Perl 6 - the future is here, just unevenly distributed

IRC log for #pr-challenge, 2016-01-05

| Channels | #pr-challenge index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:08 kid51_ joined #pr-challenge
01:16 rocamador joined #pr-challenge
02:12 gryphon joined #pr-challenge
02:25 Zoffix joined #pr-challenge
02:51 vroom joined #pr-challenge
03:42 veryrusty joined #pr-challenge
08:51 mephinet joined #pr-challenge
09:23 amalia joined #pr-challenge
09:53 veryrusty joined #pr-challenge
10:18 veryrusty joined #pr-challenge
10:37 rocamador left #pr-challenge
11:28 veryrusty joined #pr-challenge
11:36 choroba joined #pr-challenge
12:16 vroom joined #pr-challenge
12:25 kid51 joined #pr-challenge
12:28 veryrusty joined #pr-challenge
13:20 neilb joined #pr-challenge
13:21 BooK neilb: I'm looking at your lpw talks, and that got me thinking
13:21 neilb oh no ...
13:21 neilb always dangerous!
13:22 BooK the first thing is that metacpan could show how upstream (in the CPAN river) a distribution lies
13:23 BooK there might already be an issue for that
13:24 BooK the logarithmic scale is great, because can be a single digit :-)
13:25 BooK k for dependents < 10^k (so 0 actually matches the sea)
13:25 BooK I thought about something else, more related to the PRC, but I forgot it :-(
13:26 BooK ah yes, tagging
13:26 neilb yeah, there is already a metacpan issue for showing river position
13:28 BooK tags could either be in the META information, with some "x_tags" : [ ... ] (but that means stable dist would never get tagged), or in metacpan (but then there's a risk of chaos, although that can probably be mitigated)
13:29 BooK maybe that discussion should be in an email to cpan-workers
13:30 BooK actually, metacpan could probably use both (META tags + user-assigned tags) to compute some composite tags
13:32 neilb tags are a bit of an old chestnut. every now and then they come up, but no-one’s come up with even a basic model that everyone else says “yeah, that works!”
13:33 BooK yeah
13:33 BooK it does not have to be perfect at the first attempt
13:36 BooK metacpan can be a nice place to experiment
14:03 vroom joined #pr-challenge
14:03 BooK neilb: can you point me to the metacpan issue about the river?
14:14 neilb BooK: will do, later
14:19 deven joined #pr-challenge
14:33 BooK thanks
14:46 vroom joined #pr-challenge
14:50 monsieurp neilb: hey.. u there?
14:51 neilb yeah — meant to send you an update.
14:51 monsieurp :)
14:51 monsieurp cool thanks
14:51 neilb I’ve got comaint on the dist, and he’s going to give me commit bit on his repo
14:51 monsieurp perfect
14:51 neilb so do your PR, and then I’ll make sure it gets handled, if he Murase-san doesn’t :-)
14:53 vroom joined #pr-challenge
14:59 gryphon joined #pr-challenge
15:30 Lee joined #pr-challenge
17:10 neilb joined #pr-challenge
17:52 ether BooK: what sort of things are you picturing being in x_tags?
18:11 gryphon joined #pr-challenge
18:45 BooK ether: I guess that's the crux of the issue :-)
18:47 BooK that's actually why I think experimenting on metacpan is worthwhile
18:47 BooK instead of trying to figure out what the tags could be, let the cpan crowd tell us
18:49 thrig netfullauto_tag_0.999999999999999
18:49 ether but in a general way? what sort of things would go in there?
18:49 ether were you thinking of river position?
18:49 BooK no, it's a different thing
18:49 BooK tags like: web, network, webservice, fast, well-documented
18:50 BooK both about what it does and how it is made
18:50 ether there's already a 'keywords' field in meta v2
18:50 BooK I guess, there's a "tag cloud" notion, where tags have a certain weight
18:50 ether https://metacpan.org/release/Dist-Zilla-Plugin-Keywords
18:51 ether https://metacpan.org/pod/CPAN::Meta::Spec#keywords
18:51 ether I've been putting keywords into my dist meta for years :)
18:51 BooK maybe the keywords in META should be considered differently
18:52 BooK as if comes from the distribution authors
18:52 BooK ether: I'm sure I have known about keywords, and forgot
18:52 BooK I'll look at your dist for inspiration and start adding some
18:53 BooK do metacpan/sco show those keywords?
18:54 ether no
18:54 ether a tag cloud thingy would be a great way to browse distributions
18:57 BooK and would not require author participation :-)
18:57 BooK I think it's worth hiding underrepresenting tags, and showing keywords prominently
18:58 BooK (also they could duplicate each other, since tags are basically users voting on keywords)
20:30 veryrusty joined #pr-challenge
20:50 jb360 joined #pr-challenge
21:28 neilb A review of last year’s challenge: http://neilb.org/2016/01/05/cpan-prc-2015-review.html
22:14 deven joined #pr-challenge
23:01 BooK neilb: nice post
23:02 kid51 joined #pr-challenge
23:02 neilb ta! took way too long to write. probably should have written a more concise summary, but I couldn’t help it :-)
23:05 * kid51 just read http://neilb.org/2016/01/05/cpan-prc-2015-review.html
23:12 BooK neilb: could you point me to the code you used to compute the river-of-cpan numbers?
23:15 neilb Book: not right now. I can tell you how I did it. Or if you wait a bit, it will be part of my CPAN dashboard code, which *will* be on github :-)
23:15 neilb Decided I really should get on and get the dashboard done!
23:15 BooK yay
23:15 BooK I'd also think http://deps.cpantesters.org/ should be mentioned when talking about the cpan river
23:15 neilb aye
23:16 BooK after all, it can give you some numbers about the pollution level (even though the site cleary states they don't make much sense)
23:55 jberger joined #pr-challenge

| Channels | #pr-challenge index | Today | | Search | Google Search | Plain-Text | summary