Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2015-02-01

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

All times shown according to UTC.

Time Nick Message
02:47 ilbot2 joined #pdl
02:47 Topic for #pdl is now Install PDL: http://pdl.perl.org/?page=install  | Book: http://pdl.perl.org/content/pdl-book-toc.html | Mailing list: http://pdl.perl.org/?page=mailing-lists | Pasting: http://scsys.co.uk:8001/pdl | Channel is logged by ilbot2: http://irclog.perlgeek.de/pdl/today
06:11 estrabd_afk joined #pdl
06:36 pdurbin joined #pdl
08:24 estrabd_afk joined #pdl
08:32 estrabd_afk joined #pdl
09:26 estrabd_afk joined #pdl
10:15 opkick [pdl] wchristian pushed 1 new commit to master: http://git.io/FAYU
10:15 opkick pdl/master 50790e6 kmx: missing comma kmx--
12:30 sivoais joined #pdl
19:09 mohawk someone should make a PDL::Tiny that is simply a 100-line mega-naive API-compatible native perl implementation
19:10 mohawk it would be interesting to benchmark that against current PDL
19:10 mohawk more usefully, however, it could be ported to perl6
19:11 mohawk and that benchmark against the perl5 ::Tiny and ::Current would be very interesting
19:11 mohawk sivoais, sound tempting? (the ::Tiny part, at least)
19:30 sivoais mohawk: sshhhh... I already had plans for that
19:30 mohawk better get on it before i beat you to that as well :-)
19:31 sivoais my current goal is finish up IPerl. I'm writing the test suite for that
19:32 sivoais PDL needs better plotting support and IPerl is probably the easiest way to to get that
19:36 mohawk what's the problem with its current support?
19:36 mohawk i'm assuming this is in P::Graphics::P[GL]plot?
19:39 mohawk sivoais ^
19:39 sivoais mohawk: the different plotting libraries just aren't as nice as those from other tools
19:40 mohawk libraries = ones supported in PDL?
19:41 sivoais mohawk: yeah
19:42 sivoais mohawk: compare the plots you have in PDL with these <http://bokeh.pydata.org/en/latest/>
19:42 sivoais and compare the APIs
19:42 mohawk so the way forward would be updating the PDL support API?
19:43 mohawk could you help me by showing a like-with-like compare?
19:43 sivoais mohawk: the fastest way forward is to use all the browswer tools for plotting
19:43 sivoais mohawk: sure
19:43 mohawk and also a URL showing the side-by-side APIs?
19:43 sivoais eeeh, that doesn't exist
19:43 mohawk partly because i'm knee-deep in fixing Inline::Pdlpp
19:44 mohawk and partly because this can go as a feature request ticket once perfected
19:48 sivoais mohawk: ok, compare <https://github.com/PDLPorters/pdl/wiki/Plotting-with-PDL::Graphics::Gnuplot#2-d-plots>
19:48 sivoais with <http://bokeh.pydata.org/en/latest/tutorial/solutions/gallery/lines.html>
19:48 sivoais not only is bokeh interactive, but it gives you an object that represents the plot
19:48 sivoais so that you can choose to actually plot it later by using show( plot_object )
19:49 sivoais but also do more complicated things like have subplots
19:49 sivoais show( HBox( p1, p2, p3 ) )
19:49 sivoais ^ arrange 3 plots horizontally
19:50 sivoais But if I improve IPerl, all of this can be available in the browser because I can just take the JS backend of Bokeh and give it a Perl interface
19:50 mohawk ok
19:50 mohawk is it worth filing a ticket with the above info?
19:50 sivoais :-D
19:50 sivoais sure, maybe just link the IRC log
19:50 mohawk i think better is to actually excerpt it right down
19:51 mohawk by all means also link to the log
19:51 sivoais yeah, summarise it
19:51 sivoais also, there is a real need for all the bioperl folks to have something like <http://webglmol.sourceforge.jp/index-en.html>
19:51 mohawk i'm suggesting making it be complete, but a summary
19:51 mohawk if that distinction makes sense
19:52 sivoais mohawk: I'll go ahead and make the ticket
19:52 mohawk nice work
23:59 mohawk neilb says he'll fix the PDL permissions issue shortly
23:59 mohawk another benefit of using IRC

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