Wikipedia:WikiProject Chemistry/IRC discussions/17 February 2009


 * [2009-02-17 17:00:43]  Hi all.
 * [2009-02-17 17:00:44] Hello! Glad you could make it!
 * [2009-02-17 17:01:08] PC, is it your birthday today?
 * [2009-02-17 17:05:05]  And then things went awfully quiet.
 * [2009-02-17 17:05:24] I didn't think he was THAT old! Well, maybe we should get down to business.  We now have the Excel file up to the first 1000, and I plan on focusing on filling in gaps now.
 * [2009-02-17 17:05:55] So I can complete my first 100, list the problems etc.
 * [2009-02-17 17:05:58]  Meaning adding missing articles to wikipedia?
 * [2009-02-17 17:06:27]  Or verifying matches?
 * [2009-02-17 17:07:23] ali_as: No, I won't have that amount of time, to write new articles, unfortunately. I'd like to do some of the dye intermediates in time, but for now I'll just verify on WP
 * [2009-02-17 17:07:41]  Ok, that makes sense.
 * [2009-02-17 17:07:46] -->| ChemSpiderman (n=tony@c-98-231-156-242.hsd1.md.comcast.net) has joined #wikichem
 * [2009-02-17 17:08:12] The 1.0 project will be busy in the next month, as well - that's one reason why I wanted to reach this milestone before the storm hit
 * [2009-02-17 17:08:15]  It wouldn't be too helpful to create WP articles just because we have CAS numbers for them anyway.
 * [2009-02-17 17:08:45] ali_as: I agree, but a few of them are actually in a book on dye chemistry that I'm reading
 * [2009-02-17 17:08:58] |<-- Beetstra has left irc.freenode.net ("Bye Bye")
 * [2009-02-17 17:09:23] Hello ChemSpiderman - glad you could come!
 * [2009-02-17 17:09:32] -->| Beetstra_ (n=djbeetst@Wikimedia/Beetstra) has joined #wikichem
 * [2009-02-17 17:09:33] =-= Mode #wikichem +v Beetstra_ by ChanServ
 * [2009-02-17 17:09:36]  Cool.
 * [2009-02-17 17:10:13] CSM is the main person who got things really started on this project - but he's been very busy of late weaving his various webs
 * [2009-02-17 17:10:18]  hi everyone
 * [2009-02-17 17:10:22] =-= Beetstra_ is now known as Beetstra
 * [2009-02-17 17:10:39]  Hi Tony, I'm Alex/Ambix.
 * [2009-02-17 17:10:53]  Hi Alex.
 * [2009-02-17 17:11:29] ChemSpiderman - ali_as has done a lot of the online validation work from the Excel - I think he's covered a range ofat least 400 out of the 1000
 * [2009-02-17 17:11:40]  lot of work!
 * [2009-02-17 17:11:52] Yes, but very useful
 * [2009-02-17 17:12:14]  for sure
 * [2009-02-17 17:12:47] Many of this first 1000 are fairly basic compounds, so I'm glad to see their CAS#s turning green
 * [2009-02-17 17:13:23]  I think a lot of the effort involved could be reduced by semiautomating. I'm not sure exactly how.
 * [2009-02-17 17:14:22]  we've done some semi-automation to this stage but eventually it's eyeballs
 * [2009-02-17 17:15:55]  Checking the structure is the only step that really needs a human being. A lot of the energy I put into checking goes into adding {cascite} and putting the perm link into the index.
 * [2009-02-17 17:16:13]  yes..that's the big one...structure checking
 * [2009-02-17 17:16:25] ali_as: I find that the slowness of growing the Excel file is largely my "fault" - I'm doing a lot of things like creating important redirects, linking from related pages, fixing blatant errors, assessing, etc. That effort is really helpful to the project, because I'm unlikely ever again to do a search for 1-Iodonaphthalene (yes, it's unlisted but mentioned on one page)
 * [2009-02-17 17:16:33] <ChemSpiderman> but necessary otherwise the CAS number is for the wrong structure
 * [2009-02-17 17:16:44] <Physchim62> Hi CSM!
 * [2009-02-17 17:16:48] <ChemSpiderman> hi
 * [2009-02-17 17:16:56] We're doing the structure checking when we validate the CAS#
 * [2009-02-17 17:16:57] <Physchim62> sorry, I was judt on the phone
 * [2009-02-17 17:17:16] We thought you were just evading the issue of a birthday
 * [2009-02-17 17:17:59] <ali_as> If it could be brought down to a program that displays 2 structures next to eachother with a yes and no box most of the job could be done in very little time. It may not be feasable to get it down to this.
 * [2009-02-17 17:18:34] ali_as - yes, that sounds like it would be great
 * [2009-02-17 17:18:50] <Physchim62> for some reason, my father wanted to wish me happy birthday, so I couldn't just say "I'm in the middle of an IRC discussion!"
 * [2009-02-17 17:19:16] <ali_as> I'm looking at imacros currently, some perl scripts for reading SDF files and anything else. It's all new to me but it may lead somewhere.
 * [2009-02-17 17:19:17] <Physchim62> ali_as, that sounds like it would be complicated!
 * [2009-02-17 17:19:32] <Physchim62> all the same, go for it!
 * [2009-02-17 17:20:03] Good luck - I find the SDF files very hard to work with, though CSM is very comfortable with them
 * [2009-02-17 17:20:21] <Physchim62> I find the SDFs fine for 98% of the time
 * [2009-02-17 17:20:30] <Physchim62> and hell for the other 2% ;)
 * [2009-02-17 17:20:41] <ChemSpiderman> better than 80/20
 * [2009-02-17 17:20:50] <Physchim62> indeed
 * [2009-02-17 17:21:08] <ali_as> Finding a decent free sdf browser has been difficult and none of the ones I have do everything I really need.
 * [2009-02-17 17:22:50] <ali_as> But between them I can get things done.
 * [2009-02-17 17:23:00] ali_as found some problems of corrupted structures in the Union SDF file we've been using, so we're proposing that we use the latest CAS SDF alone. Would this be acceptable to folks?
 * [2009-02-17 17:23:25] <ChemSpiderman> what's "corrupt"?
 * [2009-02-17 17:24:08] I mentioned this in my email - linoleic acid vs linoelaidic acid
 * [2009-02-17 17:24:19] 60-33-3
 * [2009-02-17 17:24:35] <Physchim62> MDL file not matching CAS name
 * [2009-02-17 17:24:44] <ChemSpiderman> more examples?
 * [2009-02-17 17:25:10] linoelaidic acid isn't even in the CAS file, but somehow when we did the Union process, the double bonds got switched cis to trans and "matched" incorrectly
 * [2009-02-17 17:26:07] <ChemSpiderman> ok..any other examples?
 * [2009-02-17 17:26:13] I can't think of any other explanation for that
 * [2009-02-17 17:27:01] <Physchim62> endrind and dieldrin, which we've half sorted out but I'd like your opinion on it
 * [2009-02-17 17:27:08] One example that we've found, even in 500 we've looked at, is enough to make me wary of using it for validation purposes
 * [2009-02-17 17:27:24] <Physchim62> veratridine
 * [2009-02-17 17:28:33] <ChemSpiderman> what's the question PC?
 * [2009-02-17 17:28:45] <ChemSpiderman> (CSM cleans linoleic on CS)
 * [2009-02-17 17:29:43] <ali_as> Endrin and Dieldrin are a different set of problems I think PC, might be best to bring it up later.
 * [2009-02-17 17:30:52] <NormWork> What I miss? (reading scroll)
 * [2009-02-17 17:30:59] <ali_as> Norm :)
 * [2009-02-17 17:31:09] Just discussing issues with SDF files
 * [2009-02-17 17:31:49] <ali_as> The union file may have 'issues'. We are retreating to the sanctity of the CAS SDF file.
 * [2009-02-17 17:31:59] So I just thought if we stick with the SDF as supplied by CAS, we can say that the structure has been validated against the CAS structure.
 * [2009-02-17 17:32:13] <Physchim62> veratridine and others are problems of structure display,which make visual verification very difficult. if you could do a connection verification on some of these it would be helpful
 * [2009-02-17 17:32:49] Since we're doing a CAS# validation anyway, it seems logical anyway. When we do phase two - IUPAC names, InChIs etc, we will need CSM's SDF
 * [2009-02-17 17:33:03] <Physchim62> I certainly support keeping as closely to the CAS file
 * [2009-02-17 17:33:11] And any conflicts that we find can be resolved at that time
 * [2009-02-17 17:33:34] <Physchim62> a conflict is a conflict to be resolved: a match is a match
 * [2009-02-17 17:33:59] <ChemSpiderman> and what do we do when there are problems with the CAS file?
 * [2009-02-17 17:34:07] <ChemSpiderman> structures in there we disagree with?
 * [2009-02-17 17:34:18] <ChemSpiderman> in terms of name-structure pairs
 * [2009-02-17 17:34:33] <Physchim62> the practical problem is, "how do we know if there are problems with the CAS file?"
 * [2009-02-17 17:34:43] <ChemSpiderman> yup
 * [2009-02-17 17:34:58] I recommend that if we find a difference between the CAS structure and the WP structure, we check that against the Wikipedia SDF
 * [2009-02-17 17:35:03] <Physchim62> second, practico-polical problem: what do we do about them?
 * [2009-02-17 17:35:18] <Physchim62> walkerma, won't work
 * [2009-02-17 17:35:22] And log it in the "Problems" section
 * [2009-02-17 17:35:32] PC- explain
 * [2009-02-17 17:35:34] <ChemSpiderman> Remember ajmaline?
 * [2009-02-17 17:35:34] <Physchim62> the various files are too intertwined
 * [2009-02-17 17:35:51] <ali_as> The CAS file seems reliable, I thought I'd found 3 errors and they turned out to be bugs in the viewer program I was using.
 * [2009-02-17 17:36:06] You mean, we all pass the same corrupted information around between each other?
 * [2009-02-17 17:36:16] <Physchim62> remember the "First 500" file? CAS sent us a file with a duplicated entry, two different CASRNs!
 * [2009-02-17 17:36:51] <Physchim62> walkerma, I mean that we risk to be doing CAS's error correction for them
 * [2009-02-17 17:37:06] <ali_as> I don't see that as a risk.
 * [2009-02-17 17:37:30] PC- there can be all sorts of reasons for multiple CAS#s; in some cases, their file even lists that "This old one has now been superseded by this new one)
 * [2009-02-17 17:37:35] <ali_as> If we spot errors and they get to fix them then they get something back, I'm happy with that.
 * [2009-02-17 17:38:08] ali_as - I agree with that. I don't think we're seeing a lot of things wrong with the CAS file, are we?
 * [2009-02-17 17:38:20] With the NEW CAS file
 * [2009-02-17 17:38:21] <ali_as> eg, tubocurarine really needs redrawing for clarity.
 * [2009-02-17 17:38:29] <ChemSpiderman> anyway...we can go around in circles on this one :-) I'm fine with working on the CAS SDF only but the problem I have is structure depiction
 * [2009-02-17 17:38:36] <Physchim62> ali_as, it is for the good of human knowledge, yes, but we cannot use the CAS file as a "gold standard" if we're also amending it at the same time. Our validation criterion will become based on a circular argument. The best we can do is not validate until the situation is sorted out
 * [2009-02-17 17:38:38] <ali_as> I don't know of any errors in the new file.
 * [2009-02-17 17:38:43] <ChemSpiderman> YES..exactly...some of their structures are horrible
 * [2009-02-17 17:39:05] <ali_as> Some of the structures are not legible.
 * [2009-02-17 17:39:08] <ChemSpiderman> And redrawing will be very difficult for some structures
 * [2009-02-17 17:39:29] As I understand it, the structure representation depends not just on the original file, but on the software reading it, is that right CSM?
 * [2009-02-17 17:39:35] <Physchim62> CSM, exactly, but that shouldn't be an insoluble problem, no ;)
 * [2009-02-17 17:39:49] <ChemSpiderman> remember ajmaline?
 * [2009-02-17 17:39:57] <Physchim62> don't I just!
 * [2009-02-17 17:40:06] <Physchim62> did we ever sort that one out?
 * [2009-02-17 17:40:12] <ali_as> This has been mentioned twice, whats with ajmaline?
 * [2009-02-17 17:40:33] <ChemSpiderman> a natural product with lots of stereochem
 * [2009-02-17 17:40:49] <Physchim62> a very complex natural product
 * [2009-02-17 17:41:20] <ChemSpiderman> we didn't even get internal agreement on how to represent that
 * [2009-02-17 17:41:27] <ChemSpiderman> but different than CAS
 * [2009-02-17 17:41:56] <NormWork> The illustration in WP is equivocal, though the stereochemistry is not.
 * [2009-02-17 17:41:56] <Physchim62> my God! that has exactly the same problems as we are discussing about Endrin!
 * [2009-02-17 17:42:07] <NormWork> Worse, but yeah.
 * [2009-02-17 17:42:21] <ChemSpiderman> they are hard
 * [2009-02-17 17:43:00] <Physchim62> myself and CSM spent several hours discussing the structure of ajmaline about this time last year
 * [2009-02-17 17:43:09] I was away at the time. I'll raise the issue of that one with CAS, if one of you can email me what you found
 * [2009-02-17 17:43:26] It was in May/June, I was at my mother's
 * [2009-02-17 17:43:49] <Physchim62> I can't remember if the conversation even came to a result…
 * [2009-02-17 17:44:19] <ChemSpiderman> I've had the exchange with CAS already
 * [2009-02-17 17:44:30] <ChemSpiderman> they looked into it for me. I will find the emails
 * [2009-02-17 17:44:30] <Physchim62> I have many many structure file labelled "ajamaline01" and so on!
 * [2009-02-17 17:44:43] Anyway - I think we can agree that we may find a few things wrong, but these are likely to be "deep" problems - ones embedded in CAS for decades
 * [2009-02-17 17:44:51] <ali_as> The problems with that are worse than with Endrin and the 'Necker' type illusion is not resolved.
 * [2009-02-17 17:44:59] <NormWork> When you expand it, it's less equivocal, but it is anti-obvious, particularly the stereochemistry of the OH on the right.
 * [2009-02-17 17:45:13] <ChemSpiderman> the answer from CAS was basically "we are comfortable with our representation"
 * [2009-02-17 17:45:55] <ChemSpiderman> but that's because they use in THEIR systems. In our system it will have incomplete stereo
 * [2009-02-17 17:46:05] <ChemSpiderman> they don't use SDF
 * [2009-02-17 17:46:05] <ali_as> Oh, I just looked at that in wikichem.sdf Thats nuts.
 * [2009-02-17 17:46:13] Ha! Still, I'd be surprised if there are many such problems
 * [2009-02-17 17:46:19] <ChemSpiderman> they use CXF
 * [2009-02-17 17:46:32] * ali_as looks up CXF.
 * [2009-02-17 17:46:57] <ChemSpiderman> it's an example...so, back to the discussion (CXF is Chemical Exchange Format but they don't share with ANYONE...it's proprietary)...
 * [2009-02-17 17:47:06] <ChemSpiderman> sure...use the CAS SDF
 * [2009-02-17 17:47:15] <ali_as> Ahh.
 * [2009-02-17 17:47:17] Ironic that it has the word "Exchange" in it....
 * [2009-02-17 17:47:27] <ChemSpiderman> it's fine...but when we come to generating InChIs and SMILES all bets are off for structures like that
 * [2009-02-17 17:47:49] <Physchim62> even for InChIs?
 * [2009-02-17 17:47:52] <ChemSpiderman> yes..I have said that many times...who's the exchange with :-)
 * [2009-02-17 17:47:59] We should log all problem structures like that, right
 * [2009-02-17 17:48:01] <ChemSpiderman> yes espcially for inChis
 * [2009-02-17 17:48:03] ?
 * [2009-02-17 17:48:12] <Physchim62> especially for InChIs!!!
 * [2009-02-17 17:48:13] <ali_as> The SDF itself contains a matrix representation of the structure though, can't this be used to generate other forms?
 * [2009-02-17 17:48:43] <ali_as> More human friendly forms for example.
 * [2009-02-17 17:48:44] <ChemSpiderman> the stereo has to be properly encoded
 * [2009-02-17 17:48:49] <ChemSpiderman> into the molfile
 * [2009-02-17 17:48:53] <ChemSpiderman> it's an extreme
 * [2009-02-17 17:48:59] <ChemSpiderman> we just have to be conscious of it.
 * [2009-02-17 17:49:08] <ChemSpiderman> it's the 99.5%/0.5%
 * [2009-02-17 17:49:19] <Physchim62> walkerma, we cannot be sure of the stereo, we are literally unably to verify it with what CAS are giving us
 * [2009-02-17 17:49:41] <Physchim62> yeah, I'd say it's less than 1% of structures
 * [2009-02-17 17:50:14] One important fact needs to be raised here: In the current XML file there are around 7800 entries, but the SDF only contains 6206 of these.  They comment that the remainder are ones for which they could not generate a MOLFILE
 * [2009-02-17 17:50:32] CSM - what do they mean by that?
 * [2009-02-17 17:50:37] <Physchim62> if what CAS are giving us is what they've got, it's different from … we don't know in the end
 * [2009-02-17 17:51:01] <ChemSpiderman> they are talking CXF advantages over SDF
 * [2009-02-17 17:51:12] <Physchim62> the April file had several CASRNs which were mixtures of well difined compounds.
 * [2009-02-17 17:51:39] <ChemSpiderman> CXF is better with polymers, organometallics etc but I'm surprised that the difference is that large - there are also minerals etc
 * [2009-02-17 17:52:01] <ChemSpiderman> not everything is a structure of course...
 * [2009-02-17 17:52:05] The email in December said: "As with the version we sent this past Spring, this only contains those
 * [2009-02-17 17:52:07] substances we could generate a molfile for, so it is not the full
 * [2009-02-17 17:52:07] <Physchim62> could someone email me a link to the XML file?
 * [2009-02-17 17:52:09] collection."
 * [2009-02-17 17:52:32] It's available as txt on pluto
 * [2009-02-17 17:52:33] <ali_as> The one on pluto PC?
 * [2009-02-17 17:52:51] <Physchim62> ali_as, yes that one!
 * [2009-02-17 17:53:35] <Physchim62> I'm not sure I have the tools to treat it, but it would be interesting to see what has been left out.
 * [2009-02-17 17:54:21] <Physchim62> While I've been doing the inorganics, I've kept a lift of numbers that CAS *hasn't* given us
 * [2009-02-17 17:54:43] Such as...?!
 * [2009-02-17 17:54:47] <Physchim62> maybe some of them are in there
 * [2009-02-17 17:55:18] <Physchim62> walkerma, you read the interim report: mostly hydrates which are more usually found than the anhydrous salts
 * [2009-02-17 17:55:44] <ali_as> http://pluto.potsdam.edu/wikichem/images/4/4f/CAS-CommonChemistry2008.xml.zip
 * [2009-02-17 17:55:49] <ali_as> I think.
 * [2009-02-17 17:55:49] <Physchim62> or other compounds with a very simple connection like that
 * [2009-02-17 17:55:52] <ChemSpiderman> have to get off IRC in 5 mins for a phone call
 * [2009-02-17 17:56:03] <ChemSpiderman> will watch...won't type anymore.
 * [2009-02-17 17:56:22] Ah yes; I'm sure we can request those, especially if they are common - they'd be very amenable, I'm sure
 * [2009-02-17 17:56:49] <Physchim62> I'm sure, that's why I'm taking the time to record them ;)
 * [2009-02-17 17:57:13] <Physchim62> we already have the numbers for most of them, from commercial suppliers and often from official bvodies
 * [2009-02-17 17:58:32] <Physchim62> OK, "most" should be "a lot", but the rule I've used for the wishlist is:
 * [2009-02-17 17:58:39] <ali_as> MSDS sheets and such I assume.
 * [2009-02-17 17:58:48] They may have removed them when they got rid of mixtures
 * [2009-02-17 17:58:54] <Physchim62> those compounds we've already mentioned in articles where CAS has already given us data
 * [2009-02-17 17:59:07] <Physchim62> ali_as, even simpler, catalogues!
 * [2009-02-17 17:59:22] <ali_as> Ahh, like sigma, yes they have CAS.
 * [2009-02-17 17:59:49] <ali_as> Actually the sigma catalogue is the only book I have with CAS numbers in, figures it would be a book trying to sell me something.
 * [2009-02-17 18:01:29] <Physchim62> ali_as, http://www.sigmaaldrich.com/
 * [2009-02-17 18:02:01] <Physchim62> they're not perfect, but they are as good as WP!
 * [2009-02-17 18:02:04] <ali_as> I've looked at that, I asked martin if that was an acceptable source for verifying other numbers.
 * [2009-02-17 18:02:10] <ChemSpiderman> there are issues with that too... I've seen examples where Sigma have the incorrect structure associated with a CAS number
 * [2009-02-17 18:02:17] <Physchim62> at the moment, no.
 * [2009-02-17 18:02:22] <ChemSpiderman> ok..just watching now
 * [2009-02-17 18:02:36] <ali_as> But it's not trusted, so not in the same catagory as the CAS list.
 * [2009-02-17 18:02:48] <Physchim62> exactly
 * [2009-02-17 18:03:13] It's common to use catalogues to fill in blank entries in Chemboxes, but I wouldn't dare consider such things reliable
 * [2009-02-17 18:03:24] <ali_as> I had another wild thought about citing CAS, how about [CAS] like any other citation?
 * [2009-02-17 18:03:31] <Physchim62> it's a source, there are plenty like it, but it doesn't have the same status (for CASRNs) as the data from CAS
 * [2009-02-17 18:03:50] ali_as: explain
 * [2009-02-17 18:04:37] <ali_as> When you cite something in text you get a superscript [1].
 * [2009-02-17 18:04:43] * Beetstra reads back .. sorry guys, was busy with Ir and Rh ..
 * [2009-02-17 18:04:59] <Beetstra> Happy Birthday Physchim62!
 * [2009-02-17 18:05:18] <Physchim62> Dank je Dirk
 * [2009-02-17 18:05:22] <ali_as> We could have a superscript [CAS] rather than turning the CAS green and then it's compatable with the existing wikipedia format.
 * [2009-02-17 18:05:35] <Beetstra> :-)
 * [2009-02-17 18:05:59] <ali_as> ie you wouldn't have to explain to anyone what it was supposed to mean or where the validation comes from.
 * [2009-02-17 18:06:15] <Physchim62> ali_as, the problem, and it is not *absolute* is that people don't like too many superscripts beside figures with superscripts.
 * [2009-02-17 18:06:44] <Physchim62> so the chembox has evolved to avoid those sort of refs
 * [2009-02-17 18:07:14] If we can get that to work within a chembox, so the article still prints correctly (a common problem) it would be worth considering, but it usually creates problems to do such thins
 * [2009-02-17 18:07:18] things
 * [2009-02-17 18:07:30] I have to be gone for a moment
 * [2009-02-17 18:07:36] <Physchim62> we can change that assessment of opinion, of course,, but that's where the present system come from
 * [2009-02-17 18:07:46] <ali_as> Ok.
 * [2009-02-17 18:09:06] <Physchim62> obviously: change opinion = not impossible by any means, but not easy or quick
 * [2009-02-17 18:09:18] <NormWork> Look, what we do doesn't have to be perfect, we just need to indicate what things verify and what things are unverified.
 * [2009-02-17 18:09:27] <ChemSpiderman> check your email for ajmaline feedback
 * [2009-02-17 18:09:34] <ali_as> I'm not trying to push for a huge change, I just think about this from time to time.
 * [2009-02-17 18:09:34] <NormWork> "Never do to day what you can put off unitl tomorrow"
 * [2009-02-17 18:09:36] <Physchim62> and that is our current philosophy
 * [2009-02-17 18:09:37] <Beetstra> NormWork hits a sensitive part there ..
 * [2009-02-17 18:10:04] <Beetstra> How do we in a box indicate that we have a verified CAS, but not a verified melting point ..
 * [2009-02-17 18:10:06] <NormWork> WP improves by evolution, not by "intelligent design"
 * [2009-02-17 18:10:44] <ChemSpiderman> (CSM on hold so still typing)..let's do what we can...mark the major challenges and return to them later
 * [2009-02-17 18:10:46] * NormWork was considering invoking his physical chemistry class to verify data, actually.
 * [2009-02-17 18:11:05] <NormWork> Hey, CSM, glad to meet you!
 * [2009-02-17 18:11:16] NormWork: Get them to verify it in the lab!
 * [2009-02-17 18:11:25] <Physchim62> NOOOOO!
 * [2009-02-17 18:11:38] <ChemSpiderman> Hi NormWork...we COULD consider putting together a "game" and getting people to work on the challenges with us. See http://www.chemspider.com/blog/the-spectral-game-as-an-educational-tool.html
 * [2009-02-17 18:12:11] * Beetstra wonders what walkerma wants NormWork to verify .. LD50 of HCN??
 * [2009-02-17 18:12:39] <ChemSpiderman> We can ask people "Is this the structure of Taxol? if not why not?"
 * [2009-02-17 18:13:15] <ChemSpiderman> then compare those structures with those in the CAS file. Crowdsourced validation....a bigger version of US doing it
 * [2009-02-17 18:13:19] <Beetstra> It would give double work, ChemSpiderman ..
 * [2009-02-17 18:13:32] <ChemSpiderman> probably yes
 * [2009-02-17 18:13:43] <Physchim62> I know I always bring up the same example, but Talk:Cyclohexanone. Which poor students are you going to make measure the melting point of cyclohexanone= And, more importantly, how would you know they've done it properly?
 * [2009-02-17 18:13:43] But it might be more fun!
 * [2009-02-17 18:13:45] <Beetstra> Because I am afraid that we still would have to check them again
 * [2009-02-17 18:14:13] <ChemSpiderman> Yup..'fraid so
 * [2009-02-17 18:14:25] <Physchim62> ChemSpiderman, interesting, I'm just in the process of suggesting that for the mainpage article on WP as well
 * [2009-02-17 18:14:40] <ali_as> I wonder if you can use a zone refining method on cyclohexanone.
 * [2009-02-17 18:14:59] <Physchim62> ali_as, I don't think it would work
 * [2009-02-17 18:15:19] <ChemSpiderman> but if 10 people tell you you're a horse..saddle up...so if 20 people say "this is how we'd draw ajmaline" then....
 * [2009-02-17 18:15:23] <ali_as> Ok, why not?
 * [2009-02-17 18:15:23] <Physchim62> the uncertainty is due to supercooling, not impurities
 * [2009-02-17 18:15:37] <ChemSpiderman> anyway...it;s adding more complexity and we don't need it
 * [2009-02-17 18:15:55] <Physchim62> CSM, a fine word if ever there wa one
 * [2009-02-17 18:15:58] <Physchim62> was
 * [2009-02-17 18:16:08] <ali_as> Supercooling usually isn't an issue where you have solid contacting liquid.
 * [2009-02-17 18:16:35] Some of our members can't handle anything as complex as the word "was" so I don't think it'll work
 * [2009-02-17 18:16:43] <Physchim62> ali_as, but which of the dozen or so papers had the correct conditions?
 * [2009-02-17 18:17:32] <Physchim62> indeed, if there are no more topics to be discussed…
 * [2009-02-17 18:17:43] <ali_as> I'm not in a position I can read any of the papers I expect. Interesting problem though.
 * [2009-02-17 18:18:16] <Physchim62> ali_as, it is a very interesting problem, you're quite right, and one we haven't solved yet ;)
 * [2009-02-17 18:18:37] <ali_as> SOmetimes there is a cunning solution, like the method of determining the temperature of maximum density of a liquid.