Wikipedia:WikiProject Chemistry/IRC discussions/15 Jan 2008
--- Log opened Tue Jan 15 11:04:03 EST 2008
11:04 <+walkerma> H2Cr=CrH2 !
11:04 <+dmacks> haha!
11:05 <+dmacks> I'd say it's maybe more likely they mean the benzopyran idea?
11:05 <+ChemSpiderMan> I'm loking at IUPAC rules ...B.3, R9.1
11:05 <+ChemSpiderMan> Llet me find you the link online
11:05 <+Rifleman_82> hahahaha
11:05 <+Rifleman_82> hi dmacks
11:05 <+dmacks> hello
11:05 <+walkerma> We're still waiting for Itub and Arcadian. Henry Rzepa can't come till later, if at all. I say we get started, OK?
11:06 <+Rifleman_82> itub is here i think...
11:06 <+dmacks> okidoke.
11:06 -!- mode/#wikichem [+o Rifleman_82] by ChanServ
11:06 <+ChemSpiderMan> fine with me...the link is here: http://www.acdlabs.com/iupac/nomenclature/93/r93_691.htm
11:06 <+walkerma> Shall I attempt to moderate?
11:06 <@Rifleman_82> please
11:06 <+dmacks> good idea.
11:06 -!- mode/#wikichem [-o Rifleman_82] by Rifleman_82
11:06 <+ChemSpiderMan> yes please
11:06 <+kelson> dmacks: hello
11:06 <@Beetstra> Good plan, Walkerma
11:06 <+pm286> yes please
11:06 <+Rifleman_82> invited itub... if he's not away i guess he'll be here
11:07 <+walkerma> OK, I'd like to ask ChemSpiderMan to explain what he's been doing, and where we stand with the chemicals list
11:07 <+ChemSpiderMan> How many people on this chat have had the chance to look at the PDF reports I've been generating..that will help
11:07 * dmacks has briefly
11:07 * Beetstra had a look .. briefly
11:08 * Rifleman_82 had a look at the old version
11:08 <+ChemSpiderMan> ok...what I've done is take the dumps provided by walkerma..
11:08 <+ChemSpiderMan> used them as the basis to review the data on WP
11:08 <+CheMoBot> user:Petermr has edited monitored page Wikipedia talk:WikiProject Chemicals - diff - (+131)- summary: /* IRC discussion on using Wikipedia chemistry pages to provide chemical data */
11:09 <+ChemSpiderMan> I imported the files into a software app andused a script to create a URL link to each record in WP from the title
11:09 <+ChemSpiderMan> I did some basic text searches to remove "stuff" (polite word used)
11:10 <+ChemSpiderMan> Then set to work to review the records "one-by-one"
11:10 <+CheMoBot> user:Petermr has edited monitored page Wikipedia talk:WikiProject Chemicals - diff - (+83)- summary: /* IRC discussion on using Wikipedia chemistry pages to provide chemical data */
11:10 <@Beetstra> CheMoBot quit
11:10 -!- CheMoBot [n=beetstra@69.37.50.156] has quit ["Mayday! Mayday! .. going down!"]
11:10 <+dmacks> Quick question: "dumps provided by walkerma" are...?
11:10 <+ChemSpiderMan> people still there?
11:10 <+walkerma> Yes
11:10 <@Beetstra> Yes, all here
11:11 <+walkerma> Except for the bot
11:11 <+Rifleman_82> haha okay
11:11 <+Rifleman_82> was gonna kick the bot
11:11 <+ChemSpiderMan> Its been eyes applied to records. I have checked structure drawings, systematic names, consistency of links to PubChem
11:12 <+walkerma> [The dumps are a list of all the chemicals and drugbox articles]
11:12 <+ChemSpiderMan> I am at letter "M" in the file (but already knocked out X,Y,Z) and "TRYING" to knock out three letters per day
11:12 <+dmacks> As determined by...transcusion of an infobox, manual additon to a cat?
11:13 <+ChemSpiderMan> I have access to Name to structure conversion software and name generation software...I managed Nomenclature software in my previous role
11:13 <+ChemSpiderMan> I have three software packages in my hands and I use all of them as necessary
11:13 <+ChemSpiderMan> The report I am generating for WIkipedia is one of many
11:14 <+ChemSpiderMan> I generate a report regarding errors in the software I am using and forward it to the development team(s)
11:14 <+ChemSpiderMan> I check consistency between emolecules, PubChem and ChemSpider
11:15 <+ChemSpiderMan> I am at the letter M and am down to 6700 records in the DB at present.
11:15 <+Rifleman_82> quick question
11:15 <+Rifleman_82> how do you do it ... so quickly?
11:15 <+ChemSpiderMan> I foresee that by the time I finish I will be down to about 5500-6000 total
11:15 -!- itub [n=tubert@lalo.chemie.unibas.ch] has joined #wikichem
11:15 -!- mode/#wikichem [+v itub] by ChanServ
11:15 <+Rifleman_82> do you have something to automate the process or is it a slow manual comparison?
11:15 <+Rifleman_82> hi itub
11:15 <+itub> hi
11:15 <+itub> lots of people today!
11:15 <+ChemSpiderMan> Quickly...interesting question. It's called "Xmas vacation" really...working until 2am most nights when my family went to bed
11:16 <+Rifleman_82> your dedication is impressive!
11:16 <+ChemSpiderMan> Some things are fast...simple structures are easy. The process....
11:16 <+walkerma> He's also REALLY fast at using the software and eyeballing structures
11:16 <+ChemSpiderMan> My lunacy irritates my wife...
11:17 * ChemSpiderMan ChemSpiderman reflects on the "rants" when cold feet crawl into bed...
11:17 <+ChemSpiderMan> So, the process.
11:17 <+ChemSpiderMan> Open the record in the desktop software and click the link to open up the WP page
11:18 -!- pm286 is now known as petermr
11:18 <+ChemSpiderMan> Two monitors or split screen on laptop while watching cheesy movies
11:18 <+ChemSpiderMan> since the title of the article in general is the name of the compound click one button in the desktop software to generate the molecule for comparison
11:19 <+ChemSpiderMan> (I could do this in batch mode for the whole file but don't have that software)
11:19 <+ChemSpiderMan> Now...eyeball for differences. Then, see if the systematic name on WP will convert to the structure on WP
11:20 <+ChemSpiderMan> Validates the systematic name very well...caught many errors...
11:20 <+Rifleman_82> your software can generate structures, given a systematic name?
11:20 <+ChemSpiderMan> Validates "is the name consistent with the structure" NOT that the structure is the right structure
11:21 <@Beetstra> You check the structure by eye, I suppose .. do you catch wrong ones in that way?
11:21 <+ChemSpiderMan> Validate the CSA number as much as I can...I don't have access to any CAS databases and this is why I aask for people to help me validate these
11:21 <@Beetstra> (I know that you can't possibly know all molecules)
11:21 <+ChemSpiderMan> Oh yes...very easy to catch differences by eye
11:22 <+Rifleman_82> you don't trust emolecules?
11:22 <+ChemSpiderMan> Rifleman : yes, the software can generate a structure given a systematic name
11:22 <+Rifleman_82> (re: cas database)
11:22 <+ChemSpiderMan> as well as trivial names
11:22 <+Rifleman_82> is that the paid version of chemsketch 10?
11:22 <@Beetstra> CAS is difficult, there are cases where the different databases have different molecules
11:23 <+ChemSpiderMan> Also, since I have access to systematic names and trivial names on ChemSpider I validate there too
11:23 <@Beetstra> I recall Physchim62 ranting about that
11:23 <+Rifleman_82> Beetstra: yeah, i get that sometimes
11:23 <+ChemSpiderMan> The paid version of ChemSketch 10 (actually 11 now) only has limited capabilities...it does NOT convert names to structures
11:23 <+Rifleman_82> emolecules may give fragments for the same cas...
11:23 <@Beetstra> I have had problems with pubchem/emolecules .. not sure sometimes what is what
11:24 <+ChemSpiderMan> One comment here...I suggest NOT validating CAS against eMolecules, PubChem or Chemspider
11:24 <+Rifleman_82> they're no better than us?
11:24 <+ChemSpiderMan> ALl of us have the same problems.
11:24 <+ChemSpiderMan> The CAS numbers come from depositors
11:24 <@Beetstra> Well, CAS should be checked against CAS.org .. the problem there is .. they want money if you want more than .. say .. 20
11:24 <+ChemSpiderMan> They are not careful with the CAS numbers being correct
11:25 <+ChemSpiderMan> Yes...I am afraid you are right
11:25 <+walkerma> I could talk to ACS about this if you like
11:25 <@Beetstra> OK, what about InChI and SMILES?
11:25 <+ChemSpiderMan> But I would rather that the CSA numbers were not hyperlinked to any external dbs
11:25 <+dmacks> Are the CAS-number mistakes simple typos, or "wrong compound" entirely?
11:25 <+ChemSpiderMan> wrong compound
11:25 <+ChemSpiderMan> or multiple compounds
11:25 <+dmacks> 'k
11:26 <+Rifleman_82> dmacks: the way we find cas numbers for compounds... such mistakesshouidn't be surprising
11:26 <+ChemSpiderMan> A CAS number associated with a neutral is different than a CAS number for a salt
11:26 <+dmacks> yeah
11:26 <+Rifleman_82> some sort of checksum?
11:26 <+Rifleman_82> i mean not checksum but numbering convention?
11:26 <+ChemSpiderMan> it is common to have confusion within the article
11:26 <+ChemSpiderMan> The article name can be for a salt
11:26 <+ChemSpiderMan> The structure is the neutral
11:26 <+ChemSpiderMan> The name is the salt
11:27 <+Rifleman_82> happens for drugs very often... where you have the valerate etc
11:27 <+ChemSpiderMan> The CAS number is the neutral
11:27 <+ChemSpiderMan> The SMILES is for "that girl in the corner smiling at me"
11:27 <+Rifleman_82> haha
11:27 <+Rifleman_82> ok
11:27 <+ChemSpiderMan> Okay..InChIs
11:27 <+ChemSpiderMan> My opinion is take them off for now
11:27 <+ChemSpiderMan> Here's why...
11:28 <@Beetstra> they are ugly?
11:28 <+ChemSpiderMan> Everyone I have seen (NOT many) is "broken" with a line break
11:28 <+Rifleman_82> we can leave them in but disable the field for the moment
11:28 <+dmacks> They make screen layout a mess?
11:28 <+ChemSpiderMan> Same issue on PubChem..
11:28 * dmacks likes that, Rifleman_82
11:28 <+ChemSpiderMan> When an InChI on Pubchem needs to be converted you have to copy, remove spaces etc
11:29 <+ChemSpiderMan> I say wait for InChI KEys..
11:29 <+ChemSpiderMan> And here's my commitment to this team...
11:29 <+Rifleman_82> what about doing away with gifs and pngs and using some sort of mol file, with the molecular formula, smiles, etc. generated on the fly?
11:29 <+ChemSpiderMan> When the file is "finished" then I will return the following associated with each structure
11:29 <+Rifleman_82> not aware of what is technically available here and now but... i think it'll make sense
11:29 <+itub> on the fly generation can be ugly
11:29 <+ChemSpiderMan> If they can be generated....
11:30 <+Rifleman_82> generation of smiles and inchi?
11:30 <+ChemSpiderMan> Hold on...let me get to your suggestion...
11:30 <+itub> no, the figures
11:30 <@Beetstra> JMOL is busy with that .. but still ..
11:30 <+itub> generating the smiles and inchi is certainly possible
11:30 <+Rifleman_82> figures can be rotatable... ball stick wireframe etc
11:30 <+Rifleman_82> i'll let ChemSpiderMan get tto his point
11:30 <+itub> I think the inchi could be included but hidden in a way that people who want it can acess it
11:31 <+itub> maybe with a [show] button
11:31 <+ChemSpiderMan> So, each structure will have: 1) structure 2) InChIKey 3) IUPAC Name 4)SMILES AND...
11:31 <+ChemSpiderMan> I already worked with Walkerma to embed InChIs into structure images (png)...
11:31 <+ChemSpiderMan> I know that InChIkeys, InChIStrings, SMILES, molfile can ALL be membedded into images
11:32 <+ChemSpiderMan> embedded
11:32 <+ChemSpiderMan> the question is about searching them....
11:32 <+ChemSpiderMan> Walkerma can point you to histroicaldiscussions about this..
11:32 <+Rifleman_82> too tedious to put them in the image description page on wiki?
11:32 <+walkerma> We need to make sure that the embedded info gets into the article page rather than the image page
11:33 <+walkerma> So a Google search finds the article
11:33 <+ChemSpiderMan> If I return an SDF file to you, or some other agreed upon format then you could "bot" them into fields
11:33 <+walkerma> ChemSpiderMan, I think this is what we need to do for now
11:33 <+ChemSpiderMan> I AGREE
11:33 <+walkerma> Can we write such a bot?
11:33 <+dmacks> Wiki question: Can arbitrary pages be transcluded, or only pages in the Template: namespace?
11:34 <@Beetstra> Everything can be transcluded
11:34 <+Rifleman_82> arbitrary pages can be transcluded
11:34 <+Rifleman_82> mainspace pages can be transcluded with a {{[[Template::sodium chloride|:sodium chloride]]}}
11:34 <+Rifleman_82> for example
11:34 <+Rifleman_82> you use the ":"
11:34 * ChemSpiderMan going quiet now until there are questions
11:35 <+dmacks> So the embedded image data could be extracted onto the image talk page, and then that talk page be transcluded into the infobox.
11:35 <+walkerma> Sounds interesting!
11:35 <+Rifleman_82> but...
11:35 <+Rifleman_82> you'll want all this to be commented out?
11:35 <+dmacks> <noinclude>
11:35 <@Beetstra> That would contain to much information
11:35 <+ChemSpiderMan> The data can be in the image (but there is development work to be done there) but the SDF file is an SDF file with the data available now..
11:36 <+ChemSpiderMan> You can 'bot in what you need
11:36 <+Rifleman_82> so google can find them but you don't see them? <noinclude> will not be transcluded
11:36 <+dmacks> Er, <includeonly>
11:36 <+Rifleman_82> i think it'll display ... or you can hide them in <!-- -->
11:36 <+walkerma> Can we write a bot to handle the SDF file?
11:37 <+Rifleman_82> but then each page will balloon... and there might be some issues about transclusion and server load?
11:37 <+ChemSpiderMan> There needs to be a way to generate the images in batch mode from the SDF file to generate images for the whole database for you to deposit with the 'bot
11:37 <+ChemSpiderMan> NOT necessarily a WP problem...
11:37 <@Beetstra> What about putting the full identifiers on <pagename>/Data_page for ALL compounds .. ??
11:38 <+Rifleman_82> ' /datapage is seen as a new page, not a sub page
11:38 <@Beetstra> There the width of a box is not a problem
11:38 * dmacks was just about to ask that as a more general philosophica qurstion: what's the feel in WP for using complex inclusions? I dislike the idea of having to bot and re-hard-code the same info in multiple places vs having all data somewhere and extracting as needed.
11:38 <+Rifleman_82> and the use of this "trick" of transclusion is frowned upon
11:39 <+dmacks> Rifleman_82: the whole template: namespace is shifting towards doing this though (the /doc game) I think.
11:39 <+walkerma> [ChemSpiderMan: I think we'll have to let this discussion happen - it is of importance within WP, how we get data into the chembox]
11:39 <@Beetstra> We don't have to transclude the datapage .. google finds the datapage, which is one link away from the real page
11:39 <+Rifleman_82> dmacks: yes, i'm aware... but still frowned upon in the mainspace
11:40 <+Rifleman_82> that's the impression i got from #wikipedia-en-admins after i tried something funny
11:40 <+dmacks> Gotcha.
11:40 <@Beetstra> The ugly InChI is a problem .. we could add a field 'RealInChI', which is not displayed, but is the one that is correct (using the original InChI as the display one)
11:40 <@Beetstra> That is easy to code into {{chembox new}}
11:40 <+Rifleman_82> dmacks: check the history of Rules of basketball
11:41 <+walkerma> Look at http://en.wikipedia.org/wiki/Tributylphosphine#External_links the best InChI system we have now
11:41 <+dmacks> So the data page would be the "actual" data, and then bot would copy certain fields into the main-page?
11:41 <+dmacks> Rifleman_82: Ah yeah, I vaguely remember that.
11:42 <+walkerma> Beetstra: That RealInChI is a good one
11:42 <+Rifleman_82> having a bot copy it over seems... a bit repetitive... it'd be nice if everything from mw to inchi and systematic name can be generated from the molecule itself
11:43 <+ChemSpiderMan> It can
11:43 <+ChemSpiderMan> It will include formula and Mw
11:43 <+dmacks> So "the structure" is the primary piece of data?
11:43 <+ChemSpiderMan> All will be generated automatically and part of the file...can include monoisotopic masses for MS people if you want them
11:44 <+ChemSpiderMan> Yes...the structure is the primary piece of data from me...but the name is the primary piece of data for the article
11:44 <+walkerma> MS - On the data page
11:44 <@Beetstra> oops
11:45 <+dmacks> So the data page would have lots of data auto- (or bot-) generated from the image?
11:45 <+ChemSpiderMan> No, from the structure...the connection table
11:45 * Beetstra should use preview
11:45 <+dmacks> Yeah, "generated" including "extracted from image embedded data"
11:45 <+Rifleman_82> does this mean we're importing someone's database wholesale
11:45 <+Rifleman_82> ?
11:45 <+ChemSpiderMan> Yes...the one I am building
11:46 <+Rifleman_82> ah okay
11:46 <+walkerma> We're reimporting our own database!
11:46 <+Rifleman_82> any copyright issues there?
11:46 * kelson think about http://de.wikipedia.org/wiki/Vorlage:Personendaten
11:46 <+ChemSpiderMan> What I recieved from WP was a list of article names
11:46 <+Rifleman_82> we're exporting our database, tidying it up, reimporting them?
11:46 <+ChemSpiderMan> "kind of"
11:47 <+ChemSpiderMan> What you've handed me is a list of article names
11:47 <+Rifleman_82> how about maintainability? the next time we want to validate them, do we have to go through them all again? or can we flag those without changes? or check only changed parameters?
11:47 <+Rifleman_82> and that kinda means data is only as valid as the last check/
11:48 <+ChemSpiderMan> What I'm doing is handing back structures, intrinsic properties, "algorithmically-generated" names, InChIs, SMILES etc
11:48 <+dmacks> Having the data be on a separate page would help with this kind of maintainability (eliminates article text changes, so monitoring "changes to the page" clearly means the data has changed.
11:48 <@Beetstra> Regarding the RealInChI, see the code of http://en.wikipedia.org/wiki/Tributylphosphine
11:49 <+Rifleman_82> dmacks: will there be a greater barrier to the lay user, to editing and filling out the chembox?
11:49 <+ChemSpiderMan> If the structure is correct and the algorithms to add atomic weights isn't broken, and inChI generation doesn't change and,and,and then it won't need changing. Never say never, but, in my opinion, a big step ahead
11:50 <+ChemSpiderMan> What you shouldn't allow, in my opinion, is the editing of intrinsic properties extracted from the molecule that is displayed...
11:50 <+dmacks> Rifleman_82: could have an "edit" button on the infobox that goes to the data-page (like many navboxes have)
11:50 <+ChemSpiderMan> I saw examples where the molecule drawn didn't maytch the mass or the formula
11:50 <+Rifleman_82> Beetstra: that's nice... but looks timeconsuming unless you're going to get a bot to arbitrarily add linebreaks after a few chars
11:50 <+Rifleman_82> dmacks: fair enough
11:51 <+ChemSpiderMan> the MF and Mw are derived FROM the structure, not independent of it
11:51 <+walkerma> I think we should have another IRC meeting once we have the complete SDF to look at, would that be OK?
11:52 <+walkerma> I think I'd like to move on to ask Peter about his work.
11:52 <+walkerma> And his plans
11:52 <@Beetstra> Mw can be generated automatically in chembox new, use 'C = 1 | H = 4' i.s.o. 'Formula = CH<sub>4</sub>'
11:52 <+petermr> hi - reay when you are...
11:52 <+ChemSpiderMan> BTW...look at triphenylphosphine and the IUPAC name...
11:52 <+ChemSpiderMan> sorry tributylphosphine..
11:52 <@Beetstra> err .. and its molecular weight, which is off ..
11:53 <+petermr> ok - do you want me to strat?
11:53 <+petermr> == start
11:53 <+walkerma> Would that be OK?
11:53 <+Rifleman_82> please go ahead peter
11:53 <+ChemSpiderMan> yes
11:53 <@Beetstra> Yes Peter, please
11:53 <+Rifleman_82> we'll sort out the finer points another time and place
11:54 <+Rifleman_82> i mean of the last thread
11:54 <+petermr> thanks to all of you for the work you have put in
11:54 <+petermr> there are 2-3 issues. One, which I mention initially is that we plan to apply for short-term funding to develop a RDF-based version of the WPChem to support repositories
11:55 <+petermr> I mentioned this to walkerma in mail.
11:55 <+petermr> will have to get the grant in soonish - all I'm doing here is flag it up
11:56 <+petermr> I see WPChem as a semantic resource which will be used in conjunction with chemical literature and that is the main theme of the grant
11:56 <+petermr> the second issue is related to what you have been talking about - how should the data be maintained.
11:56 <+petermr> Sorry for typos
11:57 <+petermr> I think there has to be a mechanism whereby the data can be checked and edited by bots and passed back into WP
11:57 <+petermr> That's why I was asking walkerma about BNF (a formal spec for the info boxes)
11:57 <+petermr> please interrupt if this doesn't make sense
11:57 <+ChemSpiderMan> AGreed...for example, with the new InChI generation will come new InChIkeys
11:57 <+dmacks> Wouldn't be hard to write the grammar for it.
11:58 <+petermr> good. I have found ca 4 different info bixes for chemistry and I suspect there are some more. Is there a clear policy on what is currently supported and whether the old ones will be refactored?
11:58 <+ChemSpiderMan> Also, new nomenclature rules change names etc. So, the structure will need to be available to WP to boss to bots.
11:58 <+petermr> the vision is that a bot can extract all the data from the box and do its own thing witn it
11:59 <+ChemSpiderMan> and pass back and repopulate?
11:59 <+Rifleman_82> petermr: there will be only one infobox very soon
11:59 <+Rifleman_82> we're clearing the remnants... about 80 or so left
11:59 <+Rifleman_82> one infobox for chemicals
11:59 <+Rifleman_82> one infobox for elements
12:00 <+petermr> Then there is the questions of the actual way information is entered. I downloaded all the extant pages (through the list of chemicals )
12:00 <+Rifleman_82> one infobox for drugs which i hope will be subsumed into the chemicals box at a later date
12:00 <@Beetstra> Rifleman_82, wait, 2 for chemicals, {{chembox new}} and {{drugbox}}
12:00 <+Rifleman_82> that's about it
12:00 <@Beetstra> OK
12:00 <@Beetstra> :-)
12:00 <+petermr> in XML... And although I can parse the XML the data in the infobox were rather variable
12:00 <+petermr> OK I will go with just chembox new
12:00 <+walkerma> [Rifleman has run a bot through the remaining articles recently to update all the chemboxes]
12:01 <+petermr> the main probelsm with the data were (from memory) character encodings (can be awful), lack of consistency in units, difficulty of parsing annotations in values (e.g. 200 (decomposes))
12:01 <+dmacks> Would be good to sanitize the data somehow, both for display/MOS/etc purposes and to make it easier to parse.
12:01 <+Rifleman_82> my search doesn't catch those weird ... manual html boxes, nor manual tables using wikisyntax though
12:02 <+petermr> it would b e nice to have slightly more structure
12:02 <+petermr> (not chemical structure)
12:02 <+dmacks> :)
12:02 <+walkerma> Structure? In Wikipedia?
12:02 <+Rifleman_82> petermr: look at {{drugbox}}. they've had a bit more success at it
12:02 <+Rifleman_82> boiling_high and boiling_low to give a boiling point range for example
12:02 <+petermr> structure in the infobox.
12:02 <+Rifleman_82> where's physchim? not coming?
12:02 <+petermr> :agree with rfileman
12:03 <+walkerma> No, his (PCs) internet is down I think
12:03 <+petermr> (how do I enter a comment)
12:03 <+dmacks> /me
12:03 <+dmacks> "/me agrees with rfileman"
12:03 <@Beetstra> technically, for boilingpoints we have BoilingPt and Boiling_Notes .. but not used consistently
12:03 <+petermr> OK - anyway this is a first pass - I shall work with what I have got - ca 1000 pages
12:04 <@Beetstra> And advanced, even BoilingPtC, BoilingPtK and BoilingPtF
12:04 <+Rifleman_82> sheesh
12:04 <+Rifleman_82> no rankine?
12:04 <+petermr> I suggest separating units from quantities. Is there any reason to HOLD quantites in different units (display is a different question)
12:05 <+petermr> it leads to errors and lack of consistency
12:05 <+dmacks> Back to personal pet peeve: duplication of info. Those are not independent values, so why entering them distinctly instead of generating all from one?
12:05 <+dmacks> Yeah, what petermr said.
12:05 <+petermr> exactly. then someone alters one MPt and fails to edit the other
12:05 <+ChemSpiderMan> yes...agreed...it's all conversions
12:05 <@Beetstra> No, there is no reason, except for a historical one (it needed to be done that way in the old chembox)
12:05 <+Rifleman_82> so why don't we just use degrees C? that's by far the most common
12:05 <+Rifleman_82> and generate from there?
12:05 <+Rifleman_82> but then we get problems with significant figures
12:06 <+petermr> being dictatorial suggest SI units for all basic info unless agreed exceptions
12:06 <@Beetstra> That is what BoilingPtC does :-p
12:06 <+Rifleman_82> agree with petermr
12:06 <+Rifleman_82> petermr: another form of inconsistency is in densities
12:06 <+Rifleman_82> some in g/mL (liquids), solids in g/cm^3
12:06 -!- kelson [n=Kelson@fw-vianetworks.minick.ch] has quit [Read error: 104 (Connection reset by peer)]
12:06 <+petermr> I can see that it may be useful to report the original reported values asis but this is rather special
12:06 <+Rifleman_82> some quoted in kg/m^3, which seems to be preferred for gases
12:06 <+dmacks> Regarding data & conversions, {{drugbox}} already does this.
12:07 <+Rifleman_82> and in chemicals produced in high volume
12:07 <+petermr> anyway I am sure that you will eveolve a consist mechanism
12:07 <+petermr> an intermediate goal is then to create a RDF version of WPchem which can be quried using SPRARQL
12:07 <+Rifleman_82> we can just standardize with g/cm^3 since it works equally well for solids, liquids, and gases? the recent debate about cm^3 and mL notwithstanding
12:08 <@Beetstra> For boiling/melting points it is not too hard, kill the unit and use the advanced correct field
12:08 <+petermr> ==SPARQL
12:08 <+petermr> like DBPedia
12:08 <+petermr> here you can ask natural queries without needing a database
12:08 <+petermr> WP becomes the database
12:09 <+petermr> it works very well for smallish numbers - e.g. low thousands of entries is not aproblem
12:09 <+petermr> but can access any of the indexed fields
12:09 <+Rifleman_82> we have 3000 and change of chemicals in wp, not counting drugbox
12:10 <+petermr> so goal is to create WPChem in RDF. Example of single compound is shown on Wikipage
12:10 <+petermr> OK - I will hack drug box now I know it works
12:10 <+petermr> each compound is a node in RDF graph and has names, properties, formula, etc hanging off it
12:11 <+dmacks> (which wikipage?)
12:11 <+petermr> created by: XML=> box => CML => RDF
12:11 * petermr goes to look
12:12 <+petermr> on [[1]]
12:12 <+petermr> there is a link to my blog entry which has a RDF graph of a WP entry
12:13 <+petermr> has ca. 20 nodes
12:13 <+petermr> We have slightly modified the structure since
12:13 <+petermr> each arc is a URI in a CML dictionary entry
12:14 <+petermr> so all properties are semantic
12:14 <+ChemSpiderMan> Peter...I don't see any graph on that page
12:14 <+petermr> of course the dictionary entries could be derived from WP and this is the best way where they exist
12:14 <+ChemSpiderMan> This is the Microsft eChistry page?
12:15 <+ChemSpiderMan> excuse typos
12:15 <+dmacks> Ah found it: http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=890
12:15 <+petermr> The graph is a topological graph
12:15 <+ChemSpiderMan> thx
12:15 <+petermr> it consists of nodes linked by arcs
12:16 <+petermr> each arc is subject->predicate->object
12:16 <+petermr> water->hasBoilingPoint->373
12:16 <+petermr> and hasBolingPoint->hasUnits->K
12:17 <+petermr> or something similar
12:17 <+petermr> in that way all the information can be recorded
12:17 <@Beetstra> Pooh .. yes, that is also a way ..
12:17 <+petermr> RDF is now accepted as the approparite way to manage Web 2.0 things
12:17 <+Rifleman_82> what's CML? is it an acceptable format ?
12:18 <+petermr> It's accepted by the ACS, RoyalSocChemistry, NIH
12:18 <@Beetstra> So we need a BoilingPt and a BoilingPtUnit .. not like what it is now in e.g. Tributylphosphine (just changed it to BoilingPtK)
12:18 <+petermr> It's emitted by ChemDraw , ACD
12:18 <+petermr> see [Chemical_Markup_Language]]
12:19 <+petermr> something like that
12:19 <+ChemSpiderMan> ACD/labs CML suppotr has never been validated
12:19 <+ChemSpiderMan> not sure about CHemDraw
12:19 <+dmacks> Beetstra: From a maintainability POV, having the data "actually" entered in C seems preferable?
12:20 <+petermr> let's not get into the validation discussion. No formats are validated
12:20 <+Rifleman_82> so imperial college is handling the development of this format?
12:20 <@Beetstra> If we do that, dmacks, then we don't need a unit field ..
12:20 <+petermr> it's not critical at this stage.
12:21 <@Beetstra> i programmed all three (C, K and F) for ease of conversion, all three were used
12:21 <+dmacks> Ah, gotcha Beetstra
12:21 <+petermr> there are complex properties that require structure such as BPt at given Pressure
12:21 * petermr asks walkerma how we are for time
12:21 * Beetstra starts to think about going home ..
12:21 <+Rifleman_82> we might's well do with one... much more convenient; we can ignore bpt at reduced pressure? just quote atmospheric
12:22 <+petermr> many compounds decompose before atmospheric
12:22 <@Beetstra> Some boiling points are simply not known at atmospheric .. or may not exist
12:22 <+walkerma> I think we're OK, if people can hang on. I'd like to ask Petermr what he needs us to do
12:22 <+dmacks> (/me okay with going longer)
12:22 <+walkerma> I'd also like him to say what he plans to do with the data collection
12:22 * ChemSpiderMan okay with time
12:22 <+petermr> OK. First I am clear that the boxes are being cleaned up.
12:23 <+petermr> that's very valuable.
12:23 <+walkerma> But let's try and finish fairly soon
12:23 <+petermr> it would be useful to have agreement on how information can be extracted to and written from boxes automatically
12:23 * Rifleman_82 notices that it's 1+ am here
12:24 <+petermr> then we can write software that is maintainable
12:24 <@Beetstra> Petermr, you need a bot-account for that ..
12:24 <+Rifleman_82> beetstra's your man for awb scripts... he can hand it to me and i can run it on chem-awb
12:24 <+petermr> this also allows machine validation of consistency
12:24 <+Rifleman_82> or anyone else can get a bot account, it's not a big deal
12:24 <@Beetstra> AWB is limited .. you need 'real' programming languages for this
12:24 <+petermr> what is AWB
12:24 <+walkerma> http://en.wikipedia.org/wiki/Wikipedia:AutoWikiBrowser
12:24 <+Rifleman_82> WP:AWB
12:25 <+petermr> I agree we need real languages. I have a fairly good parser for correct chembox
12:25 <@Beetstra> AWB is autowikibrowser, script editor, but mainly for find-and-replace .. which you can make as complex as you want
12:25 <@Beetstra> using regex ..
12:25 <+Rifleman_82> okay, then the next step will be fore petermr to get a bot account. you can see WP:BAG
12:25 <+petermr> Next.. character encodings - can we find a way of using only ASCII. Not easy, but characters really foul up machines
12:26 <@Beetstra> But if you need data from external databases it is better to write it properly in perl/python/c/etc
12:26 <+Rifleman_82> what sort of characters screw things up?
12:26 <+Rifleman_82> · °ree; ?
12:26 <+petermr> At present I am working with what I have already got and will return if I need to work regularly with a bot
12:26 <+dmacks> Weak objection petermr: WP is primarily a display engine, so it needs to display well, then it's up to our parsers to read that.
12:26 <+petermr> yes, those can be problems. It's worse when people paste in higher encodings from word
12:27 <+petermr> this is the classic differentce between semantics and display. Maybe both are required. Maybe a adatbase offline which is transformed
12:27 <+dmacks> "It's all just data", so easy enough to regex replace them for internal processing in [whatever you're doing].
12:27 <+petermr> it will also make your display more consistent anyway
12:28 <+petermr> then I wanted to apply for the grant and will simply say something like:
12:29 <+petermr> I have discussed the use of WP-Chem in RDF with the WPedians and there are aware of what we plan to do.
12:29 <+petermr> (You may later add positive words if you feel this merits it)
12:30 <+petermr> the main thrust to to create a chemical ontology which supports publications such as theses and papers
12:30 <+Rifleman_82> to make wp the definitive source for chemical information?
12:30 <+walkerma> So chem articles will link to our pages?
12:30 <+petermr> I have also mentioned the value of RDF and alerted you to the likelihood that I will be using a version of WPChem
12:31 <+petermr> Yes to both of you
12:31 <+Rifleman_82> no need to link... can search?
12:31 <+Rifleman_82> or you're going to start assigning wikichemids? ;)
12:31 <+petermr> I have told the world that I see WP becoming the primary reference source for (elementary) chemistry
12:31 <+walkerma> Rifleman82: You could be reading a Chem Communications article, see a blue link, and just click
12:32 <+petermr> At present I assume that each entry has a unique ID in the XML. But that doesn't allow linking to article
12:32 <+petermr> (I ignore version IDs at this stage)
12:32 <+petermr> is there a wikichemid?
12:33 <+petermr> as opposed to a page ID? I think it would be useful to have wikichemID
12:33 <@Beetstra> IIRC, each page has an ID, which is linked to a pagename, a page-move is merely a change of the name in the name-database ..
12:33 <+Rifleman_82> haha... i think that'd be redundant...
12:33 <+Rifleman_82> i was comparing a wikichemid with pubchemid
12:33 <+Rifleman_82> and cas no and einecs and a whole bunch of other arbitrary identifiers
12:33 <@Beetstra> WikiChemID would be difficult to control, 'there can be only one' ..
12:33 <+Rifleman_82> we can just use cas numbers
12:33 <+petermr> yes - a wikichemid would be similar to the use of cas and enices
12:34 <+Rifleman_82> and redirect where necessary - D and L can redirect to the correct article
12:34 <+petermr> You can't use CAS where it doesn't exist. I don't know where that might be.
12:34 <+ChemSpiderMan> but at least open to anyone to review
12:34 <+walkerma> We can't use CAS, they're too ambiguous too
12:34 <+ChemSpiderMan> absolutely agreed
12:34 <+Rifleman_82> what's the problemw ith cas?
12:34 <+dmacks> Do we really need Yet Another ID? vs canonical-SMILES, or InChi or somesuch?
12:34 <+Rifleman_82> i think they'll work for 90 % or so...
12:35 <+Rifleman_82> problems would be for ... raney nickel which isn't well-defined...?
12:35 <+walkerma> InChIKey?
12:35 <+petermr> I see a distinction between WP on the one hand (fairly small, common chemicals) and ChemSpider or Pubchem - very large
12:35 <+petermr> WP Chem covers a differeht constiuency
12:35 <+ChemSpiderMan> InCHiKey will have problems for many chemicals - orgamometallics, polymers, inorganics
12:35 <+petermr> InChI will not work for large numbers of non-organic compounds
12:36 <+walkerma> True
12:36 <+dmacks> good point
12:36 <+petermr> There has to be an authority-based Id for named substances
12:36 <+ChemSpiderMan> agreed
12:36 <+petermr> what is InChI for gasoline?
12:36 <+Rifleman_82> it's not a chemical compound
12:36 <+Rifleman_82> :P
12:36 <+petermr> but I think there is a CAS number
12:36 <+ChemSpiderMan> ABout $3.20 :-)
12:36 <+Rifleman_82> so it's not a chemical, no chembox
12:37 <+petermr> OK, what is INcHI for glucose?
12:37 <+ChemSpiderMan> what form?
12:37 <+Rifleman_82> look at it this way, all chemicals will have a cas, if there's a cas there need not be a chemical
12:37 <+Rifleman_82> but that's good enough for us?
12:37 <+dmacks> So this database will include materials and mixtures (performance/industrial chemicals), not just fine chemicals and pure substances?
12:37 <+Rifleman_82> and wikipedia can redirect all the various cas numbers to the glucose article?
12:37 <+petermr> The only problem with CAS is ownership and independence. It's a good system
12:38 <+Rifleman_82> heh... can persuade ACS to license CAS under creative commons or something? :)
12:38 <+ChemSpiderMan> we come ack to CAS...who's going to get the "right one"?
12:38 <+Rifleman_82> "right one"?
12:38 <+ChemSpiderMan> the right CAS number for the article
12:38 <+walkerma> I think we'll need to discuss this at length another time. I think there are issues with all of these formats
12:38 <+ChemSpiderMan> the material, compound, structure, mixture
12:38 <+petermr> there is no "right one" - but this is a long discussion. Different people have clear disagreements about chemicals
12:38 <+ChemSpiderMan> exactly.
12:39 <+ChemSpiderMan> CAS won't work
12:39 <+petermr> exactly. Our RDF work will highlight exactly what agreement and what disagreement there is between WP and other common sources such as MSDS
12:39 <+Rifleman_82> how about using the names?
12:39 <+ChemSpiderMan> I agree that a WikipediaID may be of value but why not the name
12:39 <+Rifleman_82> all common variants should redirect to the correct article
12:39 <+petermr> names are just as bad. What is the formula for "snow"
12:39 <+ChemSpiderMan> there isn't one
12:40 <+ChemSpiderMan> but is there an article?
12:40 * Beetstra is leaving .. I'll be here on other times when people want to discuss things
12:40 <@Beetstra> bye!
12:40 <+Rifleman_82> night dirk!
12:40 <+ChemSpiderMan> g'night
12:40 -!- Beetstra [n=djbeetst@Wikimedia/Beetstra] has quit ["Bye Bye"]
12:40 <+dmacks> Problem here seems to be trying to force 1:1 mapping of "some identifier" to "some chemical"
12:40 <+ChemSpiderMan> http://en.wikipedia.org/wiki/Snow
12:40 <+petermr> names are not unique and are unlikely to be so. there is a formaul for "snow" in pubchem. It is correct and different from what you might expect.
12:40 <+Rifleman_82> oh?
12:41 <+ChemSpiderMan> cocaine
12:41 <+petermr> So the only absolute resuolution os for WPedians to agree that certain entries have certain WPids
12:41 <+ChemSpiderMan> also, berries, nose candy, pimp powder etc
12:41 <+Rifleman_82> if we need to have WPIDs, we should have them for all
12:41 <+Rifleman_82> we could be the authority?
12:41 <+petermr> RDF can support a many:many mapping and we are using that to find out how serious the problem is for - say - 5000 common compounds
12:42 <+Rifleman_82> but curation will fall on someone's or some peoples' shoulders
12:42 <+petermr> yes - WP will become the authority - whether it wants to or not!
12:42 <+petermr> any successful collection becomes an authority
12:43 <+walkerma> We are already the de facto authority for chemistry students under the age of 18.....
12:43 <+petermr> at present the only thing common to almost all common source is CAS
12:43 <+dmacks> What's wrong with having the ChemComm article link going to a list of compounds (all the different protonation forms) if the article-author wasn't specific enough in his linking?
12:43 <+petermr> exactly
12:43 * Rifleman_82 has failed students for plagiarizing WP... even the [edit] tags
12:43 <+dmacks> me too Rifleman_82
12:43 <+walkerma> haha
12:44 <+petermr> ideally the author should indicate what they are linking to - a precise name, a generic name, etc.
12:44 <+dmacks> Got one that happened to copy a paragraph that had been vandalized, didn't notice a "This is gay!" sentence in the middle of the paragraph.
12:44 <+Rifleman_82> hahaha
12:44 <+walkerma> One problem we have is that we on WP organise by article, not by chemical compound
12:45 * petermr asks whether we have covered most of it or want some more
12:45 <+Rifleman_82> i remember marking lab reports... all of them insisted that methyl salicylate was a "rubificient"... something about blocking light...!?
12:45 <+petermr> but the infobox is percompound?
12:45 <+Rifleman_82> ideally each compound has its own article
12:45 <+Rifleman_82> but look at the article on carbodiimide
12:45 <+walkerma> Consider tartaric acid: We have one page but there are many CAS nos and InChIs
12:45 <+Rifleman_82> and cresol
12:46 <+petermr> These are DIFFICULT problems withy no simple answer. maybe there should be multiple infoboxes
12:46 <+walkerma> R,R; S,S; meso, unspecified, racemic mixture, etc
12:46 <+ChemSpiderMan> I think you're back to the article name as the primary WP key
12:46 <+petermr> there is a major problem between macroscopic and microscopic.
12:46 <+dmacks> yeah, again ChemComm author might not care, but also might.
12:46 <+walkerma> Using the article name as the primary WP key would be the most natural way for Wikipedians
12:47 <+petermr> that is why CML uses substance for macro and molecule for micro
12:47 <+Rifleman_82> orgsynth used to list cas numbers beside substances
12:47 <+dmacks> So whatever system or unique id we're gonna promote needs to account for both types of things.
12:47 <+petermr> CAS numbers are designed to describe substances. They sometimes work for molecules
12:48 <+ChemSpiderMan> mol/sdf is not as capable as CML for supporting substance...it is why I have not been able to deal with many of them
12:48 <+petermr> InChIs are designed to work for molecules.
12:48 <+ChemSpiderMan> and SMILES
12:48 <+petermr> InChIs sometimes work for pure substances
12:48 <+ChemSpiderMan> they are all limited in the same way
12:48 <+petermr> SMILES <--> InChI
12:48 <+ChemSpiderMan> similar way
12:49 <+walkerma> OK, I think we have plenty to ponder. Should we call it a day/night? And perhaps we can have this discussion on wiki?
12:49 <+petermr> The Merck handbook deals with substances
12:49 <+petermr> BTW I see WP as amjor competitor to Merck
12:49 <+petermr> == major
12:49 <+dmacks> yup to both.
12:49 <+Rifleman_82> and crc handbook :)
12:50 <+walkerma> Is everyone OK with me making a transcript of this discussion public?
12:50 <+Rifleman_82> sure
12:50 * petermr waves bye to all left and thanks for discussion.
12:50 <+ChemSpiderMan> yes
12:50 <+Rifleman_82> petermr
12:50 <+petermr> yes to public
12:50 <+dmacks> walkerma: Sure.
12:50 <+Rifleman_82> before you go
12:50 <+Rifleman_82> last question
12:50 <+petermr> sure...
12:50 <+ChemSpiderMan> I need a couple of moments from anyone who's interested please
12:50 <+walkerma> OK
12:50 <+Rifleman_82> should this database stay on wiki? or should we split it off to something off-wiki
12:50 <+walkerma> Good point
12:50 <+Rifleman_82> or on wiki but off-of the article space
12:51 <+Rifleman_82> vandalism is a problem, but can be fixed
12:51 <+Rifleman_82> i'm more worried if to make it efficient the syntax will not be easily figured out for the drive-by chembox filler
12:51 <+dmacks> If on-wiki, then separate page.
12:51 <+Rifleman_82> as in, the syntax will be so complicated
12:52 <+walkerma> We need to allow drive-by users to add information, then we with bots & AWB go and format the info
12:52 <+walkerma> If needed
12:52 <+Rifleman_82> we can have a tag asking them to add it to the talk?
12:52 <+Rifleman_82> if needed of course
12:53 <+Rifleman_82> anyway, if there're no responses to my comments we can sleep on it
12:53 <+Rifleman_82> ChemSpiderMan had something to say?
12:53 <+ChemSpiderMan> yes
12:53 * dmacks not sure how much more complicated the syntax would be than cheminfobox now.
12:53 <+ChemSpiderMan> please look at: http://en.wikipedia.org/wiki/N-Cyclohexyl-2-aminoethanesulfonic acid
12:54 <+Rifleman_82> yup
12:54 <+ChemSpiderMan> click on the CAS number
12:54 <+dmacks> okay...
12:54 -!- Kelson [n=Kelson@77-57-4-193.dclient.hispeed.ch] has joined #wikichem
12:54 -!- mode/#wikichem [+v Kelson] by ChanServ
12:54 <+ChemSpiderMan> and http://en.wikipedia.org/wiki/N-Acetylmannosamine
12:54 <+Rifleman_82> sodium salt, free acid
12:55 <+ChemSpiderMan> click on the CAS number
12:55 <+ChemSpiderMan> My opinion....CAS numbers should not link out
12:55 <+ChemSpiderMan> it adds confusion when there are many "forms" of the molecule with the same CAS
12:55 -!- carl-m [n=cbm@wikipedia/cbm] has joined #wikichem
12:56 <+ChemSpiderMan> salts, with/without stereo, incomplete stereo
12:56 <+walkerma> Hi Carl, Kelson, we're just finishing up here
12:56 <+ChemSpiderMan> The CAS in the ChemBox should be for the structure drawn
12:56 <+petermr> this was raised on the talk pages earlier I think
12:56 <+walkerma> ChemSpiderMan: This is bad - they would have different CAS nos anyway
12:56 <+petermr> as to where if anywhere the CAS should link
12:57 * Rifleman_82 wonders if we can start adding IR 1H, 13C, F, P, etc. NMR, other spectra... and complement/compete with SDBS and NIST webbook
12:57 <+Kelson> walkerma: :)
12:57 <+Rifleman_82> if we're going to be the authority, we needn't bother with external links
12:57 <+ChemSpiderMan> Yes...there may be many CAS numbers for the sAME structure
12:57 <+walkerma> NMR etc: Yes we should, but not today...!
12:57 <+Rifleman_82> linkfarms and other search engines will be linking to us instead
12:57 * petermr agrees
12:57 <+dmacks> yup
12:57 * petermr about the links
12:57 <+ChemSpiderMan> But using the CSA numbers to link out to many structures is different
12:58 <+ChemSpiderMan> One structure > Many CAS but I don't think Many structure > ONE CAS
12:58 <+ChemSpiderMan> anyhow...worth a talk page discussion.
12:58 -!- petermr [n=chatzill@arcturus.ch.cam.ac.uk] has quit ["ChatZilla 0.9.80 [Firefox 2.0.0.11/2007112718]"]
12:59 <+Rifleman_82> yes... many structure -> one cas
12:59 <+walkerma> Yes, we have a lot to discuss on wiki
12:59 <+Rifleman_82> tartaric acid?
13:00 <+walkerma> exactly - tartaric acid must have at least 5
13:00 <+ChemSpiderMan> Rifleman...re spectra...they are being populated on ChemSpider. You might want to consider another ID...a ChemSpider ID in your box :-)
13:00 <+ChemSpiderMan> http://www.chemspider.com/docs/Uploading_Spectra_onto_ChemSpider.htm
13:00 <+Rifleman_82> chemspider: thanks... i've got a few spectra... from aldrich bottles
13:01 <+ChemSpiderMan> I'd welcome them...we can chat offline. u have my email?
13:01 <+ChemSpiderMan> walkerma..we're done maybe?
13:01 <+Rifleman_82> what's the copyright for these spectra? okay to upload on wiki? to discuss the finer points of NMR for example
13:01 <+Rifleman_82> ChemSpiderMan: yeah, i still have it from the chemsketch discussion a year ago
13:01 <+Rifleman_82> i'm done, we can discuss again
13:01 <+ChemSpiderMan> most are OPEN data
13:01 <+Rifleman_82> should we set a date for next week?
13:02 <+ChemSpiderMan> with me or the group? With me...sure
13:02 <+Rifleman_82> need not be next week
13:02 <+walkerma> Yes let's end it there I think. Should we meet at exactly one week after this first meeting? Rifleman, can you stay up?
13:02 <+Rifleman_82> i was thinking an open invitation for all
13:02 <+dmacks> works for me.
13:02 <+Rifleman_82> i can stay up... it's 2 am now, but i start work at 10 am
13:02 <+walkerma> I have office hours, but otherwise OK
13:02 <+Rifleman_82> so 12-2 my time is fine
13:03 <+ChemSpiderMan> am I invited back?
13:03 <+Rifleman_82> so.. why don't we meet here next tue, same time
13:03 <+walkerma> You're one of us now aren;t you, ChemSpiderMan?
13:03 <+ChemSpiderMan> If so...11am is good.
13:03 <+ChemSpiderMan> Sounds like fun...
13:03 <+Rifleman_82> no obligation, whoever wants to come can come
13:03 <+Rifleman_82> and... we can always hang out here
13:04 <+Rifleman_82> like beetstra and itub do
13:04 <+ChemSpiderMan> if not lunacy... I hhave added Chromene to ChemSpider... :-)
13:05 <+Rifleman_82> martin you might be interestedhttp://stats.grok.se/en/200712/diphosphines
13:05 <+Rifleman_82> a tool to check how often an article is viewed
13:05 <+Rifleman_82> may be helpful for your wp v 1.0 or whatever it is
13:05 * ChemSpiderMan ChemSpiderman leaving
13:05 -!- carl-m is now known as carl-m|away
13:05 -!- carl-m|away [n=cbm@wikipedia/cbm] has left #wikichem []
13:05 -!- ChemSpiderMan [n=ChemSpid@c-68-33-151-242.hsd1.md.comcast.net] has quit []
13:06 -!- You're now known as dmacks_away
13:06 <+Rifleman_82> night people
13:06 -!- Rifleman_82 [n=Rifleman@wikipedia/Rifleman-82] has quit []
13:07 <+walkerma> Thanks Rifleman! OK, I suggest to Kelson and carl-m that we relocate to #wikipedia-1.0
13:07 <+walkerma> Bye!
13:08 -!- walkerma [n=chatzill@cpe-74-71-213-87.twcny.res.rr.com] has left #wikichem []
15:47 -!- carl-m [n=cbm@wikipedia/cbm] has joined #wikichem
15:49 -!- Kelson [n=Kelson@77-57-4-193.dclient.hispeed.ch] has quit ["Client exiting"]
15:49 -!- carl-m [n=cbm@wikipedia/cbm] has left #wikichem []
15:52 -!- Physchim62 [n=Physchim@unaffiliated/physchim62] has joined #wikichem
15:52 -!- mode/#wikichem [+v Physchim62] by ChanServ
15:52 -!- Physchim62 is now known as PC62|away
15:55 -!- PC62|away is now known as Physchim62
--- Log closed Tue Jan 15 16:37:27 EST 2008