Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-06-17

| Channels | #bioclipse index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
07:16 Gpox joined #bioclipse
07:56 jonalv joined #bioclipse
08:00 egonw joined #bioclipse
08:43 pasmo joined #bioclipse
08:51 egonw joined #bioclipse
08:56 egonw moin
08:56 zarah privet egonw
08:56 zarah egonw: You have new messages. Write '/msg zarah @messages' to read them.
08:56 egonw @msg
08:57 zarah egonw: 18 h 2 m 16 s ago, jonalv said what is the difference between the targets "2.0" and "2.0 release" and "2.1" and "2.1 release" on pelezilla?
08:57 egonw jonalv: no clue
08:57 jonalv egonw: fair enough
09:12 jonalv uhm guys how do I intall sample data in Bioclispe 1.2? :)
09:16 egonw don't remember...
09:16 egonw I think they should automatically show up as 'projects
09:16 egonw '
09:30 jonalv (bioclipse 1.2)--
09:31 jonalv it crashed :(
09:32 * jonalv @lunch
09:49 CIA-53 bioclipse: lc69 * r11378 /bioclipse2/trunk/plugins/net.bioclipse.​ds.signsig.bursi/exec/signatures-linux: Changed to a static version of the linux binary. It has been compiled on Ubuntu 9.04. The 32-bit version. Openbabel version 2.2.1 was statically built and then statically linked to the signature binary.
10:58 CIA-53 bioclipse: goglepox * r11379 /bioclipse2/trunk/plugins/net.bioclipse.c​dk.ui.sdfeditor/src/net/bioclipse/cdk/ui/​sdfeditor/handlers/ExtractMolecules.java: Copy properties when extracting a single molecule
10:58 CIA-53 bioclipse: goglepox * r11380 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui.sdfeditor/src/net/bioclipse/cdk/ui/sd​feditor/editor/SDFIndexEditorModel.java:
10:58 CIA-53 bioclipse: Added add/removePropertyKey()
10:58 CIA-53 bioclipse: For working with availableProperties
10:59 CIA-53 bioclipse: goglepox * r11381 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.ui.sdfeditor/src/net/bioclipse/cdk/ui/sdfe​ditor/business/MoleculeTableManager.java: Only save properties in avaiable properties on model
10:59 CIA-53 bioclipse: goglepox * r11382 /bioclipse2/trunk/plugins/net.bioclipse.cdk.u​i.sdfeditor/src/net/bioclipse/cdk/ui/sdfedito​r/editor/MoleculeTableContentProvider.java: Added removeColumn()
10:59 CIA-53 bioclipse: jonalv * r11383 /bioclipse2/trunk/plugins/net.bioclipse.core/sr​c/net/bioclipse/managers/MonitorContainer.java: fixed typo in log warning
10:59 CIA-53 bioclipse: goglepox * r11384 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui​.sdfeditor/src/net/bioclipse/cdk/ui/sdfeditor/ (3 files in 2 dirs): Only do t readProperties if availableProperties has not ben initialized
11:00 CIA-53 bioclipse: jonalv * r11385 /bioclipse2/trunk/plugins/net.bioclipse.co​re.tests/src/net/bioclipse/managers/tests/ (2 files): moved voidJobString() and voidJobIFile() into JavaManagerMethodTest since running things as jobs and returning the job will not be done in JavaScript. This completes the fixing of bug: #1228
11:00 zarah bug #1228 | http://tinyurl.com/mmezoe
11:19 CIA-53 bioclipse: goglepox * r11386 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui.sdfeditor/src/net/bioclipse/cdk/ui/sd​feditor/handlers/RemoveColumnHandler.java: Added RemoveColumnHandler
11:19 CIA-53 bioclipse: goglepox * r11387 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.ui.sdfeditor/src/net/bioclipse/cdk/ui/sdfe​ditor/business/MoleculeTableManager.java: Save only properties from availableProperties
11:20 CIA-53 bioclipse: goglepox * r11388 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk.ui.sdfeditor/plugin.xml:
11:20 CIA-53 bioclipse: Added Remove Column command
11:20 CIA-53 bioclipse: Solves bug #1206
11:20 zarah bug #1206 | http://tinyurl.com/llw6vr
11:22 jonalv @pz 1319
11:22 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1319
11:29 steinbeck joined #bioclipse
11:32 CIA-53 bioclipse: carl_masak * r11389 /bioclipse2/trunk/plugins/net.bioclip​se.scripting.ui/src/net/bioclipse/scr​ipting/ui/views/JsConsoleView.java: [JsConsoleView] added 'apropos' command
11:32 jonalv Gpox: I think you can mark
11:32 jonalv @pz 862
11:32 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=862
11:32 egonw @masak we should have a link from the console to the HTML in net.bioclipse.docs
11:32 jonalv Gpox: as fixed
11:33 egonw argh... no SkypeIn in NL
11:37 jonalv egonw: I am looking at
11:37 jonalv @pz 1006
11:37 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1006
11:38 jonalv egonw: how should the charge be printed and how do one get it? (maybe I could add it)
11:41 egonw the MFmanipulator toString method should take care of that
11:46 egonw jonalv: pin
11:46 egonw +g
11:46 jonalv egonw: pon+g
11:46 egonw I am pondering about unit testing features
11:47 egonw e.g. doing coverage testing of features
11:47 jonalv egonw: what featues?
11:47 egonw no, eclipse 'features' :)
11:47 egonw next up in the hierarchy after plugins
11:47 egonw those things
11:47 egonw just before product
11:48 jonalv egonw: intresting
11:48 egonw you know
11:48 jonalv egonw: how would this work?
11:48 egonw plugin-feature-product
11:48 egonw well...
11:48 egonw the feature-test (OMG) would have a suite with all tests for all the plugins it includes
11:49 egonw it would also need a way to find all plugins in a feature
11:49 egonw the Java API way
11:49 egonw then it would try to find matching test plugins
11:49 jonalv egonw: hm aha
11:49 egonw and see how well it is tested...
11:49 jonalv so some fancy reflections?
11:49 egonw does this make sense?
11:49 egonw right
11:50 jonalv egonw: I just wonder if this is well spent time?
11:50 egonw yes, it's automated project management
11:50 jonalv egonw: it does not sounds liek a trivial thing to do...
11:51 egonw it's easier then processing MDL molfiles
11:52 jonalv koey...
11:52 jonalv I like the idea
11:52 egonw what about those test suites?
11:52 egonw AllFoo...
11:52 egonw which are now in net.bioclipse.test
11:52 egonw for all stuff...
11:52 egonw that's really easy
11:52 jonalv egonw: yes what about them?
11:52 egonw does that make sense?
11:52 egonw suites for features?
11:53 jonalv egonw: I am not 100% sure about how it would work
11:53 egonw can you come down
11:53 jonalv you want to generate the class files represetning the suites somehow?
11:54 masak joined #bioclipse
11:56 egonw jonalv: no, manual hacking
11:56 egonw like now
11:56 egonw but
11:56 egonw from their we can cool stuff
11:56 egonw I have been doing that for the CDK for a long time now
11:57 CIA-53 bioclipse: goglepox * r11390 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.ui.sdfeditor/src/net/bioclipse/cdk/ui/sdfe​ditor/business/MoleculeTableManager.java:
11:57 CIA-53 bioclipse: Update the progressmonitor more often for calculateProperty()
11:57 CIA-53 bioclipse: Fixes bug #1319
11:57 zarah bug #1319 | http://tinyurl.com/lyk2ty
12:20 CIA-53 bioclipse: goglepox * r11391 /bioclipse2/trunk/plugins/net.bioclipse.​cdk.jchempaint/src/net/bioclipse/cdk/jch​empaint/editor/SWTMouseEventRelay.java:
12:20 CIA-53 bioclipse: Only block Ctrl-Click on mac
12:20 CIA-53 bioclipse: Fixes bug #1289
12:20 zarah bug #1289 | http://tinyurl.com/krtqdn
12:26 steinbeck_c joined #bioclipse
12:26 masak egonw: reading http://chem-bla-ics.blogspot.com/2​009/06/no-pdfs-really-do-suck.html with interest.
12:26 zarah masak's link is also http://tinyurl.com/nkwea7
12:27 egonw masak++
12:27 masak egonw: as much as PDFs do suck, they're really useful.
12:28 egonw oh, sure... but not in scientific communication of facts
12:28 masak even scientists want their article to look good (and the same) on all platforms.
12:28 masak that's the allure, I think.
12:29 steinbeck_c left #bioclipse
12:29 egonw masak: and that's where this scientific is wrong
12:29 egonw he should not worry about looks
12:29 egonw but about facts
12:29 egonw he's a scientist
12:29 egonw not a movie star
12:29 masak that's a very idealistic stand, I think.
12:30 egonw thank you
12:30 masak if we all concerned ourselves about the facts, science would be better off.
12:30 egonw I think so too
12:30 egonw and better science means less deaths
12:30 masak then there's reality, where money has to be fought over, and impressions matter.
12:30 egonw less war
12:30 egonw better living quality
12:30 egonw end of global warming
12:31 egonw you know...
12:32 masak egonw: I hope your dreams come true. but I also hope we won't have to compromise too much on the platform-independent WYSIWYG strictness that PDF promises.
12:33 CIA-53 bioclipse: goglepox * r11392 /bioclipse2/trunk/cdk-externals/trunk/org.​openscience.cdk.control/src/org/openscienc​e/cdk/controller/AddBondDragModule.java:
12:33 CIA-53 bioclipse: Do not create bond beetween itself when draging
12:33 CIA-53 bioclipse: Fixes bug #1325
12:33 CIA-53 bioclipse: [jchempaint-primary] r14618
12:33 zarah bug #1325 | http://tinyurl.com/me4cxm
12:34 egonw masak: I know OS/X sucks... but have you seen many WISIWYG problems with HTML+CSS ?
12:34 egonw :)
12:35 egonw Gpox++
12:36 masak egonw: the answer is not only "yes", but "where do I begin?"
12:36 egonw LAMP ?
12:36 masak HTML/CSS is simply not made to look the same everywhere at all times.
12:36 masak what does LAMP have to do with anything?
12:36 egonw Perl being your expertiece :)
12:36 masak that's a technology for server-side serving of stuff.
12:36 masak I'm talking about clinet-side display of articles.
12:37 egonw ah, your browser sucks?
12:37 egonw yes, that can happen
12:38 masak egonw: browsers don't, and fundamentally can't, agree on the interpretation of HTML and CSS.
12:38 egonw why not?
12:38 masak the Web of today is still tag soup, not XHTML.
12:38 masak (those two are two different problems, not the same one)
12:38 masak egonw: because the specs aren't clear enough. they are too complex.
12:39 egonw @bug 1275
12:39 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1275
12:41 masak egonw: the closest thing to press-perfectness with HTML/CSS that I know about is Håkon Wium Lies "Prince". but that's proprietary.
12:41 masak the rest of the world of HTML/CSS clients is in a sorry state.
12:43 masak and that's a fitting description even _before_ taking older versions of Internet Explorer, still the most deployed web browser out there, into account.
12:44 masak so, all in all, I'm not surprised researchers choose PDF if they want their material to look good.
12:44 masak doing the same in HTML is certainly possible, but maybe an order of magnitude more work.
12:56 egonw right
12:57 egonw then we are back to scientists who should not care about looks but about science instead
12:57 egonw masak: if there is enough demand, I am sure people will come up with a working HTML+CSS
12:57 masak egonw: that's Prince. it costs money.
12:59 egonw no, getting your paper published is cheap :)
12:59 masak I'm all for the semantic revolution. I hope it comes soon.
12:59 egonw good :)
12:59 masak I'm just saying I understand the need to guarantee how the article will look on paper.
12:59 masak PDF does that.
12:59 masak and it does a really good job, at that.
13:00 egonw yes, which is why others like HTML+CSS
13:00 egonw because thay like communicating facts
13:00 egonw caring about looks should be a big CV item
13:00 masak HTML+CSS is probably fine for publishing purely on the web.
13:00 egonw now I know a good question for taking job interviews :)
13:01 CIA-53 bioclipse: jonalv * r11393 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/domain/CDKMolecule.java: layout
13:01 masak I'm not sure how it would go over trying to hand a dead-tree publisher a HTML file.
13:01 egonw Firefox.print?
13:02 egonw but then you got the hamburger again
13:02 CIA-53 bioclipse: jonalv * r11394 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/domain/CDKMolecule.java:
13:02 CIA-53 bioclipse: Don't use sun.reflect.generics.reflectiveO​bjects.NotImplementedException. (In
13:02 CIA-53 bioclipse: fact don't use and sun.reflect classes at all since they are known to cause
13:02 CIA-53 bioclipse: problems on some platforms.) There is a perfectly fine Exception --
13:02 CIA-53 bioclipse: java.lang.UnsupportedOperationException that can be used instead.
13:02 egonw they .porint is the source of all evil
13:02 egonw .print() == .666()
13:03 egonw OK, let's take this twitter :)
13:03 masak :P
13:04 * jonalv just realised that egonw and masak are discussing something intresting
13:04 masak jonalv: quick summary: egonw things HTML+CSS are good enough for publishing, I'm saying they're not even close.
13:04 jonalv I would say you are right
13:04 egonw well, I particularly think PDF is evil
13:04 masak well, nuh.
13:05 jonalv pdf is really good at that though
13:05 masak it is.
13:05 egonw so is bitmaps
13:05 jonalv egonw: bitmaps does not scale
13:05 masak well, they don't scale as well.
13:05 jonalv :)
13:05 egonw so, let's just publish PNGs
13:05 masak PNGs are bitmaps.
13:05 masak egonw: please don't draw the discussion down to strawman arguments.
13:05 egonw yes, who says it wasn't
13:06 jonalv pdf is not bitmap
13:06 egonw masak: you changed the subject
13:06 jonalv pdf can contain bitmaps but that is not the same thing
13:06 egonw I said PDF does not allow proper communication
13:06 egonw HTML+CSS is much better at that
13:06 egonw you started to care about looks
13:06 egonw PNG can contain CML
13:06 masak egonw: I agree with your point about communication.
13:06 jonalv egonw: I would say neither format is doing a good jib for what we want
13:06 jonalv |job
13:07 masak egonw: therefore, that's not what we're discussing. we're discussing the looks part.
13:07 masak egonw: I'm not saying looks is 100%, but you seem to imply that it's 0%, and I don't agree.
13:07 egonw HTML+CSS looks are good enough for scientific publishing
13:07 egonw it's plain text!
13:07 egonw all the rest is bitmaps anyway
13:07 masak egonw: also, I'm saying that other researchers who also don't agree will be more successful because they care about looks.
13:08 masak egonw: that might be so, but PDF already does the looks part really well. HTML+CSS will be a step back.
13:08 masak egonw: one that not everyone will be willing to take.
13:08 egonw which is why people use Word and not Tex?
13:09 jonalv egonw: you will not get me to leave Tex. I don't want what I ahve spent blood and tears on to look like crap
13:09 masak egonw: partly that, and partly inertia.
13:09 steinbeck joined #bioclipse
13:09 masak also, TeX does have a much higher learning curve than does Word.
13:10 masak it's a tool for the curious intelligentia.
13:10 jonalv is it just me that thinks that the pdf file should not be confused for data?
13:11 jonalv I think both are needed
13:11 jonalv the pdf file is something that is meant to be read on paper
13:11 jonalv I think we are always going to need that
13:11 masak jonalv: ack. a good first step would be (in the field of bioinformatics, at least) to give an URL to a git repo with all your research scripts and input data.
13:11 jonalv It's not so much a case of either this or that but that two things need to complement eachother
13:11 masak then we would already be halfway towards a semantic worls.
13:12 masak s/worls/world/
13:14 egonw no, git != semantics
13:15 masak egonw: true.
13:15 masak but it's a way to enable other researchers to duplicate the research.
13:18 jonalv egonw: is cdk10 used somewhere these days?
13:19 CIA-53 bioclipse: jonalv * r11395 /bioclipse2/trunk/outdated/net.biocli​pse.cdk10.jchempaint/src/net/bioclips​e/cdk10/business/CDK10Molecule.java:
13:19 CIA-53 bioclipse: Don't use sun.reflect.generics.reflectiveO​bjects.NotImplementedException. (In
13:19 CIA-53 bioclipse: fact don't use and sun.reflect classes at all since they are known to cause
13:19 CIA-53 bioclipse: problems on some platforms.) There is a perfectly fine Exception --
13:19 CIA-53 bioclipse: java.lang.UnsupportedOperationException that can be used instead.
13:20 egonw jonalv: not sure... still awaiting answer from miguel and stefan on that
13:20 CIA-53 bioclipse: jonalv * r11396 /bioclipse2/trunk/outdated/net.biocli​pse.cdk10.jchempaint/src/net/bioclips​e/cdk10/business/CDK10Molecule.java: quick and dirty fix to keep CDK10Molecule compiling
13:20 * jonalv makes some quick fixes to keep it compiling... :)
13:22 CIA-53 bioclipse: jonalv * r11397 /bioclipse2/trunk/plugins/net.bioclipse.core/​src/net/bioclipse/core/domain/IMolecule.java: Added javadoc for enums in Property
13:33 CIA-53 bioclipse: jonalv * r11398 /bioclipse2/trunk/outdated/net.bioclipse.cdk10​.jchempaint/src/net/bioclipse/cdk10/business/ (CDK10Manager.java CDK10Reaction.java): quick fixes to keep cdk10 compiling
13:58 Gpox joined #bioclipse
14:02 jonalv egonw: do you know something that needs to be done?
14:15 egonw look at my bug list
14:16 * jonalv is looking at all bugs...
14:16 jonalv :)
14:16 jonalv I just wondered if you had something special to recommend...
14:17 CIA-53 bioclipse: carl_masak * r11399 /bioclipse2/trunk/ (7 files in 2 dirs):
14:17 CIA-53 bioclipse: [net.bioclipse.ui] added 16/32/48-px win icons
14:17 CIA-53 bioclipse: Chose to not use the .ico format, since it was difficult to produce and
14:17 CIA-53 bioclipse: edit on my platform (even though Gimp reads it fine). Instead went for six
14:17 CIA-53 bioclipse: .bmp images. Had to put them directly in the plugin directory, because
14:17 CIA-53 bioclipse: Eclipse refused to acknowledge they existed if I put them in icons/
14:23 egonw masak: sorry, but can you give me one more time the link to the excellent git write up?
14:24 egonw cannot find it in my bookmarsks
14:24 egonw egonw--
14:24 egonw for bad bookmark tagging
14:24 masak egonw: this one? http://www.newartisans.com/2008​/04/git-from-the-bottom-up.html
14:24 zarah masak's link is also http://tinyurl.com/cf6fjv
14:24 egonw yes
14:24 egonw ! thanx
14:24 masak or this one? http://tomayko.com/writings/the-thing-about-git
14:24 zarah masak's link is also http://tinyurl.com/69xpm2
14:24 masak or this one? :) http://www-cs-students.sta​nford.edu/~blynn/gitmagic/
14:24 zarah masak's link is also http://tinyurl.com/2oq2ur
14:24 egonw the bottoms up one
14:25 masak they're all very good, and I recommend them often to a lot of people.
14:25 masak but yes, I think I recomment the first one most often.
14:25 steinbeck joined #bioclipse
14:25 egonw http://www.newartisans.com/2008​/04/git-from-the-bottom-up.html
14:25 zarah egonw's link is also http://tinyurl.com/cf6fjv
14:47 masak egonw: your blog front page does not validate.
14:47 masak it claims to be XHTML 1.0 Strict, which is a pretty wild claim.
14:47 masak it contains 750 validation errors.
14:48 masak being a Semantic Web proponent and having a blog that doesn't validate (and by such a wide margin) could be seen as hypocritical in some sense. just thought I'd let you know.
14:52 masak by the way, this is (or is at least related to) what I meant by "tag soup" earlier. the Web of today is so much not XHTML that it isn't even funny. even so, large parts of it claim to be.
15:01 egonw @masak yes, the blogger templates are really bad...
15:01 egonw @tell masak yes, the blogger templates are really bad...
15:01 zarah Consider it noted.
15:01 egonw @tell masak but they use macro's, so I can fix it unfortunately
15:01 zarah Consider it noted.
15:01 egonw @tell masak but it should indeed not suggest XHTML strictness
15:01 zarah Consider it noted.
15:01 egonw @tell masak was not aware it claimed that, thanx for the bug report :)
15:02 zarah Consider it noted.
15:02 egonw @tell masak actually... where is that claimed??
15:02 zarah Consider it noted.

| Channels | #bioclipse index | Today | | Search | Google Search | Plain-Text | summary