Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2014-11-15

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

All times shown according to UTC.

Time Nick Message
16:44 chm joined #pdl
16:48 chm plicease, mohawk: Really happy to see all the progress with Alien::Base and the CheckLib::FFI work
16:48 chm wonderful!
16:51 chm mohawk: I'm hoping to have some time for PDL work coming up and
16:51 chm would like to clean up the mess I made of the ExtUtils::Depends merge.
16:52 chm My load average since August has been pretty much max'd out
16:52 chm Hope for some time as the holidays come through.
16:55 jberger_ joined #pdl
16:55 chm plicease: Sorry I meant FFI::CheckLib
17:00 jberger joined #pdl
17:20 jberger joined #pdl
17:22 jberger joined #pdl
17:29 jberger mohawk: I finally got around to reading your slides
17:29 jberger cat herding indeed :-)
18:34 chm cat herding++
20:13 jberger oh btw, I should mention
20:13 jberger part of the reason I have been so absent is that I have been working at a python shop since I defended
20:13 jberger it has been all I could do to keep my nose in #mojo and make some example blog posts from time to time
20:14 jberger but as of yesterday I have accepted a new job doing Perl and indeed Mojolicious
20:14 jberger so while I still may not be using as much PDL as I did when I was in my Ph.D. if I'm back in the language more, who knows?!
20:34 pdurbin jberger: how did python treat ya?
20:34 jberger I'm not a big fan
20:35 jberger I guess all in all its ok, but I'm happy to be going back to Perl
20:37 pdurbin what were the pain points?
20:38 jberger I found that the whitespace thing had its worries
20:38 jberger deeply indented code became hard to grok when dropping out of a few blocks at once
20:39 jberger the worst actual thing was the fact that methods are just attributes
20:39 jberger so you can stomp on methods by assigning to an attribute of the same name
20:39 jberger (that bug took all day to find)
20:40 jberger I find list comprehensions unreadable, especially if doing nested data structures
20:41 jberger variables are function scoped rather than block, tripped on that a couple times
20:42 pdurbin these all seem like valid complaints. I guess you weren't flying: http://xkcd.com/353/
20:42 jberger closures over "scalars" make the scalar immutable
20:43 jberger the type system usually just causes things to crash rather than prevent bugs (that's opinion)
20:43 pdurbin so the strong typing is nice when you get an exception in dev but not so nice when stuff blows up in prod
20:44 jberger the problem is that you still can't declare input and output types, so you have to either check them manually or trust that no one changes other apis
20:44 pdurbin maybe you'd like haskell :)
20:45 jberger I keep thinking I would, but I haven't had enough reason to try to get deep enough into the syntax
20:46 pdurbin I found this interesting: https://joeyh.name/blog/entry/no_longer_a_perl_programmer/
20:49 jberger pdurbin: it is more interesting than the title implied
20:49 jberger oh, it does remind me, the python lambda keyword is rediculous too
20:51 pdurbin one line lambdas
20:51 jberger there are other restrictions too
20:51 jberger you can't do assignment
20:51 jberger can't use the return keyword
20:52 jberger perl's variable scoping and ease of use closures are its hidden killer features
20:53 pdurbin easy to take for granted
20:56 pdurbin jberger: congrats on the new job, by the way :)
20:56 jberger pdurbin: thanks

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