Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2017-11-03

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

All times shown according to UTC.

Time Nick Message
01:47 gordonfish joined #metacpan
02:06 castaway joined #metacpan
02:12 castaway joined #metacpan
02:56 ilbot2 joined #metacpan
02:56 Topic for #metacpan is now https://github.com/metacpan/metacpan-api/wiki/MetaCPAN-V1-announcement-and-migration-plan | MetaCPAN Developer VM https://github.com/metacpan/metacpan-developer | Chat logs available at http://irclog.perlgeek.de/metacpan/ | Can't find your module on MetaCPAN? https://metacpan.org/about/missing_modules
03:16 castaway joined #metacpan
04:22 castaway joined #metacpan
04:53 castaway joined #metacpan
05:12 castaway joined #metacpan
05:51 castaway joined #metacpan
06:33 nakiro joined #metacpan
08:44 castaway joined #metacpan
08:48 neilb joined #metacpan
08:49 castaway joined #metacpan
09:03 Relequestual joined #metacpan
09:13 neilb joined #metacpan
10:54 castaway joined #metacpan
11:00 castaway joined #metacpan
11:27 karjala_ joined #metacpan
11:42 castaway joined #metacpan
11:50 Guest69 joined #metacpan
11:59 castaway joined #metacpan
12:23 ranguard cute... https://www.fastly.com/blog/diff-edge-serverless-cloud-functions/
13:22 ranguard I do wonder if we shouldn't stick our .tar.gz files into S3 - as well, the our sever infra could be lighter and cpan.mc.org could be S3+fastly
13:23 ranguard something to discuss at metahack
13:23 castaway joined #metacpan
13:31 haarg the api server needs continued direct access to the files
13:32 haarg it doesn't seem like serving the files has been a particularly heavy load
13:59 castaway joined #metacpan
14:10 ranguard haarg: I guess downloading would be a little overhead, but could store unzipped as well as zipped in S3 - that would then remove the overhead of unzipping - either way having in S3 would be nice... for the backpan especially
14:37 jberger what is the advantage? are you tight on storage?
14:38 jberger I especially hesitate to fracture an app like this into partially S3/serverless when there is a notion of people self-hosting it
14:38 jberger (also, I'm pretty skeptical of serverless in general)
15:00 castaway joined #metacpan
15:01 ranguard jberger: I guess advantage is we can't run out of storage (I'm not happy with our current level of monitoring) - doesn't have to replace what we have, but also means someone could run the API just the ES dataset (from a S3 snapshot) and not have to have a full copy of the backpan...
15:01 ranguard but yea, might be a solution looking for a problem :)
15:02 ranguard jberger: I'd be interested to discuss serverless with you, see where you feel most skeptical :)
15:31 jberger I think partially what you say there is my concern, that it is a solution looking for a problem
15:31 jberger I think that its ability to automagically scale for you is at least offset by the amount of tooling needed to make it work
15:31 jberger and then your application is written in a way that is absolutely vendor-locked
15:32 jberger which I'm always hesitant to do, especially when it is to a closed-source commercial product
15:32 jberger my boss doesn't love that I use so many of the postgres-specific features, because of vendor lock
15:32 jberger but at least postgres is free and open source
15:37 trs also, Pg is awesome. :-)
15:39 ranguard heh
15:45 haarg s3 isn't such a hard lock in
15:46 haarg i'm just not sure if it really buys us anything, and there's definitely a bit of stuff we'd need to change
15:58 karjala_ joined #metacpan
16:00 jberger s3 might not be lock in since there are alternatives for bucket storage
16:00 jberger I'm not sure if there are serverless alternatives that compare with amazon's offerings
16:01 jberger but the coordination aspect is still something
16:02 mst there's plenty of things that emulate the s3 API as well
16:10 oalders https://twitter.com/PerlTricks/status/926452133586751488
16:12 jberger wow, really?!
16:13 oalders It would be weird if that were the case
16:13 Grinnz I'm sure that's not true
16:13 haarg there's one that is marked as first for me, in the 9 results that url gives
16:13 jberger makes me think I should have rushed to release Mojo::Chrome, I could have had his month round up all to myself!
16:14 Grinnz just looking at recent uploads i found https://metacpan.org/release/App-mysqlinfo
16:14 oalders :)
16:15 haarg he doesn't say how he's actually querying
16:15 haarg just gives a bare url that isn't useful without an actual query
16:15 oalders I've just asked him about that
16:15 jberger maybe someone ought to reply with a useful query?
16:16 oalders I assume he's been running that same query monthly, so I'd like to see the actual code.
16:16 oalders Maybe that's a bad assumption on my part
16:17 haarg probably just wait to see what he actually provides
16:22 castaway joined #metacpan
17:44 castaway joined #metacpan
17:48 neilb joined #metacpan
18:11 castaway joined #metacpan
19:58 neilb joined #metacpan
20:33 gordonfish joined #metacpan
20:53 neilb joined #metacpan
20:54 neilb_ joined #metacpan
21:16 jacoby joined #metacpan
22:13 castaway joined #metacpan
22:15 brunoramos joined #metacpan
22:37 neilb joined #metacpan

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