Monday, 2010-04-19

*** isforinsects has joined #cc00:11
*** JoiIto has quit IRC00:28
*** isforinsects has quit IRC00:35
*** Bovinity has joined #cc03:10
*** papyromancer has joined #cc03:31
*** JoiIto has joined #cc03:33
*** wormsxulla has quit IRC03:34
*** wormsxulla has joined #cc03:50
*** JoiIto has left #cc04:18
*** JoiIto has joined #cc04:52
*** JoiIto has quit IRC05:07
*** JoiIto has joined #cc05:16
*** Bovinity has quit IRC05:38
*** JoiIto has joined #cc05:44
*** Kaetemi has quit IRC06:04
*** sama has quit IRC06:11
*** JoiIto has quit IRC07:24
*** sama has joined #cc08:06
*** FHaag has joined #cc10:23
*** ea4xgr has joined #cc11:32
*** ea4xgr has quit IRC12:27
*** ea4xgr has joined #cc12:57
*** tvol has joined #cc13:22
*** nkinkade has joined #cc15:14
*** oshani has joined #cc15:28
*** Kaetemi has joined #cc15:30
*** oshani has quit IRC15:40
*** nkinkade has left #cc15:46
*** nkinkade has joined #cc15:47
*** tvol has quit IRC15:49
*** bzzz has joined #cc15:50
nkinkadeparoneayea: Do you happen to know if there is a way to select "None of the above" on the chooser via a query parameter?15:52
nkinkade... sort like lang=?15:53
nkinkadeI can select other jurisdictions by passing &jurisdiction=nl, for example, but not the Unported.15:53
*** sama has quit IRC15:55
*** oshani has joined #cc16:13
*** akila87 has joined #cc16:15
*** bzzz has quit IRC16:18
*** nathany has joined #cc16:22
*** michi_ has joined #cc16:27
*** tvol has joined #cc16:40
ea4xgrnathany: hi. I've started to rewrite the validator (= extractor) from scratch. I'm using my own hosting for this.16:40
ea4xgr is the git repository16:41
ea4xgr is the Web application (runs the source code found in the repository)16:42
ea4xgrI've also played with the VirtualBox OSE and Ubuntu 9.10 (Minimal and Standard) in order to see how much space does the current version of the validator require16:45
ea4xgrit turned out that the required Ubuntu packages are 79.7 MB (in case of the minimal installation) and 29.0 MB (in case of the standard installation); the installed validator.git is 83 MB more16:46
*** akozak has joined #cc16:46
*** mralex has joined #cc16:54
*** michi_ has quit IRC17:06
*** michi_ has joined #cc17:06
*** tvol has quit IRC17:22
*** michi_ has quit IRC17:34
*** wormsxulla has quit IRC17:40
*** balleyne has joined #cc17:43
nathanyea4xgr, interesting; I wonder why it takes more space when on a standard installation17:46
nathanynot sure why that'd impact the dependency graph17:46
ea4xgrnathany: standard takes 29+83 (where 83 are mostly the eggs)17:46
ea4xgrminimal 79.7+83; these are the figures are came up with17:47
nathanyea4xgr, oh, i see; you're just counting the packages that need to be installed (79.7 for minimal)?17:47
nathanygot it17:47
nathanyea4xgr, you might look at using something light-weight like webob for your request/response objects; it can help clean things up so you're not creating your HTTP headers by hand :)17:50
ea4xgrnathany: I assume you've taken a look at extractor/ARCHITECTURE17:55
ea4xgrnathany: my objective now is to use the standard Python 3 library to the maximum extent, even if it would result in, say, 50% more code17:56
ea4xgrnathany: should some parts prove really impractical to write from scratch, a dependency will be used17:57
*** wormsxulla has joined #cc17:57
* ea4xgr is rewriting to Python now18:01
nathanyea4xgr, just read it18:08
ea4xgrnathany: I think starting from a Web app is a wrong way to go18:08
nathanyea4xgr, i thought that's why you split it into libvalidator v. validator last time around, so that the web app bits were isolated18:09
ea4xgrnathany: more like: CLI, Web service, Web application, GUI (or GUI before to the Web service stage)18:09
ea4xgrnathany: yes, but I'm referring to user/machine interfaces now18:09
nathanyi see18:10
ea4xgrnathany: we encourage people to use XHTML+RDFa, but the current validator does not event state license information that way18:10
nathanyi'm going to go out on a limb and say that writing 50% more code instead of using a well known dependency is not what i'd consider a best practice18:10
nathanyea4xgr, right, the existing implementation is not perfect18:11
ea4xgrnathany: the other way round we end up in 100 MB Web app boiling down to two files that are real contribution (one almost trivial - the Web app validation controller)18:11
ea4xgrnathany: I think pyRdfa may still be essential, but this whole Pylons/Genshi/mysql (?) dependencies are way off18:12
nathanymysql is pulled in via rdflib, iirc, so if you want to pull license information from the RDF, you may still get that in the graph18:13
ea4xgryes, required packages are libmysqlclient16 and mysql-common...18:13
nathanyea4xgr, required by... ? it might be interesting to use one of the dependency analysis tools to see what python deps are actually needed18:14
nathany(and by what)18:14
ea4xgrI've manually installed each required (Ubuntu) package one by one to see what's going on18:14
ea4xgrI think I've got the log somewhere18:15
nathanyah, right, librdf18:17
ea4xgrnathany: would it be possible to use rdflib instead of librdf?18:17
nathany(not rdflib)18:17
nathanyea4xgr, possibly; iirc rdflib has weaker SPARQL support, so that might be an issue18:17
nathanyi haven't looked closely in a while18:18
ea4xgrplease note that we end up using rdflib in the eggs regardless of the python-librdf dependency18:18
nathanyrdfadict may declare it as a dependency18:18
nathanyare you including gcc and git-core in your size analysis?18:18
ea4xgrgit-core - yes, as the required package in both minimal and standard; gcc - only in minimal, since it ships in the standard18:19
nathanyi see18:20
ea4xgrexcluding git-core and gcc, we have 13.6+5.3+7.0=25.9 MB18:20
ea4xgr(for Ubuntu packages alone)18:21
ea4xgr(for standard Ubuntu)18:21
ea4xgrnathany: cc.licenserdf requires rdflib 2.4.0 or newer18:22
ea4xgrI'm going to check rdfadict18:22
ea4xgryes, it also requires rdfadict18:23
ea4xgrand then cc.license requires python-librdf18:24
ea4xgrso, basically, cc.license requires both librdf and rdflib (through cc.licenserdf)18:25
ea4xgrwhy so?18:25
nathanyea4xgr, iirc we use rdflib where possible because it's pure python and easily installed as an egg; we use librdf in cc.license because we need fuller SPARQL support18:25
ea4xgrpyRdfa is also dependent on rdflib18:27
ea4xgrare there any alternatives to rdflib and librdf when it comes to SPARQL?18:30
ea4xgrmaybe fyzz?18:30
nathanythose are the two i'm most familiar with; not familiar with fyzz18:31
nathanylogilab seems borked right now, can't read up on it18:31
ea4xgrthere's an exemplary code at the bottom18:33
nathanylooks like it only handles parsing, though, right?18:33
ea4xgrwell, in fact they're post parsing, as they get to the abstract syntax tree18:33
nathanyright, but it doesn't actually return a result set was my point18:34
ea4xgryes, you're right18:34
ea4xgra kind of library that could power rdflib or librdf, not end-user apps18:35
nkinkademralex: That England-Wales flag looks bizarre. :-)18:35
nkinkadeBut I'm going to stay out of that one.18:35
mralexnkinkade: i know. ;)18:36
nkinkadeHey mralex, speaking of flags ... Mike mentioned that he thought you had a mock-up of an "international" flag.  Is that right?18:39
mralexnkinkade: yeah, i'm working on it18:39
nkinkadeAh, cool.18:39
mralexalso turns out the UN flag doesn't scale well.18:40
ea4xgrhtml5lib+pyRdfa (html5lib dep.)+rdflib (no dep.) = 998 KiB in total18:45
*** akila87 has left #cc18:47
ea4xgrwe're using cc.license to provide some extra information about the extracted CC license (most likely stated by an URI); I dunno if this is essential, since by simply following this URIs, the user would see the deed which states everything we do now and more (title, permissions w/ icons, conditions w/ icons, link to legal, etc.)18:49
*** MarkDude has quit IRC18:56
ea4xgrnathany: do python-dev and libxslt-dev dependencies come from rdflib, librdf, or from something else?18:56
ea4xgrnathany: I recall that I've got a compilation error due to the lack of these packages18:57
ea4xgrI'll check one more time18:57
*** stas has joined #cc18:59
nathanyea4xgr, librdf, probably19:00
nathanyif you're compiling c extensions you need to have the python headers installed19:00
*** FHaag has quit IRC19:03
ea4xgrnathany: it's rdflib, not librdf19:09
*** FHaag has joined #cc19:09
*** d00b has joined #cc19:28
ea4xgrso rdflib is not pure Python either, since it relies on bison to obtain the SPARQL parser19:41
ea4xgrmaybe if they replace the Bison parser with fyzz, the software would become a pure Python library (w/ no reliance on gcc)19:43
ea4xgrfyzz uses a lexer and a parser generated by ANTLR and a Python antlr3 runtime19:49
*** sama has joined #cc20:06
*** michi_ has joined #cc20:17
*** luisv has joined #cc20:18
*** oshani has quit IRC20:29
*** michi_ has quit IRC20:35
*** oshani has joined #cc20:43
*** oshani has quit IRC21:00
* ea4xgr has finished porting into Python 2.6 and Python 3.121:13
*** oshani has joined #cc21:32
*** jduro has joined #cc21:43
*** FHaag has left #cc21:47
*** balleyne has quit IRC21:55
d00bnathany, the missin 1 is flash , m i rite ??21:56
nathanyd00b, yup :)21:56
ea4xgrnathany: thanks for the tip re: git. do you want me to reply via mail to such letters (if you expect reply at all) or on Freenode?22:03
ea4xgrd00b: hi! what's up?22:03
nathanyea4xgr, no preference, just figured it was easier to forward that commit log to you so you had some more context :)22:04
d00bea4xgr, I'm fine u doing ??22:04
ea4xgrd00b: fine, thank you :)22:05
ea4xgrnathany: I do receive any git-related e-mails, maybe that's because a) you're a git admin b) I've provided a fake e-mail address to git?22:05
nathanyea4xgr, they go out to an email list, cc-commits (I can find the signup if you're interested)22:06
ea4xgrnathany: anyway, I've been experimenting much with git lately, so I've done things like reset --hard, commit --amend, push origin +master, etc.22:06
ea4xgrI was not aware that my changes result in any e‐mails22:06
nathanyea4xgr, right; those are all really powerful commands, good to know how to use them, but when you've already pushed they can cause problems22:07
d00bea4xgr, how's ur gsoc work is going on ? tuf competition dis time and even tougher for me22:07
nathanyi don't mind the emails, just trying to offer some feedback on best practices; fwiw :)22:07
ea4xgrd00b: the GSoC is not started yet, so I cannot be doing any GSoC work, but I do work on the project that I've proposed to extend under the GSoC umbrella22:08
ea4xgrnathany: thank you, your feedback is much appreciated :)22:08
ea4xgrd00b: “even tougher”? why's that? I remember reading your CV and I was utmost impressed—fwiw, my “achievements” are nowhere near yours, not to mention that I'm like three years older than you; I know that bright future is awaiting for you; keep up the good work :)22:11
ea4xgrnathany: why does this mailing list: have an awesome interface, whereas looks so humble?22:13
nathanyea4xgr, because cc-metadata has custom html to make it look like our website did in 2004 :) (hence why there are two jurisdictions under "icommons")22:14
ea4xgrah, ok :)22:14
d00bea4xgr, my project proposal is clashing with other guy, I really worked hard, but still It all depend on mentors, gsoc slot n other stuff , I can just hope for the best n I already started my work, so that I don't left any stone unchew22:15
ea4xgrd00b: ok, I see your point; it's great that you've already started22:17
*** bassel has joined #cc22:17
ea4xgrd00b: do you publish the current work anywhere (cc git or cc svn)?22:18
d00bea4xgr, nopes.. I read the entire code from previous year gsoc ...will start shifting the code to OOo 3.x from tomm itself..22:19
d00bea4xgr, I lost 7 days in between for my GRE :( :(22:19
ea4xgrgood idea22:19
ea4xgrthe Graduate Record Examination?22:20
d00bea4xgr, yes22:22
ea4xgrd00b: where does your summer break start?22:23
ea4xgrd00b: 'cause GSoC is tailored to the US summer break, you know22:24
ea4xgrd00b: for instance, summer break in .pl is July 1st-Aug 31st (or Sep 30th if you've passed all the exams in June)22:24
d00bea4xgr, I will apply for US next year, and rite now I'm free22:25
ea4xgrsounds like a plan22:25
d00bea4xgr, long term plan , let see whether I'm successful or not, I will invest my gsoc money for applying to US Univ. (If I got selected :) )22:26
ea4xgrI say, this could be one of the greatest investments of your life22:28
ea4xgrgsoc money in .pl is what the stipend that a doctoral student gets from the government (to support himself) for 11 months in total22:29
d00bea4xgr, let c how it turns out to be, all depend on GSOC for my funding22:30
d00bea4xgr, and at the same time gsoc has lifted the no of contributers 4 open source to great extend :) :)22:30
ea4xgrI wish that was true, but, as far as I know, although this is the main goal of Google (when it comes to GSoC), they've failed in this regard (in general)22:32
ea4xgrI'm referring to post-GSoC students-contributors22:32
ea4xgrin general, students do not stick to the organisation after they get the last instalment, sad but true22:33
ea4xgrnot to suggest that I was in anyway different in these terms22:34
ea4xgrs/anyway/any way/22:36
d00bea4xgr, what's ur thesis for ur doctoral studies ??22:36
ea4xgrI develop an architectural style for a Web application that would be targeting thin and thick clients architectures at the same time22:37
d00bea4xgr, great thesis.. so how much u completed, I think u started ur work on 2007 , so whn ur thesis defense ??22:38
ea4xgrto put it simple (though inaccurate), thin client would be the one w/ a JavaScript engine disabled, so no Ajax etc.22:38
ea4xgrthe goal is the following: a Web developer writes an application for a thin client according to this architectural style I develop; then he/she runs a compiler and obtains a thick client (Ajax-powered) version of the application without any additional effort/programming22:39
ea4xgrI can discuss this in any detail, if you want to (maybe better on XMPP since it's unrelated to Creative Commons)22:40
d00bea4xgr, I  will certainly talk 2 u on ur thesis..22:41
d00blater on :) :)22:41
ea4xgrea4xgr: the doctoral studies last 4 years in Poland, I hope to defend the thesis at the end of those 4 years (which is September 2011 in my case)22:41
d00bbest of luck frm my side :) :)22:41
ea4xgrthank you very much :)22:41
*** bassel has quit IRC23:05
*** ea4xgr has quit IRC23:23
*** nathany has quit IRC23:30
*** stas has quit IRC23:55

Generated by 2.6 by Marius Gedminas - find it at!