--- Day changed Fri Apr 10 2009 | ||
kanzure | hm, the OCR isn't working for the 1970s Nature papers | 00:06 |
---|---|---|
fenn | hehehe.. the-tan, scientology mascot | 01:01 |
fenn | please tell anonymous | 01:02 |
genehacker | scientology mascot? | 01:08 |
genehacker | oh i get it | 01:09 |
-!- PeerInfinity [n=someone@216.36.180.162] has quit [] | 01:37 | |
-!- genehacker [n=chatzill@wireless-128-62-96-128.public.utexas.edu] has quit [Read error: 110 (Connection timed out)] | 03:29 | |
-!- genehacker [n=chatzill@wireless-128-62-96-128.public.utexas.edu] has joined #hplusroadmap | 04:18 | |
-!- any25439579 [n=someone@75-120-4-84.dyn.centurytel.net] has joined #hplusroadmap | 04:58 | |
-!- genehacker [n=chatzill@wireless-128-62-96-128.public.utexas.edu] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"] | 05:12 | |
-!- katsmeow-afk [n=someone@75-120-34-77.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 05:16 | |
-!- any25439579 is now known as katsmeow-afk | 05:48 | |
-!- jm|space [n=jm@p57B9EA8E.dip.t-dialin.net] has joined #hplusroadmap | 06:47 | |
-!- jm [n=jm@p57B9E73D.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] | 07:04 | |
-!- any55560268 [n=someone@75-120-4-84.dyn.centurytel.net] has joined #hplusroadmap | 07:07 | |
-!- katsmeow-afk [n=someone@75-120-4-84.dyn.centurytel.net] has quit [Nick collision from services.] | 07:20 | |
-!- any55560268 is now known as katsmeow-afk | 07:21 | |
kanzure | hurray, it's now separated into somewhat reasonably sized folders | 08:34 |
-!- jm|space [n=jm@p57B9EA8E.dip.t-dialin.net] has quit [Read error: 113 (No route to host)] | 08:59 | |
* kanzure needs to hunt down "An information-bearing seed for nucleating algorithmic self-assembly" in the March 24th issue of PNAS | 09:08 | |
kanzure | oh boy. now diybio is turning into a job forum. ugh | 09:26 |
kanzure | http://news.slashdot.org/comments.pl?sid=1194963&cid=27530751 "Oh! I get it! Nice work." | 09:53 |
-!- samrose [n=samrose@adsl-99-147-182-70.dsl.lgtpmi.sbcglobal.net] has joined #hplusroadmap | 10:36 | |
bkero | kanzure: are employers trolling it, or kids who just graduated? | 11:34 |
fenn | it was just two posts that ought to have been off-list | 11:35 |
kanzure | true, but there has also been past occurences | 11:37 |
kanzure | let's all post our resumes, yay jobs for everyone | 11:37 |
kanzure | have any of you checked out mendeley yet? | 11:39 |
kanzure | I want to try out the group sharing feature | 11:39 |
fenn | i looked, didnt see anything interesting | 11:40 |
kanzure | http://heybryan.org/books/papers/microfluidics/microfluidics.20090410.bib | 11:44 |
fenn | how does it 'extract' metadata anyway? | 11:44 |
kanzure | well after having it completely fail to OCR the Nature publications, I'm thinking now that maybe it doesn't | 11:45 |
kanzure | it pays special attention to anything that has a DOI number | 11:45 |
kanzure | it completely fails (like everything else) in reading 'fl' and 'fi' next to each other in the titles in the paper | 11:46 |
kanzure | a clear sign that it's OCRing. | 11:46 |
fenn | of course it wouldn't do OCR | 11:46 |
fenn | oh, it does? | 11:46 |
* kanzure nods | 11:46 | |
fenn | uh, why | 11:46 |
kanzure | it would be amazingly useful if I could train it on datasets and draw rectangles of where a title is on a page that it doesn't know how to process.. | 11:46 |
kanzure | what? | 11:46 |
kanzure | because people just keep PDF files on their hard drives and they don't have the metadata | 11:47 |
kanzure | so it extracts it. | 11:47 |
fenn | so you download this application that runs and it uploads the metadata to the website? | 11:48 |
kanzure | yes, but it has local export options as well | 11:48 |
kanzure | it wants to synchronize paper collections between multiple computers and possibly different people ("group sharing") | 11:49 |
-!- PeerInfinity [n=someone@216.36.180.162] has joined #hplusroadmap | 12:08 | |
-!- PeerInfinity [n=someone@216.36.180.162] has quit [Client Quit] | 12:09 | |
-!- PeerInfinity [n=someone@216.36.180.162] has joined #hplusroadmap | 12:09 | |
fenn | uh.. no drag and drop, i can only add one document at a time? | 12:16 |
fenn | oh there it goes | 12:16 |
-!- PeerInfinity [n=someone@216.36.180.162] has quit [] | 12:18 | |
fenn | that 'do you want to save metadata?' dialog is really annoying | 12:29 |
fenn | Referencer is a GNOME application to organise documents or references, and | 12:30 |
fenn | ultimately generate a BibTeX bibliography file. | 12:31 |
fenn | oh this is handy http://www.zotero.org/support/pdf_fulltext_indexing | 12:32 |
fenn | i want a tagging metadata index with full text search | 12:35 |
fenn | not more hierarchical crap.. i can do that with a filesystem | 12:35 |
kanzure | are you "josef blousef" | 12:38 |
fenn | ja | 12:38 |
fenn | supposedly zotero does "Automatic synchronization of collections among multiple computers." | 12:39 |
fenn | so why is mendeley interesting? | 12:39 |
fenn | why not just make a torrent with all the individual pdf's | 12:40 |
fenn | people can pick and choose what they want, or get the whole thing | 12:40 |
kanzure | mostly for the automatic metadata extraction even if it does somewhat suck | 12:40 |
kanzure | sucks less than tesseract | 12:40 |
fenn | can't you just rename the files with the paper title? | 12:40 |
kanzure | I do | 12:41 |
fenn | it doesn't do OCR btw | 12:41 |
kanzure | it's also very painfully slow though to switch between reading a paper to figure out authors, publication, etc., to the file rename utility (whatever I happen to be using) | 12:41 |
fenn | i sniffed the binary, it doesnt have any ocr code | 12:41 |
fenn | well, you wouldnt do it by hand | 12:41 |
fenn | does bibtex include the filename? | 12:42 |
kanzure | at least zotero grabs metadata at point of download :) | 12:42 |
kanzure | no | 12:42 |
kanzure | erm, wait | 12:42 |
kanzure | phone | 12:44 |
fenn | the problem with these things is they don't play well with existing organization schemes | 12:45 |
fenn | so every time you switch metatada managers you have to basically start over | 12:46 |
kanzure | wtf is wrong with this woman | 12:47 |
kanzure | she calls me because she's throwing shit out of my room back at the death-house | 12:47 |
kanzure | and she can't be bothered to do a full inventory? jebus | 12:47 |
kanzure | anywho, the idea with zotero is that at least you can capture metadata at the time of download | 12:48 |
fenn | your stuff doesn't matter because you're dead anyway, right? | 12:48 |
kanzure | I would like to implement something with zotero such that the downloaded PDFs are put into a tar file next to a dot bib or metadata thingy. | 12:48 |
kanzure | oh right | 12:48 |
kanzure | but srsly. she scares me. | 12:48 |
kanzure | is it reasonable to expect to be able to randomly call somebody and then go through their entire room asking them what can be thrown away? | 12:49 |
fenn | uh, no comment | 12:49 |
fenn | 'at time of download' means there's some microformatting on the download page? | 12:50 |
kanzure | well, it looks like zotero has some custom download templates for scraping certain publisher websites | 12:50 |
fenn | blah | 12:50 |
kanzure | yeah :/ | 12:50 |
kanzure | but zotero might have that critical mass variable for it | 12:50 |
kanzure | i.e., with enough people using it, there's enough scraper-templates already implemented, or something | 12:50 |
fenn | the great thing about standards is that nobody uses them | 12:51 |
kanzure | hm, a large grasshopper just got on my window, 12 floors up | 12:51 |
fenn | they can fly | 12:51 |
fenn | plagues of locusts | 12:52 |
kanzure | does their flying ability automagically scale with body size? | 12:52 |
fenn | no | 12:53 |
kanzure | I think I invited you to a group in mendeley | 13:12 |
kanzure | http://heybryan.org/books/papers/microfluidics.20090410.bibliography.html | 13:24 |
fenn | oh, i'm already over it :\ | 13:25 |
fenn | i have enough projects to work on without reading obscure science papers | 13:26 |
-!- samrose [n=samrose@adsl-99-147-182-70.dsl.lgtpmi.sbcglobal.net] has quit [Remote closed the connection] | 13:27 | |
kanzure | heh | 13:33 |
* kanzure waddles off to turn in paperwork by hand, because he's living in the stone age | 13:34 | |
-!- fenn_ [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has joined #hplusroadmap | 13:57 | |
-!- fenn [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has quit [Nick collision from services.] | 14:01 | |
-!- fenn [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has joined #hplusroadmap | 14:02 | |
-!- fenn__ [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has joined #hplusroadmap | 14:02 | |
-!- fenn__ [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has quit [Client Quit] | 14:04 | |
-!- fenn_ [n=fenn@cpe-72-177-52-180.austin.res.rr.com] has quit [Read error: 110 (Connection timed out)] | 14:14 | |
-!- cis-action_ [n=cis-acti@c-66-31-201-117.hsd1.ma.comcast.net] has joined #hplusroadmap | 15:01 | |
-!- any65752339 [n=someone@99.194.249.112] has joined #hplusroadmap | 15:44 | |
-!- katsmeow-afk [n=someone@75-120-4-84.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 16:03 | |
bkero | waddler | 16:03 |
-!- xp_prg [n=root@c-24-130-14-113.hsd1.ca.comcast.net] has joined #hplusroadmap | 16:14 | |
-!- wrldpc [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has joined #hplusroadmap | 16:19 | |
-!- cis-action [n=cis-acti@c-66-31-201-117.hsd1.ma.comcast.net] has joined #hplusroadmap | 16:33 | |
-!- cis-action_ [n=cis-acti@c-66-31-201-117.hsd1.ma.comcast.net] has quit [Read error: 104 (Connection reset by peer)] | 16:33 | |
-!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has joined #hplusroadmap | 16:39 | |
-!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has quit [Remote closed the connection] | 17:06 | |
kanzure | you take that back | 17:12 |
bkero | But I have evidence! | 17:12 |
kanzure | wtf mac, wtf | 17:14 |
kanzure | now when you want to talk to diy people you have to email nine different lists? | 17:14 |
kanzure | diybio-nyc, diybio, diybio-boston, diybio--seattle, diybio-sf, diybio-announce, diybio-chicago | 17:14 |
* fenn wonders if that was directed at cis-action or not | 17:15 | |
cis-action | oh | 17:15 |
cis-action | was that a faux pas | 17:16 |
cis-action | maybe I should have just used the main diybio list. | 17:16 |
cis-action | I got a little excited, I admit it. | 17:16 |
-!- wrldpc_ [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has joined #hplusroadmap | 17:17 | |
kanzure | well, if you want, I'll send all future emails to all mailing lists | 17:17 |
kanzure | it's not a big deal for me, but most people hate crossposting | 17:17 |
kanzure | I also don't know why anyone has taken up interest in the automatic safety protocol generator/instruction work re: MSDS and skdb | 17:17 |
kanzure | but that's another rant :-p | 17:17 |
fenn | i've been thinking about it | 17:18 |
cis-action | hey, I'm interested | 17:18 |
kanzure | cis-action: do you remember it? | 17:18 |
kanzure | fenn: but you don't count because nobody listens to you (or me) | 17:18 |
fenn | well i havent really talked about it either | 17:18 |
cis-action | hmm, what I remember is wanting something like the cc licensing wizard to give you a safety badge for your digital lab notebook / website by asking a series of safety questions | 17:18 |
cis-action | obviously there would be liability issues there. | 17:19 |
kanzure | cis-action: that's one idea, but here's specifically what the emails were about | 17:19 |
kanzure | in particular, given diybio-certified "protocols" like pcr.xml, safety instructions could be automatically generated from a standard knowledge base of safety | 17:19 |
kanzure | even up to and including things like community checklists or something for various get-together occassions. I don't know how people would want to work that out | 17:19 |
fenn | "safety" in gen-engineering goes beyond actual safety precautions | 17:20 |
kanzure | but the big deal is getting people to help work on protocol encoding in XML format | 17:20 |
fenn | it's more like "containment" | 17:20 |
kanzure | that's true | 17:20 |
kanzure | a basic series of questions (FAQ) isn't enough IMHO- it's a good start but we already have that | 17:20 |
kanzure | it should be tailored to the experiment/instructable/thingy | 17:20 |
cis-action | Personally, I like it, but I'm afraid it might be conceptual overhead when what we need right now is something simpler | 17:20 |
fenn | in my mind the only way to contain bacteria is with BSL ratings | 17:20 |
kanzure | what overhead? | 17:21 |
cis-action | If you think this would be a productive and fruitful avenue for helping address the safety concerns outsiders have with garage labs, please make it happen. | 17:21 |
kanzure | well, I already began to do it but nobody really seemed interested in helping as I said | 17:21 |
kanzure | I spent a few hours on pcr.xml and nobody corrected my obvious errors | 17:21 |
fenn | you can't have a safety wizard unless you already have a protocol wizard in place | 17:22 |
* kanzure nods | 17:22 | |
kanzure | wizards are easy to write- it's really not that much overhead- the only bottleneck is in encoding the information and knowledge | 17:22 |
kanzure | maybe you could hook us up with somebody with some time and a good knowledge of protocols? | 17:22 |
kanzure | and a good knowledge of XML. | 17:22 |
kanzure | I mean, since it didn't really work when I posted to diybio | 17:22 |
fenn | doesn't have to be xml | 17:22 |
cis-action | It could be more general-purpose. NIH guidelines don't look at protocol-level of detail to classify work as bl1, 2, etc | 17:22 |
kanzure | computer-parseable is a mouthful | 17:23 |
fenn | formalized | 17:23 |
kanzure | but then they will think of beauracracy | 17:23 |
fenn | it doesnt even need to be computer parseable is the point | 17:23 |
kanzure | hm? | 17:23 |
fenn | massaging well structured text into XML-ish format is not difficult | 17:24 |
kanzure | that's parseable | 17:24 |
fenn | it can be delegated to a volunteer | 17:24 |
cis-action | half of the value comes from asking people to think about the safety concerns for themselves. If it's all automated, people might actually become less safe because they use less common sense | 17:24 |
fenn | people have no common sense | 17:24 |
cis-action | but I have to go now. I would recommend looking at the contributors to protocols on OWW and finding someone who has put a lot of time into basic protocols | 17:24 |
cis-action | bye | 17:24 |
fenn | ciao | 17:24 |
kanzure | :-/ | 17:24 |
fenn | oh nice, OWW actually has protocols | 17:25 |
kanzure | yes | 17:25 |
kanzure | in the typical overview-materials-procedure-notes-references format | 17:26 |
kanzure | hm, you know, if I was a masochist | 17:26 |
fenn | you'd massage that structured text into an xml-ish format? | 17:26 |
kanzure | I might be inclined to believe that steps in a procedure could be parsed for physical units | 17:26 |
fenn | uh, of course they can | 17:26 |
fenn | that's the whole point | 17:26 |
kanzure | but but it's never quite clear the relation between the physical units and the other words | 17:26 |
fenn | "electroporate - 12.5 kV/cm, pulse number of 10, puse interval of 500 ms, plasmid DNA concentration of 25 ng/ul" | 17:27 |
fenn | these are basically arguments to a function | 17:27 |
kanzure | "2 volumes EtOH" <- what the *fuck* | 17:27 |
kanzure | http://openwetware.org/wiki/DNA_Precipitation | 17:27 |
fenn | that's badly written from any perspective | 17:27 |
kanzure | oh, you know what would be even better | 17:28 |
kanzure | hrm, okay I see now | 17:28 |
kanzure | we will have to define function signatures | 17:28 |
fenn | yeah i think so | 17:28 |
kanzure | and then steps in a procedure formatted in a specific way will be parsed so that some parts are arguments and the other is the function name | 17:28 |
fenn | add A to B | 17:29 |
fenn | mix(x) | 17:29 |
fenn | same crap different domain | 17:29 |
kanzure | well in particular I'm thinking of those tagging interfaces where they do character-by-character suggestion | 17:29 |
kanzure | thus our 'protocol wizard' for parsing this shitty dataset | 17:29 |
kanzure | "did you mean ____ ?" | 17:29 |
fenn | protocol creation wizard | 17:29 |
* kanzure nods | 17:30 | |
-!- wrldpc [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has quit [Read error: 110 (Connection timed out)] | 17:30 | |
fenn | i think there should also be another one for following the protocol | 17:30 |
kanzure | the reason I haven't written one is because I don't know enough about protocols | 17:30 |
kanzure | I've read many of them, but that doesn't help me to figure out exactly how to write a good creation wizard | 17:30 |
fenn | excuses excuses | 17:30 |
kanzure | yeah I agree | 17:30 |
kanzure | "did you mean to add water to acid, not acid to water? <paperclip>" | 17:30 |
fenn | i dont know the answer because i havent solved the problem yet | 17:30 |
fenn | hmm | 17:31 |
fenn | other way around | 17:31 |
kanzure | the paperclip is the mischevious symbolifier | 17:31 |
kanzure | so why isn't there a general "create an instance of an XML file from this given DTD" wizard app interface thingy? or anything for "creating an instantation of something that would conform to this given BNF grammar" | 17:32 |
fenn | because you dont know the search terms | 17:32 |
kanzure | "AutoTag makes it easy for non-technical people to create XPath select statements " | 17:33 |
kanzure | isn't this what code generation is about? | 17:33 |
kanzure | although that's mostly translation. | 17:33 |
kanzure | 17:36 < sproingie> sure, just run the wizard wizard wizard and you'll get a wizard wizard for your wizards | 17:37 |
kanzure | why did I bother with #python? surely I knew I was going to get a zen answer | 17:38 |
bkero | kanzure: I put a wizard in your wizard so I can easy while you're easing. | 17:39 |
kanzure | you know, if I go ask the perlmonks, they'll give it to me in a single line | 17:39 |
* kanzure still likes txt2regex, a nice wizard. but for regexps. | 17:41 | |
kanzure | fenn: so how about this? a few lists would be made: physical quantities, equipment, materials; you just type in the procedures, or some preliminary notes; at the end, the wizard will ask you whether or not there is conservation of some material (in terms of stuff that you have to deal with), etc. | 17:42 |
fenn | ok so basically you need to specify a grammar and then prompt the user for valid choices from that grammar | 17:43 |
kanzure | right | 17:43 |
fenn | conservation? | 17:43 |
fenn | i notice protocols leave out a lot of information | 17:43 |
kanzure | yes | 17:43 |
fenn | like "add this TO a 50mL beaker" | 17:43 |
-!- cis-action [n=cis-acti@c-66-31-201-117.hsd1.ma.comcast.net] has quit [Read error: 110 (Connection timed out)] | 17:44 | |
kanzure | and then at the end you have no more of "this" | 17:44 |
fenn | without that info your wizard/converter will have to use common sense, which has historically been difficult to program | 17:44 |
kanzure | bleh. | 17:44 |
fenn | no you still have it, it's in the beaker | 17:44 |
kanzure | hrm | 17:44 |
kanzure | so it's hard to distinguish between empty beakers and bottles too sometimes | 17:45 |
kanzure | I heart perl: http://search.cpan.org/search?query=wizard&mode=all | 17:45 |
fenn | if you're doing automatic safety checking, you need to have a chemical reaction simulation | 17:45 |
fenn | right? | 17:45 |
kanzure | I think safety starts with cleanliness | 17:45 |
-!- any65752339 is now known as katsmeow-afk | 17:45 | |
kanzure | but chemical reaction checking should be implemented | 17:45 |
kanzure | (perhaps as a future thingy to implement as a plugin) | 17:46 |
fenn | no xml creation wizard tho | 17:46 |
fenn | ah well, i'm sort of over DTD's anyway | 17:47 |
kanzure | http://search.cpan.org/~jwied/Wizard-0.1006/ | 17:47 |
kanzure | Text, TextArea | 17:47 |
kanzure | gah, those are defined by the HTML DTD | 17:47 |
kanzure | why isn't this abstracted | 17:47 |
fenn | what does Wizard do? | 17:47 |
kanzure | I don't know, but they do stuff like: "use Wizard(); my $wiz = Wizard->new(%attr); my $form = $wiz->Form(%attr); $wiz->Run($form);" | 17:48 |
kanzure | I'm not sure what benefit this adds.. | 17:48 |
fenn | # Wizard - A Perl package for implementing system administration | 17:48 |
fenn | # applications in the style of Windows wizards. | 17:48 |
kanzure | there's also a shell version though | 17:48 |
kanzure | http://search.cpan.org/~jwied/Wizard-0.1006/lib/Wizard/Shell.pm | 17:48 |
kanzure | oh look | 17:49 |
kanzure | http://search.cpan.org/~oesterhol/Term-Screen-Wizard-0.56/Wizard.pm | 17:49 |
kanzure | "PROMPT=>VARNAME, LEN=16 chars; ONLYVALID=>regexp here" | 17:49 |
kanzure | that's close. | 17:49 |
kanzure | ooh, what if onlyvalid can be sent to a function? | 17:50 |
kanzure | that would be really nice. | 17:50 |
kanzure | http://search.cpan.org/~zim/Tk-XML-WizardMaker-0.91/WizardMaker.pm "Tk::XML::WizardMaker - easy way to build the Software Assistants and Installation Wizards based on XML description." | 17:52 |
kanzure | hm, not quite. | 17:53 |
kanzure | http://pypi.python.org/pypi?%3Aaction=search&term=wizard&submit=search | 17:55 |
-!- nsh [n=nsh@host217-43-230-27.range217-43.btcentralplus.com] has joined #hplusroadmap | 17:55 | |
kanzure | nsh: know of any wizard wizards? | 17:56 |
kanzure | hm, yep, Term::Screen::Wizard allows validator functions | 17:58 |
kanzure | hm, that grasshopper is still there | 17:59 |
kanzure | I'm not sure what to ask the user | 18:08 |
kanzure | do we care about where to put stuff? | 18:08 |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap | 18:21 | |
-!- wrldpc [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has joined #hplusroadmap | 18:40 | |
-!- wrldpc_ [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has quit [Read error: 110 (Connection timed out)] | 18:45 | |
kanzure | http://www3.interscience.wiley.com/browse/?type=CURRENT_PROTOCOL | 19:53 |
kanzure | this is interesting: http://www.protocol-online.org/latest_breakthroughs.html | 19:54 |
kanzure | I have some of those papers | 19:54 |
kanzure | cis-action: you around? | 19:59 |
kanzure | cis-action: I was wondering if you could tell me what EXACT is | 19:59 |
cis-action | yes | 19:59 |
kanzure | http://groups.google.com/group/diybio/msg/e9acc67ca97312da | 20:00 |
cis-action | More context please. | 20:00 |
kanzure | where did EXACT come from? it's not exactly easy to search for | 20:00 |
kanzure | that link. | 20:00 |
kanzure | hm, there's this I guess: http://bioinformatics.oxfordjournals.org/cgi/content/full/24/13/i295 | 20:02 |
kanzure | oh. http://www.aber.ac.uk/compsci/Research/bio/dss/EXACT/ | 20:03 |
cis-action | hmm ok good | 20:04 |
cis-action | I was searching for it | 20:04 |
cis-action | so it was mentioned in that bioinformatics article? I just scanned it again and missed it. | 20:05 |
kanzure | " | 20:05 |
kanzure | Examples of how parts of the Competant Cells protocol could be implemented in a programming language for validation/simulation purposes." | 20:05 |
kanzure | Larisa Soldatova just earned a new fan | 20:06 |
cis-action | Ha ha. I like the human-friendly syntax | 20:06 |
* wrldpc is looking for website with .rar text book rapidshare links .. bio, robo, w/e | 20:10 | |
kanzure | wrldpc: http://heybryan.org/docs/ check the .txt file with a list of book sharing sites | 20:11 |
wrldpc | ty | 20:11 |
kanzure | also, see http://heybryan.org/books/ since I probably have everything already anyway | 20:12 |
kanzure | I like the list of pre-conditions | 20:12 |
fenn | the haskell is the best parseable protocol i've seen so far | 20:31 |
fenn | i probably would have chosen "From" and "To" instead of "Start" and "End" | 20:32 |
kanzure | I'm not able to tell whether or not that haskell script (the first one (not Actions.hs)) generated that yeast protocol example file | 20:33 |
kanzure | I think it did, but if so, I'm not accustomed to how haskell works- I see no main() | 20:33 |
kanzure | I see a runState() actually | 20:33 |
fenn | which example file? | 20:34 |
kanzure | the yeast competent cells file | 20:34 |
kanzure | http://www.aber.ac.uk/compsci/Research/bio/dss/EXACT/yeast_Competent_cells_v1.txt | 20:34 |
fenn | no | 20:35 |
fenn | i dont know what that is | 20:35 |
kanzure | it looks like the output log of the file | 20:35 |
kanzure | see Actions.hs, there's a logging flag or somethign | 20:35 |
fenn | it would generate lines like this "Incubate " ++ name ++ " in " ++ equipment ++ " at " ++ (show temp) ++ " degC, " ++ (show rpm) ++ "rpm in order " ++ goal | 20:35 |
kanzure | instead I see: "experiment action: incubate \n object: competent cell culture flask\n equipment: shaking incubator \n rpm: 200 \n temp: 30C \n time: 30 mins" | 20:37 |
fenn | it does seem like that's in some kind of format | 20:37 |
fenn | you'd think they would say what format.. | 20:37 |
kanzure | I like the list of equipment at the end of the yeast_Competent_cells_v1.txt file | 20:38 |
kanzure | pre/post conditions are awesome too. | 20:38 |
fenn | meh | 20:39 |
kanzure | the extra \n's at the end make me wonder | 20:39 |
fenn | those should be automatically generated | 20:39 |
kanzure | right | 20:39 |
fenn | i dont want to have to go through and count the number of different beakers | 20:39 |
kanzure | (that's what I had to do anyway..) | 20:39 |
fenn | it should just count the number of beaker references | 20:39 |
kanzure | I mean, that's what it ended up turning into | 20:39 |
kanzure | erm, you should separately label all of your objects | 20:39 |
kanzure | and then declare the labels at the beginning | 20:39 |
fenn | oh pff | 20:40 |
fenn | did you learn that in C++ class? | 20:40 |
kanzure | no | 20:40 |
kanzure | you just need to distinguish between beakerA and beakerB in your protocol, is all I mean | 20:40 |
fenn | so what if some evil hacker inserts a step where you need a clean spork right in the middle of your protocol | 20:40 |
kanzure | error undefined reference? wizard failed. | 20:40 |
fenn | no, it's defined right there, myspork = Spork() | 20:41 |
kanzure | did you add the myspork object to @listofobjects? | 20:41 |
fenn | no, why would i do that | 20:41 |
kanzure | because you're *not* an evil hacker? | 20:41 |
fenn | sez who | 20:42 |
kanzure | just because you can read the wireshark man pages doesn't mean you're a leet hax0r | 20:42 |
fenn | anyway do you see what i'm getting at | 20:42 |
kanzure | no, would Spork's constructor add itself to the global list? | 20:42 |
fenn | um, maybe | 20:42 |
fenn | i think if you instantiate a physical object it should get counted somehow | 20:43 |
fenn | whether that's at creation or deletion doesn't really matter | 20:43 |
fenn | in C++ does a derived object call its parent class's constructor? | 20:45 |
kanzure | only if you don't provide your own constructor | 20:50 |
kanzure | or something | 20:50 |
kanzure | it's different with a virtual class (?) | 20:50 |
* fenn is writing a DXF parser for some odd reason | 20:54 | |
kanzure | huh, that grasshopper wasn't dead | 20:58 |
kanzure | just decided to stay put for 8 hours? | 20:59 |
kanzure | I like its style. | 21:00 |
kanzure | doesn't OCC have a DXF importer? | 21:01 |
fenn | yes | 21:04 |
kanzure | hm, I think I can do "all that" iff I am allowed to constrain the domain to volumetric-related lab protocols (i.e., moving volumes of liquid around) | 21:10 |
kanzure | what percentage or how many protocols would that give me access to? I figure it would give me at least a few | 21:10 |
kanzure | pipetting grammar heh | 21:10 |
kanzure | because volumes are easy to represent with pq | 21:12 |
kanzure | (and check/evaluate/expressions tags conditional checkies thingers) | 21:12 |
kanzure | you know, that really is a good domain constraint problem-wise I mean for skdb | 21:22 |
kanzure | what other physical quantities are you going to have, really? besides volumes. | 21:22 |
kanzure | maybe some temperatures, or something, but that's not a quantity | 21:22 |
kanzure | erm, I don't know how to say that. | 21:22 |
fenn | temperature is a physical quantity | 22:05 |
fenn | python is pissing me off | 22:06 |
fenn | i can't figure out how to do a simple pointer | 22:06 |
kanzure | why are you using a pointer? | 22:32 |
kanzure | http://lookherefirst.wordpress.com/2007/12/19/python-copy-by-reference-copy-by-value-shit-it-depends-on-what-you-copy/ | 22:32 |
kanzure | http://www.wellho.net/solutions/python-copying-an-object-copy-the-reference.html | 22:33 |
* fenn grumbles something about alligators and swamps | 22:34 | |
fenn | i dont know how to do that in this case | 22:35 |
fenn | i have a line defined by start and end points | 22:35 |
fenn | gah | 22:35 |
fenn | http://pastebin.ca/1388579 | 22:36 |
fenn | if i try to get a reference to myline.start.x i just get the value | 22:37 |
fenn | if i try to set parameters[' 10'] it just changes the entry in parameters | 22:37 |
fenn | when i want to change myline.start.x | 22:38 |
fenn | i'm sure this is super interesting | 22:39 |
fenn | i bet it would work fine if i had a set() method | 22:41 |
fenn | but that gets old after the zillionth time | 22:41 |
kanzure | fenn, did we ever put some thought into the wizard interface to help package maintainers use skdb to find compatible components? I was always thinking of something like apt-cache search, and just adding parameters as we add new ways to search the metadata, does that still fly? | 23:21 |
-!- any26898923 [n=someone@75-120-21-158.dyn.centurytel.net] has joined #hplusroadmap | 23:23 | |
fenn | dunno. brain fried. bbiab or tomorrow | 23:29 |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [] | 23:37 | |
-!- katsmeow-afk [n=someone@99.194.249.112] has quit [Read error: 110 (Connection timed out)] | 23:40 | |
-!- genehacker [n=chatzill@wireless-128-62-99-202.public.utexas.edu] has joined #hplusroadmap | 23:41 | |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap | 23:57 | |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [Client Quit] | 23:58 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!