Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2013-02-07

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

All times shown according to UTC.

Time Nick Message
00:15 Molaf_ joined #pdl
06:04 sivoais_ joined #pdl
15:10 Molaf__ joined #pdl
17:01 run4flat Just posted a rant on blogs.perl.org: http://blogs.perl.org/users/david_mertens/2013/02/3m-says-perl5-needs-a-new-major-version-number.html
17:05 sivoais_ I'm no good at hype, but I do agree with you.
17:06 sivoais The problem I see is that people are instantly dismissive of Perl based on antiquated notions that are mostly second-hand.
17:09 sivoais I interact with other students at uni daily and when it comes to discusses languages, there's nobody else that talks about Perl. The only way I've shown Ruby and Python users that Perl is still around is by showing them code that works. I wow them by showing how clean it is to use CPAN modules and how quickly I can get new things written.
17:10 sivoais s,discusses,discussing,
17:11 run4flat I work in a Python lab
17:11 run4flat I would love to be able to show to my labmates how awesome Moo is
17:11 run4flat but it's additional
17:11 run4flat (and it doesn't help that I don't know it)
17:12 run4flat But if there was a new major version number, I think that would go a *long* way to at least *starting* the discussion
17:13 lungching genehack, in his perl oasis keynote, talked about instead of doing "cool stuff in perl" just do "cool stuff"
17:14 run4flat lungching, how does that help?
17:14 lungching I don't know if he meant it this way, but I took it as, if you keep doing cool things people like eventually they'll get curious as to how you did it
17:14 lungching then they're hooked
17:15 sivoais Yeah. It leads to the thinking: "wow, you did that in Perl?"
17:15 lungching where as, if you start out "this is in perl" people might dimiss it without finding out what it is
17:15 run4flat oh, OK
17:16 sivoais And then the next thought would be: "why not in $LANG?". And then you can go into the discussion.
17:16 lungching yup, that's my thinking, at any rate
17:16 lungching well, any way, lunch time! :)
17:17 run4flat o/
17:30 jberger the problem is person-to-person small-time hype is too slow, Perl is bleeding users faster than that
17:31 jberger I do what I think is great work in Perl, yet of my officemates, who sit just feet from me, only one of the three has started to learn it
17:31 jberger in the meantime, the other two have picked up MatLab
17:31 jberger why not Perl+PDL
17:31 jberger I don't know
17:32 jberger probably something they read, or someone they talked to in the hallway
17:32 sivoais One thing most people don't know is that every MATLAB install comes with a copy of Perl.
17:32 jberger but if I can't get new users in my own room to try Perl its not good enough
17:32 run4flat Don't get me wrong, I think that Matlab and NumPy have much deeper tool chests than PDL
17:33 run4flat there really is a shortage of software
17:33 jberger my point isn't MatLab or PDL even
17:33 jberger the problem is hype
17:33 jberger Perl is seen as ancient
17:33 run4flat yeah
17:33 jberger for no good reason
17:33 run4flat sivoais, if only we could tap into that
17:33 run4flat "Have Matlab? Try giving Perl a spin"
17:33 jberger why not bump the version number? IT COSTS P5P LITERALLY NOTHING
17:33 run4flat actually, developing a Matlab extension to pass data between Matlab and Per would be cool
17:34 run4flat The only thing I can see is that Larry has moral objections to 7
17:34 jberger and there would be free press about it
17:34 run4flat so, fine, bump it to 8
17:34 jberger pardon my french, but on this F*** Larry
17:34 jberger we needed larry to sort out the signatures thing, and instead Peter Martini left with his fully written code and went home
17:35 jberger over something that just needed Larry to pick A or B
17:35 jberger p5p couldn't come to an agreement.
17:35 jberger and larry couldn't be bothered to come make a little (yet important) ruling
17:36 run4flat I don't know Larry's prerogative, but I like the guy
17:36 jberger if he wants a say in what Perl does with its version numbering, he should at least take a cursory interest in perl
17:36 run4flat sure
17:36 jberger I like him too
17:36 jberger but he doesn't work on Perl anymore
17:37 jberger he works on a language that might some day exist and might bear a slight resemblence to perl
17:37 run4flat First of all, I mostly agree
17:38 run4flat but, from the point of channel governance, I don't like the language you used
17:38 run4flat Larry is a great guy. Without him we wouldn't have Perl5 as it is today
17:39 run4flat I lament that he hasn't stepped in and wielded his BDFL hat
17:39 run4flat but we ought not curse the man
17:39 run4flat He has made very few requests of Perl5 governance since leaving, but one of them was specifically that Perl never use the version number 7
17:40 jberger run4flat, sorry about the language
17:40 run4flat jberger, thanks
17:40 jberger its rather out of character for me, especially to say of another persno
17:41 jberger Larry, should you happen to read this, I apologize to you too
17:41 jberger but if you DO read this, we really need you
17:41 run4flat which brings me to wonder
17:41 run4flat did you ever send him an email about it?
17:41 jberger I'm sure he got a dozen emails about the signatures thing
17:42 jberger I think he must actually be actively determined not to take a side
17:42 run4flat He is a very nice guy. I wonder if enough emails would convince him to at least drop by occasionally
17:42 run4flat I think nobody framed it as, "Think of the good your ruling would do here."
17:42 jberger perhaps thats true
17:44 run4flat though that really puts things into perspective
17:44 run4flat what good---for the world---would Larry's ruling do here?
17:44 jberger actually I want it more for the signatures thing
17:44 jberger a guy actually wrote the code
17:44 run4flat yeah, that's what I was thinking of, too
17:45 run4flat If Larry is anything like me, he probably finds it all too easy to get wrapped up in Perl stuff
17:45 jberger and over a difference of what it should do in a few cases, any of which were easy to implement, the project got scrapped
17:45 run4flat and maybe Perl6 was his chance to unwrap
17:45 jberger and I wouldn't be surprised if few people EVER try to submit something of that scope to P5P again
17:46 run4flat apart from p5mop?
17:46 jberger think about it
17:46 run4flat I think p5mop is going to be a very big deal
17:46 jberger lets say p5mop (which is stalled btw)
17:46 run4flat I know it's stalled
17:46 jberger lets say get gets to 85%
17:47 jberger and it gets down to the nitty gritty details, which will likely be similar to the ones signatures faced
17:47 jberger and the committee on p5p can't decide on that 15%, the current trend is to grind the submitter until they give up
17:48 run4flat but I will jump and and help
17:48 jberger so with p5mop at lets say 30-ish %, are you surprised that Stevan decided to try something that wouldn't be subject to p5p?
17:48 run4flat and somebody else will jump in and help
17:48 jberger I'm not
17:49 jberger doesn't matter, if it can't get added
17:49 run4flat but enough people care about this
17:49 jberger they cared about signatures
17:49 run4flat really?
17:49 run4flat did you see blogs about it?
17:49 jberger enough to put there stake in the ground for the thing they wanted
17:49 jberger and so did other people
17:49 jberger until the trenches were dug
17:49 jberger and there it sat
17:50 jberger yeah, it wen't out to the blogosphere, but it was too technical to see the level of discussion that something like version numbering can prompt
17:50 jberger this was more like 3 against 5 ish size
17:51 jberger version numbers are everyone against everyone else
17:51 jberger actually split into 4 camp (that I can see)
17:51 jberger run4flat, are you subscribed to p5p mailing list
17:51 jberger ?
17:51 run4flat I *un*subscribed
17:52 run4flat it's too easy to get involved in discussion of minutia
17:52 jberger I joined in order to comment
17:52 run4flat when I have absolutely no intention of writing code
17:52 jberger but I haven't seen my email on the archive
17:52 jberger I don't know if it went in or not
17:52 run4flat The problem with p5p is that it's scope is too big
17:52 run4flat *everything* regarding Perl's core runs across that list
17:53 run4flat and it's all too easy to think, "Oh, I have a useful opinion about that."
17:53 jberger indeed
17:57 jberger actually, I'm probably ok
17:57 jberger I have gotten a couple other replies now that also aren't on the archive page yet
17:57 jberger it probably recaches the pages hourly or somthing
17:57 jberger something
18:01 run4flat yep, yours hit the archive
18:05 jberger hmmm, I'm still not seeing it
18:05 jberger but I do see it one google's mirror
18:05 jberger https://groups.google.com/d/msg/perl.perl5.porters/IjbVnFNayBY/ldjnLtGSlv4J
18:05 jberger thanks
18:05 run4flat http://www.nntp.perl.org/group/perl.perl5.porters/2013/02/msg198270.html
18:34 lungching https://github.com/MoeOrganization/moe
18:35 lungching a reimplementation of perl5 written in scala :)
18:35 lungching stevan little started it, I think
18:37 sivoais there's also <http://perl11.org/>
18:38 sivoais But I don't know if there enough round tuits to make something remarkable happen.
18:39 jberger lungching, Perl in scala breaks XS and thus breaks ... #pdl
18:43 sivoais Of course, stepping away from the main codebase and trying things anew has helped reinvigorate Perl 5 development in the past.
18:47 jberger has anyone seen: http://perl11.org/p2/
18:47 jberger does it work?
18:48 lungching jberger: my understanding was they were going to leave XS like stuff out
18:48 jberger yes but it has an ffi
18:48 jberger or claims to
18:49 lungching at any rate it seems really really crazy to me :)
18:49 lungching but we are perl programmers and that seems to be our modis operandi
18:50 lungching but I brought it up because of the discussion on changing the perl version number
18:50 lungching they went and changed the perl name ;)
18:56 jberger if what run4flat (and others) have said is true, that Larry won't allow Perl to increment its major version number, then its the only remaining choice
18:56 jberger looks like p2 is reini urban
18:57 jberger and it looks only slightly implemented at this point

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