptability?, is that a word?) of the subpage tabs.

I don't have any immediate thoughts about the categories. They should all go in the appropriate workgroups (for the parent article), no? J. Noel Chiappa 20:39, 26 April 2008 (CDT)

Probably right here. I will add the workgroups too. I was trying to avoid a situation where we end up having hundreds of different categories with only a few (single?) members. Chris Day 11:17, 28 April 2008 (CDT)
A couple of comments:
  • I take your point about the mis-spellings. Having the list of defined subpages might help with that, though; if people see "Isotopes" listed, they might not create "Isotope".
  • I don't think it's a big problem if we have a category for every subpage type, even if it has a few entries. What's the problem there, anyway?
  • It's too bad there's no way to automatically put all those categories into a super-category automatically, so it would be easy to keep track of them - but to do that, you'd need to edit the category pages. What you could do, to keep track of them, is call then all "Per-article-Subpage-Category-<foo>", where <foo> is the individual subpage name - that way Special:Prefixindex/Category:Per-article-Subpage-Category would find them all easily. That would at least allow you to find the Isotope/Isotopes of the world - and then you could use the category to track down the rogue suppages, and fix them. You'd have to manually peruse the list of subpage types to find the problems - but you'd have to have a human in the loop anyway, for that.
I think this is getting there... J. Noel Chiappa 21:46, 28 April 2008 (CDT)

Now that I think about it, the categories should be added in a template which is always included, and does nothing else. That way, if someone screws up Template:<tab>_footer, you at least get the cats right. Would there be any need for a 'default' footer template (if no Template:<tab>_footer is defined), at that point? J. Noel Chiappa 22:41, 26 April 2008 (CDT)

But wouldn't this defeat the whole point of having a generic template, i.e. "many users here prefer not to interact with mediawiki "? If we go this route then I think we should just go with the low speedbump that allows the versatility but gives allows some control of the tab names to reduce potential errors and redundancy (via similar tab names). Chris Day 11:17, 28 April 2008 (CDT)
I didn't follow this. Were your comments in response to "the categories should be added in a template which is always included, and does nothing else", or in response to "Would there be any need for a 'default' footer template"?
Were you saying 'yes, let's put the category-adding in a separate template that is always added, and does nothing else - so the users can't mess it up - and then also require a Template:<tab>_footer to be defined, as a low-speedbump to slow down creation of new types of subpages'? J. Noel Chiappa 21:46, 28 April 2008 (CDT)

Noel, sorry if my replies above are a bit rambling. This whole proposal is still tumbling around in my mind. Please keep adding your ideas though, as discussing this helps me cement the idea. I've already thought of things that were not even on my radar screen before answering your points above. Chris Day 11:51, 28 April 2008 (CDT)

No problem - I know how that 'half-formed thoughts' thing goes! J. Noel Chiappa 21:46, 28 April 2008 (CDT)


OK, I have to finish copyediting Faraday first. Hopefully I'll be done with that tonight - it's starting to feel like a life sentence.. :-( J. Noel Chiappa 16:11, 24 April 2008 (CDT)

I would have thought you'd be charged by his magnetic personality to write and write and write? Chris Day 18:13, 24 April 2008 (CDT)
Funny person. NOT! But it was really neat to learn about him (I never knew that much); turns out the dude was a really, really major scientist. I'd now rank him below only Einstein and Newton, and on a par with a handful of others. J. Noel Chiappa 20:39, 26 April 2008 (CDT)
What amazes me about that era is their versatility. For example, Robert Boyle was the first to see biological cells (even if dead and only the cell walls in cork) and coined the term cell. CD
Heck, what about da Vinci - one of the great artists of all time, and also did other stuff! But yeah, it must have been cool to be a scientist back then - so much to be discovered, so much virgin territory! I feel like I missed out twice; back in the late 40s / early 50s in computer science, if you were breathing and had a pulse, you could make a major discovery! Still, we'd have missed out seeing where it all went, and seeing the amazing stuff we've found out by today. J. Noel Chiappa 21:53, 28 April 2008 (CDT)


Chris, I like the MSDS tab and an isotopes one might be good too. I do wonder how few people know what the MSDS is, so we might start out the top of the page with "Material Safety Data Sheet" for Cadmium, and inlude a date so we know how old the safety data is, and possibly a source. I note that other sites gives links to MSDS pages so the information is timely. David E. Volk 16:15, 24 April 2008 (CDT)

I agree MSDS is not exactly great either and the whole MSDS subpage needs a lot of work stylistically. MSDS is , however, better than catalogs. Chris Day 18:11, 24 April 2008 (CDT)
If you had per-tab-type headers, you could include Material Safety Data Sheet in the header template for the MSDS subpage. There, I think I've answered my question above; optional subpage header/footers are a good idea! J. Noel Chiappa 20:44, 26 April 2008 (CDT)

Also, regarding the question marks showing up on the tabs, I don't like the look of it one bit, but I do realize how useful it is for getting guidelines. They make it look like our software is choking on a missing variable. I wonder if you can use two lines on the tabs, with a very small font (help) or (guide) directly under the text. David E. Volk 16:15, 24 April 2008 (CDT)

I have to say that I'm not keen on the question marks or the hash (#) marks we had prior to that. I feel purple ? at the end should probably be enough, possibly leading to and image map jpg of a typical set of tabs with labelling too. I have experimented with smaller fonts and double height and no combination is satisfactory. Chris Day 18:10, 24 April 2008 (CDT)
I think that the average person, seeing the ?, will not think of it as the outcome of a programming error. Maybe a hover that says "Help" would be useful, though? J. Noel Chiappa 20:44, 26 April 2008 (CDT)
I was not thinking of it from the perspective of a coding error but rather a space issue. In clusters with many tabs they will run off the screen to the right. Less space in the tabs means more tabs can be accommodated at the top. Will people use the CZ:Subpage links that much? One thing is that a red ? link will indicate when an article specific subpages function has not been defined. Chris Day 11:34, 28 April 2008 (CDT)
Good point about the width. Is there any way to force the "?" underneath the tab name? That would save on the width. J. Noel Chiappa 21:55, 28 April 2008 (CDT)

Thinking some more. Above I was discussing that the subpage categories could be a minimum requirement for creating an article specific subpage. But actually, the minimum requirement should be to create a "CZ:Article specific subpage" description of the subpages role. In this way we will have a record of ALL the article specific subpage names being used and this should help us catch spelling error at the source and reduce redundancy when creating article specific subpages. This is the low speedbump that needs to be in place to keep things consistent. While the generic footer and header templates are useful our goal should probably be to have specific header and footers as these new subpage types are created. Chris Day 11:40, 28 April 2008 (CDT)

Motto bug

Hi Chris, thanks for taking a stab at that motto bug. I think it is caused by too large (with inclusions) of a template. Which quick fix? Of course, I could be wrong. --Larry Sanger 10:39, 25 April 2008 (CDT)

You're exactly right. It also explains why the workgroup todo lists collapse too. The post expand size is huge (about 4.7Mb for one page I checked). There is no easy fix that I can see but I will think this over a bit. It should be possible to only include what is used. Chris Day 12:02, 25 April 2008 (CDT)

Subpage header design

Since now everything on the site has a "flat" appearance, any chance of making it so that the subpage header also appears "2d" as opposed to the beveled table look it has now? --Robert W King 13:53, 25 April 2008 (CDT)

Awesome Template:Codewink. I was going to adjust the colors too but everything I chose was too bright. --Robert W King 14:31, 25 April 2008 (CDT)
What you call dull is actually neutral. You go away from neutral and the colour becomes the centre of attention. Exactly what we need to avoid. Chris Day 14:33, 25 April 2008 (CDT)

SubpagesSub-workgroups proposal or resolution


I noted in the Forums that Larry Sanger asked you to submit a proposal or resolution to the Editorial Council regarding the creation of sub-workgroupssubpages. I would be happy to review or critique (or whatever) that proposal or resolution when you have drafted it. Please feel free to call upon me for help. - Milton Beychok 12:16, 26 April 2008 (CDT)

Did you mean per-article subpages, or sub-workgroups (or subgroups or sub-groups or whatever their formal name is)? J. Noel Chiappa 20:41, 26 April 2008 (CDT)
I am pretty sure that Milton is talking about the sub-workgroups. I have two irons, three if you count the elemment infobox, in the fire at the moment which makes discussions on this page a little confusing. Chris Day 11:41, 28 April 2008 (CDT)

Feature request

Hey Chris, I think we should display at least the metadata (if not the unused subpages) by default. Also, it would be very handy to be able to click on a datum and be taken to the page immediately to edit it--but that's too much to ask for at this point. But more easily accessible "edit" links for the metadata would be useful, anyway. --Larry Sanger 12:23, 26 April 2008 (CDT)

There is a direct link - and a direct edit link - to the Metadata in the header on the article's talk: page. (The "M" button goes there, and there's a text edit link.) My guess at the thinking as to why (since I wasn't there for the discussion, but it's a design call which I agree with) was not to burden our ordinary users with the internal CZ mechanisms. It's only one extra click away - do we need to make it more accessible? J. Noel Chiappa 20:20, 26 April 2008 (CDT)
Noel's interpretation of history is correct. Originally the metadata button was on every page, I reduced that to the minimum number. As Noel points out there is already a link at the top of the talk page to the edit window, is it too subtle? The goals was from the talk page one could either directly edit or just view the raw metadata page. I could make the button go directly to the edit page. Just let me know which is preferable. Chris Day 11:45, 28 April 2008 (CDT)

Re: Displaying metadata, yes, I can change that to be show by default. Chris Day 11:49, 28 April 2008 (CDT)

Oh, that's what Larry meant by "display at least the metadata ... by default". Blah. Looks ugly, to me - a huge header I usually don't care about. If I want to see something in the metadata, I click on the show button - it operates entirely on the local machine, so it's fast. What the devil is the point of having a show/hide button if it displays by default? What is someone supposed to do, click the hide button every time they go to a talk: page? J. Noel Chiappa 20:52, 28 April 2008 (CDT)

Elem Infobox

FYI:I tweaked the text size specifier in the infobox as well. It was ok in IE but firefox didn't like the font size...--David Yamakuchi 16:23, 27 April 2008 (CDT)

{{Physical properties}}

Chris/Richard/Noel, would one of you be able to take a look at Phosphorus/MSDS#Physical__Properties and help me figure out how to get rid of the extra whitespace in the Mass and Electronegativity cells? I'm stumped. It looks like it's coming from the line breaks in between lines in {{Physical properties}}, but when I remove them, the table stops recognizing the new rows for some crazy reason...#^%$!!!. It seems like it's possibly something so simple someone with a fair amount of wikitable knowlege will scoff at it, but sadly, that aint me. So, if you have a couple of minutes to spare, please...scoff away :-) --David Yamakuchi 21:42, 27 April 2008 (CDT)

Never mind. I got it!--David Yamakuchi 23:08, 27 April 2008 (CDT)

Show/hide on Talk: page

Any particular reason you took out the "collapsed" in {{Unused subpage list}} and {{Checklist22}} (not the greatest name, BTW)? I liked them a lot better collapsed out of the way. J. Noel Chiappa 20:41, 28 April 2008 (CDT)

Never mind - I just figured out what Larry meant above. J. Noel Chiappa 20:55, 28 April 2008 (CDT)

Check out Talk:Yangtze Patrol, With IE, the Wikipedia notice is ovewritten by the subpages header. The Wikipedia notice didn't used to get zapped like that. What happened? J. Noel Chiappa 23:30, 28 April 2008 (CDT)

Don't see it on my broswer and not sure what the wikipeia notice is either. But I know why it happens. I had to drop a work in progress but plan to finish off tonight. Chris Day 23:35, 28 April 2008 (CDT)
Also, I guess there's some joke with {{Checklist22}} I'm missing - is it a Catch-22 joke? (It's late, I'm tired!) I just thought you'd gone through Checklist1, Checklist2, etc and created one more variant by doubling the number! J. Noel Chiappa 00:28, 29 April 2008 (CDT)
Ah. Light dawns over famous Massachusetts fishing port, as they say... J. Noel Chiappa 00:34, 29 April 2008 (CDT)
?? Block Harbor is in Rhode Island, I thought, not Massachusetts? :-) J. Noel Chiappa 00:44, 29 April 2008 (CDT)
Yes - I was making a (clearly lame :-) joke back at you, with the 'Block Harbor' thing. Yes, it is Marblehead I was referring to. I completely missed the Charlie Brown reference... sigh, it's late; plus to which I'm not a big CB fan. Definitely a cultural lacunae, or something; Schultz was definitely really good. (Although 'Pogo' might have been a tad better.) Maybe it was too subtle for me - I'm more a 'Calvin and Hobbes' kind of person... J. Noel Chiappa 00:51, 29 April 2008 (CDT)
Sarcasm? J. Noel Chiappa 00:56, 29 April 2008 (CDT)
Sounds like you need sleep too! :-) J. Noel Chiappa 01:01, 29 April 2008 (CDT)
Yes, it works fine now. J. Noel Chiappa 01:47, 29 April 2008 (CDT)

Header footer stuff

I did reply to you about the header/footer templates, etc (see section above); not sure if you'd noticed that. J. Noel Chiappa 01:47, 29 April 2008 (CDT)

Hi, I finally had a chance to think about your comments, and replied over there, at User talk:J. Noel Chiappa#header footer update (it was easier to answer your questions in situ, rather than back and forth on two pages, in case you're wondering what motivated that :-). J. Noel Chiappa 19:39, 29 April 2008 (CDT)
Excellent point about the [[Category:Experimental subpages]] (about how they'd have to get them both right). Now why didn't I think of that? :-)
I'm still a little confused about how "group-AS-subpagename .. represents a reduced set"... Oh, wait a minute (light bulb finally goes on :-) - you mean "<workgroup>-AS-<newsubpage>"? Ah, now I get it! Hmm, I'm not sure I see much use for that? Why would the <bar> workgroup want to be able to see the <foo>-type subpages in their workgroup? I wouldn't worry about that one - because if we ever discover a need for it, it's trivial to edit the template, and all the group-specific categories will be populated after some database lag. Unless we see a need for it right now, I'd leave it out for the moment.
About the proposal - I don't think the Editorial Council (the formal one) will want too much technical detail. The technical detail would probably be more for a CZ: documentation page. (Actually, it could just get added to CZ:Article-specific subpages, no?) J. Noel Chiappa 21:01, 29 April 2008 (CDT)

Is this a permanent change on the article Talk pages?

Hi, Chris: The Talk pages suddenly unhide the MetaData Page and the Checklist page. Is this going to be permanent? The Hide buttons work but only until I have left the Talk page. When I return again, those pages are unhidden again.

For what it is worth, I vote for leaving them hidden all the time. - Milton Beychok 12:01, 29 April 2008 (CDT)

removal of approval

Has removed= always been there? Or did I miss something recently? --D. Matt Innis 12:24, 29 April 2008 (CDT)

news on whether strings will be added?

I don't even know who to ask or where to push, but it's my hope that we could get it soon. I suspect it will open up a lot of possibilities we never even considered previously.--David Yamakuchi 13:27, 1 May 2008 (CDT)

I'd like to see this too. Any ideas on how we could possibly make this happen? I've tried leaving notes on a couple of dev's talk pages (Zach, I think, and someone else), but nada. Can Stephen Ewing install it? J. Noel Chiappa 17:12, 3 May 2008 (CDT)

Dunno...lets ask...--David Yamakuchi 18:50, 3 May 2008 (CDT)


Hi Chris, thanks for the help on the new Catalog colors. I will try to remember to find this article when I need to see the formatting for doing something similar. Hayford Peirce 12:14, 2 May 2008 (CDT)

Thanks for help

It's not only that I'm learning the differences from Wikipedia, but learning the differences between the way I think things ought to work. :-)

Perhaps there could be some collaboration between military and biology workgroups on swarming behavior. There are quite a few publications examining the relationships; I have a pleasant mental image of a general pausing to consult with an army ant. Howard C. Berkowitz 16:26, 2 May 2008 (CDT)

Recipe template needs some minor tweaking

Hi, Chris. If you look at the new bearnaise recipe at you'll see that the template needs some tweaking in the "howmany=" and "what=" areas. The problem is that the way it is now set up is that it is generally for "servings" -- and the numerical number shown is LARGE. But, as I wrote the bearnaise sauce recipe, and looked at the Times and Julia, it became clear that the result I want to give is "cups", not "servings". I tried to do this in various ways, but all of them were either bad or not very good. (If I just substitute "cups" for "servings", then the "3/4 to 1" is ridiculously large.) Deleting "howmany" certainly didn't work.

Ideally, what I'd like to be able to do here is something that ended up looking like:

3/4 to 1
cup, about 4 to 8 servings

Is there any way to fiddle this? Many thanks! Hayford Peirce 18:36, 2 May 2008 (CDT)

Subpage default contents

I notice that the links to non-existent Related-Articles/Biblography/etc in the subpages navigation bar (and also any non-existent per-article subpages) don't offer the preload of the blank page (with {{subpages}} already there). Would it bloat the code greatly to do an ifexists, and if the target is not there, make the link a preload edit? That would probably help the forgetful / less technically steeped... J. Noel Chiappa 09:25, 4 May 2008 (CDT)

I too was wondering about whther to do this but it slipped my mind. More than once probably. I don't think it would b a problem size wise. Chris Day 09:28, 5 May 2008 (CDT)

MSDS subpage

It would seem to me that unlike Isotopes (which only element pages would have), MSDS pages would potentially be found on all chemical compound pages, which is a vastly greater number. So should MSDS be a 'standard' subpage? (BTW, there are a whole bunch of backlogged suggestions for standard subpages, including Quotations, Glossary, etc... the system for approving them seems to be wedged/non-existent - see CZ:How to add a new subpage type.) J. Noel Chiappa 16:47, 4 May 2008 (CDT)

Yes MSDS is probably a more general example since every checmical under the sun would have such a page. It may well be more suitable as a standard page. I was thinking about what types of subpages we will have. I remember Larry early on suggesting we would have hundreds and I think there is no reason why this will not be the case. With so many, however, it might make more sense to have a hierarchy. There will be some used by all pages (we have these as defualt now). There are some that could be used by all pages, like catalogs. There will be others that are more specific to a particular workgroup such as MSDS. And finally ones that will only be used on a small set of articles (this is what we want tabs for). I have been wondering what to do with the workgroup specific ones like MSDS which currently fall between the stool with regard to our current set up. Do we really want MSDS to be seen as an unused subpage option on every talk page? But, I agree we do not want to have to define a MSDS tab in the metadata for every chemical. I don't have an answer, yet, but it would be nice if from the standard set of subpages only appropriate ones show up in the unused subpages list of any given articles talk page. Chris Day 09:40, 5 May 2008 (CDT)
Well, you could always define "workgroup-specific-subpages"! :-) I.e. have templates of the form "Template:subpages-<group>", and the template which displays the unused subpages on the talk page would pull the cat(s) out of the metadata, pull up the appropriate "Template:subpages-<group>" template(s), and go through the list in there and add them to the displayed list. One problem: if two different WG's both have a subpage, it might get displayed twice. I can't think of any easy way to fix that that's not really ugly. (E.g. define "Template:subpages-<group1>-<group2>", for cases where cat1 and cat2 are both specified. With ~30 workgroups, and something 30! combinations [can't be bothered to work out the exact math :-] that's too many to be feasible.)
As for the backlog, I think you have a great point that experience with per-article subpages will give us a lot of info. I have yet to look at the proposal page (been doing catchup), I'll make sure that point is included as a plus for PA subpages. J. Noel Chiappa 10:34, 5 May 2008 (CDT)
What kind of hierarchy could one construct, though, other than 'all/workgroups/specific'? Are you saying construct some new, distinct, hierarchy to organize articles into, and associate sets of subpages with nodes in that new hiearchy? Because I can't see any other existing information we can 'mine', to suggest additional subpages past the 'all' set, other than wgroups. J. Noel Chiappa 17:31, 5 May 2008 (CDT)
I have not thought it through at all. Chris Day 18:01, 5 May 2008 (CDT)
With workgroup specific subpages you could also basically offload some of the administration of the subpages and names and such to the WG Editors, who would presumably have the best insight into the way subpages of their articles should be arranged.--David Yamakuchi 19:39, 5 May 2008 (CDT)

Talk pages

Well, yes and no. As the project gets bigger (as Wikipedia did), it's simply too much to keep track of, particularly when you're leaving messages on lots of personal talk pages, and unlikely to go back to a random person's talk page for a different message. You either have to i) check for a reply manually, or ii) add the page to a (usually bloated) watchlist to watch that way. Of course, i) CZ is still pretty small, and ii) I'll be coming over here all the time anyway, soin your case either page works fine! J. Noel Chiappa 10:21, 5 May 2008 (CDT)


Yep, Mr. G. now shows up in the right place. I'll remember to make a minor change to the article next time I correct an ABC=. Thanks!

Formatting problems

Hi Chris, just applied the rpl template to the List of biology topics, which thus became inaccessible. Is there a technical limit to the number of templates that can be used on a page? I think there were about 800 instances of rpl on this one. Please take a look. Thanks! -- Daniel Mietchen 04:57, 6 May 2008 (CDT)

Thanks for checking, but for me, List of biology topics remains inaccessible, and Natural Sciences seems to be at the edge, too. To me, this looks like the use of templates should be discouraged for a while (which would be a pity), or some of these technical limits shifted. Can't comment on their suitability, though. -- Daniel Mietchen 09:46, 6 May 2008 (CDT)
I made a mistake and was looking at the core article page. Now I see you said the List of biology topics. That page is now huge so i'm not surprised the tempalte crashes. I can't open it easily either so it has certainly passed the pre expand size, probably post expand size too. That template does add a lot of extra stuff to a page so i agree if it is used it needs to be in a more streamlined version or not at all. Chris Day 10:07, 6 May 2008 (CDT)
The main reason why I created the rpl template as a combi of pl and the r is that I wanted to make the Core article pages more attractive, thereby hopefully stimulating further contributions. The pl info on the status of an entry should be kept, while the def part was meant to act as a teaser. Without it, an appealing redesign of the core article pages would have to take an entirely different approach. Instead of dropping the def part, I would rather opt to limit it to, say, 100 characters. If possible, I would in parallel lift the limits for a few of these core article lists but I have no idea what other process might be effected by this. Is there any page that describes the expansion process in detail and provides insights into why the limits have been set the way they were? As for the low impact of the def removal on Natural Sciences, I guess that's simply due to very few of the articles listed there actually having a non-empty def. -- Daniel Mietchen 10:55, 6 May 2008 (CDT)
As you told me before, List of biology topics is not in use by any workgroup any more, so it is dispensable in principle. The reason I use it still is that it provides a good playground for core articles and similar lists (as we see today). However, this testing could also be achieved with a split list. Since List of biology topics is currently the only page causing problems with the original rpl template, I would thus opt for keeping the def part in the template (perhaps trimmed down, if you see any further potential for that) and splitting lists that reach close or beyond the expansion limits. Such splitting makes core article lists more easy to handle, and the defs provide them with greater appeal for potential contributors, I guess. -- Daniel Mietchen 11:53, 6 May 2008 (CDT)
This sounds good. Chris Day 12:04, 6 May 2008 (CDT)
OK - I split List of biology topics and reverted rpl. I'll leave the core article splitting for you. -- Daniel Mietchen 12:24, 6 May 2008 (CDT)
Yes, with the raw code, it's better and perhaps less resource-intensive. Thanks. -- Daniel Mietchen 12:34, 6 May 2008 (CDT)
Thanks for refining the template and trimming it down. I think it does a good job now. -- Daniel Mietchen 02:00, 7 May 2008 (CDT)
Level4.jpg is not used now and thus not included in the documentation of Template:Rpl. -- Daniel Mietchen 03:00, 7 May 2008 (CDT)
OK, it is used, and thus included in the documentation. -- Daniel Mietchen 03:08, 7 May 2008 (CDT)

Subpage for def?

Not really on topic, but certainly related: I was wondering whether it would make sense to include the definition page of an article into its subpages and, step II, to display the def in case of an empty main article page. What's your take on that, and where should I ask for such things to be considered by the relevant people? -- Daniel Mietchen 12:00, 6 May 2008 (CDT)

Larry set up the def template and the related pages format. I'd start by asking him. Chris Day 12:03, 6 May 2008 (CDT)

I believe I thought about doing that when I first set up the def template, but the problem is that subpages can't act as templates. Only Template: namespace pages can act as templates.

But what you might be saying is that we should display the contents of the def template (i.e., from the template namespace) when there is no other content to include. There is no need to actually move the definition content from the template namespace to subpages, even if that could be made to work. Still, logistically, I am not sure how much sense that suggestion makes; I mean, it's possible to do that, but it takes quite a few keystrokes just to set up subpages.

What might make sense, however, is to make a new Definition subpage tab, without actually putting the definition on a subpage. That would allow people to click on and view definitions directly from the subpage tabs, and then edit them seamlessly from within the Template: namespace. Then it would "feel" like you were editing a subpage, even though you aren't. Then, the idea is, when you are at article XYZ, if the entire contents of XYZ is {{subpages}}, then the definition template displays. (This is another thing I remember considering and rejecting, but I completely forget why I rejected it; probably just because the definition didn't seem like important enough information to make a whole tab about. Hmm...)

That would also encourage people to write those definitions, which would make Related Articles subpages work better. But the problem it wouldn't solve is getting people to set up a topic with subpage apparatus (i.e., start article with subpages). You can get people to write lots of definitions, but then wouldn't it be another whole bunch of trouble to get them to set up subpages for each of the new pages?

But there seems to be a way to solve that problem. Hmm, this could work.

  1. You know how, when you look at Related Articles definition lists like this one, you see the usual red links when there are no articles available? Well, first, you could use the r template to color those red links green.
  2. Then if you follow the green link, you come to a page with just {{subpages}} prefilled in the text box (together with some commented-out instructions). You just save that without making any changes.
  3. Next, we (Chris) would have changed the subpages template so that, if there is no metadata template for the topic but there is a definition template, then we (1) display the definition (in large type, perhaps); (2) include a message below the definition, saying, "This is just a definition taken from the definition template. The article itself is blank. Please click here to start a new article on this topic. (This page is not included in our article count.)"
  4. Finally, if you look back on the Related Articles page, the green link now appears the usual soothing blue.

OK, having said all that, the point of this rigamarole would be to prefill lots and lots of pages with bare definitions, rather than have nothing at all. I am not convinced that that would be such a great idea. Do we really want a lot of definition-length pages masquerading as real CZ articles? Can you explain why you think any such thing would be very useful?

I suspect that what we need more than any of these technical fixes is a Big Initiative to get people to write lots and lots of stubs. We had CZ:Stub Week, which was a success, I think... --Larry Sanger 13:33, 6 May 2008 (CDT)

To me, defs with empty main articles are, like stubs, explicit invitations to contribute, especially if the invites come from core article lists like that for Natural Sciences. As for the {{subpages}} structure, I did not yet quite understand why it is not created automatically when the initial edits of the article are saved (perhaps with some prompting to fill in the AE/BE fields and so). After all, saving preformatted text of those subpages unchanged (as currently required) is certainly something that could be automated, right? -- Daniel Mietchen 02:00, 7 May 2008 (CDT)
Two more points on this: (1) If we keep defs separate from main articles, we will have to do redirects for both. Is there a way to couple the two such that if one has a redirect, the other will get the same (perhaps with the main overriding in case of conflict)? (2) Small initials seem to be a problem, too (e.g. Template:Def life which had to be redirected to Template:Def Life). -- Daniel Mietchen 07:29, 7 May 2008 (CDT)

Definitely we want to automate the creation of the subpage apparatus; but there is no easy way to do that, as far as I have been able to ascertain. We might (probably do) need to create a new MediaWiki extension.

I also agree that it's a shame that we have to keep defs separate from main articles. If you can figure out a way to keep them together, great! --Larry Sanger 08:47, 7 May 2008 (CDT)

Larry the trick is to use the colon in front. For example, we currently have {{Def Brain_plasticity}} which transcludes as: "Template:Def Brain plasticity" but this can be on a subpage (as an example I just created Brain_plasticity/Definition).
While {{Brain_plasticity/Definition}} gives us: "Template:Brain plasticity/Definition"
Adding a colon, as in {{:Brain_plasticity/Definition}}, transcludes as: "

The ability of the brain to adapt to new situations, e.g. by learning or neurogenesis.".

I had forgotton that this problem was the reason we had the def page as a template. It is also the reason we have the metadata as a template. With regard to the metadata, though, it still has to be on a template since the colon trick does not work on the sub-subpage. But for the definition I don't believe that template will be used on the sub-subpages so we could move all the definitions to a subpage.
This would be very preferable, in my opinion, since with the current system the template can be on the wrong page. For example, depending on whether one writes {{r|life}} or {{r|Life}} the template will be named either Template:Def life or Template:Def Life. This is a royal pain. If the definitions are on a subpage it will not matter if the first word is capitalised or not. For example, both {{:Brain plasticity/Definition}} and {{:brain plasticity/Definition}} will transclude the same page (unlike the life example above).
Sorry I didn't think of this earlier. Maybe someone can write a script for a bot to move all the definitons to a subpage? Chris Day 10:27, 7 May 2008 (CDT)

Continuing from above. I just modified the {{r}} template to a new version, {{ra}}, which can use a subpage version of the definition instead of the Template:Def version. For example:

{{ra|Brain plasticity}} gives: Template:Ra {{ra|brain plasticity}} also gives: Template:Ra

Chris Day 11:17, 7 May 2008 (CDT)

Wow! If you hear a big D'OH emanating from Ohio, that would be me.

Yes. Let's do this. Absolutely no reason not to, and some excellent reasons to do it.

Definitely a bot would be work, but let me ask the tech guys if they can do this at the database level. Seems to me they could. Simply rename anything in the form Template:Def Foo to Foo/Definition. I'm pretty sure they'll be able to do it. Just let us ( know when you want them to throw the switch. -- We'll also have to rewrite CZ:Definitions and CZ:Related Articles and Template:R as appropriate. --Larry Sanger 13:19, 7 May 2008 (CDT)

I see no reason why we can't throw the switch right now. I have already rewriten Template:R (see Template:Ra ). When we throw the switch I'll just update Template:R and its notation to be correct. Likewise rewriting the CZ:Definitions and CZ:Related Articles page should be trivial. At a later date (soon though) the bot can migrate the Template def pages to their new home. Chris Day 13:48, 7 May 2008 (CDT)

Yeah, sure--as long as e.g. Biology/Related Articles doesn't break. Also, we might have to wait for an answer from the tech guys to see if they can transfer stuff automagically. Also, I'd like to make the Def pages (and maybe we should title it just "Def"--I'm not sure--it's shorter, but "Definition" is clearer) default, like Bibliography, etc. That's because such pages will always be useful. We'll also want a subpage template-generated explanation of the function of definition pages on the definition pages themselves. --Larry Sanger 15:25, 7 May 2008 (CDT)

def and redirects

Hello again Chris, please take a look at the following:


as opposed to


Is there an easy way to fix this such that the two sets look identical? Same problem for {{ra}}, by the way.

Besides, I noticed that Category:Approved Definition Subpages contains the def of Life yet not Biology - any idea? -- Daniel Mietchen 11:32, 8 May 2008 (CDT)

Another request

Hi again Chris. What would we do without you? Thanks! Anyway, here's something else. Now that the definitions are visible up there, I think we need to make it slightly easier for people to start Related Articles pages. To do that, I think we should make the red links to RA pages prefill with a page with some instructions about how to fill out the page. See CZ Talk:Related Articles for proposed text for that prefill. --Larry Sanger 21:09, 7 May 2008 (CDT)

I was thinking along those lines too. That will be an easy change. Chris Day 21:11, 7 May 2008 (CDT)

Aleta considers her "options"

Hi Chris--while working with the optional article-specific tabs on the metadata page, I ran into a snag. I see you fixed the abc at Moose (dog actor) but I tried to give him a credits subpage (he has mixed media, not a filmography, not works...) but instead of coming out Moose (dog actor) > Credits it just came out


and it wants its own metadata page.

Me, or a bug?

Aleta Curry 04:11, 8 May 2008 (CDT)

Flagging long response on my talk page

Didn't know if you were watching the page, if I should put the content on your user page, or what -- let me know what is preferred.

While I do hope to get more interaction here, the sense of editing and authoring just my own material, not under WP constraints, is the intellectual equivalent of a long hot shower after mucking out the catastrophic failure of a litter box (this is a 15 cat, 4 dog, 1 squirrel household, in which the dogs and squirrel think they are cats). Howard C. Berkowitz 10:34, 9 May 2008 (CDT)

anchorencode versus urlencode

Neither is perfect for turning something like "Foo bar#Zap" into a URL, i.e. urlencode turns " " into "+", so article titles with spaces in them don't work. (Does work to get to [[Foo bar]]? I thought it didn't, but maybe I'm confused.) anchorencode works OK for that (i.e. turns "Foo bar" into "Foo_bar"), but converts "#" to ".", etc. J. Noel Chiappa 21:23, 9 May 2008 (CDT)

Oh, wierd. does the wrong thing, but does the right thing. So presumably would work too, if you didn't want an edit link. How idiosyncratic. J. Noel Chiappa 21:41, 9 May 2008 (CDT)

Just bizarre. And thanks for the info. Chris Day 22:38, 9 May 2008 (CDT)


I noticed in your delayed coking defintion you started it with the words "Delayed coking". But remember in the context of the


  • Delayed coking [r]: A petroleum refining process that converts heavy residual oils into petroleum coke and other byproducts. [e] will be seen as:

Delayed coking: Delayed coking is a petroleum refining process that converts heavy residual oils into petroleum coke and other byproducts. [e] [r]. Chris Day 22:52, 9 May 2008 (CDT)

Chris, does that mean that the article titles must not be used in the definitions? If so, I will have to redo all of the definitions I have added so far. Perhaps you should broadcast that fact. - Milton Beychok 01:29, 10 May 2008 (CDT)

Definition location

I'm coming into this late, but.... why not put the definition into the metadata? Why put it into a subpage (with all the attendant overhead)? J. Noel Chiappa 22:18, 7 May 2008 (CDT)

That's possible too. It could go at the very top with the pagename field so it is more obvious. The most obvious advantage is that we do not have to have the subpage template in the <noinclude></noinclude> tags at the top of the definition subpage. It's just as easy to do it that way. We still have a def subpage in such a scenario, but possibly better to just have at the top of the checklist on the talk page? Chris Day 22:53, 7 May 2008 (CDT)
Sorry, I didn't quite understand that last sentence? (And sorry I was offline a few days; slightly sick.) I see Larry wasn't too enthusiastic about having it on the metadata page? Are there any technical advantages that you have thought of since your first note on my talk page to having it in the metadata? You said something about "it makes the coding a lot easier and simpler", which given the complexity of the subpage stuff, is not to be sneezed at. Does this still seem true? If so, I think we probably ought to ask on the Forums (I haven't gotten over there yet to see what I missed, maybe it's been raised there) to see if people can cope with a definition in the metadata. J. Noel Chiappa 08:02, 10 May 2008 (CDT)
Okay, tuppence' worth:
This bugs me:
Write definition here (maximum 100 characters), for use on Related Articles pages. Don't include the term defined in the definition itself. (Delete this note after reading.)
What happened to the nice yellow box that popped up on the main article page when you created the article and it said something like "you have to click here to write the definition, c'mon, click, go ahead, you can do it"
And then you click and it's really easy, you're on a def. page and you just click again and it's save you're done.
But that include/subpages </no include bizzo - my heavenly stars. I know it says 'delete this note after reading'--delete the code as well? leave the code?
Can't you put it back the other way? Please pretty please. Aleta Curry 21:50, 11 May 2008 (CDT)
Edited to add: okay, see, here's the problem. I can't even show you what I see on the screen that's upsetting me because when I clicked enter it shows what the code produces, not the code itself. sigh. Aleta Curry 21:54, 11 May 2008 (CDT)

Aleta, I know what you're seeing but what part is confusing? You're not sure which bit to leave and which to delete? What would make it easir? The little red def tab is equivalent to the nice yellow box just more subtle. The talk page does have a box asking for a definition and a link. Either way you get the "</no include bizzo - my heavenly stars", so I'm not sure that the addition of a box is going to help you through this. What do you recommend the text in the edit box should read? Bear in mind that <noinc1ude>{{subpgaes}}</noinc1ude> code (note I replaced the l with a 1 so this code can be seen in the text here) has to be part of the definition subpage in the same way that {{subpages}} has to be at the top of all subpages, otherwise the definition will not appear correctly when used by the {{r}} template on the related articles subpage. Also bear in mind there needs to be instructions to remind authors not to state the word/s being defined as part of the definition other wise these words will be repeated when used in the {{r}} template on the related articles subpages. Chris Day 00:02, 12 May 2008 (CDT)

I just noticed that your chronicles of Narnia talk page does not have a to-do list on the talk page with a reminder to start a definition page. It was supposed to. I'll have to track down if I made and unintended edit that shut off the to-do list part. I know i was playing around with a few idea and I may have forgotton to reactivate the to-do list part of it. However, you'll find the link in the to-do list is no different than just pressing the red def tab, just so you know. Chris Day 00:10, 12 May 2008 (CDT)

If you put the definition in the metadata, you wouldn't have the <noinc1ude> stuff... :-) Are you positive about having it in a subpage? Weren't there technical advantages to having it in the metadata? J. Noel Chiappa 10:55, 12 May 2008 (CDT)
Yes, this is why it is better in the metadata. The reason it is better on a subpage is that any definition that is not already part of a cluster would have to live on an orphan metadata page. Which is more confusing for those that hate code, a small amount of code at the top or all the unused metadata?. Is there an easier way then either of these? I think the real answer is that we need a user friendly interface for the metadata as soon as possible. Then the definition could live there no problem at all. I'll think about this some more. Especially now that I am seeing people have a problem with the code in the definition subpage. Maybe we should just go the whole hog and use the metadata as a home?
The technical advantage of having it in the metadata is that is that all the code is simpler and I only have to have one to do list template rather than the two I use now. Actually this is a trivial thing and I could probably have one now but it is easier to have two. But as you know, if we keep adding templates it is harder to maintain. Chris Day 11:54, 12 May 2008 (CDT)

Thanks for what you did to my user page. That is really cool!

Chris: That template {{pl}} is really cool and helpful as well. Thanks, Milton Beychok 12:58, 12 May 2008 (CDT)

I know you are very busy ... but please don't forget that subworkgroup initiative that you started. Thanks again, Milton Beychok 12:58, 12 May 2008 (CDT)

You can thank Robert King for that template. I have not forgotton. Chris Day 13:00, 12 May 2008 (CDT)
Not me, check the history; I think it was Joe Quick! --Robert W King 10:14, 17 May 2008 (CDT)

Trying to grasp "definitions"

Hi, Chris, thanks for the long explanation. Does the following fit what you are saying:

  • We have an article called Tennis right now.
  • We create another article called Grips
  • Grips is then listed in the "Related Articles" tab of Tennis
  • If one goes to the "Def" tab of Grips, one could then write:
    • Continental -- the racquet is held the same way for all strokes
    • Eastern -- the racquet is held in the so-called hammer fashion and is different for both the forehand and the backhand
    • Western -- the racquet is held in a manner completely opposed to that of the Eastern grip

Or some such?

On the other hand, now that I think about it, the article called Grips, would, by definition, be about the different types of grips and tell how they differed. Therefore the above short list of definitions should more properly be on the Def. tab of Tennis? Hayford Peirce 13:33, 12 May 2008 (CDT)

Okie, I see what you've done. I'll go to Grips and put in some info.... Hayford Peirce 14:30, 12 May 2008 (CDT)
Grips will be a new article when I get a mo. Right now it redirects to Grips (tennis), okay, I can see why, but we won't let it stay that way.
Now, someone explain to me why "tennis grips" lives at Grips (tennis) instead of redirecting to Tennis/catalogs/grips??
Aleta Curry 22:49, 12 May 2008 (CDT)
Aleta, you make my head hurt, myte! And I'm wye too old to start developing headaches! Hayford Peirce 23:04, 12 May 2008 (CDT)
Why not both, a catlog of grips in tennis and articles on grips? I know nothing about tennis so I have no clue if there is enough for an article on grips. My sole reason was to demonstrate the concepts behind the {{r}} template and its usage in the context of Hayfords example above. Chris Day 08:13, 13 May 2008 (CDT)
Yes, there's definitely enough for a long article on Grips -- there's a fairly long one at Wikipedia called Grip (tennis) at [2]. I wrote a lot of the original article several years ago, so I can port over my contribution and edit and update it. Gimme a little time. Then you can do whatever you think is necessary about the Definitions. Hayford Peirce 10:54, 13 May 2008 (CDT)
What, me? Making your head hurt? Never! I'm just still tryin' to work it all out, mate. But look, you and Chris have just solved it. There's enough for grips (how people hold their cricket bats, motorbike handles, horse reigns AND tennis rackets), [[grip(s){tennis)]] and the catalogue Tennis>catalogs>grips. Is it MOI fault you tennis people make things so goshdarned complicated? You can't just hold the b* thing in your hand whatever way is most comfortable? Why does this not work as a concept??? Aleta Curry 19:52, 13 May 2008 (CDT)

Hey Chris, Check out Talk:1 CE. It is in the "Needs definition" category and the "Articles with definitions" category. Hmm... --Larry Sanger 19:20, 12 May 2008 (CDT)


Hi Chris, thanks for your message! I have entered the forum and left a short note for Nick. I can't tell whether I will be able to contribute much in the short term, though. Bye! --Paul Schächterle

More definitions stuff

Okay, so I just used the To Do. thingy with the <>/// stuff in it. Nothing blew up, so I suppose it's all right.

This has embolden me to experiment further. Based on one of your comments somewhere, CD, I decided to try to find a "standalone" definition, by which I mean one that I created *before* we had to definition tab. I found one at Black Beauty. If you look at the cluster talk page, it says that the definition needs to be moved (I wasn't game to try that.)

Here're my questions:

1. That blurb that says that the definition needs to be moved; is that now generated automatically, or do we have to be on the lookout for lost definitions and mark them as needing to be moved to the cluster--linked to the cluster--clusterised--subpaginated--whatever it is?

2. Should definitions begin with a capital or a small letter? (Just that it'll look nicer on a related articles page if they're all done the same way.)

3. If someone wants to walk me through how to move the definitions (takes deep breath and tries to look brave) I'll give it a go moving lost definitions when I find them.

Aleta Curry 21:17, 13 May 2008 (CDT)

Hi Aleta, I'll just jump in here to reply.

  1. It is created automatically via Template:Def to finish, no need for manual interaction unless any errors occur.
  2. As per CZ:Definitions#Format_of_the_definition_itself, start with a capital letter.
  3. Dunno where exactly you get stuck, so I would suggest you just give it a try with Scotland and report here so that we can improve the documentation if you experience or sense problems.

Daniel Mietchen 02:01, 14 May 2008 (CDT)

Okay, thanks Daniel. I've taken your suggestion, but I've had to stop part of the way through, otherwise I won't remember all the points I want to mention. So I'll comment now, then continue.
When I first looked at the ToDo note, I saw this: Scotland/Definition subpage (use reason modified Def template). Remember to place <no-include>{{Subpages}}</no-include>
that's only part of it, of course, the part I copied, because there was no way I was going to remember exactly what bits of code belonged, and I (rightly) guessed that by the time I got to whatever page was displayed following my actions, the instructions for what to do next would be gone.
So that page gave me a blank box in which to put the new location in. Well, Template: something or other is in there by default, but of course you don't want to put a template location, you want to put ArticleName/Definitions (I think). Thought briefly about whether or not I needed Scotland/Definitions, Scotland/Definitions, Scotland/definitions, or Scotland > Definitions. Went with Scotland/Definitions, that seems to have done the trick.
But, now you're at ToDo page which says something like, don't forget to speedy delete template Scotland whatever it was called.
Sorry, I forgot to mention: putting reason 'modified template' isn't a good idea in my opinion. I'm not modifying the template, I'm moving a definition from one place to another for a specific reason, namely that the def template was created before clusters and the Article name definition was. Modifying/editing the template says to me that I'm doing something computerwise with the template itself, and when this bit of cleanup is over noone is going to remember what the--? we were doing. So I put "moved definition"; something like that.
Okay, I'm now back to the part where I've just moved the definition and I'm being prompted to speedydelete the old definition at the old def template (there's a redirect there right now, naturally). Problem is, I now have to copy the code and do that, which I won't do, because I still have my wits about me and I remember that my copier is holding the code to be placed at the new definition space, and I a) don't want to forget to go do that and b) don't want to forget what to put in when I get there and now in addition c) I have to remember to speedy delete the old template, although I don't know where the speedy delete code goes, even if I remember what the speedy delete code is. Also, I have to remember, I think, that the code for the subpages goes on the Talk page of Scotland/Definition, right? So NOW I have to use a word processor to hold the words and code I need. This would be a pain in the neck if I also was trying to keep new wording of a definition in my head, but fortunately I'm not doing that right now.
Okay, I'll keep going and let you know what happens.
Aleta Curry 19:05, 14 May 2008 (CDT)
Back again. Okay, I put in the code but it doesn't look right. I seem to remember that this is because I have to take the dashes out. Washing machine is ready to empty and I need to go recycle the rinse water. When I come back, I have to remember to: fix the dashes and speedy delete...the something--hopefully I'll remember what when I'm watering the plants. Oh it's the old definition template, that's it. bye. Aleta Curry 19:21, 14 May 2008 (CDT)
Back yet again, and Noel's completed that which I didn't finish. Glad to see that even experts can make code errors, hee, hee.
This works pretty well, boys, I do think that it presumes some knowledge, though, as I say above.
Aleta Curry 20:57, 14 May 2008 (CDT)
Just changed "use reason 'modified Def template'" to "use reason 'as per modified Def template'". Hope this is clearer. -- Daniel Mietchen 01:59, 15 May 2008 (CDT)

Daniel thanks for alll the improvements to the def to finish template and the answers above. Regarding the small or capital letter to start a definition I belieive I have confused you Aleta. In my ignorance I was starting them with a small letter and now realise it should be a capital. Chris Day 02:53, 14 May 2008 (CDT)

No, in all fairness, you didn't really confuse me, I would probably have had the question anyway--if you didn't do it, I or someone else would have. I think such problems happen (for me at least) because 1) I wouldn't remember where CZ:Definitions was, and 2) all those [[CZ:]] pages are of course still being written, so something could've changed from the last time a girl looked at it.
I'm just seeing you reply of 11th May (is it possible that if someone makes a comment on a different matter *before* I've seen a reply, the reply no longer shows up on my watchlist? Is it possible that I'm totally distracted and/or half blind?) Anyhoo:
Yes, you got it: I was confused because I had no ToDo list, that's what I was calling my nice little yellow box.
Aleta Curry 20:57, 14 May 2008 (CDT)

Core article lists: Splitting badly needed

Hi Chris, when fiddling around with {{rpl}}, I noticed that CZ:Core_Articles/Natural_Sciences is off limits again - independent of my modifications to rpl. I do not know what Barry changed there, as neither page nor history would display in my browsers, but I suppose he did extend the use of rpl to maths, and this crashed the page. I thus assume that it's time to split the core article pages into smaller units. I can't do it on my own, but if you need help, let me know. -- Daniel Mietchen 04:51, 14 May 2008 (CDT)

Splitting is finished for natural sciences. Please take a look, comment and continue with the other disciplines as you see fit. Also, ToA did not yet work on {{rpl}}. -- Daniel Mietchen 06:33, 14 May 2008 (CDT)
I was playing around with the template to try and reduce the size but the problem is the deinfitions soon add up. With every definition defined it will be impossible to have so many links on one page. I see you started the split. I'll think about how we want to organise it with respect to the header. Chris Day 09:10, 14 May 2008 (CDT)
I don't want to interfere with your thinking process but I would really like to have the icons back in {{rpl}}, as these are amongst the best things we have for effective progress monitoring on Core articles. -- Daniel Mietchen 10:45, 14 May 2008 (CDT)
Icons will definitely be back. It was just a quick way to reduce the size. Chris Day 10:48, 14 May 2008 (CDT)

There is still a problem with some redirects: {{rpl|Cell Biology}} gives

  • Cell Biology: The study of the components of cells and their interactions. [e],

which currently displays as

*  Cell Biology: 
  1. REDIRECT Cell biology/Definition [e] .

Apart from that, {{rpl}} looks fine now, ready to be used generically for all lists of CZ:Core_Articles, but perhaps best on the split versions nonetheless. -- Daniel Mietchen 05:52, 15 May 2008 (CDT)

As for the design of Core Article presentation and navigation, I have several related ideas but before deciding on which ones to elaborate, I would like to have some more info on how people arrive at CZ (an update to CZ:Top Google search queries would be good) and on how Core Article structures are implemented elsewhere. In short, I would split the current CZ:Core Articles such that CZ:Core_Articles#Article_lists (expanded, hierarchically structured and with well-formatted brief definitions) goes into CZ:Core Articles (list) (or so) and everything above (rephrased, e.g. getting rid of the point system and explicitly allowing topics to be listed at several Core sites, with consequent use of disambiguation) into CZ:Core Articles (concept). I also consider creating a Category:Core Articles, for which checklisting would be compulsory. Besides, I think having a link to CZ:Core Articles (list) right above the Live articles link in the left navigation frame would be a good choice in the long term. I'll leave you here until next week, though, and I am also hoping that the disambiguation initiative will be launched at least (and possibly passed and implemented) before the Core Articles are to be featured prominently on the site. -- Daniel Mietchen 09:12, 15 May 2008 (CDT)

I have started out here. -- Daniel Mietchen 11:55, 19 May 2008 (CDT)

Line breaks

Umm, I have to look into this a bit more, but my first guess is that this is a bug in the pre-processor (the chunk of MediaWiki that does all the template expansion), or something like that. An experiment you can perform is to transport this stuff to Wikipedia (do it all as subpages of your User: page - they won't like you testing stuff in Template:) and see if you get the same result. If not, because Wikipedia is running the new preproc, and we still have the old one, I wouldn't worry about it overmuch, as it will go away 'soon'.

Oh, "no line break when I transclude the definition subpage within an expression (Please correct my lingo here ...)" - that's exactly correct. Very odd bug - it would have made more sense if it was the other way around! Somehow the code in the first instance is inserting the line break... Have you tried looking at the HTML output to see if there's a clue there? J. Noel Chiappa 11:58, 14 May 2008 (CDT)

Actually, I realized I was wrong above; it's not that the normal code is inserting a line break, it's that the expression code is eating the line break. Which makes a little more sense... Odd that the new preprocessor has the same bug...
As to which would be better, I don't think it makes a huge difference. I'd go with the simpler one:
<noinc1ude>{subpages}</noinc1ude>definition here
simply because more complex things seem more likely to confuse people. J. Noel Chiappa 13:32, 14 May 2008 (CDT)
I agree, thanks for the second opinion. Chris Day 16:59, 14 May 2008 (CDT)
So when I edit a /Definition now, should I move the text up to the same line as the {subpages}? And should there be a space before the start of the definition, or just run it straight in there? (Also, if you have time, thoughts on the name of the templates, categories, etc for the dab stuff would be good.) J. Noel Chiappa 10:36, 15 May 2008 (CDT)

Defs links

I notice the little "[e]" link in the defs on the talk page goes not to the edit page/tab for the defintion, but to the definition itself. To me, it would make more sense for "[d]" to go there, and for "[e]" to take you directly to the edit tab. J. Noel Chiappa 19:13, 14 May 2008 (CDT)

A suggestion

Chris, I am embarassed to admit that after creating 45+ articles, I finally became aware of the {{r}} template and that it was supposed to be used in the Related Articles subpages. When I looked at a number of articles by others, I found that I was not the only one that did not use the {{r}} template.

Now, I am engaged in playing catch up with my 45 articles and it is one heck of a grind!! I would like to suggest that all of the Related Articles subpages automatically include a sentence or two at the top of the page about using the {{r}} template, including a link to CZ:Related Articles. I now know that the ? mark on the Related Articles subpage tab already does that, but I had never paid attention to it ... and apparently others have also ignored it. - Milton Beychok 00:33, 15 May 2008 (CDT)


Thanks Chris - well caught! --Mal McKee 13:51, 15 May 2008 (CDT)

Definitions meet disambigs

Cool! J. Noel Chiappa 15:20, 15 May 2008 (CDT)

Wrote that before I read my talk page; I saw what you'd done with the reel disambig page in RecentChanges, and immediately appreciated it!
The only comment I have is that 'here in the future' Tux, since there is a Tux (disambiguation), will be a redirect to Tux (Linux), or some such disambiguated name. So your proposed code of 'if {BASENAME}/Metadata exists, give the basename's definition, not the dab message' wouldn't work. If you wanted to do that, you'd have to see if {BASENAME} pointed to "{BASENAME} (disambiguation)" or not - and I'm not sure if you can do that in a template. But I'm not sure if it's the end of the world if {{r|Tux}} did the dab thing, so that you'd have to write {{r|Tux (Linux)|Tux}} to get the Tux definition. J. Noel Chiappa 15:32, 15 May 2008 (CDT)
Just trying to cover some of the bases that we're currently stuck with, such as Tux. So much of this template coding is accommodating the exceptions rather than the rule. For that reason alone we should try and avoid having exceptions to this disambiguation idea. Chris Day 15:36, 15 May 2008 (CDT)

Writing definitions

Chris, I'm somewhat puzzled by the new concept of definitions. I didn't follow the forum discussions on it, so it came somewhat unexpected to me when I noticed them for the first time a couple of days ago. I tried my hand at a few, thinking that the article title was understood to be the first word or first clause of the definition. That is why I started lowercase. You changed one definition to start with uppercase. Now my question is: are there any style rules for writing definitions (in addition to the less than 100 words rule)? --Paul Wormer 10:10, 16 May 2008 (CDT)

Chris, having just finished adding definitions to 45+ articles (which took hours), I would like to point out that the Definition subpage asks us to use no more than 100 characters (not words) and that limit is highly unrealistic. It ought to be revised to at least twice that amount. Milton Beychok 12:07, 16 May 2008 (CDT)

But the approving editor is not editor in both of these workgroups...

See Crystal Palace at the top: "Article approved by an editor (see the talk page) of the History Workgroup and Architecture Workgroup"

The wording is incorrect. It was approved by an editor, who has to be an editor of *at least* one of the listed workgroups; and the History and Architecture Workgroups are responsible for the article. Could you fix that?

TIA! --Larry Sanger 10:25, 16 May 2008 (CDT)


I had made precisely that point in the proposal, in "Open issues"! :-) Sigh, I should take another run at the people who didn't like that idea at first sight, to see if time has mellowed them at all. J. Noel Chiappa 12:04, 16 May 2008 (CDT)

Brain wave on definition line breaks

Hey! I know! What we do is have a template (I'd go for {{Def}}) which does some meaningless conditional, and tell people that if they want to transclude a definition, they should do so via that template, which will eat the line break. Plus to which {{Def|Foo}} is less characters than {{Foo/Definition}}! :-) J. Noel Chiappa 12:10, 16 May 2008 (CDT)

Time to introduce you to more CS jargon! :-) If you have some operation that does one thing, and as a side-effect does something else, sometimes we do that operation and throw away the main result (or, as you did in {{Def}}, do something useful with it) just to get the 'side-effect'.
This one's a great example. The amusing part is that it's so obvious in retrospect, but neither of us thought of it for several days! My familiarity with side-effects is probably what eventually gave me the nudge... J. Noel Chiappa 12:30, 16 May 2008 (CDT)
Umm "The trigger for me ... realising that a specific template was actually quicker than writing the whole .. term. ... The side effect occurred to me when I was writing {{Def}}" - you mean we both came up with the same idea at the exact same time, and independently?!? (Because I'd just read your plaintive recent note to Larry, and as I was doing that, the penny dropped!) How funny!
Yes, good to see all the defs. One more thing CZ has... J. Noel Chiappa 13:23, 16 May 2008 (CDT)

More line-feed-ology: check out CZ:Definitions#R template! Notice how you get a line feed before the '*'? Wierd! (Or is that deliberate, to space out entries on See Also pages? J. Noel Chiappa 13:42, 16 May 2008 (CDT)

Chris, please explain

Chris, I don't understand why you deleted the definition I had created for a catalogue page:

04:38, May 16, 2008 Chris Day (Talk | contribs) deleted "Air pollution dispersion modeling/Catalogs/Compilation of air pollution dispersion models/Definition" ‎ (content was: A compilation and brief discussion of the many air pollution dispersion models developed worldwide.' (and the only contributor was 'Milton Beychok'))

I want to include that catalogue page in the "Related articles" subpage of a number of articles. Therefore I created the definition for the catalogue page. Is there some other way? Or what? I would appreciate some guidance. Milton Beychok 13:24, 16 May 2008 (CDT)

/Approval page

Hey, can we suppress prompting to create the Approval page? Now that we have /Definition as well, it's just that much more work to move a cluster; one more page to tag for deletion, etc. /Approval's only are needed when an article is Approved, and the Constable can surely create it when needed? Now, that reminds me, time to go write CZ:How to move an article... J. Noel Chiappa 15:09, 17 May 2008 (CDT)

Good point. I'll try and track those prompts down. Chris Day 15:49, 17 May 2008 (CDT)

And yet more on Definitions

Geez! Do you know how long it took me to find the + sign???!!!

Someone fix the skin, I'm begging you.


Sorry for (probably) writing a 'plain-as-the-nose-on-your-face' thing down, but I just want to make sure you guys are cool with this.

As far as I can tell, I will still have to use {r|name} template, right? Like, if I'm typing on a Related Articles page, and I want to list something for which we don't yet have an article? I realise that that will create more standalone definitions which will have to be moved to articlename/definition, soooo....I guess my guestion is, go ahead and use {r|name} anyway, or type some other %&*%^! code in, or--here's a thought--can you start an article with a definition, even if there's no article? (Keeping in mind that a one- or two-sentence definition does not a stub make, I mean.)

Am I making sense? Aleta Curry 16:39, 18 May 2008 (CDT)

Aleta, you can simply start writing, no need to move things again and again. And definitions without articles live here. -- Daniel Mietchen 12:03, 19 May 2008 (CDT)

See User talk:Aleta Curry#Formatting troubles. {{r}} seems to be inserting an extra new line before the '*' it inserts. Can we get that removed? (And if removing it breaks some pages somewhere that expect it to be there - ARRRGHHHH!!!!!)

And speaking of inserting a '*', that was a non-optimal design choice; it means users can't use that template with a '#', or in some other way - they get the '*' whether they like it or not. I suppose you could move the guts of R out into a new template that does everything else (which people could call directly if they want), and have R call it after inserting a *, but I think a lot of things use R now, so it might be too inefficient. The other choice, copying the guts of R over, mean we now have two almost-identical templates we have to update in sync. Blagh. J. Noel Chiappa 18:02, 19 May 2008 (CDT)

Oh, here's another one. {{r}} includes an edit link for existing definitions, but {{def}} doesn't. I think probably having {{def}} not include one is the right thing (minimal functionality - the same thing I was saying above), but maybe we could have an {{edef}} one that does have an edit link, would that be useful? J. Noel Chiappa 20:49, 19 May 2008 (CDT)

Ah! Clever! Very clever! I like it. "indent" is probably as good as anything else - I might have gone with "fmt" (for "format"), a few less characters to type, but it's all good. One question - when you do the "indent=|" thing (i.e. a null argument), the line of output comes out in a box, of the kind you get when there's a leading ' ' or tab on the line before the link to the term - is the code such that there is always a space in there? If so, that's ugly. How about changing the code to require any space to be part of the "indent=" argument (if that argument it given)? That way, you can (should you want) get the line with nothing in front of it at all. J. Noel Chiappa 07:31, 20 May 2008 (CDT)
No, no, no, not default to an nbsp - don't add anything except what's given in the argument. So the Wiki markup emitted if the argument is given as "" would be:
'[[linked term]] definition'
thereby leaving it up to the user to insert whatever formatting they wanted. So it could be '*' or '* ' or ':* ' or whatever they want. I.e. the code would be:
'{{{fmt|* }}}[[{{{1}}}]] {{:{{{1}}}/Definition}}
or something like that. J. Noel Chiappa 07:44, 20 May 2008 (CDT)
I'm being slow this morning! :-) {{r}} doesn't have a second argument at the moment, right? How about we make the formatting an (optional) second numbered (i.e. non-named) argument? J. Noel Chiappa 07:52, 20 May 2008 (CDT)
OK, so I'm officially a non-awake moron. I did look at the documentation, I promise! I just didn't see it. So either we can make it a third argument (in which case the code will have to be written to work with a null second argument, i.e. {{r|Foo||:*}}, because we don't want to require people to type the second one), or we go back to names.
BTW, don't forget that rather than writing:
{{#if: {{{1|}}} | {{{1}}} | something else }}
you can just say
{{{1|something else}}}
in most cases. (If you need to distinguish between an undefined parameter in a template, and a blank one, you need to use #ifexpr.) I know you do use this, but if you see the other one in a template, you can rip it out, it will make the template smaller and simpler. J. Noel Chiappa 08:17, 20 May 2008 (CDT)

I tried {{{3|* }}}{{#ifexist: etc and it works fine. Note the space after the '*', and none elsewhere (although you could take it out - "*[[" will emit the same output as "* [["). But check out User:J. Noel Chiappa/Sandbox, and then look at the source. Is that wierd behaviour, or what? :-) I think it must be that thing with an extra newline being emitted - it must be coming out after the leading '*' and before the second '*' (in the middle cases), i.e. right at the start of the template invocation. That's why the ones with the '**' (at either end) work, the new line is not splitting the two '*'s apart. In those middle ones, the reason the naked bullet is coming out right up on the left margin is that it sees:




and of course then we get two separate bullets, fully left-justified. Dont ask me why the **{{r|link||}} case works, because it should be seeing:


if my theory about when the 'rogue' new line is appearing is correct, but clearly it's not appearing then! J. Noel Chiappa 08:49, 20 May 2008 (CDT)

Oh, that is truly wierd! Why the difference in the Zap1 behaviours between the two cases?!?! J. Noel Chiappa 09:54, 20 May 2008 (CDT)
Nevermind - I just realized you took out the null third arg in the second Zap1. J. Noel Chiappa 09:56, 20 May 2008 (CDT)
No biggie - {{r|Name||**}} and similar all work fine - you just leave the second arg out if you don't want to change it to something else, and everything works fine. See the Sandbox. J. Noel Chiappa 10:00, 20 May 2008 (CDT)
As predicted, getting rid of the ' ' in {{{3|*}}} worked fine - no difference in behaviour from before. Whether you put it in ({{{3|* }}}) or not is purely up to how you think the template code looks when edited. J. Noel Chiappa 10:20, 20 May 2008 (CDT)
Umm, sure, stuff like {{r|Foo (bar)||Foo}} will work; it takes whatever is in the third arg and sticks it at the front of the line.
I'm not sure which will be used more; a lot of people might like to hide the disambiguator ("(bar)"), in things like {{r|Foo (bar)|Foo}} (which is why that second arg got added). I don't think the "||" will confuse people - once they see an example, they'll get it pretty quick, I think. The problem with switching it is that you'd have to track down all the places which use {{r|Foo (bar)|Foo}} and change them; probably not worth the hassle/work. And then you'd have to explain to people that to get the old {{r|Foo (bar)|Foo}} behaviour, you now have to say {{r|Foo (bar)||Foo}}; i.e. you've still got the "||" issue with them, plus you have to get everyone used to the new invocation sequence. Best to let the dog lie... :-) J. Noel Chiappa 10:28, 20 May 2008 (CDT)


Chris: did you get my email concerning the Editorial Council? Please either reply or put a message on my user page. Many thanks, Martin Baldwin-Edwards 01:46, 20 May 2008 (CDT)

Ha! In reply, I should say that it is possible to sit in the back row and merely vote, if that is what you want to do. We really need the Council to be composed of active editors and authors, so your presence would be most helpful. Can you email confirming acceptance? Many thanks. Martin Baldwin-Edwards 01:55, 20 May 2008 (CDT)

Moving definitions to new template

I just moved the Highland games definition to the new template. I used the To do link on the talk page of the Highland games article. The instruct said to place some "noinclude" language on the new Definition template. There was no way to do this using that link except to first create the new Definition template (by moving the old one) and then, after the move, editing the new template to include the requested language. This seems convoluted.

Anyway, the new template seems to work just fine without the "noinclude" language. It would be nice if there were a more direct way to do this as I have about a hundred or more definitions (old template) to move.

James F. Perry 07:49, 20 May 2008 (CDT)

Okay, got your message about the new Def template. I'm gearing up for a major definition moving session sometime later this week. James F. Perry 08:03, 20 May 2008 (CDT)


Hey, thanks! You didn't have to do that! I had a list (abstracted from Constellations, and then munged in my programmable editor :-), and was working my way down them, and my eyeballs were falling out of their sockets, so I called it a night, and figured I'd finish it off tomorrow (today). Unlike you, I can't function on 4 hours of sleep! :-) J. Noel Chiappa 08:01, 20 May 2008 (CDT)

Oh, and thanks for the /Approval thingy too... J. Noel Chiappa 10:01, 20 May 2008 (CDT)

rpl and pl

Yes, I saw this page, and it inspired me to watch out for something like rpl. -- Daniel Mietchen 01:29, 21 May 2008 (CDT)

Of interest

See a comment on Daniel's talk page: User talk:Daniel Mietchen#Definitions of ambiguous (and so disambiguated) terms --Larry Sanger 10:18, 21 May 2008 (CDT)

Aaaarrrrgghhhhh! My template doesn't work right when the /Definition subpage is transcluded somewhere else - BASEPAGENAME then refer to the page which is transcluding the definition. Is there an easy way to fix this? (Other than subst: - you don't know about subst: much yet, right? :-)
Also, while I'm here asking about BASEPAGENAME, etc, is there a function which, on page Foo/Bar/Zap, returns just Foo? I thought BASEPAGENAME did that, but it's more like STRIPONELEVELNAME (calling it BASEPAGENAME is so misleading). J. Noel Chiappa 12:03, 21 May 2008 (CDT)
I don't know substitute. But since you don't want to do that any way I'll think some more. But first, how will this be used on an example like Tux? Or is this only for articles that do not exist? Chris Day 12:06, 21 May 2008 (CDT)
The only way to get Foo on Foo/Bar/Zap is to call the pagename from the metadata. You have to use the format {{../../Metdata|info=pagename}} but that will only work on sub-subpages of foo. Chris Day 12:09, 21 May 2008 (CDT)
Ah. OK. Not that I needed to do the two-level strip, I was just curious (I had to read the documentation for those functions, and wondered how you did it). How do you tell you're on a sub-sub-page anyway (to do the ../../ thing)?
It turns out subst: doesn't do the right thing either; it sticks the string {BASEPAGENAME} into the text of the page that transcluded the template, so it still malfunctions; I was hoping it would evaluate that too, and save the evaluated form, but it doesn't. I need to go melt my brain by trying to understand subst better, I guess.
Or is there another way to refer to the pagename of the file that contains a given expression, and not any transcluding page? Let me read that documentation first, I guess....
For Tux, see Daniel's page. J. Noel Chiappa 12:22, 21 May 2008 (CDT)

All the templates in the subpages hierarchy get their information from the top template {{subpages}}. In that template there are two sets of defined fields (only one is used for any given page). One set that will work on sub-subpages and another set that will work on all other pages. The test for the type of page is relatively simple.

{{#ifexist:{{{{BASEPAGENAME}}/Metadata|info=pagename}}|Use fields that work on subpages and basepages| }}
{{#ifexist: {{../../Metadata|info=pagename}}| |Use fields that work on sub-subpages}}

Only one of those expressions can exist on any given page, thus we can differentiate the level in the hierarchy to some extent. Chris Day 12:31, 21 May 2008 (CDT)

Ah, so. Clever. I was going to suggest something else, but it won't work. Blasted templates, they are so limited! J. Noel Chiappa 13:41, 21 May 2008 (CDT)

Moving pages - URL arguments

Hey, check out this, and then to see it in action click on this - please don't click the "Move page" button! For a full list of the possible arguments, you need to look at the source here. (Let me know if you can't work them out - they are just after where it says "function MovePageForm" - I don't know PHP, the language this is written in, yet - guess I need to learn it :-). J. Noel Chiappa 14:20, 21 May 2008 (CDT)

The third one is double dutch to me. But the
in the first one looks promising. Do you want to try it on the {{Move def}} template? Beware the Template:Def's that are associated with a redirect. They are the purple ones at Category:Move def. We should probably do those manually. Chris Day 14:25, 21 May 2008 (CDT)
Ah, if you click on the second link, (here), you can see it in action over here, with some arguments you might find useful.
I will go through the third one and produce a list of all the arguments to the 'Special:Movepage?' form. Some of them won't work for me, but might for you (e.g. 'wpDeleteAndMove').
I wonder if we could do something that moves a cluster with this? Can we have a button, that, when clicked, invokes several URL's? And I wonder if there's an argument that suppresses the move confirmation page (so users don't have to click 'Move page' N times)? I'll have to look... still, gotta be something useful we can do with this!
I'll look at Move def - not sure exactly what you mean by the redirect stuff, but I can probably work it out. J. Noel Chiappa 14:39, 21 May 2008 (CDT)
Don't worry about the fact that if someone clicks both 'move' and 'cleanup', the page no longer shows up as purple; if someone else clicks on 'move', they will get an error message. (Course, we do lose knowing which ones have been done, since they aren't redirs any more, but I figured it was worth it to automate the cleanup.) Anyway, this should make it really easy - a couple of clicks to move a page. Course, we still have to add the {{subpages}} by hand. Also, I couldn't figure out a way to include the ~~~ in Template:DeleteRedir - can you?
I'd say don't worry about the ~~~. Nice job, I like the idea that this might be able to help cluster moves too. I'm going to give your handiwork a test run now. As far as the ones being done is concerned, that should not be too much of a problem as long as I delete them on a regular basis. Fortunately I have that tool at my disposal, I suppose, for this very reason. Chris Day 15:39, 21 May 2008 (CDT)
I could have sworn that I saw that preload work at one point today - maybe I'm confused, I've been doing so much fiddling around. Oh well.
I've been looking for something that allows me to preset the edit summary (for the cleanup); there is a wpSummary argument, but I think it only works when you construct a URL which will actually save a modified version of a page. Which of course would be better - one fell swoop.
I found Manual:Parameters to index.php and am looking at it now... J. Noel Chiappa 16:03, 21 May 2008 (CDT)
No, I think I was just confused. I moved Stavanger/Definition‎ and I thought it worked, but I see (looking at documentation) that it couldn't have. Oh well.
As for the other arguments (like wpSummary), I think some of them can only be passed inside an HTTP transaction, not in the URL. Bummer. J. Noel Chiappa 16:52, 21 May 2008 (CDT)
Check on both of those. Here is the source, with URL supported functionality in the function "getContent". According to that, you can indeed only do a preload if the page doesn't exist (although it looks like you might be able to use preload if "section=new"). The supported arguments to &edit in URL's are 'section', 'preload', 'undoafter' and 'undo' - that's alllll, fffffolks. For the rest (which includes just about everything), you have to do an HTTP POST operation, which means Javascript. So we can almost certainly write a "move cluster" function in JavaScript, and add it to the skins. OK, I know what I'll be learning soon... :-) J. Noel Chiappa 21:52, 21 May 2008 (CDT)

Umm, your latest changes broke {{Move def}}. If the related article page exists, we don't get the move/cleanup links. I could probably fix it, but it looks a little hairy - if you know what you did, it's probably easier/faster for you to do it. J. Noel Chiappa 17:43, 21 May 2008 (CDT)

I thought about GNU/Linux/Definition for a while. The thing is that it had a different definition from Linux, so I decided that it could live as a stand-alone definition. I suspect that Linux will eventually wind up as a redir to Linux (kernel), becauase we have a Linux (disambiguation). Typical naming non-standardness.... J. Noel Chiappa 08:53, 22 May 2008 (CDT)
Yeah, once I figured out the delete link, I realized the speedy notice was a waste of time - but I saw you'd already deleted the step telling people to add it. Not that I planned no doing many, at that point - I don't have access to delete! :-) J. Noel Chiappa 10:03, 22 May 2008 (CDT)

Dab box

That's listed as one of the open points in the proposal. There are good arguments both ways: one is the navigation tab bar jumping up and down that you raise, the other is that people should see that notice first in case they are on the wrong page. I personally don't care much - I'm all about function, not form! J. Noel Chiappa 10:21, 22 May 2008 (CDT)

For now, I'm onlying putting {dabbox} on articles where [basename] points at the article, not all disambig'd articles. That's anothr of the open points (whether to put that header on all, or just the one (if any) that [basename] points at. (Too bad we don't have strings - I wouldn't have to pass in 'basename'. I wonder if I should have made those default arguments, instead of named? I'd make purpose #1, and basename #2, so if someday we get strings we could drop the second one.) J. Noel Chiappa 11:41, 22 May 2008 (CDT)

I'm going to try to get people to use {{dabhdr}} in dab pages, so they have a consistent look (which we can easily change, if we want), but that's in round 2 of dab stuff - round 1 is mechanisms, 2 is policy (e.g. what do dab pages look like). I like the concept of using {{r}} in dab pages - that would be something else to take on at that point. J. Noel Chiappa 16:05, 22 May 2008 (CDT)

A heads up though, Larry said somewhere, I forget that the disambiguation should not be definitions of the term, just signposts (my term). i just used you {{dabhdr}} template, that will work well. Chris Day 16:08, 22 May 2008 (CDT)
Wonder why he doesn't want that? Well, he might not want separate text definitions, but he might be OK with {{R}}, given that it uses /Definitions. We'll have to find out what his reasoning is - I don't always agree with him, but he's usually got a good point. J. Noel Chiappa 16:14, 22 May 2008 (CDT)
He usually has reason and will listen to reason too. It's not a done deal but as he might listen to the arguments for it. Personally I have not really thought it through. Chris Day 16:19, 22 May 2008 (CDT)

Pipe in rp template

So this bar (I don't know if I must see it is an "or" as a Unix "pipe") works in the rp template! I didn't know that, thank you.--Paul Wormer 10:45, 22 May 2008 (CDT)

See User:Paul_Wormer entries: "energy" and "Faraday's law", where I naively called the template with too many arguments.--Paul Wormer 11:10, 22 May 2008 (CDT)
You are fast, it is fixed already.--Paul Wormer 11:11, 22 May 2008 (CDT)

Constellations move

I just realized we can pull the same kind of thing for moving the constellations that you have for Defs. (Clever idea, BTW!) So that should make for somewhat less work for John... J. Noel Chiappa 16:35, 22 May 2008 (CDT)

Awww, how nice of you to think of me! =P John Dvorak 16:38, 22 May 2008 (CDT)


Thanks for your changes. I thought that blue and purple would not the best color choice (Link). Is there a possibility to test my changes on such templates without annoying everone else if it does not work? Alexander Wiebel 08:40, 23 May 2008 (CDT)


Hi Chris, was wondering if you could get rid of This for me. I made it a long while back but let it go to rack and ruin; Not even sure if should get its own article in that format anyway. Cheers. Denis Cavanagh 11:29, 23 May 2008 (CDT)

Ok, thanks. Denis Cavanagh 11:42, 23 May 2008 (CDT)

Please do! Any advice or tips you have would be great! Denis Cavanagh 12:27, 23 May 2008 (CDT)

Template: Def

You are definitely 'da man'! J. Noel Chiappa 14:54, 23 May 2008 (CDT)

BTW, I am starting to think that creating a zillion "{Basename}/Definition" pages is probably non-optimal, since there are going to be a bazillion of them. I think we should make it the rule to always get to definitions through an accessor function (i.e. template), and we can code all the templates (e.g. {{Def}}) to check for the existence of "{Basename} (disambiguation)", and issue the error message. We can easily track down the one's we've created so far with Special:Whatlinkshere/Template:Dabdef. Sounds like the right move? J. Noel Chiappa 15:25, 23 May 2008 (CDT)
Definitely the right move there is no obvious reason why anyone would create one. If it was created, it could not be seen using the {{r}} or {{rpl}} templates and their like anyway. I agree we should just delete them during a disambig move.
I assume you saw i am trying to make such moves a lot easier with {{Move}} in conjunction with a moveto field in the metadata template. In this way the computer can find all the pages that need to be move automatically and guide through a sequence that makes sense. It might take me a while to build it and trouble shoot it though. Chris Day 15:33, 23 May 2008 (CDT)
I think that's a great idea. Editing the metadata isn't the most user-friendly interface, but it's better than nothing. Hey, idea: the template that displays the metadata template could include a 'Move cluster' button when moveto exists; that way, they edit the metadata, hit 'Save page', and the next screen they see has a 'Move cluster' button right there. Too bad there isn't some way to use <inputbox> to do start the ball rolling - or is there?
One thing you might not be able to automate: if there are subpages to Catalogs, there's no easy way to find them in a template, right? (If we coded this in JavaScript instead, and made it part of the skin, we could call [Special:Prefixindex] and pull the results apart...) J. Noel Chiappa 15:43, 23 May 2008 (CDT)
I already figured out the subsubpages problem, as in I don't know what to do about it. The prefix thing was the only easy way i could come up with (but by necessity will involve a more manual approach; i can't do java). All in all this will be a tool for the more experienced. As it is, it is the more experienced that are doing these moves anyway so , if nothing else, it will save us some time.
The metadata button idea etc. is good, once the code is in place a prettification like that will be useful. I'll think about if it is possible to utlise input, I'm not sure if it will be possible. Chris Day 15:47, 23 May 2008 (CDT)
Thinking about this more, it might be better to get all the functions (delete, move, edit etc.) coded for the various different pages and just have them ALL appear at once. If it's a tool for the experienced it is less important that the commands appear sequentially. Just thinking a loud. Chris Day 15:54, 23 May 2008 (CDT)
Yah, that's probably the right thing; having them all there in one place will really speed up moving a cluster (for the cognoscenti :-). But they only need to show up on the metadata page, yeah? You can right-click/open-new-window on them one after another (as needed), with the metadata last on the list; the last one should open it for editing when done with the move, so you can change the name and remove the 'moveto'. J. Noel Chiappa 20:17, 23 May 2008 (CDT)

Move template, etc

Sorry about that - didn't realize you were working on Mars with it! J. Noel Chiappa 14:54, 23 May 2008 (CDT)

No problem. I got distracted when moving doing venus. I will carry on with the test article. Chris Day 14:56, 23 May 2008 (CDT)
When {Foo} is getting moved to "{Foo} {disambiguator}", and we're setting up a "{Foo} (disambiguation)" page, I think "Talk:{Foo}" should always be linked to "Talk:{Foo} (disambiguation)", because any discussion about "{Foo}" (e.g. whether to set "{Foo}" to point to the dab page, or a particular meaning) ought to happen on the dab page's talk page, so all that sort of stuff gets discussed in one place. Discussion about the content of a 'preferred' (i.e. claimant of the redirect from the basename "{Foo}") meaning should happen on that meaning's "Talk:{Foo} {disambiguator}" page, of course.
Yes, I normally do delete the redirect from "Talk:{OldName}" (along with all subpages); I always check [Special:Whatlinkshere] for all of them, and if anything links there, I update it. If there was some page with like a bazillion links, I might make an exception, but otherwise I try an always clean up. J. Noel Chiappa 15:34, 23 May 2008 (CDT)

On pages with no metadata, they get a "Move metadata page" prompt. J. Noel Chiappa 07:50, 24 May 2008 (CDT)


Iupiter is the Roman equivalent of Zeus, IIRC. J. Noel Chiappa 17:16, 23 May 2008 (CDT)

Oh, {{dambigbox|<usage>|<basename>}} is the way to go; {{dabbox}} is a redirect for fewer characters to type. J. Noel Chiappa 17:33, 23 May 2008 (CDT)

help needed in French words in English

Hi, Chris and Noel (alphabetical order!), could one or both of you take a look a the French words in English article and see if you could figure out a scheme to index the list of words with a TOC so that we can go directly to the A's, for instance, or the T's. We're now at the point where there are so many words that it's beginning to become quite tedious to scroll up and down a gazillion times a day. Many thanks! (PS, you don't have to index all the letters -- just the first couple, so that I can see how you did it: I'll then take care of all the rest....) Hayford Peirce 12:00, 24 May 2008 (CDT)

Hi, I would like to use the CZ logo on my web site, with a link to the CZ homepage on it, to help spread the word on Citizendium. You can see it on - is that OK? If I'm not allowed to use it, I'll take it down asap.

Disambig formatting

Please comment here. Thanks! -- Daniel Mietchen 08:18, 26 May 2008 (CDT)

I thought {{disambig}} was supposed to be a footer template. One more thing to sort out, sigh (I'd been putting off what a dab page should look like until the mechanics thing was through the system, but maybe I should get moving on the way they look, too.) But yes, no reason to have two separate header templates. J. Noel Chiappa 15:47, 27 May 2008 (CDT)

Core Article structure

Hi Chris, here are some more thoughts on Core Articles (I always write "should", but read this as recommendations):

  1. General:
    • CZ:Core Articles should be split into CZ:Core Article policy (for policy and guidelines) and CZ:Core Articles (for contents; that what currently is at CZ:Core Articles#Article lists)
    • Displaying the state of the current article via the pl family should be made a (default?) preference to be set by registered users, others should always see the non-pl variant. This may enhance collaboration within CZ, without hindering the user experience of those not logged in.
  2. Presentation:
    • The Article overviews should be organized as gallery grids rather than lists (e.g. like Biology/Gallery, but with two hierarchical levels):
    • Six columns (for the six supercats) and one row for each workgroup per supercat
    • One logo per grid element, representative of the workgroup, labeled with the title of the workgroup, e.g. Biology (alternatively, the title could be displayed when hovering over the image); this means that all workgroups should have a logo
    • A click on the image or label should lead to the lower-level gallery of articles within the scope of the relevant workgroup (allowing for duplications, e.g.Protein to be listed in both CZ:Core Articles/Biology and CZ:Core Articles/Chemistry).
    • Especially the core articles should be enriched with multimedia materials, e.g. animations or lectures on the subject
    • The overall design should consider the special needs of blind, colour-blind or similar groups of people
  3. Logos:
    • All CZ:Workgroups should have a logo (image or pictogram) that should be used on all articles within their scope (at least in the main namespace).
    • There should be a coherent way of transforming those logos into an even more "active" version during the relevant CZ:Workgroup Weeks.
    • {{r}} on Related Articles pages should display, in addition to what it currently does, the logo of the workgroup(s) of the linked item, excluding the current workgroup
    • Subfields within a workgroup will be hard to grasp in pictograms, so their logos should be image-only
    • On CZ:Workgroups, the icons should be rearranged such that the icon belonging to the current supercat is displayed in the first row of the table, while no icons should be displayed at the place currently filled by {{Workgroup navigation}}, but the individual workgroups listed below should be accompanied by their logo
  4. Duplications
    • Duplicate entries of core articles across fields should be encouraged (e.g. Biochemistry, Biophysics and Cell biology could all list Cytoskeleton as being within their core)
    • Articles belonging to hybrid subfields like Biochemistry or Biophysics should be listed in both parental cats, using the same logo (perhaps with some kind of colour inversion to highlight the change of perspective)
  5. Disambiguations
  6. Outlook:
    • The Core Articles structure should be adaptable for use by CZ:Eduzendium which mostly necessitates finding non-written (and thus, at least after a while) non-core articles
    • At least the two superficial layers of the presentation hierarchy (see above) should be arranged such that children can easily find their way through it.

So far. -- Daniel Mietchen 11:58, 26 May 2008 (CDT)

Template:Biology open house goes well into the direction of points 2 and 3. Thanks for that! -- Daniel Mietchen 04:47, 27 June 2008 (CDT)

More use of abc= field

See Talk:Diplomacy, U.S., Timeline‎ - we should be making more use of the abc= field. How about if we had a version of R that took as input the article name, but spat out a bullet using the abc field as the text of a piped link. I.e. it would generate: * [[Article|{{Template:Article/Metadata|info=abc}}]] I'd like to get rid of the inconsistency where articles have names like "Guadalcanal, Battle of" but we don't have "Faraday, Michael". The abc= field, if used more, seems a way to meet all these goals. J. Noel Chiappa 18:49, 27 May 2008 (CDT)


I have favoured deleting "{Foo}/Definition" (which would contain {{dabdef}}) when "{Foo} (disambiguation)" exists for a while now. (I forget where I said that). So I'm perfectly in agreement with you on that - we should always get to /Definition subpages through templates, which can check for "{Foo} (disambiguation)" and give the appropriate message.

For finding existing "{Foo}/Definition" pages which call {{dabdef}}, use Special:Whatlinkshere/Template:Dabdef - that will show them all, no need for the cat. J. Noel Chiappa 19:44, 28 May 2008 (CDT)

Good point about /Definition pages which don't use {{dabdef}}.
About the categories not being updated properly thing - that's been a bug on Wikipedia for a long time, so I'm not sure it's just CPU cycles. There used to be a problem with the internal structures not keeping enough data to allow cagegories to be updated properly - see this bug report and this one. They claim to have fixed it, but as you can see by this and this it's still happening. Not sure why - probably some bug where the task gets lost, or dies before it's completely finished, or something. J. Noel Chiappa 13:30, 29 May 2008 (CDT)
Another good reason not to use categories and try to get related articles pages to take off. Along those lines, I've been wondering whether we should be using templates in related articles. For example, after changing the name of Pluto to Pluto (dwarf planet) I had to change many different RA subpages to point to the correct definition rather than the disambiguation page. Maybe we should have terms that are usaully grouped together, such as the planets in a universal template? Or is that just getting to obscure? As RA subpages develop we will be seeing a lot of redundancy, although, a lot of effort is already saved by using the {{r}} template since all definitions get updated automatically. Chris Day 14:00, 29 May 2008 (CDT)
My other category pet hate is that we have to see articles full name. This is OK for articles but a nightmare now we are using subpages since there is so much repetition in the entries. Recently, I played with the idea of having all subpage categories on the talk page, rather than on the actual subpage, to simplify the category lists (I tested with Catalogs, see Category:Biology Catalogs along with Talk:Dog). It works great, in that there is a simplified list but the link does not link to the category which makes the idea a non starter. So my question is there any way to get a different name listed in the category but link to the real page. The equivalent of a pipe link. If not i think we should probably just abandon the categories since they are visually a mess and don't even update correctly. Chris Day 14:11, 29 May 2008 (CDT)
Sorry, didn't follow that "the link does not link to the category"? Which link? The link in the category? It goes to the article, not the talk page, but that's the right thing, in some sense. Or did you mean "it links to the article, not the catalogs subpage"? Yeah, I can see that's non-optimal.
There is no way, AFAIK, to link to an article but have a differentname show up in the category list. The [[Category:Foo:Bar]] syntax, AFAIK, only affects the alphabetization - the article's ordinary name is still listed. But let me look at the source, and see what I see.
I hear you about the categories. Yeah, with templates, they don't work so hot... very complicated stuff, to get all the linkages right when a template is updated. (See the BugZilla links...) J. Noel Chiappa 16:31, 29 May 2008 (CDT)

Move to Catalog

OK. I see you did it. I'm sure it's a good thing. What is the catalog and what should I do with it (pointers are fine)? Howard C. Berkowitz 22:50, 28 May 2008 (CDT)

Disambigs and writing levels

Hi Chris, inspired by a recent discussion due to disambig-initiated page moves, I am thinking about an easily implementable mechanism that would allow to use the disambiguation criteria (those usually in parentheses in disambig page names) coherently as a determinant for the level at which the article should be written. This would allow for, say, Root (plant) and Root (botany) to coexist and to be written for different audiences (in terms of length, structure, specialist language, further reading etc.).

To get this system work, we would need a kind of (hierarchical) tag cloud (or list of categories) from where the disambig tags (or categories) have to be selected for each article, ideally by means of the future easy-to-use metadata editor. On Root (disambiguation), we could then do something like *[[Root (plant)|Root, general audience icon]], **[[Root (botany)|Root, more specialized audience icon]] and so on, interlinked such that the intro of the more specialized article should start at the level that the more general one reaches at its end, from where it would be referenced under "Further reading". This might be a means to avoid friction on talk pages about the target audience or about how to name articles, at least for the two first levels.

For specialist level articles possibly listed at a disambig page (they could of course always go to some sub- or sub-sub-pages, which represent a hierarchy already), I would drop the need to have article titles containing the term "root" (or something pronounced or written similar) but suggest coherent use of the second-level tag for the article names instead, as in [[Apical meristem (botany)|Apical meristem]]. Once disambiguation becomes necessary there, too, we could easily and flexibly adapt to that, e.g. by splitting Apical meristem (botany) into [[Apical meristem (botany, root)|Root apical meristem]] (as opposed to [[Apical meristem (botany, shoot)|Shoot apical meristem]]). I would like to avoid using too specialized tags, e.g. Rhizology, and so the tag cloud should be well-limited.

I am sure we cannot decide such things on our own, but before discussing it broadly, I would like to get your opinion, especially from a Core Article and Workgroup Week perspective. Thanks. -- Daniel Mietchen 04:27, 29 May 2008 (CDT)

Update: Another discussion on article naming here. -- Daniel Mietchen 04:30, 29 May 2008 (CDT)

I need some time to think about this. I did see it, so you know. Another thing I was thinking was that the wikilinks could some how be configured to show the definition when you hover the mouse over the link. This would be more useful than showing the title. Chris Day 11:08, 29 May 2008 (CDT)
  • Take your time (consider things like proper names as something that might cause problems or call for exceptions to the rule which was more category-inspired), also on the Core Article structure - these topics are well worth some second or third thought.
  • Well-dosed hovering could be useful on multiple occasions, and so could image maps (e.g. to unfold the Core Articles from the hierarchically organized gallery grids mentioned above). Daniel Mietchen 12:30, 29 May 2008 (CDT)
It was the biology montage that got me thinking of image map a long while ago and your ideas above rekindled that memory. The extension was only recently installed so we had been unable to play with it until this year. I agree with your well-dosed thought, we need to be careful to maintain simplicity where possible and only use these tricks when there is a real gain to the users. Chris Day 12:38, 29 May 2008 (CDT)
Another place where hovering might be useful (and inviting) is red links which could show the corresponding definition (if it exists, but we could work a bit more systematically on that). This would probably lower the threshold to start an article, especially since it can be assumed that those who contributed to writing the definition might also be willing to help out with the article itself. Do you see an easy way to do that? -- Daniel Mietchen 08:57, 1 June 2008 (CDT)

Talk redirects

Hey, those redirects from [[Talk:{Foo}]] to [[Talk:{Foo} (disambiguation)]] are deliberate - they are there so that any/all discussion about a disambiguated terms takes place at [[Talk:{Foo} (disambiguation)]], not (potentially) spread over both [[Talk:{Foo}]] and [[Talk:{Foo} (disambiguation)]]. J. Noel Chiappa 07:19, 2 June 2008 (CDT)

Here's the list of ones that got deleted:

if you want to restore them - although I suppose I could always just recreate them. J. Noel Chiappa 07:34, 2 June 2008 (CDT)

Ah, I had forgotten that was part of the plan. Chris Day 10:03, 2 June 2008 (CDT)

Template documentation

Oh, I hadn't gotten that far yet! (The thought had occurred to me, but I decided it was too hard to solve 'quickly', so I hurriedly pushed it to the back of my mind. :-) Once I get this thing working, then I can decide how this relates to {{newtemplate}}. J. Noel Chiappa 11:13, 2 June 2008 (CDT)

We should standardize on either /Doc or /doc for the documentation subpage. Most seemed to use /doc, so I went with that. You have a couple that use /Doc:
but since they are transcluded through direct naming, I didn't fuss with them. Also, I found these two:
which seem to be lost; i.e. for templates that have been deleted - any idea what's up with them? J. Noel Chiappa 10:55, 3 June 2008 (CDT)

Hi, any chance I can get you to look through the table section at CZ:Templates#Subpages and break it up into sections (e.g. ones for subpages headers and footers would be one obvious group to pull out)? I've been doing a lot of looking at templates and creating new folders, and have them (at least, the ones that were listed there - there seem to be a lot of new ones that aren't listed) all somewhat better sorted out now. J. Noel Chiappa 10:55, 3 June 2008 (CDT)

Hey, thanks for all the hard work there - I didn't expect it to be that much time for you!
As to your bugs: The extra </noinc1ude> at the bottom is because there's an extra </noinc1ude> at the end of {{Subpages/doc}}. The other one, with the headers, that's kind of wierd. Let me investigate a bit more... J. Noel Chiappa 14:09, 3 June 2008 (CDT)