Re: post-release fun, episode 1
Hello Tibor et al, guess you have also found this link, courtesy of a Google alert for 'cds invenio': http://www.ohloh.net/projects/invenio Enjoy, Ferran
Re: post-release fun, episode 1
On Wednesday 02 April 2008 12:36:10 Ferran Jorba wrote: Hello Tibor et al, guess you have also found this link, courtesy of a Google alert for 'cds invenio': http://www.ohloh.net/projects/invenio Enjoy, Ferran Hi Ferran, was me who have added Invenio to Ohloh for experimenting (since we're an OpenSource project) :-) But Tibor's statistics are definitively more accurate and geared to the real structure of the project (e.g. division in modules, testsuite). But it Ohloh has good features in order to discover implicit links between CDS Invenio and the surrounding software environment... Bye! Samuele -- .O. ..O OOO
Re: post-release fun, episode 1
Hi Samuele, was me who have added Invenio to Ohloh for experimenting (since we're an OpenSource project) :-) Good, then But it Ohloh has good features in order to discover implicit links between CDS Invenio and the surrounding software environment... Bye! Samuele Or comparing it with others: http://www.ohloh.net/projects/compare?metric=Codebaseproject_0=EPrintsproject_1=CDS+Invenioproject_2=DSpace http://www.ohloh.net/projects/compare?metric=Activityproject_0=EPrintsproject_1=CDS+Invenioproject_2=DSpace http://www.ohloh.net/projects/compare?metric=Contributorsproject_0=EPrintsproject_1=CDS+Invenioproject_2=DSpace So far about Java verbosity... Thanks, Guido :-) ! Ferran
Re: post-release fun, episode 1
Hello Tibor, [...] P.S. I do agree with your Java remark; my favourite example is anonymous functions: Myself, with some white hairs in my beard, Java verbosity reminds me this so much: 001000$control optimize 001100 identification division. 001200 program-id. 001300 staglist. 001400 author. 001500 Ferran Jorba. 001600 installation. 001700 UAB. 001800 date-written. 001900 January 1991. 002000 Rewritten and translated into English, August, 1992. 002100 Change default-tags, 31 Jan 1994 002200 remarks. 002300 This program takes the output of MARC records 002400 from MARCPRT or HOLDPRT and reformats the output 002500 to one physical record per field or tag, for wide- 002600 paper printer (132 chars). 002700 The output record consists on the control number 002800 (BIB-ID, AUTH-ID or HOLDINGS-ID), level information, 002900 tag number, indicators and the textual data. 003000 It also allows the selection of a the set of tags 003100 to be printed. 003200 003300 003400 environment division. 003500 003600 configuration section. 003700 003800 source-computer. 003900 HP-3000. 004000 object-computer. 004100 HP-3000 sequence is hpascii. 004200 004300 special-names. 004400 hpascii is standard-1. etc..., before the meat begins. For me, Java is the new Cobol, necktie included ;-( Ferran
Re: post-release fun, episode 1
Hi gang: On Wed, 02 Apr 2008, Samuele Kaplun wrote: http://www.ohloh.net/projects/invenio was me who have added Invenio to Ohloh for experimenting (since we're an OpenSource project) :-) But Tibor's statistics are definitively more accurate and geared to the real structure of the project (e.g. division in modules, testsuite). Ohloh-like stats are nice (e.g. I've been using StatCVS privately), but they have some inconvenience such as not being able to distinguish between generated files (e.g. big RDF) stored in CVS, or give hints as to various modules and test suites as Sam said. So while we can have some overall fun with Ohloh (and you guys can even rate our project there ;-) as well as on Freshmeat http://freshmeat.net/projects/invenio/), please stay tuned for further episodes of the post-release fun series with more inside-look numbers... I just hope to get around to writing them one of the evenings finally... Best regards -- Tibor Simko ** CERN Document Server ** http://cds.cern.ch/
Re: post-release fun, episode 1
Hi Ferran: On Wed, 02 Apr 2008, Ferran Jorba wrote: http://www.ohloh.net/projects/compare?metric=Codebaseproject_0=EPrintsproject_1=CDS+Invenioproject_2=DSpace So far about Java verbosity... Thanks, Guido :-) ! Dunno how codebase is calculated WRT renaming and branching and whatnot throughout the CVS history, but the DSpace 1.5.0 release has much less code than that: $ find dspace-1.5.0-src-release -name *.java -exec cat {} \; | wc -l 189872 $ find dspace-1.5.0-src-release -name *.jsp -exec cat {} \; | wc -l 24159 when compared to: $ find cds-invenio-0.99.0 -name *.py -exec cat {} \; | wc -l 149323 to list just the main languages. And I can easily produce opposite trend numbers if need be: $ find dspace-1.5.0-src-release -type f -exec cat {} \; | wc -l 282665 $ find cds-invenio-0.99.0 -type f -exec cat {} \; | wc -l 647401 thanks to the PO files, RDF files and friends, e.g. this single one: $ wc -l cds-invenio-0.99.0/modules/bibclassify/etc/HEP.rdf 119383 cds-invenio-0.99.0/modules/bibclassify/etc/HEP.rdf Only to show that all these tools are to be taken with a big grain of salt. ;-) P.S. I do agree with your Java remark; my favourite example is anonymous functions: - Python: lambda x: k*x - Java: new Callable() { public Object call(Object x) { return x.times(k) } } Best regards -- Tibor Simko ** CERN Document Server ** http://cds.cern.ch/