Misplaced Pages

Revision as of 19:27, 12 February 2015

This page is for discussion about format, layout, content (parameters), and application howto of the Chemical infoboxes. Discussions older than several months, and now outdated, are archived. Planning and progress of implementation of these infoboxes to the numerous articles is discussed and (where possible) organized in the Chemicals WikiProject, part of the Chemistry WikiProject.

Chemistry Template‑class

This template is within the scope of WikiProject Chemistry, a collaborative effort to improve the coverage of chemistry on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ChemistryWikipedia:WikiProject ChemistryTemplate:WikiProject ChemistryChemistry Template This template does not require a rating on Misplaced Pages's content assessment scale.

Archives

Archive 1 Archive 2 Archive 3 Archive 4 Archive 5 Archive 6 Archive 7 Archive 8 Archive 9 Archive 10 Archive 11 Archive 12 Archive 13

This page has archives. Sections older than 90 days may be automatically archived by Lowercase sigmabot III when more than 5 sections are present.

Solubility in water

In "Solubility in water", water links to water; I suggest it should be changed to properties of water, e.g. in {{Chembox Properties}}:

] → ]

Christian75 (talk) 21:42, 10 November 2014 (UTC)

For {{Chembox}}, {{Chembox Properties}} is the single place to edit indeed (AFAIK). Note that {{Infobox drug}} has this data element too. -DePiep (talk) 23:20, 7 December 2014 (UTC)

 Done in {{Chembox Properties}}: . No changes in {{drugbox}} by this. DePiep (talk) 23:36, 7 December 2014 (UTC)

• Also added (in the sandbox now) in {{infobox drug}}. Talked here. -DePiep (talk) 15:02, 18 December 2014 (UTC)

As I have stated here, I think it would be better to change the text to "solubility in water". Solubility is as much a property of the solute as it is of the solvent. Furthermore water is a common term that does not need to be linked. Boghog (talk) 16:09, 18 December 2014 (UTC)

Good point, fine with me. Christian75, what do you think? -DePiep (talk) 17:09, 18 December 2014 (UTC)

I do not agree that water is a common term in regards to chemistry. Because is has technical dimensions that are specifically relevant to the topic. But antoher target could be water solubility which is a redirect aqueous solution. Christian75 (talk) 11:49, 2 January 2015 (UTC)

That's a better link. I am undecided. Note that {{infobox drug}} follows this thread. Since I updated that box yesterday, it now has water unlinked. But depending on the outcome here we'll adjust that of course. In the end, {{Chembox}} and {{infobox drug}} will be the same in this. -DePiep (talk) 12:23, 2 January 2015 (UTC)

I agree that aqueous solution is a much better target for the link. My preference would be the piped link "]" as "water solubility" is slightly ambiguous (i.e., water can act as either a solvent or a solute). Boghog (talk) 00:08, 3 January 2015 (UTC)

We are closing in ;-). You mean to label it:

1. Solubility in water

2. Solubility in water

3. Solubility in water

-DePiep (talk) 11:13, 3 January 2015 (UTC)

#1 is simplest (least number of links). #2 is bad style (two adjacent wikilinks so it is not immediately apparent if there is one or two links). #3 is most accurate but somewhat redundant. I am leaning toward #3. Boghog (talk) 11:46, 3 January 2015 (UTC)

I used #1 "Solubility in water" in {{drugbox}} for its relaxed link view. One link should do. {Chembox} will follow. -DePiep (talk) 09:38, 9 January 2015 (UTC)

•  Done in {{Chembox}} too. I consider this closed. -DePiep (talk) 16:12, 6 February 2015 (UTC)

Request, page break for each CAS No

Would it be possible for each CAS No (and indeed for each value for every 'identifier') to start on a new line? Currently the text wraps, which leads to problems when the CASNo_Comment option is used, or even worse problems when it's used for some enties but not other. For example: Copper(II)_sulfate - I know the formatting behind chembox, so I can figure out what referes to what, but our readers most certainly can't. --Project Osprey (talk) 13:24, 17 December 2014 (UTC)

Yep, I will go into the sandbox shortly. Each CASNo to get their own datarow; makes a nicer list anyway. (I don't exactly understand what you mean by "worse problems", but could be solved by this already). Any other identifiers you are thinking of? -DePiep (talk) 13:39, 17 December 2014 (UTC)

I've only seen the problem with CAS No's as its more common for compounds to have several of them (hydrates etc) but I certainly haven't been everywhere so it may also effect pubchem, chemspider etc. This might create a new issue though; some people have been using <BR> to try and sort out the formatting themselves (i.e. Cobalt(II) chloride). --Project Osprey (talk) 14:00, 17 December 2014 (UTC)

Seems like a good idea to me, just insert a <br /> before the next ones. For the ones that already are 'fixed', they will probably still be better than the ones without, and that will clear out in time (maybe someone can figure out a search-trick to find them and repair them by AWB?). --Dirk Beetstra 19:12, 17 December 2014 (UTC)

1. Aha, hadn't seen that indexes like |CASNo1=, |CASNo2= already exist: . |CASNos= and |CASOther= is for the unstructured leftovers only.

2. Indeed, the <br/> should be (and so will be) added by {{chembox}}. Internally, Chembox will check & remove any <br/> that was added with the article input. I expect, that would have broken the CAS-link so it won't have happened that often. -DePiep (talk) 20:46, 17 December 2014 (UTC)

I'll take on the CAS numbers first. Question: currently we can enter 6 (|CASNo= and indexes 1–5 as in |CASNo3=). Today Magnesium sulfate uses them all, so I suspect that limit may be too low for some others. I can quite easily change that now. What number of CASNo's should we cover? (=max number of CASNo's in a chembox, with no need to use |CASOther=)? Note: this number should also cover each other possible identifier: SMILES, InChI, etc. In other words, the one max number to cover (say 8) must do for each and every indentifier. If there are 8 CASNo options possible/needed, and 11 SMILES, the number should be 11 for all. (Background reasoning on request). -DePiep (talk) 11:27, 18 December 2014 (UTC)

|CASNo6= and |CASNo7= added to the sandbox (that's 8 controlled CASNo's max). Building full support in {chembox} and by the bot. -DePiep (talk) 15:20, 18 December 2014 (UTC)

• (edit conflit) Good question - the answer it appears is lots. Hydrates seem to be the most common reason for a page needing multiple CAS No's. The highest hydrates I've found on chemical pages are octadecahydrates (Aluminium sulfate and Chromium(III) sulfate). Assuming that CAS No's exist for each distinct hydrate, we would need a maximum of 18 (say 20 to make it a round number)... I admit that does seem like a lot, is 20 possible? --Project Osprey (talk) 15:31, 18 December 2014 (UTC)

Possible yes, practicable no. I'll build for indexes 0–9 (that's 10) all right. 0–7 (that's 8)

In my grander design, all identifiers and properties can have all these indexes (0–9 then): SMILES, InChI, PubChem, and also MeltingPt, ... . Then, any editor can make them correspond in one article: CASNo7=SMILES7=PubChem7=MeltingPt7 etc. Next step, an extra parameter |use_index=yes can be added and voila, every id has its sub-dentifier added ("R, S, (dodecahydrate)"). (For images there is another route). So that is why I want this index for all those params, and how to use it.

Now when we start this before having an chembox module (in Lua), this is nigh impossible for 20 indexes. We'd have to hardcode all 20 params for each (today there are 500 params already). But the Lua module can easily look for & find an index when it is present in the article (not by hardcoding the number, but by looking for it in the used parameters only). That Lua chembox I'm thinking of is not available yet (don't expect it this year ;-) ).

So, meanwhile we can start serving 10 indexes in core parameters, the editor can add & check carefully. More can be added in the |CASOther=, |CASNos= freetext parameters, uncontrolled. (I note that the two links only have two CAS registry numbers added, so there is no pressing need right? No editor is actually waiting for this to add 11–18 ;-) ). I must say that I do know about data structures, but I have no deep knowledge of chemical data definitions; I'm learning.

This way, ten should do for now. Questions left? -DePiep (talk) 17:05, 18 December 2014 (UTC)

Make that eight max. Is huge already, and we'd better have the principle in there widespread (less deep). -DePiep (talk) 17:45, 18 December 2014 (UTC)

No, there's no urgent need for 18 or 20. If it were easy to do then we could future-proof ourselves, but as it is not then 8 should be plenty. Thank you for this. Project Osprey (talk) 22:03, 18 December 2014 (UTC)

What I am mainly afraid of here is massive table-creep. 18 CASNo's!? Of most compounds that have more than 5 CASNo's, only 2-3 are really important (e.g., generally: the water-free variety, the water-full variety, and sometimes a stable one inbetween. I know, MgSO4 could have 8 varieties (0-7), and one could possibly force to make all 8 (well, it looks like s.o. did ..) .. but I guess that about 3 of them have hardly any 'real life' meaning). I would really consider to have a not-too-high max built in, and let the rest be covered by the CASOther or CASNos fields. Remember, the page is about MgSO4, not about MgSO4.5 H2O, and in some cases that page may have it's own Misplaced Pages page (some hydrates are so important that they have their own page next to the parent salt-page). --Dirk Beetstra 09:56, 20 December 2014 (UTC)

In a way, see it that the individual fact should pass our notability guidelines as well, ánd it should be verifiable that it really exists (is it a mixture of 6 MgSO4 and 1 MgSO4.7 H2O, or 8 distinct units of MgSO4.1H2O ...). Especially for CAS, it is not in their task to check how reliable the data is that assigns a certain formula to it. --Dirk Beetstra 10:01, 20 December 2014 (UTC)

Sure, but that would be a Feature Article, where everything & their neighbors is clear out. Until that quality level, there might be a period where the inclusion/exclusion of substances is not stable or clear, the expected separate articles do not exist yet, etcetera: the editor might need some extra CAS mentionings. Or, more simple: the {chembox} could be applied into a List of MgSO4-related chemicals (my layman's example begs some belief, but the important word is: "List"). Or a List could be made from a bunch of stub individual articles.

In general, IMO the quality of edits should not be enforced by such a chosen template-suffocation. (That would be waaaay to difficult to document. The documentation saying: "If you need more than 5 CASNo's, you must create more articles" - ouch). In this situation, I prefer that grey-area substances & CASNo's are added, up for improvement (reasoned removal) later on. A removal could be into a new article too.This also covers the 'notability' check mentioned: needed is indeed, but not by forbidding addition; a notability talk can follow.

As for using |CASOther= for more related CASNo's: better not. Any CASNo worth entering or maybe worth it (the grey area), should be under control (bot). CASOther is for non-controlled data, outside of article topic (like |CASOther=See also ] for 1234567-89-0). Or any other prose that does not need control.

(I note that I propose in the code a change of behaviour: all entered CASNo's are shown, there is no requirement to fill in |CASNo= first. All CASNo's are equal-ly important in this. And another simplification of documentation btw).

In short, I propose to not use a low number in the template to 'enforce' a quality somehow someway. Inclusion/exclusion of a CASNo for quality reasons is part of the wiki edit process. |CASNOther= shold be used for data that does not need a check, only. A wiki List article does not have a number restriction at all (so could have many CASNo's). With this, 5 is too low, ~8 I like. -DePiep (talk) 21:22, 21 December 2014 (UTC)

• Project Osprey, see below #here. Took some time, worth it ;-). -DePiep (talk) 23:06, 26 January 2015 (UTC)
•  Done Project Osprey. Piece of cake really. Copper(II) sulfate looks good right away, without edit :-) Have a nice edit. -DePiep (talk) 16:15, 6 February 2015 (UTC)

Do we need both CASNos and CASOther?

We have parameters |CASNos= and |CASOther= that can have input as free text(unchecked, unstructured). I don't see any benefit having two. I propose to merge them, and deprecate |CASNos=. (Technically: CASNos is removed from documentation, and categorised for edit . When not used any more in articles, it can be removed from code). DePiep (talk) 20:08, 22 December 2014 (UTC)

I think it comes from a merging of fields, following a new format. Something archaic. Choose one, deprecate the other, merge their use and hope people forget. For some fields we had an autocategorisation into 'Category:Chemical_infoboxes_with_misplaced_or_deprecated_parameters', this could be added to that until that cat becomes empty. --Dirk Beetstra 20:39, 22 December 2014 (UTC)

Yes, as you describe. I'm already using that (and sibling) maintenance cat to list such to-edit pages. -DePiep (talk) 09:17, 23 December 2014 (UTC)

• Deprecated: CASNos → use CASNoOther. For all these indexed params.

-DePiep (talk) 18:45, 8 February 2015 (UTC)

Whitespace in {Chembox}: a standard

In article code, whitespace can be very helpful & distracting. Irregularities are distracting & confusing.

For {{Chembox}} I propose to use whitespace one way Y:

{{Chembox
| Name = Amonia
| ImageFile =
| Section1 = {{Chembox Identifiers
| ChemSpiderID = 217
| ChemSpiderID_Ref = {{chemspidercite|correct|chemspider}}
| StdInChI = 1S/H3N/h1H3
| StdInChI_Ref = {{stdinchicite|correct|chemspider}}
  }}
}}

That is

Use single space after parameter pipe, before/after =symbol when value present

But: inner section-closing }} on newline and with double space (to align with the opening Sectionparameter).

Also: No space at end of line. No empty whitelines in between.

This does not have any effect on presented result.

This is to prevent iregualrities like N:

{{Chembox| Name =Amonia
|ImageFile1= 
| Section1= {{Chembox Identifiers
|  ChemSpiderID = 217
|   ChemSpiderID_Ref = {{chemspidercite|correct|chemspider}}
|  StdInChI = 1S/H3N/h1H3
| StdInChI_Ref = {{stdinchicite|correct|chemspider}}
 }} }}

Any comments? -DePiep (talk) 11:50, 22 December 2014 (UTC)

The space after the pipe is unnecessary, and it is more usual to see equals signs aligned:

{{Chembox
|Name             = Amonia
|ImageFile        =
|Section1         = {{Chembox Identifiers
|ChemSpiderID     = 217
|ChemSpiderID_Ref = {{chemspidercite|correct|chemspider}}
|StdInChI         = 1S/H3N/h1H3
|StdInChI_Ref     = {{stdinchicite|correct|chemspider}}
}}
}}

which aids readability. However, editors may use any form; it's not clear what you're proposing, It would be impractical to impose one style on all editors, and it would not be acceptable to make mass edits solely to change the use of whitespace. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:30, 23 December 2014 (UTC)

All spaces are unnecessaryu; I'd like to have the indented inner "}}" to show the sttructure, aligninh the "=" looks good for the editor but is a hell to maintain (esp in AWB). I did not write about imposing or mass-edit, but about voluntary keeping the same standard. -DePiep (talk) 11:06, 3 January 2015 (UTC)

I think this is going to be very dependent on the editor. Personally I like a variation of option II above:

{{Chembox| Name =Amonia
| ImageFile1= 
| Section1= {{Chembox Identifiers
|  ChemSpiderID = 217
|  ChemSpiderID_Ref = {{chemspidercite|correct|chemspider}}
|  StdInChI = 1S/H3N/h1H3
|  StdInChI_Ref = {{stdinchicite|correct|chemspider}}
  }}
}}

Main template parameters with one space, the in-section parameters with 2 spaces (but this may be my programming-instinct at work). In the end it does not really matter, except for being able to see what is in where. I think this is why I made CheMoBot putting two spaces before the _Ref parameters and 1 space before the 'verified'-parameters in the body of the template. I don't have any preference in any form. --Dirk Beetstra 11:21, 23 December 2014 (UTC)

I know the programmers habit to indent. But here it ends up in a chaos, also because other indent-rules are applied (like indent the nested {{ }} section). -DePiep (talk) 11:06, 3 January 2015 (UTC)

New: Odor threshold value

I think that odor threshold value is a very important piece of information (see Odor detection threshold). Various sources are available:

• http://www.timestream.com/vgh/downloads/NO2articles/Ruth%201986.pdf
• http://www.ncbi.nlm.nih.gov/pubmed/8472679
• https://books.google.ca/books?isbn=093262734X
• chemical-specific publications in various journals

What do you think? Mintz l (talk) 00:50, 3 January 2015 (UTC)

Dunno, but what would a data row look like? (label text & link, expected/fixed unit)?

Odor treshold | 7×10 ppm

-DePiep (talk) 11:02, 3 January 2015 (UTC)

Chemical infoboxes

Many articles with chemical infoboxes such as Teixobactin look really bad, with the infobox taking up 50% or more of the screen width (in this case, more than 100% of the screen width, and the thing doesn't scroll either, so overflow over the edge of the browser window is not readable) Is there some way to make all infoboxes scroll automatically when the exceed the display area of an article? (ie. before they start impinging on the sidebar that says "Main Page", "Contents", etc) (Better, if they start scrolling when they take up 50% of the content area width) -- 65.94.40.137 (talk) 01:35, 8 January 2015 (UTC)

The Teixobactin infobox is formatted correctly and of modest width as viewed on my setup; however it is as described above when the "InChI" section is expanded. Is this a recent change? Forcing the value to wrap would solve this. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:55, 8 January 2015 (UTC)

InChI text wraps for me (viewing in IE). Project Osprey (talk) 11:44, 8 January 2015 (UTC)

No recent changes. As Andy describes, with me it is too wide when show (which is not the default/initial page status). (Firefox). AFAIK, long ago it was made this way keep the string 'findable', it is an identifier. -DePiep (talk) 11:55, 8 January 2015 (UTC)

... but in mobile view, it opens bad (show not hidden). -DePiep (talk) 11:57, 8 January 2015 (UTC)

Working on this. I'm optimistic. -DePiep (talk) 12:28, 8 January 2015 (UTC)

 Fixed (I think) by cropping the File:Chemical_Structure_of_Teixobactin_based_on_Nature_Article.svg figure with Inkscape to remove surrounding white space . Boghog (talk) 14:42, 8 January 2015 (UTC)

Boghog Not fixed I think. The problem is that the very long InChI string (525 characters) is not wrapped. (this was done to keep this searchable web-wise, it is an identification). You can check this by unfolding the InChI data row, or by checking the "mobile view" (a link on the very very bottom of the article). With me, it was still tooooo wide.

I am changing that in {{chembox}}, but it needs testing before spreading it into 10,000 articles.

Meanwhile, I have added spaces in the long InChI string. That allows wrapping in this article. There is a lot of edit activity, so a nice page will help. After my solution change, I'll have another lookj here. (btw, your image edit looks like an improvement in itself, it caught my eye before). -DePiep (talk) 15:19, 8 January 2015 (UTC)

The problem is that there is no way to force-wrap a long string. It will only wrap at certain characters, but there is no way to say that it should wrap randomly at a certain max-width in the boxes. That is why it was hidden, but on systems without javascript it defaults to unhidden. Manually breaking the string is a solution, but I am afraid whether search engines would be able to find the full string if it is broken, and InChI is a string that one would Google for (some online databases have that hardcoded on their identifiers to enable finding other sources/databases). It is a recurring theme, where we either have to find a solution, or implement the one that has the least collateral damage (either a few viewers get a very broken page, or some viewers can't find the page). --Dirk Beetstra 16:17, 8 January 2015 (UTC)

I wouldn't say that the addition of spaces will be too much of a problem as regards to searching for the InChI in Google - its known that Google doesn't handle long InChIs - this was the specific reason for the introduction of the InChIKey. A much bigger issue is that the spaces actually corrupt the InChI - if anyone wants to use the InChI - to generate a structure - search a database or any other task that the InChI is useful for you have to fix it, in such long strings it may not be easy to see that there are spaces causing the issue. --The chemistds (talk) 16:44, 8 January 2015 (UTC)

Still, if the InChI is broken on-wiki, even searching a sub-part of the InChI may not show up, and what determines what InChI is long - a 15-character InChI gives a 21-character-string (including the 'InChI='), which may for some pages already define as 'long'. Also using the InChI as an entrance into a search might end up in problems (external sites may not understand a 'broken' InChI. I am thinking, is it an idea to write a module that force-wraps long strings by just breaking them every #-characters without addition of hyphens? --Dirk Beetstra 07:37, 12 January 2015 (UTC)

Can't we do that with CSS, rather than code? That would leave the underlying text unchanged. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:26, 12 January 2015 (UTC)

I am working on this and yes in css. And I am optimistic. Within a few days I'll put up a proposal on this page, to invite scrutiny. Note that, with other improvements in {{chembox}}, this involves 30+ templates in the {chembox} suite (30+ are in /sandbox development now, all related).

Halfway-results are in the /testcases pages for those interested, but in my proposal I'll ask for specific checks (like: check browsers, check mobile/desktop). In other words, you can help by checking my proposals, in a few days. -DePiep (talk) 10:48, 12 January 2015 (UTC)

I think we have tried things before like that, the problem is that some InChIs do not 'break' according to grammatical rules. We have been unable to set tablewidth (with Java(script) turned off) for Wiki-tables - on computers which have that turned off, it simply does not 'break' and shows the string in one continuous line (and the 'hide'/'show' function does not work, so it is standard shown), widening the table to page-width (or even wider and giving a horizontal scroll-bar). But lets see, it has been some time since this was last tried to be resolved. --Dirk Beetstra 11:01, 12 January 2015 (UTC)

By the way, with code I did mean LUA-module to enforce a max length/width. --Dirk Beetstra 11:02, 12 January 2015 (UTC)

•  Done. The long InChI now line-break within the chembox, without a need for extra code (no spaces etc needed). Especially mobile view shows good now. Meanwhile, hot article Teixobactin has moved to a {{drugbox}}. -DePiep (talk) 16:18, 6 February 2015 (UTC)

Missing links

• What should Atmospheric OH rate constant link to, or what does it mean? Used in Phenyldichloroarsine ( = 1.95E-12 cm/molecule·s). -DePiep (talk) 07:37, 9 January 2015 (UTC)
• A link for EU Index? Is not Dangerous Substances Directive (67/548/EEC) (EU Classification) -DePiep (talk) 09:26, 9 January 2015 (UTC)
• 3DMet? Tetrahydroharmine has http://www.3dmet.dna.affrc.go.jp/html/B00115.html broken link. More. Remove? -DePiep (talk) 09:30, 9 January 2015 (UTC)

New: Identifier linking to MetaboLights database

The MetaboLights database (http://www.ebi.ac.uk/metabolights/) was created by the European Bioinformatics Institute (EBI) with the aim of filling a gap in bioinformatics resources for storing and searching data from metabolomics studies. It is already the fastest-growing database at the EBI and is the recommended resource for deposition of metabolomics data by several leading journals including EMBO journal and Nature Scientific Data. The addition of a new Chembox link to the MetaboLights database would be of considerable value, providing not only structural information but also details of metabolomic studies where the compound has been reported including the study species, the metabolic pathway(s) involved in the production of the compound, reference spectra (NMR, MS) and literature references. See, for example MTBLC16530 (3-methyl-2-oxobutanoic acid).ChEBI Namrata (talk) 16:25, 14 January 2015 (UTC)

I have no judgement on whether it should be in. But for the case of adding it, can you describe or propose:

• label text & links (in lefthand side in the infobox)
• |parametername?= and data value to be added (by article editor)
• Structure in resulting data value = article page result (visible text+link expected)
• Any other notes expected? (reference, unformatted freetext, ...)
• Preferred position in the Chembox: under which header, above/below which other data.

-DePiep (talk) 01:26, 15 January 2015 (UTC)

3-methyl-2-oxobutanoic acid is in article alpha-Ketoisovaleric acid (you know what: I copied the InChI from that website, pasted and let the wikiSearch do the job. Bingo). -DePiep (talk) 01:33, 15 January 2015 (UTC)

Bleulink MetaboLights database or MetaboLights expected. -DePiep (talk) 02:36, 15 January 2015 (UTC)

Just a note, if this parameter is to be added, please put it in the ChemBox as with all identifiers, the whole set of them, with possibility for bot verification, comments etc. We should however first see it's notability (articles should be created and stick). --Dirk Beetstra 03:16, 15 January 2015 (UTC)

Agree. Didn't see yet what new data that database adds (next to combining existing data). The link in Chembox be a (automated) <ref>erence? I hope the future article brings clarity. -DePiep (talk) 09:57, 15 January 2015 (UTC)

As mentioned above, MetaboLights is the first open access, general-purpose open-access repository for metabolomic studies that acts both as an archive for metabolomics studies and a reference database for annotated metabolites. Metabolic profiling is an important tool for an insight into a biological functioning and perturbations induced due to various endogenous and exogenous factors. Linking a Misplaced Pages chemical or drug article to the corresponding MetaboLights entry would provide details about the occurrence of the compound at species level, the basic unit of classification, the Reference Spectra and associated metadata, biological roles, cellular locations, concentrations, and raw data from the metabolic experiments. See another entry, for example MTBLC16765, tryptamine, gives information about the different species in which the compound could be found, metabolic pathways in the respective organisms, spectral details including NMR and Mass and additional literature references. Additionally, the associated metabolic studies would also allow users to compare and analyse metabolomics data for different species on different platforms and techniques along with details about other metabolites that could be found under similar conditions.ChEBI Namrata (talk) 14:30, 15 January 2015 (UTC)

Yes. Why not create MetaboLights? And what kind of link should it be, from {chembox}? Why is this not a <ref> kind of link? WP:External links confirmed? -DePiep (talk) 14:54, 15 January 2015 (UTC)

You would envisage that an identifier for MetaboLights (e.g MTBLC16765) in Chembox/Drugbox would directly link to the corresponding entry in the MetaboLights database analogous to the existing identifiers such as KEGG and ChEBI giving the additional details. Although the name "MetaboLights" itself might initially seem to apply to a restricted (metabolomics) community, the inclusion of spectroscopic reference data and experimental procedures in "MetaboLights" could be considered useful for a much broader user community including the organic and biochemical chemists.ChEBI Namrata (talk) 15:40, 15 January 2015 (UTC)

I'd be inclined to start by setting it up as a parameter in Wikidata, and to adding values to appropriate items there. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:56, 15 January 2015 (UTC)

Andy, fuck off of talks and works I am doing. First you have to apologise or retract your other threads you disrupted & PA'ed. After that you maybe could contribute. Only then. -DePiep (talk) 00:04, 8 February 2015 (UTC)

re ChEBI Namrata: Big chance that KEGG will be removed some time per WP:EL and WP:LINKFARM. Skipping twice the request to at least create a wiki article about that database, does not support its suggested importance. -DePiep (talk) 18:00, 15 January 2015 (UTC)

Creating section 'Other names' in {Chembox}

 Done -DePiep (talk) 16:21, 6 February 2015 (UTC)

|OtherNames =
|IUPACName =
|PIN =
|SystematicName =

Pantothenic acid, DDT

Names
IUPAC name 1,1,1-Trichloro-2,2-bis(4-chlorophenyl)ethane
Preferred IUPAC name 3-propanoic acid
Systematic IUPAC name 3-propanoic acid
Other names 2-Ketoglutaric acid alpha-Ketoglutaric acid 2-Oxoglutaric acid Oxoglutaric acid
Identifiers
CAS Number	123-45-6
Except where otherwise noted, data are given for materials in their standard state (at 25 °C , 100 kPa). Infobox references

• PairSizeSet

I propose to change the presentation of the other names in {{Chembox}}. They can have their own section (-header). To the right is a demo version, with more in the testcases.

Changes

• An infobox Section header is added Other names. It shows right below the images, where the names are already.
• All line width is limited for all names to ~infobox width (max row width is 22 em). This affects long names. No more expanding into the lefthand screen side. Of course, this is a very important change (current situation could be called page destroying).
• The order is changed into OtherNames, IUPACName, PIN, SystematicName. This prevents repeating names like 'Other' in subheaders, see next point.
• Subdivision: the existing four data entries have their own sub-section. With the new order, there is no need to repeat section name "Other names" or "IUPAC name". The more below, the more specific a name is. IMO this is a clear structure, and makes a natural reading flow. Also when some data & headings are not present (see the variants in testpage4).
• The subheader "IUPAC names" (-s) is pluralised when multiple names are detected. The editor can also enforce this by using |IUPACNames= (-s).
• See below for the "hide/show" option. "show" is default.
• Names are left-aligned.

Not changed

• The four name entries are not changed. Data now present in the articles is shown unedited.
• The names are right below the images.
• No data, no header. In variants.
• The |name= (infobox title, usually the article name) is not affected by this.
• Note: the   green color is used here to show it's a sandbox. It will be the soft yellow color when going live:   .

Effects

• The names, which can take up a number of lines, now are grouped under one infobox header. No suggestion as if they relate to a picture above.
• Whitespace has disappeared, esp vertically. Sometimes a 20–25% height reduction in the names block shows.
• Only one show/hide button in the names, not two four.
• All changes are checked in mobile view too. At least, this should not worsen (it doesn't).

Useful links

Testcases have the sandbox/live Chembox versions side by side (with same input). Some other changes like formatting are being developed too.

Template:Chembox/testcases4 -- Testing other names and identifiers, also long names

Template:Chembox/testcases -- General testcases

Template:Chembox/testcases3 -- All parameters used.

Comments?

Any Remarks? The focus is on the new structure and setup. The show/hide issue is minor (& comment separately below, pls). The mobile view is available via the very last link on every wikipage. You might notice other changes in {{Chembox}}, they will be opened in an other section soon. If you want to add cases to the testpages, you can ask me (or: in your pet article, preview simply with {{chembox/sandbox ...). -DePiep (talk) 02:22, 16 January 2015 (UTC)

The hide/show button

The Other names section has a button "hide/show" to collapse/uncollapse the IUPAC names (three IUPAC data entries). The top row 'Otehr names' does not collapse ever. Default setting is "uncollapsed" (show all). Collapsing can be set in a {{Chembox}} by |IUPACNames_hidden=yes (or =collapse, =hide, =hidden). Existing parameters |IUPACName_hidden= and |PIN_hidden= will have the same effect. They are kept for compatibility.

Why hide?

Per WP:accessability, information should not be hidden (do not require extra user action, like a mouseclick). So the default setting should be "show". Also, in mobile view, the show/hide option does not appear and all will be shown always. In short, an article should show good in uncollapsed form.

{{Chembox}} has added these options because many names are technicxally and very long. That is, they should be on the page for e.g. search engines, but the reader would not need to see them. This is the (old) reason to hide them. However, since the wiki is approached by mobile device —which does not "hide" at all— that argument is moot. A modern rule would be: if it must not be shown, don't add it to the infobox! -DePiep (talk) 02:22, 16 January 2015 (UTC)

Other issues

•  Done. Implemented. Long names do not break the page (they line-break within the infobox). -DePiep (talk) 16:21, 6 February 2015 (UTC)

REL, short-term exposure, and international limits

Hi! I'm here with a suggestion for a couple small additions to the chembox. We already include the permissible exposure limit from OSHA, but NIOSH publishes the recommended exposure limit, which is often different from the PEL, as well as short-term exposure limits, which also typically differ from the PEL. Another parameter would be the IDLH, the concentration where the substance becomes "immediately dangerous to life and health". I think it'd be helpful to have these as extra parameters in the box. Also, per discussion on WT:CHEMISTRY, I'm working on getting international standards so we can include those in order to be less US-centric. NIOSH has collaborating agencies all over the place, so they shouldn't be too hard to include. Are there already parameters for any of these? (I'm still learning the ins and outs of the chembox, doesn't help that I'm bad with templates.) Please let me know what you think, and thank you! Best, Emily Temple-Wood (NIOSH) (talk) 06:15, 24 January 2015 (UTC)

Just linking: the WT:CHEMICAL discussion is WT:WikiProject Chemistry#Occupational safety & health. NIOSH is National Institute for Occupational Safety and Health ("is the U.S. federal agency responsible for ..."). And to give comfort: Emily, the {{Chembox}} is complicated for every one. There are ~500 parameters ;-). A question: is there any overlap with the topic of {{Infobox drug}} aka {{drugbox}}? In the background, we have in mind that these two infoboxes could merge in the future. -DePiep (talk) 16:52, 25 January 2015 (UTC)

I can not advise on whether we should add them (I'll read & follow this thread). I'll start a subthread to explore the what-if options (what-if we add them). I note that, since we already have US-relaterd PEL, another US-related data can be acceptable. Some country-raleted data (esp in {{drugbox}}) mention the country in the datarow ('CA'). -DePiep (talk) 17:30, 25 January 2015 (UTC)

@DePiep: Thanks so much for this mockup! More below. :) Emily Temple-Wood (NIOSH) (talk) 19:04, 25 January 2015 (UTC)

Please spend time (your're payed I read :-) ) -- please spend time on the {{drugbox}} for this. Ask me if you don't see why that's important. -DePiep (talk) 19:21, 25 January 2015 (UTC)

I'm not sure what exactly you're getting at - though I will be looking at adding safety data via the drugbox. If there's something else I'm missing (only one cup of coffee today - clearly I need more) please let me know! Emily Temple-Wood (NIOSH) (talk) 19:58, 25 January 2015 (UTC)

{{Chembox}} and {{Drugbox}} are 'the same', but just have different accents from wiki history reasons. In the far future, they will be merged (same parameters, same meaning, just accents in prescriptions and so). So I suggest you take a look at how REL and IDLH would apply in sister {{drugbox}} (the same parameter treatment, full?). -DePiep (talk) 20:02, 25 January 2015 (UTC)

Ooh, didn't realize they were going to be merged. REL and IDLH would apply the same, as far as I can see. Emily Temple-Wood (NIOSH) (talk) 20:06, 25 January 2015 (UTC)

There is no Law that they will be merged, we just keep it in the back of our minds ("let's not make it more difficult to merge"). 'Algorithm' is just 'logic', what you do in your mind (and what others like me want to hear&learn from you). -DePiep (talk) 20:13, 25 January 2015 (UTC)

@DePiep: Hi, so I've been super busy the past few days and I'm just now getting back to this, sorry. Can we open a discussion to get consensus for adding these two values to the infobox? I'm happy to start an RfC or whatever. Best, Emily Temple-Wood (NIOSH) (talk) 05:17, 30 January 2015 (UTC) (moved this post to this section, is about adding yes/no. -DePiep (talk) 08:33, 30 January 2015 (UTC))

This thread is a correct proposal in itself. No RfC needed (that's for heavier policies & conflicts IMO). People can discuss & comment here. It's just that {{Chembox}} and {{infobox drug}} do not gain much talk traffic. If no one opposes, that's an outcome too. A few more days before concluding seems fine. (My personal opinion follows below). Oh, and stop saying 'sorry' for doing nothing wrong. Will make feel better — both of us. -DePiep (talk) 08:42, 30 January 2015 (UTC)

• IMO: I have looked deeper into REL and PEL, and the agency NIOSH. Although US-only, and non-legal (as PEL is), the data is well-founded enough to add. In other substances (like medicine) we already have country-specific data (esp. {{drugbox}}:"AU, CA, UK, US"). Our REL article is much older that this proposal (another canceling out of possible COI pushing). Given that we have ~150 data rows already in {{chembox}}, often with multiple data facts, REL can have the benefit side of any grey area too. Todo: exact wording is to be fleshed out (see #Would look like, below), and {{infobox drug}} aka {{drugbox}} should strongly consider following. -DePiep (talk) 08:54, 30 January 2015 (UTC)

@DePiep: I'm okay with the wording of "PEL + foo ppm" for the REL, the reason I want to include it at all is because sometimes there hasn't been a legal limit enacted for a substance by OSHA (the legal agency) when there is still a recommendation from NIOSH (the research agency). And I haven't had anything to do with the REL/IDLH/PEL articles, though I do plan on eventually working on them to make them more accessible & complete. Emily Temple-Wood (NIOSH) (talk) 18:13, 30 January 2015 (UTC)

Would look like

Without prejudging whether the data will be added to {{chembox}} (discuss above), we can explore what it should look like. First suggestions:

(This demo is developing, see talks below)

|REL=

|IDLH=2000 ppm (ethyl acetate)

|PEL= is existing (demo from ammonia).

Hazards
Permissible exposure limit (PEL) (US)	50 ppm (25 ppm ACGIH- TLV; 35 ppm STEL)
Recommended exposure limit (REL) (US)	(demo text?)
Immediate danger (IDLH) (US)	2000 ppm

Questions: where would IDLH wikilink to? A few demo substances with real data available? -DePiep (talk) 17:30, 25 January 2015 (UTC)

We have an article on the IDLH so that would be an appropriate wikilink. Most every substance that has a PEL in the NIOSH Pocket Guide has an IDLH value. For example, the IDLH value of ethyl acetate is 2000 ppm. Hope that helps! Emily Temple-Wood (NIOSH) (talk) 19:04, 25 January 2015 (UTC)

Y In demo now. More demo data please. -DePiep (talk) 19:18, 25 January 2015 (UTC)

@DePiep: Yay, more data! So in this case the REL is the same as the PEL - TWA 400 ppm (1400 mg/m). Do you want more compounds with a REL/PEL/IDLH? Emily Temple-Wood (NIOSH) (talk) 19:22, 25 January 2015 (UTC)

"REL=PEL" I do not understand, and the template won't either unless you specify how and when that notion should be made (made visible in the article). In other words: how shold the template show "REL=PEL" depending on the editors input (in the article page)? And more more simple demo's to get the data feel. -DePiep (talk) 19:30, 25 January 2015 (UTC)

@DePiep: So when the REL = PEL, I think it's totally sufficient to just put the values in separately. Here's some data for you, from just some random ones I've done so far. Emily Temple-Wood (NIOSH) (talk) 19:38, 25 January 2015 (UTC)

Allyl alcohol

Acetic acid

Ethyl acetate

• About last demo (Ethyl acetate): bad. When REL and PEL are the same --not by coincidence--, they should not be repeated. Please describe an algorithm (reasoning) on when/how to say: "PEL=REL", when relevant.

Note: the prior discussion is more important (include or not?). -DePiep (talk) 19:45, 25 January 2015 (UTC)

I'm not sure if there is an algorithm, perhaps there could be a parameter that says PEL/REL and use that when it's the same? This is all just data from the NIOSH Pocket Guide which doesn't give reasoning on why OSHA adopts the PELs they do. I'll ask around at NIOSH and see what people say. Emily Temple-Wood (NIOSH) (talk) 19:56, 25 January 2015 (UTC)

"TWA" and "ST" must be clarified, by either a link or an abbr tooltip: TWA. I'm still looking for the overlap (explaining the data repetition). -DePiep (talk) 17:41, 26 January 2015 (UTC)

TWA is the time weighted average of the exposure and ST is just short term exposure (<15 minutes, usually). The difference between a REL and a PEL is that the PEL is a guideline put in place from OSHA that must be followed, and the REL is just NIOSH's recommendation, which is usually more stringent than the OSHA PEL. Emily Temple-Wood (NIOSH) (talk) 19:21, 26 January 2015 (UTC)

No need to explain it here, it should be explained in the infobox. -DePiep (talk) 22:21, 26 January 2015 (UTC)

• I think that any acronyms should be wikilinked to something (e.g. TWA, ST, ppm etc) as their meanings are not obvious. Project Osprey (talk) 10:00, 30 January 2015 (UTC)

Agree. Or at least give them that abbr-link: TWA. (Can do that in code for REL and PEL). -DePiep (talk) 11:39, 30 January 2015 (UTC)

I don't like the repetition of PEL data at all. Could it be: REL: "PEL, and 15 ppm foo"? -DePiep (talk) 12:42, 30 January 2015 (UTC)

Does the value for REL always match that of PEL? If they are always the same then we don't need two boxes (REL could be moved inside of the PEL row - REL is a recomended limit but PEL is a legal limit, so I think PEL should 'win' if we were only keeping one). --Project Osprey (talk) 13:18, 30 January 2015 (UTC)

@Project Osprey: @DePiep: So, the REL does not always match the PEL, when they do I think just having the PEL is fine. The reason I want to have both as parameters is for when there is no PEL but there is a REL. With the abbreviations, I'm happy to change 'TWA' and 'ST' to "TIme Weighted Average" and "Short Term" but if the {{abbr}} works that seems less clunky. I'm flexible. :) Emily Temple-Wood (NIOSH) (talk) 18:13, 30 January 2015 (UTC)

(REL/IDHL arbitrary break)

• LH side text proposals "demo10":

Ethyl acetate (demo10)

Hazards
NIOSH health limits (US): Permissible exposure (PEL) Recommended limit (REL) Immediate danger (IDLH)	TWA 400 ppm (1400 mg/m) PEL, and ST 15 ppm (37 mg/m) 2000 ppm

@DePiep: Hi! I like this a lot! I've tweaked a couple of things (using {{abbr}} and linking short-term exposure limit), does it still look ok to you? Emily Temple-Wood (NIOSH) (talk) 20:38, 31 January 2015 (UTC)

Working on {{chembox}}, but not exactly PEL & REL. Plan to. Notes:

note. When the editor enters "TWA" (plain text, uppercase), the template will add the abbr. Same for ST. Don't worry

Question: Is there a general source link for REL? (i.e., valid always). OR: CAn we contruct automatically a source link for example "Ethyl acetate" by creating:

 ? Ask me if this is unclear.

Note: Expect the next input options:

|PEL=
|PEL_ref=
|REL=
|REL_ref=
|IDLH=
|IDLH_ref=

-DePiep (talk) 20:47, 31 January 2015 (UTC)

So unfortunately there isn't a general source link beyond where XXXX is the 4-number identifier. I can't find a full listing of the data, just the alphabetical listing of chemicals with a link to the full data page. (that's here). I'm okay with putting the {{PGCH}} in each one. Thanks! Emily Temple-Wood (NIOSH) (talk) 21:00, 31 January 2015 (UTC)

1. The PGCH (pocket, pdf) is general and so could be (should be?) when there is PEl/REL/ILDH input. But that's generic, and could be on the LH side automatically. (If you add it to the REL input of Ethyl acetate, that is not good referencing).

2. Is the XXXX another internal id-number niosh uses? If that is the only way, we sigh and use that. But using a CAS Reg Number for the external link would be much better.

3. If the niosh XXX (4-digit) nume is the key, we'll add input parameter to enter like |REL_ref_id=XXX. What do you think? -DePiep (talk) 00:04, 1 February 2015 (UTC)

The PGCH PDF should be the PEL/REL/ILDH general ref. As for specific refs, the 4-digit number is an internal indexing number for the PGCH, NIOSH does put CAS numbers in each entry but it's not linkable if that makes sense? Here is the list of all chemicals included in the PGCH by CAS number, if that's helpful. There's no way to get from CAS number to PGCH entry besides that list, as far as I can tell. I will admit that the UI of the NIOSH site is not as friendly and easy to navigate as it could be and I may be missing something. Emily Temple-Wood (NIOSH) (talk) 00:19, 1 February 2015 (UTC)

P.S. - Thank you so very much for your help with all of this. I'm so grateful. :) Emily Temple-Wood (NIOSH) (talk) 00:20, 1 February 2015 (UTC)

Useful answer (and we'll save the flowers until it is working ;-) ).

Priority issues (we need to define the input data first & asap, so you can enter it in the articles):

1. So there is no useful search option with known id's (like CAS number).

2. Q: Shall we use like |NIOSH_id=0260 (for editors to input)? Then the link will go directly to . If |NIOSH_id= is empty, the reference will link to the Pocket general page (search box) .

Less important issues.

3. See demo10, improved (don't worry about the line-breaking).

4. Q: Which name should we show: NIOSH, OSHA, CDC? (The most commonly known, most clear with this)?

5. Q: Is the sequence right? PEL-REL-IDLH?

6. {{chembox}} will replace TWA, ST with the appropriate text (linked, abbreviated, etc.).

7. For your help: these are pages have a PEL today: Special:WhatLinksHere/Template:Chembox_PEL (326 P). -DePiep (talk) 10:00, 1 February 2015 (UTC)

Early test at Template:Chembox/testcases#REL. -DePiep (talk) 11:22, 1 February 2015 (UTC)

• @Emily Temple-Wood (NIOSH): I am closing the discussion for Version One of REL and IDLH data. See below. -DePiep (talk) 12:03, 1 February 2015 (UTC)

@DePiep: Just to answer your questions - having the NIOSH_id parameter is super useful and not terribly onerous to input. NIOSH should be the name on the REL and IDLH since they're who sets it, not OSHA. The sequence is correct. Thanks so much! Emily Temple-Wood (NIOSH) (talk) 17:07, 2 February 2015 (UTC)

OK. Warn me if you see big errors (fancy things are for March). Could go live shortly. -DePiep (talk) 17:10, 2 February 2015 (UTC)

Will do, thanks! Emily Temple-Wood (NIOSH) (talk) 17:16, 2 February 2015 (UTC)

Oh and when entering PEL, REL data, do just type "TWA"; later the template will turn that into the link as talked here. Same for ST. -DePiep (talk) 20:48, 2 February 2015 (UTC)

Can do. Is it live now? Emily Temple-Wood (NIOSH) (talk) 00:01, 3 February 2015 (UTC)

• @Emily Temple-Wood (NIOSH): Please check the tests at Template:Chembox/testcases#REL. You can add and change demos there!

Input in the {{Chembox Hazards ... section:

|PEL =
|REL =
|IDLH =
|NIOSH_id = <!-- 4-digit e.g. 0260 -->
|NIOSH_ref = <!-- any reference; use <ref>...</ref> tags -->

If there are no major issues, it will go live in a few days. -DePiep (talk) 11:04, 5 February 2015 (UTC)

@DePiep: Hi, I tried using it here but I think there's something weird with it. Could you please take a look? Emily Temple-Wood (NIOSH) (talk) 01:57, 6 February 2015 (UTC)

Your edit fell in a crack: yesterday I removed all testing code (also the "use /sandboxes" settings), so that today code can go live. In this interval, {{Chembox}} /sandboxes do not work as expected (are no use for testing). That is what you saw! I just did an individual extra test on this demo you made, and it looks OK -- input shows as expected. (side topic: I fixed a temperature calculation here). And now: showtime. -DePiep (talk) 07:21, 6 February 2015 (UTC)

•  Done. Please start all issues in the new section below. -DePiep (talk) 16:23, 6 February 2015 (UTC)

Closing for first version

I am closing discussion part one, to implement the first version of REL and IDLH data.

Input parameters. All are optional.

|PEL=<!-- existing -->
|REL=
|IDLH=
|NIOSH_id=<!-- 4-digit, like '0260' for ethyl acetate-->
|NIOSH_ref=<!-- any <ref>...</ref> -->

These input parameters are guaranteed to be handled and shown OK. When entering data here, the article could require a References section. An early testcase demo is here.

Please do report any bugs and errors. Do not rise formatting or showing issues: they will be discussed later (few weeks?). Meanwhile, everyone can add data!

I am closing this because we must update the big {{chembox}}, because at the moment mobile views can be broken. We must solve that asap. -DePiep (talk) 12:03, 1 February 2015 (UTC)

New issues, after version 1

• Implemented 16:24, 6 February 2015 (UTC). The PEL/REL/IDLH presentation has room for improvement. But data can be added. -DePiep (talk) 16:24, 6 February 2015 (UTC)
• Just discovered: Ethylene glycol uses template {{PGCH}} (the Pocket Guide):

| PEL = none<ref name=PGCH>{{PGCH|0272}}</ref>

which produces the reference: NIOSH Pocket Guide to Chemical Hazards. "#0272". National Institute for Occupational Safety and Health (NIOSH)..

So the NIOSH id (4-digit) is already in use on 350 pages: Special:WhatLinksHere/Template:PGCH. For now, I have no idea on how to use this. -DePiep (talk) 14:53, 7 February 2015 (UTC)

Proposal to add indexed identifiers (eg: |CASNo7=)

Note: proposed index numbers -6 and -7 will not be implemented (eg |CASNo6= and |CASNo7=). This is to limit template size and complexity. Text below shold be read accordingly -DePiep (talk) 11:11, 3 February 2015 (UTC)

Note: later, |ChEBI= was added to this list of indexed parameters.-DePiep (talk) 02:10, 1 February 2015 (UTC)

I propose to add indexed identifiers to {{Chembox}}. In short: {{Chembox}} takes indexes 1–7 for |CASNoX=, plus zero |CASNo=. The same for six seven other identifiers like SMILES.

Already indexes "0"–1–4 are used. This change will add more, and add bot-checking. Regular naming pattern. -DePiep (talk) 23:09, 26 January 2015 (UTC)

Useful links

• {{Chembox/testcases}}
• {{Chembox/testcases2}} (all indexes)
• {{Chembox/testcases3}} (all params, per section)
• {{Chembox/testcases4}} (long names & ids)

Documentation (prepared)

Described by future documentation:

Indexed CASNo identifiers

|CASNo = 
|CASNo1 = 
|CASNo2 = 
|CASNo3 = 
|CASNo4 = 
|CASNo5 = 
|CASNo6 = 
|CASNo7 = 
|CASNoOther = 

Infobox {{Chembox}} can show identifier |CASNo=. In addition, there are seven more indexed CASNo identifiers. See the list to the right. So {{Chembox}} can take and show 0–7 (eight) CASNo identifiers. Note that index number "0" (for |CASNo=) is not written or shown.

All indexed CASNo identifiers should be entered as a well-formatted CAS registry number, and nothing more. No formatting code should be added. They are shown in the infobox with a line break added (<br/>), when needed. All CASNo entries are bot-checked. So one can expect |CASNo7_Ref= in the article. There is no requirement for any sequence (it is not required to fill the numbers from 0 upwards; this freedom did not exist before, when CASNo= was required in the first place).

CASNo free text parameter

There also exists |CASNoOther=, which takes any text (e.g., an annotated list or prose). A <br/> might be needed in here. This parameter is never bot-checked.

The seven eight indexed identifiers

Overview (incomplete):
|CASNo =
|CASNo1 =
|CASNo7 =
|CASNoOther =
|ChEBI =
|ChEBI1 =
|ChEMBL7 =
|ChEBIOther =
|ChEMBL =
|ChEMBL1 =
|ChEMBL7 =
|ChEMBLOther =
|ChemSpiderID =
|ChemSpiderID1 =
|ChemSpiderID7 =
|ChemSpiderIDOther =
|InChI =
|InChI1 =
|InChI7 =
|InChIOther =
|PubChem =
|PubChem1 =
|PubChem7 =
|PubChemOther =
|SMILES =
|SMILES1 =
|SMILES7 =
|SMILESOther =
|UNII =
|UNII1 =
|UNII7 =
|UNIIOther =

Similar to CASNo, there are eight identifiers that take indexes:

|CASNo=, |ChEBI=, |ChEMBL=, |ChemSpiderID=, |InChI=, |PubChem=, |SMILES=, |UNII=

All the indexed parameters are bot-checked. (7×9=63). They also take the |...Other= parameter to enter free text, unchecked.

Use the index

When the chembox has multiple substances, use the index. Make sure that |CASNo3= and |SMILES3= are about the same substance, and |CASNo3= and |SMILES2= are not. Treat index "0" the same way:

|CASNo= is |InChI= is |SMILES=

|CASNo1= is |InChI1= is |SMILES1=

|CASNo= is not |InChI1= is not |SMILES4=

When the indexes are used this way, you can add |use_index=yes to set a flag. That lists the article for further development using the indexes. At the moment, this parameter setting has no effect in articles at all. It only serves as a maintenance categorizing help.

This still is a good idea, but it is not tracked now.

Details

Deprecations

• All -s parameters like |CASNos are deprecated. Use |CASNoOther instead
• Parameter |CASOther is deprecated (because it has an irregular name pattern). Use |CASNoOther instead

Not for images

The image numbers (like in |ImageFile1= and |ImageFileR2=) are not indexes this sense. They do the row-positioning only.

(end of documentation)

Future with indexes

In the far future, {{Chembox}} can handle indexes even more systematically, I hope. For example, any image can be tied to a CASNoN index. And that can be shown or used in the infobox. But not tomorrow :-) . We'll keep that open. -DePiep (talk) 23:19, 26 January 2015 (UTC)

Comments

So in general: the Chemboxes listing multiple forms of a compound will become longer but better formated. Chemboxes listing only one form should see no major changes?... Love it! --Project Osprey (talk) 10:56, 27 January 2015 (UTC)

Yes. (And in the InChI and SMILES data some vertical whitespace will go; less line-height. See testcases4). -DePiep (talk) 11:47, 27 January 2015 (UTC)

Ah, yes I see. How did you get around the wrapping problem? --Project Osprey (talk) 12:00, 27 January 2015 (UTC)

See Template:Chembox ChemSpiderID/format: applied word-wrap:break-word;. Don't know exactly what was wrong, just build this new data formatter from ~scratch. More about format changes in a separate post, to follow. -DePiep (talk) 12:23, 27 January 2015 (UTC)

Proposed changes to the Chembox

I have prepared a dozen or two of changes to {{Chembox}}. They are described below. There are testpages to demonstrate the differences. Please comment below. When things are stable, I can move them into life. -DePiep (talk) 23:43, 29 January 2015 (UTC)

• Testpages: Template:Chembox/testcases; /2 (All indexed identifiers up); /3 (All parameters use); /4 (Long names; with show/hide)

• Infobox title: transparent background (usually white), Same as regular infobox. So not chembox-yellowish any more.
• Images: Obsolete HTML code replaced (align="center" → style="text-align:center;"). No other changes in the images.
• Other names and IUPAC names: see #Creating section 'Other_names' in {Chembox}

Full Chembox subheader "Other names" added. Right below that is data |Other names=.

Then all IUPAC names in one show/hide box, with their subheaders: |IUPACname=, |IUPACnames= (new, plural), |PIN=, |SystematicName=

|IUPACName_hidden=y/yes/hidden/hide/collapse/collapsed to set collapsed box state (default:show, uncollapsed).

Change: the four name sets were all show/hide. Now there is one setting, for IUPAC name(s) only.

• InChI and SMILES: limited width, forced line breaks.

The (long) names are now limited to infobox width (22em). This resolves a current page-breaking disruption (long identifiers run off-page to the left hand side. In mobile view, this is always visible because mobile does not have a show/hide option).

InChI input should not have |InChI=InChi=... added (will be double in output). These articles are listed for editing.

• CASNo has indexes "blank" and 1–7. See #Proposal_to_add_indexed_identifiers_(eg: |CASNo7=)

|CASNo = 
|CASNo1 = 
|CASNo2 = 
|CASNo3 = 
|CASNo4 = 
|CASNo5 = 
|CASNo6 = 
|CASNo7 = 
|CASNoOther = 

|CASNo= now takes indexes blank–1–7 (see list to the right).

|CASNoOther= takes free text.

All indexes should have plain correct text, without formatting (then |-Other= can take any text). All the identifiers are separated with <br/>.

Seven Eight identifiers in total have this same index pattern & behaviour:

|CASNo=, |ChEBI=, |ChEMBL=, |ChemSpiderID=, |InChI=, |PubChem=, |SMILES=, |UNII=

• ATCCode: uses newline, minor reformatting.

• MolarMass can have |MolarMass_ref= (no space) and |MolarMass_notes= (with space)
• MolarMass_notes or MolarMass_ref (==> _notes in cat. space)

Deprecated: MassRound use MolarMassRound

• Solubility in water: the label will be Solubility in water. (same as in {{drugbox}}).

• Legal status, pharma: show in correct column (Data) (see Chloral hydrate; disruption)

See added bullet below.

• Main hazards: move to near top of Hazards subsection.

• Temperatures MeltingPt, BoilingPt, FlashPt, AutoignitionPt.

Four parameters (maximum) are needed to enter a complete temperature point:

MeltingPt .. MeltingPtC (+calculated conversions) .. MeltingPt_ref .. MeltingPt_notes

Or use MeltingPtF (Fahrenheit to convert), MeltingPtK (Kelvin to convert)

deprecated: MeltingPtCL/CH use MeltingPtC=10 to 20 to enter a range

• Show link and quantity symbol like: "Refractive index (n)". (descriptive, and the symbol unlinked).

• Bot-supported validation

All indexes (blank–1–7) will be bot-checked (was: the blank-indexed one only).

Only articles will appear in categories.

Punctuation of the bottom validation note (in "verify; what is")

• Visual effects

Multi-line labels (wrapped texts): less line height.

Colored section headers: removed horizontal borders.

Footer: the extra formatting row below, a few pixels high, is invisible. rm horizontal top border.

• Internal changes

{{Chembox entry}}, an intermediate datarow builder, is removed. This step is skipped.

Maintenance categories are simplified, and are {{main other|]}} (categorize articles only).

The yellow subheaders are composed in the subtemplate, no {{!}}-code in the template any more.

PArameters like |CASNo_url= are not used in any result, so removed from internal code and from documentation (no categorization of other catch)

Ouch: bot params like |CASNo_Ref= did a double (nested) calculation! First input param {{cascite}} was processed (say into a green tick+maintCat; all fine). Then internally that value of |CASNo_Ref= was processed again for a tick (no extra effect). Did rm inner one, the template result of bot param input "|CASNo_Ref = {{cascite|changes|??}}" is presented directly. For all such identifiers (see new /format subtemplates).

• Known issues

InChI and SMILES have a superfluous bottom border line (when no -Other data). Acceptable for now.

Bottom row (references) is ~2px too high (is where the row-width setting is kept).

Legal status (pharma): too much whitespace, see Chloral hydrate. ({{drugbox}} does better).

Pharma: presentation of Legal status, Pregnancy category, Licence were checked and improved. {{Drugbox}} is leading. More at the end of this section. -DePiep (talk) 11:36, 30 January 2015 (UTC)

Identifiers not indexed (=not changed, this time): |ChEBI=, |DrugBank=, ...

• Pharmacology: Legal status, Pregnancy category, Licence

These data presentations are changed. Follows {{drugbox}}. See testcases. -DePiep (talk) 11:36, 30 January 2015 (UTC)

Deprecations

MeltingPtPrefix use MeltingPt

MeltingPtCL, MeltingPtCH use MeltingPtC=10 to 20 to enter a range.

Same for MeltingPt, BoilingPt, FlashPt, AutoignitionPt.

Same for -FL, -FH, KL, -KH: Use simple range input with -F, -K.

Autoignition use AutoignitionPt (also for -C, -F, -K, _ref, _notes).

PIN_hidden use IUPACName_hidden hide/collapse/

CASNos use CASNoOther for free text. Same for all eight identifiers.

CASNo_url input is not used. rm from documentation, for all eight indexed identifiers.

CASOther use CASNoOther (name pattern; this one for CASNo only)

MassRound use MolarMassRound

PregCatCA, PregCatUK: use PregCat free text (not regulated in Au, UK). See .

Articles with deprecated parameters are categorized for maintenance editing. There will be no broken effects (no hard deletions before cleanup).

-DePiep (talk) 23:43, 29 January 2015 (UTC) (tweaked. -DePiep (talk) 09:12, 30 January 2015 (UTC))

Notes and considerations

The {{chembox}} is not a modern Lua {{infobox}} yet. This may follow. We keep in mind that {{drugbox}} could be merged with this one.

The forced newlines in long names like InChI are highly necessary. It breaks mobile pages always.

Testpages

• Template:Chembox/testcases -- General, various tests
• /testcases2 -- All indexed identifiers up (7×8+)
• /testcases3 -- All parameters used
• /testcases4 -- Long names; show/hide

It is always useful to check mobile view too (From a desktop view, that is a link at the very bottom of a wikipage).

To test your favourite substance, don't forget to use all Section sandboxes:

{{Chembox/sandbox
| Section1 = {{Chembox Identifiers/sandbox
...
  }}
| Section2 = {{Chembox Properties/sandbox
...
  }}
}}

If you ask me, I can put an article's chembox in testcase page. It will show nicely side-by-side.

Maintenance categories

• See Category:Chembox maintenance categories
• Tracked and deprecate parameters, specified in the category. No new categories, some deprecated.

• CheMoBot categories are updated (eg, {{casnocite}}).

-DePiep (talk) 23:43, 29 January 2015 (UTC)

•  Done. Implemented in today's {{Chembox}} version. 16:25, 6 February 2015 (UTC)

Comments on list of changes

Looks good in general. I'm don't really like the transparent background for the title. Would it be possible to force a line break in "Solubility<BR> in water" to improve the formatting of the columns? --Project Osprey (talk) 11:57, 30 January 2015 (UTC)

Thanks for taking a check.

Title transparent bg: this is what most enwiki infoboxes do, it is the {{infobox}} standard. It shows in lithium ({{Infobox element}}) and Aspirin ({{drugbox}}). Non-chemical: War and Peace ({{Infobox book}}). That is where I took it from (for this old artisan handcrafted wikitable infobox). There is no urgent or topical reason to use a color. Could you consider this an experience of "I'm not used to it -- yet" (happened with the chemical elements change last year).

In general, grand wikipage design(ers) tend to bring in more white(-space) and less lines & colors. So if there is no reason to deviate from standard, I'd prefer to keep it transparent (~white).

Breaking line in "Solubility<BR> in water": can do, but is it needed? For me, the list of solubilities is a strain to understand already (I'm not not familiar with the topic; two similar words &tc.). A linebreak then adds a tiny bit to this mental grasping. Plus, in my screens, it does not push the column or infobox much wider if at all. In testcase ammonia, there are some others equally wide. You want to minimize the LH column more? Or the overall infobox (now 22em minimum)? Its about personal preferences now. -DePiep (talk) 12:26, 30 January 2015 (UTC)

@Project Osprey: OK, I've added the break (Good Editorship). But I'm not convinced it should stay there. Let's take a good look at the effects. -DePiep (talk) 16:10, 30 January 2015 (UTC)

You're right; adding the line break hasn't improved anything. I think the LH column should be kept thin as that improves the readability of the RH column - however that just my personal style choice. Regarding the transparent bg - this I can live with, after a while I probably wont notice anymore. --Project Osprey (talk) 16:46, 30 January 2015 (UTC)

OK & agree. It's 22em now total, with LH at 40% (set last March, unchanged now). Text can push a column wider. You want to see something else tried? -DePiep (talk) 17:13, 30 January 2015 (UTC)

No, I'm quite happy with the additions as they stand. --Project Osprey (talk) 11:55, 2 February 2015 (UTC)

• I'm not sure about this. As proposed, the "other names" are above IUPAC name(s). Is that right? -DePiep (talk) 15:02, 30 January 2015 (UTC)

Regarding the long InChIs etc. - you say they still break on mobile? I would then suggest again to render those through LUA ASAP, otherwise we get edits manually 'breaking' the InChIs (and others) again/still (I don't know if that is done still), and then we have to do a cleanup run later to repair them all again? --Dirk Beetstra 03:34, 1 February 2015 (UTC)

No. Current live version breaks pages (bad). The new version does not break the page (good). That shows in the testcases pages esp. /testcases4, left hand side = sandbox. The max width for these is 22em (~regular infobox width)(Misunderstanding?: they do line-breaks as intended!). With me, they show OK in mobile view too (I test that by clicking the very last link on a testpage). eg ]. Tell me if you see a broken sandbox! -DePiep (talk) 10:21, 1 February 2015 (UTC)

And yes, we can do a cleanup later (for unwanted <br/> in input). But what I'm changing is huge (40 sandboxes interacting), better not add that in the same change. -DePiep (talk) 10:24, 1 February 2015 (UTC)

• I am closing the trunk for feature discussions. (Also as a note to self ;-) ). Time to implement the fix for broken pages, especially mobile views.

Please keep reporting on (testcase) bugs, errors, etc. -DePiep (talk) 12:10, 1 February 2015 (UTC)

Last minute notes

• Just only today, |ChEBI= was added to the list of indexed parameters. -DePiep (talk) 02:10, 1 February 2015 (UTC)

• Note: proposed index numbers -6 and -7 will not be implemented. This is to limit template size and complexity. (eg |CASNo6= and |CASNo7=). -DePiep (talk) 11:17, 3 February 2015 (UTC)
• Bot-managed templates like {{cascite}} are redesigned & recoded (7 verified identifiers + images). Will categorize more systematically "changed" verified fields. Required list of cats made complete (7, + 1=images). "images changed" still won't fill. Also "without CASNo" (x 7 identifiers) categories, filled by template not by bot. See new documentation at {{cascite}}. -DePiep (talk) 09:19, 6 February 2015 (UTC)
• Y Going live now. -DePiep (talk) 09:20, 6 February 2015 (UTC)
• Y Live now. -DePiep (talk) 15:24, 6 February 2015 (UTC)

Post-implementation issues

• Some maintenance categorizations are filtered incorrect.
• |DrugBank= and |ATCCode= can appear in Identifiers and in Pharmacology Sections (in the same form). From this, Category:Chemical pages without DrugBank identifier now has articles that do have a DrugBank id in Pharmacology Setion.
• Paired images (left & right) do not show nicely even in mobile view. Not broken.

-DePiep (talk) 15:30, 6 February 2015 (UTC)

• PubChem_Comment doesn't appear to be rendering in all cases (See: Periodate, Periodic acid etc) -sign?: 19:56, 7 February 2015 Project Osprey. -DePiep (talk) 01:02, 8 February 2015 (UTC)

This is a bug and will be solved in a few days. |PubChem_Comment=foo will show as expected. No need to change (edit) articles for this. -DePiep (talk) 18:55, 8 February 2015 (UTC)

Updates (new version)

Follow-up edits, in preparation:

• Make sure |PubChem_Comment=foo shows (bug, see above. All PubChem indexed comments)
• Make sure |FlashPt_ref= shows (bug)
• side-by-side images (L/R): in mobile view, the images do not show symmetric (instead, they are shifted to the left of the infobox). Had to restore deprecated code align=center in sbs-row. See esp. /testcases7images.
• IUPAC names, other names, see esp. /testcase3. Last week, these were moved into their own subsection. More changes:

1. Remove any show/hide option. No reason to hide names. And anyway, in mobile view these are shown always. Unrelated to InChI-names showing/hiding.

2. Chemboxes with IUPACNames_hidden are categorized for a check, afterwards (& remove this deprecated parameter).

3. Restore sequence: IUPAC names first, 'other names' last.

4. Layout: to present long names more gently, they start indented (not in the righthand column).

• Allow correct input like: InChI=IncChI=1/N123, but preventing double showing like InChI=IncChI=1/N123. This follows the InChI definition.
• |KEGG= indexed 1–5, as other verified parameters.
• |DrugBank= indexed 1–5, as other verified identifiers. All DrugBank input can be entered in both Identifiers and Pharmacology sections (as before).

Now all 8 verfied identifiers are indexed and completely bot-tracked for each index. They also take {{cascite}}-like templates for each indexed |_Ref=. E.g. |KEGG5_Ref={{cascite|changed}}.

• {{pubchemcite}} follows {{cascite}} pattern (CheMoBot verfification etc).
• New category: Category:Chemical articles having a data page, like Ammonia having Ammonia (data page). Used in Chembox section "Supplementary".
• Make sure deprecated parameter catsort "3" is in the right category. (code 3 = 'Section has unnamed parameter |1=').

-DePiep (talk) 16:04, 11 February 2015 (UTC)

Y Going live now. -DePiep (talk) 23:41, 11 February 2015 (UTC)

 Done -DePiep (talk) 08:05, 12 February 2015 (UTC)

Note: |PubChem_Ref= does not show. Chek, old behaviour. See also old changes in CheMoBot wrt pubchem. -DePiep (talk) 11:22, 12 February 2015 (UTC)

AutoignitionC does not show (depr). -DePiep (talk) 12:17, 12 February 2015 (UTC)

Tracking parameter errors

I would suggest that parameter errors (and some syntax errors) in the chembox are tracked in a maintenance category. The compulsory and the optional parameters may be defined. All articles with other parameters in the chembox would be found. I introduced this for Template:GESTIS, where the articles with errors are now shown in Category:Pages with errors in Template:GESTIS (not yet fixed in order to visualize). --Leyo 00:54, 5 February 2015 (UTC)

You are absolutely right, there could be (more) parameter checking in {{chembox}}. The topic is very fit for that. However, there are a few complicating factors:

• Chembox has ~500 parameters. Not all need a check, but reducing that to 200 or 100 leaves a big job still.
• Chembox uses some 150 subtemplates (ouch we say).
• The bot User:CheMoBot also tracks certain parameters for verification. Keeping up that support takes attention too (and also serves as a parameter tracking aide).
• {{Chembox}} and {{Drugbox}} share a lot of categories, subtemplates, design setups (though not consistently all), so we have to keep an eye on Drugbox params too.

This is what we used to call an "intricate template" (and that's an understatement).

{Chembox} is not in a Lua module now. Once it is in there, controlling data as you propose is much much easier.

I can note that these weeks I have been preparing many changes to many {{Chembox}} templates. When tests are stable, this will go live one of these days. This time, not many parameter checks as you propose are added. Other changes needed attention first (eg, the mobile page view can be disrupted in situations!). But I kept an eye on that goal all along. More on this is in talk sections above.

Some changes that are coming, and that might interest you, are: consistently fill categories like "articles without CAS Reg number", "articles with changed CAS Reg number".

Meanwhile (now that you are here), if you see disruptions in {{Chembox}} articles in the next week or so, you are invited to report them. My changes need all the eyes we can get. -DePiep (talk) 10:24, 5 February 2015 (UTC)

Module:TemplatePar needs to be given just a list of all parameters, which may be quite long (example). It will find e.g. misspelled parameters and syntax errors such as parameter name: instead of parameter name=.

In de.wikipedia, all chemistry-related templates have it (see maintenance categories).

If the template documentation is up-to-date, i.e. it contains all possible parameters, it's not such a hard work to add Module:TemplatePar. --Leyo 07:19, 6 February 2015 (UTC)

We have Category:Chembox maintenance categories with 40+ specific maintenance categorizations. The parameter name: mistake you mention is covered in there. However. A parameter list you point to "which may be quite long" has only 100 params. As said, for {{Chembox}} that is multiplied by 5 (500 params). Then, we have 10 subtemplates must have that param check too (that's how it is). And all that for a misspelled param? And that still would not help against unwanted <br/> in input, or a misformatted CAS NR. So, let's first clean up those 40 categories. Then, next week we'll take another look. (did you see that the German {{GESTIS}} template requires another entering of the CAS NR?) -DePiep (talk) 15:48, 6 February 2015 (UTC)

The categories in Category:Chembox maintenance categories are not the same as the ones I linked above (purely “parameter-related”), but Category:Chemical infoboxes with misplaced or deprecated parameters seems to be close. My suggestion would—by its design—cover all misspelled (or deprecated parameters), not just selected ones. --Leyo 21:27, 6 February 2015 (UTC)

Any reply to my answers? btw, I do like the German links. Inspiring. But changing your own topic again again to tell me what is wrong with {{Chembox}} -- no. I know. -DePiep (talk) 21:36, 6 February 2015 (UTC)

Tell me, honestly: which of the links I gave did you actually click & read & digest? Your answers here suggest a zero. -DePiep (talk) 21:42, 6 February 2015 (UTC)

Yes, I did not answer your concern about the number of parameters. I agree that there are many, but apart for the work of getting a complete list, I do not see a real problem here. Are you concerned about the performance, i.e. that the page rendering time would be too high?

I visited all the pages you linked. I've even fixed all type 1 errors and the 100 type 3 errors. --Leyo 23:55, 6 February 2015 (UTC)

No I am not worried about performance at all. It's just, it's complicated as in: those templates interact, and there is the CheMoBot working, and external chemistry databases definitions. And this time I did not review the image showings (almost not). Another worry is that there is little activity in the verification process. 50% of the chemboxes are redcrossed. And there is {{drugbox}} to look at, always doing things with the same stuff. Anyway, I think I made another cleanup, into simplification. -DePiep (talk) 17:30, 8 February 2015 (UTC)

• I appreciate your contributions (I first thought an error happened: empty maint cats ;-) ). I'm not replying now to this last post of yours. It happened that up into Friday I put live 45 intricate Chembox templates I was working on for eight weeks. All interconnected, me having to research a bot behaviour, chemistry habits & external links, and an inherited Chembox setup. I liked it (it works, feels good today!), but I was not in the mood to read criticism on how to do well during these days. I already clicked those German dewiki links, and there are nice & great solutions indeed. But today I take a break, doing only the edits I like. Later, I might re-engage Chembox improvements. I was thinking, Drugbox can use a leap-frog to catch up with this one. -DePiep (talk) 20:46, 7 February 2015 (UTC)

next step

Leyo, I am thinking about this: for every relevant institute, we create a top category: "Category:Chemical articles and CAS". In there are any CAS-related categories we have. It should not be about templates (not "CAS issue from {{Chembox}}", not "CAS issues from {{Drugbox}}"). It will be articles only (not anypage). Because, I grasp (I'm not a chemist) that these institutes and their databases interact heavily about chemical substances in many ways (also with wiki of course). Chemistry, medicine, hazards: everything relates. So why not have a category that gives 'outsiders' and us an overview from there: "Cat:Chemical articles and KEGG", with subs "Cat:Chemical articles that use KEGG"; "Cat:Chemical articles with a KEGG issue"? (existing categories, eg by the bot, will stay). -DePiep (talk) 17:46, 8 February 2015 (UTC)

Are you talking about my suggestion concerning the use of Module:TemplatePar above? This parameter check is only related to the formal/technical correctness, but not to the content. Hence, I would prefer to have a main Category:Pages with errors in Template:Chembox and—at least until legacy errors are fixed—subcategories for the subtemplates, e.g. Category:Pages with errors in Template:Chembox Identifiers. --Leyo 10:17, 12 February 2015 (UTC)

I understand. Useful, agree. I'm taking a break, last changes were quite a strain. -DePiep (talk) 10:28, 12 February 2015 (UTC)

Ordering Identifiers

I am ordering the Identifiers alphabetically. From random. -DePiep (talk) 19:27, 12 February 2015 (UTC)

• Template-Class Chemistry pages
• NA-importance Chemistry pages
• WikiProject Chemistry articles