Template talk:Chembox/Archive 2

Supplementary pages for major compounds
Working on the main template, I came up with an solution to the problem of having a huge table (for completeness). This is the same solution used in the chemical literature, and indeed it is Wikipedia official policy- namely, if it is getting too long, start a new page. 98% of people accessing a page on hydrochloric acid or methanol want to know basic information, not standard molar entropy etc. Let's keep obscure information out of important pages. Therefore I suggest that the table should include the main things that will satisfy 98% of users, but include links from the table to an MSDS and to a supplementary page. This supplementary page would include thermodynamic data and spectral information useful to experienced chemists, without filling up bandwidth for thousands of schoolkids etc. I am writing the table in this way- I hope to post it by this evening (USA EStime) but would like feedback on this. Walkerma 15:59, 5 May 2005 (UTC)
 * Excellent idea! This is indeed a break-through idea! I really like to see this in effect. Are you thinking about adding a link in the bottom of the infobox to a detailed infobox on a separate page? Sounds good to me. Thinking on, this means that ultimately, we'll only have ONE simple chembox, with a link to some larger table (on a separate page) if necessary. Wim van Dorst 19:16, 2005 May 5 (UTC).

New chembox
I'd like to direct comments on the new chembox here. There is a draft version for people to look at. It attempts the impossible- to be compact & easy to use, yet suitable for any chemical compound. It uses the supplement idea (see above); detailed safety data are on an MSDS page (which might just be an external link), and thermodynamic & spectral data on the supplement page. As well as general feedback, I'd like people's specific thoughts- please vote for in the main table or in the supplement: Walkerma 21:56, 5 May 2005 (UTC)


 * 1) Refractive index Aldrich etc put this in all their catalogue entries, but is it widely used these days?
 * supplement: not widely used. Wim van Dorst.


 * 1) Viscosity I put in as "liquids only, optional", but should it still be in the supplement?
 * main table: especially with the remark Liquid only. Wim van Dorst.


 * 1) Dielectric constant I put in as "solvents & dielectrics only".  I like to use it as rough guide to solvent polarity, but am I alone in this?
 * supplement: not widely used. Wim van Dorst.


 * 1) Dipole moment
 * main table: but only when the structure section is applied (which should be optional). Wim van Dorst.

For the rest I think we should make recommendations about the the Supplement page, e.g., specific names of section titles so that in the chembox you can directly link to that particular section. Notably, I strongly recommend to make only one supplement page and not several for the various sections. Wim van Dorst 08:36, 2005 May 6 (UTC).


 * Update comment: I really like the way the new chembox is developing. Excellent progress. Even the lines.... Wim van Dorst 20:01, 2005 May 11 (UTC).


 * My update comment- Thanks for updating the lines/colour, Wim. I think we can make the switch to this being the official version of the infobox on Monday May 16th. Ideally I'd like to hear comments from User:Cacycle before setting things in stone, as I'd hate to change it again too soon- one problem in the past has been the use of 73 different "standard" tables.  To encourage standardisation we may need to spend some time converting some pages to the new style.  A couple of unresolved issues:
 * 1) Will the colour for the organics still be blue as it is (in practice at least) currently?  I would favour this, though it does make the standard table a little more complicated.
 * 2) The supplement page needs to be standardised. Wim, would you be able to take the three tables and make one big table out of it?  Instead of the present |S&P| |Spectral| |Thermo| going across the page, I'd like to see one combined |S&P|Spectral|Thermo| or similar . I don't think I can do this- Wim, can you do this?  Then we can create a supplement template. Walkerma 21:51, 11 May 2005 (UTC)


 * We're working now on one chembox, which has one proper colour scheme. Personally I like it when (in the end) the two simple chembox variant have exactly the same layout (colours, lines, wikilinks) albeit in shorter form. But let's discuss that later, if necessary.
 * I see now three tables which fold on the page. If the page is narrow (or the font large), it winds up as just one long table (with extra footers), If the page is wide (or the font tiny), there are three tables next to eachother. Where do you want it differently? Just one long long table? That's easy, but is is nice? Wim van Dorst 23:08, 2005 May 11 (UTC).
 * I already did the two simple chemboxes. Wim van Dorst 14:35, 2005 May 12 (UTC). And the supplement too. Is this what you have in mind (normal wiki-layout), or might perhaps just one longer (combined) be better (like the chembox itself)? Wim van Dorst 15:27, 2005 May 12 (UTC).


 * Yes, it looks great now. I will try to do the details you mention on my talk page. Walkerma 16:11, 12 May 2005 (UTC)


 * Update: I have put the toluene page into the new style. I don't have time to write the supplement page for it, maybe tomorrow.  The table seems rather wider than it should be, and even the draft template has grown an inch or so in the last 24 hours.  Is this necessary?  I much prefer the skinnier table, it increases space for text and also reduces problems when printing articles. Please check the page and let us know what you think. Walkerma 19:49, 12 May 2005 (UTC)

Wim van Dorst asked me to comment. What I've seen till now is a great job. I have a few suggestions, although I am not a chemist nor a doctor yet (and I hope I did not say something silly ;). Where I use the word category, I mean a sub-box, a group of lines.
 * Correct links - "Ingestion" has been made a redirect to eating; the contents were moved to Wiktionary. This link should be corrected or removed, while other links should be formatted so that they bypass redirects (e.g. molecular formula; I mean put this in: molecular formula ).
 * References in the bottom. Do they point to references of the topic described by the article or of the infobox? This should be defined (add e.g. "infobox references" or something).
 * Dissociation constants. There are three articles now: dissociation constant, acid dissociation constant, base dissociation constant; and two lines in the infobox. Perhaps the articles should be merged and the lines in the infobox replaced with "Dissociation costants". Besides this, are not Ka and Kb used as dissociation constants instead of their negative logarithms pKa and pKb?
 * The order of categories. It seems more natural to me to have it in the following order(from top to bottom): image, general, structure, properties, hazards, supplementary data page, related data compounds.
 * Related category: what is: "related ?"
 * "Chemical data" line in the category "supplementary data page". That sounds somewhat strange as other things in the category also can be considered chemical data. I don't have a suggestion.
 * "Related compounds" line in the category "Related molecules". Will it not interfere with "Other cations" and "Other anions"?
 * what about the line "Biological significance"? Would it be redundant or unusable? Or perhaps "Occurrence" (where in nature, or is the substance synthetic etc.)

Thanks, happy wiki-ing. --Eleassar777 20:32, 12 May 2005 (UTC)

Overall very good suggestions, most of which I have already implemented. I'm much obliged that you as an experienced non-chemical wikipedian had a fresh view at it. Thanks. Wim van Dorst 09:52, 2005 May 13 (UTC).
 * Correct links: Done. Pfew, there were quite a lot of them.
 * References: Done. Good suggestion
 * Dissociation constants: For future work.
 * order of categories: I mainly agree and therefore move Hazards down.
 * Related: This is up to the person who fills the table. He/She is expected to be knowledgable, so no change.
 * Chemical data: Agree. The supplements page is still under construction, so no decisive change yet.
 * Related compounds: No problem here: the knowledged table filler will be able to make the organic/inorganic distinction to remove either field when inappropriate. no change.
 * Biological significance and Occurrence: Interesting. I think I rather want this in the main text.

Thanks for having a look at my proposals and implementing some of them. May I just point out two things again: a) dissociation constant - that's K, K and K, isn't it? And b) I also don't understand why properties come above structure. --Eleassar777 15:45, 13 May 2005 (UTC)


 * Thanks for your comments, Eleassar, they were very helpful & insightful, (for a "non-chemist" you seem to know plenty of chemistry!) Thanks Wim for chasing down the links, something I didn't do before or I'd never have got the draft done! Regarding the remaining loose ends-
 * Dissociation constants Indeed, the pKa (or pKb) is strictly speaking "Minus long to the base ten of the acid (or base) dissociation constant", but that's quite a mouthful to put on a data table. I've looked at a couple of books and they only ever refer to these things in the text as pKa without using words- though March's voluminous index says "pKa- see acidity constant." I would propose that we just use the term "acidity (pKa)" and "basicity (pKb)" in the table, using links for those unfamiliar with pKa.  I will change the draft table to reflect that.  Ka and Kb (the actual acid or base dissociation constant) are rarely thrown about in chemical conversations, I know organic chemists will always say, "That proton's got to be pretty acidic, around 6 (understood as pKa)."  It is used in the same way as pH is used for aqueous solutions.  They won't refer to Ka any more than people talk about the hydrogen ion concentration in the river.
 * Structure vs Properties This in effect mostly means Molecular scale properties vs Bulk properties. IMHO: Most people work with bulk matter and so things like density, BP etc are the main things people need, they should be near the top of the table, right below a drawing of the structure. I accept that the box would "flow" better with structural stuff first, but I think usage in this case trumps that.  You should be able to type in "toluene" into Wikipedia and see structure, BP and density without a lot of scrolling. Walkerma 16:48, 13 May 2005 (UTC)
 * Template wikitext vs appearance on page Some of the things you things you mentioned earlier have a lot of guidance in the code to help people through, though of course this doesn't show up on the displayed page. I really like the "related compounds" link, this is one of the strengths of the element tables, but the problem is that toluene has different related compounds than copper(I) chloride.  It's fairly open-ended because trying to pin it down to specifics makes it very clumsy. I think anyone proficient enough to be typing up this full length table should EITHER know what is meant for that particular compound, OR be able to work it out from similar pages.  To assist things there are comments in the code like for Related compounds it says "A miscellaneous heading- use for covalent inorganics;  e.g. for PCl3 you would list PCl5, POCl3, PF3, PBr3, NCl3 and AsCl3. Please omit if not applicable". Thanks again for your valuable help. Walkerma 16:48, 13 May 2005 (UTC)
 * The new infobox has been really well done and it was an honour for me to have contributed my two cents. Only dissociation constants still need to be replaced with acidity and basicity (or just plain pKa and pKb). Regards. --Eleassar777 07:28, 17 May 2005 (UTC)


 * I like what you've done with the infoboxes. Nice work!  ^_^  &#8227; &#5339;&#5505;  [[Image:Venus symbol (blue).gif|&#9792;]] [ &#5200; ] 09:28, 4 Jun 2005 (UTC)

Table resolving in various browsers: losing divider lines and border colour
As a proud co-developer of this new chembox, today I took a few minutes of a coffeebreak at work to have a look how the new chembox works out, now it is published. At work we have Microsoft Internet Explorer 6.0. And it showed the table quite differently from what I expected. (Insert here some expletives not to be written in full). The problem is in the apperent different resolving of table code by the various browsers.

 The left version is as shown in Mozilla derived browsers (which is as the table is intended), such as in Mozilla, Firefox, Epiphany, etc. The right hand is as shown in Opera. The table in the MSIE browser looks like the Opera variant, but worse. Now what? Wim van Dorst 20:51, 2005 May 17 (UTC).


 * I just took a look, as I have IE 6.0 here- I found that in the template the lines disappear in IE where they are present in Mozilla Firefox (my default). Otherwise the differences are largely superficial.  However in the toluene and HCl pages the lines are still there.  Also, if you go to print the template from IE you get the lines in there.  I notice that the resolution of you screen shot is poor- is that just an artifact of the screen shot?  My IE page for HCl or toluene looks fine on the screen. (Martin Walker asking Wim van Dorst)


 * 1) Do we lose the lines in IE only in the original template? If I go back through history on my page to the edit that says "uploaded the new template to toluene and hydrochloric acid" the lines are still missing in IE, yet they are there in all versions of the toluene page. Martin Walker

Using external links in the infoboxes
A more serious problem that I noticed is probably a general one on Wikipedia- when you go to print, any external links are written out in full. That has the effect of widening out the table to fill or almost fill the page. The toluene page looks really bad- the text is crammed into about 3 cm on the left hand side. This happens with both Firefox and MSIE. Another related problem that I'd noticed before but never addressed- take a look at phosphorus tribromide, it should look OK. Then do a print preview- and in Mozilla notice how the image goes down past the table, as it has to squeeze the page into a narrower space. Now try the same page in MSIE6.0, you will see that it prints the image right over the top of the table! I think this was in the back of my mind when I said I wanted a skinny table- but this business of the expanded links makes that go out of the window! Where should we go next?


 * 1) Are we using an incorrect format for inserting external links? If so, we need to find the correct format, and rigorously enforce that in the table, to stop the unwanted expansion.  If we are already following protocol, then should we just tell people not to put external links in the table?
 * 2) Regarding the table/image conflict, we probably need to be more careful about checking browser and print compatibilities when we first write the pages. We have some old Macs down the hall, I will check IE and Netscape on there. I know I've been bad at doing this...
 * 3) Are there other issues I'm missing with the IE version? If so, please elaborate, Wim. Walkerma 00:54, 18 May 2005 (UTC)


 * I am for discouraging external links within the table, even if it means that we have to add See also: MSDS to the bottom of every page... Physchim62 06:24, 18 May 2005 (UTC)


 * Me 2. I did remove them from the infobox of hydrochloric acid for trial, and I like the printout much better.


 * This is an unfortunate feature of Mediawiki; there is currently no way to control the display of external links. However, it seems that the behavior may change in version 1.5 &mdash; see http://bugzilla.wikipedia.org/show_bug.cgi?id=2301.  &#8227; &#5339;&#5505;  [[Image:Venus symbol (blue).gif|&#9792;]] [ &#5200; ] 09:00, 4 Jun 2005 (UTC)

I have updated the toluene page too. Any thoughts on the problem of printing tables & images together in IE? Do we just try to write narrow images? See PBr3 Walkerma 01:46, 19 May 2005 (UTC)

Wording of the Disclaimer
I reverted some latest tweaking there, so as to have the most accurate disclaimer, referring only the potential inaccuracy of 'information contained in the table' and not 'the table', referring to 'hazards', not 'safety' (that section there is called Hazards). Nonetheless, I also misread the 100.000 kPa thing, first thinking about onehundredthousand kiloPascal. So I reworded it to read 'by definition'. Wim van Dorst 22:20, 2005 May 18 (UTC).
 * The current wording of the "hazards" sentence is poor English: might I suggest "includes any information on hazards, which..." or "includes any hazard information, which..." See the essay "English chemists secretly practice German vice" in The Chemist's English by Robert Schoenfeld for a light-hearted discussion of the (mal)formation of English compound nouns ;) Physchim62 23:38, 19 May 2005 (UTC)
 * I wouldn't mind a copy, if only for bedtime reading. Would you have an on-line source? I changed the disclaimer text, btw, albeit slightly differently from your suggestions here. Please comment. Wim van Dorst 21:39, 2005 May 20 (UTC).
 * It's copyright, so you'd need a library (VCH published the edition that I have). Disclaimer text is now OK with me. Physchim62 22:23, 20 May 2005 (UTC)

Formalization?
Shall we try to formalize the three chemical infoboxes as a Wikipedia policy (set in rock and poured in concrete)? Or as a Wikipedia guideline (just the rock :-)? I think we should go for guideline. Wim van Dorst 22:40, 2005 Jun 2 (UTC). I vote festina lente. Physchim62 06:20, 3 Jun 2005 (UTC)
 * It is already a guideline (See Featured article criteria#4. I wouldn't go any further until we have resolved the current issues:
 * 1) 18 out of 27 A-Class articles have non standard tables (including 5 which are in HTML format, not wikipipe).
 * 2) Many many tables have external links, when we know that this can cause browser problems, not to mention the fragility of external links (almost all the links to "Hazardous Chemicals Database" are broken, for example)
 * 3) WikiProject Drugs uses a non-compatible format for their drugbox: we will need to compromise for articles which fall under both Drugs and Chemicals.
 * I would say guideline too. Chemistry is too full of exceptions to be able to set things in concrete. As we say in Newcastle Upon Tyne (my original home town), let's gan canny. 16:16, 3 Jun 2005 (UTC)

Then I close this discussion by simply putting the guideline template up. Wim van Dorst 19:28, 2005 Jun 3 (UTC).

Component-based infoboxes
Would it be a good idea to build the infoboxes from components, as at WikiProject Astronomical objects? You could have: and simply pick and choose which to include. This way instead of having two totally separate templates for "short" articles, and translating to a new template when the article expands, you simply grow the infobox as required. --Phil | Talk 11:27, Jun 10, 2005 (UTC)
 * chembox base
 * Everything has IUPAC nomenclature, Chemical formula, Molecular mass/Molar mass, SMILES, CAS number, Melting point, Boiling point, Density
 * chembox structure
 * chembox hazards
 * chembox related
 * chembox end


 * Hi Phil, thanks for pointing this out. I previously took notice of this technique, and it is indeed very nice. The thing is that I, for one, see the chembox as one total block of data. It is all obligatory for a good (A-Class) chemical compound article, apart from where is it explicitly indicated. There is no modularity in the information provision there. The two simple boxes are only for starter and stub articles in the WikiProject.


 * The total of three infoboxes covering it all in a straightforward subst:template installation (fast code!) are very easy to maintain and use, contrary to your proposal (six chemboxes templates, and all fields in them again as templates), with nested templates as resource grabbers.


 * So all in all, I see two major things in favour of retaining the current version of the infoboxes. Not that there aren't any improvements possible:
 * use of CSS for table makeup
 * better MSIE/FireFox/Opera support
 * etc.
 * which is already doable with the chembox header template that we use for table layout definition. Wim van Dorst 22:16, 2005 Jun 10 (UTC).

When to use full and when to use simple chemboxes
I do not agree with the current text when to use the full and when to use simple chemboxes in articles. I suggest to use the following explanation instead:


 * Chembox infobox for full articles
 * The Chembox infobox template is applicable for all chemical substances including organic and inorganic compounds. This infobox is the recommended infobox to use where detailed information is appropriate, for example:
 * Common solvents like ethanol or benzene.
 * Common reagents like acetic anhydride, methyl iodide, or hydrochloric acid.
 * Simple chemical structures like formaldehyde, acetic acid, or water.
 * Common starting compounds and building blocks like toluene, indole, or butadiene.
 * Simple infoboxes
 * Two simple infoboxes are covering the most important data for organic and for inorganic compounds where in-depth information is not available or required. These templates can be used for:
 * Not so common compounds like benzofuran or adenine.
 * Complex compounds like bromophenol blue or chlorhexidine gluconate.
 * Biomolecules like serotonin or cholesterol.
 * Natural products and secondary metabolites like ergotamine or azadirachtin.
 * Drug-related articles where currently no drugbox is in place.
 * For starting an article.
 * On pages.
 * Simple infoboxes may later be replaced by the full infobox or expanded stepwise.

I have not checked if all examples exist or contain an infobox. Feel free to add the respective articles and infoboxes - especially the acetic anhydride article is a shame :-) Cacycle 7 July 2005 19:55 (UTC)
 * Hi, Cacycle, You have used more words than originally to exactly formulate what I fully agree with. But what is the important thing that is wrong with the original text, which in my humble opinion means the same? Its brevity? Wim van Dorst July 7, 2005 20:10 (UTC). PS The acetic anhydride doesn't have any infobox yet, but menthol has a nice one. That is the reason why the former is an A-Class article in the Chemicals wikiproject, and the latter would be listed as Stub if it were in the wikiproject. WvD.


 * Speaking as someone who hated the old infobox, I have to say I now agree with the text as is. We should use the starter box for stubs and short articles, but a full-length article deserves the longer table.  The new infobox is limited to only important data- minor things are spun off onto a supplementary page (still not written for menthol, I accept).  I would argue anyway that adenine is common, and as such it deserved complete information- wouldn't the pKa or pKb value be useful?  Isn't solubility information and chiral rotation on ergotamine or cholesterol useful?  Isn't it convenient to be able to be at a page on benzofuran and have a convenient direct link to indole without wading through text?  The "non-useful" entries for complex natural products such as dipole moment, coordination geometry, related anions, etc, are simply deleted from the box, but for a simple inorganic you would delete the SMILES and the chiral rotation entries.  Is there anything in the menthol page (besides flash point) that would be inappropriate for, say, cholesterol? Heck, I even managed a supplement page (albeit rather sparse) for gold(III) chloride, not exactly common! Walkerma 7 July 2005 20:44 (UTC)


 * Sorry for not making my point clear. The current version is somehow self-contradictory: On one hand it says the full infobox is the recommended infobox to use "where detailed information is appropriate". On the other hand it continuous with "and for high quality chemicals wikipages" and "This template is recommended by the WikiProject Chemicals for all its wikipages.".


 * My point is that for many articles - even for high quality pages - the full chembox is inappropriate. We have to keep in mind that this is an encyclopedia, not a database of chemical and physical properties. For many compounds the chemistry part of the article is just a small part (e.g. see the drug articles). The readability of such articles and infoboxes would clearly suffer from chemical data deserts. The full infobox (mis)leads to listing all available data even if they are without any foreseeable significance for our readers - chemists as well as laymen.


 * It might often be more appropriate to extend the simple infoboxes, e.g. with pKa or pKb values, if this is an important property of a compound instead of automatically adding them to every compound just because the data exists somewhere.


 * This is in no way a criticism of existing tables or disrespect for the work we put into setting them up. I think at the current stage most full chemboxes are appropriate (although menthol might be a borderline case...). It is also no criticism of the full chembox structure which now looks pretty good. I just don't want that the current guideline misleads us into wasting our time with setting up full infoboxes where it is clearly inappropriate and at the cost of the article's quality.


 * Cacycle 7 July 2005 23:06 (UTC)


 * I agree that in cases where the project interfaces with other projects, we need to tread lightly- we had a long discussion with the Drugs Wikiproject regarding paracetamol which came to exactly that conclusion- paracetamol has a page principally because it is a drug, not because of its chemical properties. However in cases where the chemical aspects are the major part of the article, it is perfectly appropriate to use the full chembox.


 * I think a lot of these disagreements come from a difference in perspective. When I write a page such as gold(III) chloride, I see it as part of a Wikipedia five years from now, where every chloride in the periodic table has a full-length article. I imagine millions of chemists worldwide using Wikipedia as a valuable source of information, and millions of schoolkids and college students writing papers on things like "The Chemistry of Gold" finding useful "nuggets" of information there.  I am aware that many others see a much more limited chemical content on Wikipedia, or they focus on where Wikipedia is now.  However I have been astonished to see the pace of change.  When I wrote aluminium chloride around six months ago I think I was the only Wikipedian at the time writing lengthy articles on metal chlorides.  I joked a little later that "I will probably be bored by the time I write a page on NbCl5".  In January an article on NbCl5 would have seemed rather perverse- yet now I find that someone else has already written a significant article (rather technical IMHO) on it, and what is more this now seems perfectly appropriate, as we have about half of the metal chlorides in the periodic table covered. Or take a look at Category:Chemical_compounds_by_element to see how far we've come.  What I have seen is that as the chemistry content grows we get more chemists writing articles, and therefore the chemistry content goes up etc. etc. I have absolutely no doubt that many things like cholesterol and benzofuran will have extensive articles with full-length tables eventually, quite possibly by the end of the year.  That doesn't make it a database instead of an encyclopedia, it just makes it a bigger and better encyclopedia.  And think what an amazing resource we will have by 2010! Walkerma 8 July 2005 03:42 (UTC)
 * I agree with User:Cacycle almost entirely, so I will mention my one disagreement first! Some articles are going to stay with a short form chembox for the simple reason that the data necessary for the full form does not exist in verifiable form. Short form boxes can also always be extended to provide important information beyond that which is in the template, as I have done for amino acid articles.
 * My rule of thumb is to not to use the long form where this will be substantially longer than the article text (although I allow myself exceptions where I intend to extend the article myself, as in arsenic trioxide). The problem as to whether these articles can become A-Class is one that we set ourselves, and one that I am quite happy not to worry about yet! A short form chembox is better than no chembox at all&mdash;if this were not true, the templates would not be there! It has to be said that is substantially easier to convert a short form chembox into a full one than to insert a full form chembox from scratch.
 * As yet I have had no comments on the text in the style guidelines, where I deliberately tried to find a middle line between my position and Wim's (Wim: maybe I failed, but I did try ;)
 * Physchim62 8 July 2005 09:56 (UTC)
 * Hi, PC, I'm really very sorry to inform you that you have horribly failed in treading the middle line: in your Style Guidelines you write about the infobox exactly MY position. You missed your own position at least by a mile :-). Now, as we apparently all feel very much alike, can we get on with adding whichever chembox you consider most appropriate and creating A-Class articles? Wim van Dorst July 8, 2005 19:38 (UTC).