"Minor version Y (x.Y.z | x > 0) MUST be incremented if new, backwards compatible functionality is introduced to the public API. It MUST be incremented if any public API functionality is marked as deprecated. "
warp
:)
adhawkins
However, point 12 effectively says all number between the '.' are considered independantly. Major first, then minor, then patch.
warp
I wouldn't consider 5.1 and 5.10 to be identical when used as software versions.
adhawkins
Ok.
Sounds like in future 'schema' changes should cause a minor version bump, not a patch version as I've been doing.
That might complicate packaging some more :)
warp
adhawkins: if you choose to follow semver. you also make your own statement about what all of it means.
+can
adhawkins
Yeah.
warp
but I guess following semver would be appreciated by some groups of packagers and users.
adhawkins
But the apache one is similar. Patch level versions are for pure bug fixes only. New functionality should increase minor version.
warp nods.
I'll have to see how that affects things like Debian.
For example would libmb5 and libmb5.1 be different packages, or still called libmb5?
warp
I have no idea. I know very little about debian packaging.
adhawkins
Ok.
teuf
adhawkins: still libmb5 unless you break api/abi and want those to be parallel-installable
(which you don't want)
reosarevok
"Google scores victory: judge rules APIs not copyrightable"
Guess that's good news, since less legal BS tends to be
adhawkins
teuf: I think I agree there.
warp
reosarevok: yes, that definitely is good (though not unexpected) news.
reosarevok: I was more worried about the patent phase, but google won that as well.
teuf
adhawkins: using the 2nd number to tag schema changes, and the last one for bug fixes would be nice
adhawkins: but anything will do imo
adhawkins
Yes, that's what I think we'll do moving forward.
Just trying to work out how that fits into Debian etc.
warp
reosarevok: so I think the oracle vs. google case is now over (apart from some formalities).
teuf
adhawkins: don't worry too much about that ;)
adhawkins
teuf: Well, Debian is what I use so it'd be nice to handle it properly!
warp
I still use Debian on two machines. but the slow release cycle is now starting to annoy me, so I'm considering switching everything over to Ubuntu.
adhawkins
My main server is debian (firewall, routing, DNS, NTTP, SMTP, POP3 etc). I like the fact that I know it's just going to keep working.
I don't really need new versions mostly.
warp
adhawkins: development cycles have speed up in recent years.
adhawkins
Yeah
warp
as a webdeveloper doing javascript it is convenient that I can just 'sudo apt-get install nodejs' on ubuntu to get a working javascript prompt
(even though I don't care about node.js as a platform)
stuff like that being available faster is making ubuntu quite tempting, even on servers.
teuf
adhawkins: what I mean is that debian doesn't really care about how you version your module
adhawkins
Yeah, just been asking that on a debian channel, and as long as the SONAME stays the same, the package name should too.
So in future, if there's new functionality (schema changes and hte like) the minor version will get bumped. Pure bug fixes will just increment the patch version
doskir joined the channel
marco-oliveira
hi everyone
doskir_ joined the channel
outsidecontext joined the channel
warp
bitmap: boo!
bitmap: how are you doing with your gsoc stuff?
ijabz joined the channel
plaintext
I wrote the log sanitizing script
Does anybody want to see?
kurtjx joined the channel
gravelld joined the channel
ijabz joined the channel
gravelld
Has anyone been working on a semi-automated bot for uploading artwork to CAA? I heard a rumour there was one in the works.
reosarevok
Rumours, all!
ijabz joined the channel
gravelld
Righto
Thanks Reo
I'll go git my lunch
Ooops I mean get
reosarevok
:D
kepstin-work wishes he could use git to pull lunches.
gravelld is still trying to work out how to incorporate 'fork' into the puntastic convo
kepstin-work
fork... lunch... ? It should be trivial.
but yeah, the awesome thing about gitting someone else's lunch then by forking it is that they still have their lunch :)
got to love digital non-scarcity.
plaintext_ joined the channel
so, is it intentional that people aren't allowed to log in to classic.musicbrainz.org?
I wanted to look at the pre-ngs-conversion edit history for something for comparison purposes
but I couldn't, since login is disabled.
kurtjx joined the channel
kepstin-work welcomes navap to the canadian english translation team :)
marcooliveira joined the channel
ijabz joined the channel
ijabz joined the channel
hawke_1 joined the channel
bitmap
warp: so far so good, though I realized the javascript is a bit annoying because it involves template-y stuff that jquery doesn't do well on its own (i.e. adding/removing/changing relationships and then updating the UI)
if I could I would use something like http://knockoutjs.com/ but I realize that's departing a lot from the rest of the MB js