stanislas, Actually, I think I deleted my G+, I just have a google account
2016-01-11 01106, 2016
stanislas
gcilou: I didn't know it is deletable, wow. It must be hidden somewhere, otherwise there would be no G+ users left.
2016-01-11 01125, 2016
gcilou
Haha, yes it's in your google account settings :)
2016-01-11 01116, 2016
gcilou
It's very hidden
2016-01-11 01146, 2016
weeksio
Re: git merging, I got it....nevermind
2016-01-11 01152, 2016
Freso
Gentlecat: Hm. I hid one of the releases at https://critiquebrainz.org/user/48f2a23c-90e5-4c3… but I now realise that maybe it was actually attached to the proper RG originally but the RG has since gone (merge, deletion). I'm not sure if there's currently any way to handle that, or what I'd prefer of means to deal with it.
The US ranges between -0500 and -1000 (in standard time, not summer/DST).
2016-01-11 01132, 2016
Freso
Time is difficult, yo. :)
2016-01-11 01150, 2016
Freso
And names are difficult. And addresses are difficult.
2016-01-11 01156, 2016
Freso
:|
2016-01-11 01123, 2016
CJ_ thinks Freso is difficult.
2016-01-11 01153, 2016
gcilou
*life is difficult*
2016-01-11 01103, 2016
CJ_
C'est la vie. :)
2016-01-11 01120, 2016
Freso
C'est moi.
2016-01-11 01156, 2016
Leftmost
stanislas, points for rendering your proposal with TeX.
2016-01-11 01105, 2016
ruaok
gcilou: you in Chicago?
2016-01-11 01126, 2016
gcilou
Nope :) just Chicago time zone. I'm in Arkansas
2016-01-11 01146, 2016
gcilou
*two letters better than Kansas* as we say.
2016-01-11 01155, 2016
ruaok
ah. otherwise I would've invited you to lunch with bitmap and myself.
2016-01-11 01112, 2016
stanislas
Leftmost: I can't afford Microsoft Word, sorry :)
2016-01-11 01132, 2016
gcilou
That would've been awesome (and intimidating..) :)
2016-01-11 01135, 2016
opatel99
Libre
2016-01-11 01148, 2016
stanislas
opatel99: it's shit
2016-01-11 01148, 2016
Freso
OpenOffice :U Kappa
2016-01-11 01106, 2016
CJ_
LibreOffic.
2016-01-11 01156, 2016
gcilou
Freso, for this screencast thing, is there a way that we could include links on the youtube so when you click a certain place on the video, it takes you to for example the release style guidelines? I've seen YouTubes that had those before..
2016-01-11 01154, 2016
Gentlecat
CallerNo6: should be UTC
2016-01-11 01109, 2016
CJ_
How long until the meeting?
2016-01-11 01112, 2016
CallerNo6
Gentlecat, ah, thanks
2016-01-11 01114, 2016
Gentlecat
supposed to be
2016-01-11 01148, 2016
CallerNo6
I'll experiment later.
2016-01-11 01107, 2016
opatel99
student-student interaction doesn't seem to be as high? *embraces for rejection of idea*
2016-01-11 01136, 2016
stanislas
opatel99: No, unfortunately
2016-01-11 01155, 2016
stanislas
opatel99: There is a mention in GCI rules
2016-01-11 01106, 2016
stanislas
opatel99: But I would be happy to do that after GCI.
2016-01-11 01130, 2016
gcilou
opatel99, I would want to if it didn't break any rules.
2016-01-11 01141, 2016
opatel99
Well yeah. That is what I meant. I wouldn't want this to interfere with GCi and any other complications that may arise.
2016-01-11 01121, 2016
gcilou
opatel99, I'm in as long as I'm not completely lost in what's going on :)
gcilou: The point of the project would be to make sure nobody is lost. Primarily to just work together? Be it small/large/whatever. We could discuss this in 2 ? weeks.
2016-01-11 01140, 2016
gcilou
opatel99, gotcha.
2016-01-11 01105, 2016
LordSputnik
Leftmost: ping
2016-01-11 01104, 2016
Leftmost
LordSputnik, pong.
2016-01-11 01130, 2016
LordSputnik
<BANG>
2016-01-11 01140, 2016
LordSputnik
How's the thinking about relationships been going?
2016-01-11 01158, 2016
Leftmost
I keep thinking along the lines of constraints to ensure integrity. It seems complex and I haven't really mapped out exactly what those constraints would look like, but I haven't come up with good alternatives.
2016-01-11 01123, 2016
LordSputnik
ocharles: ping, about?
2016-01-11 01148, 2016
gcilou
Freso, sorry for all the pings. Do you have a good release for me to enter while I'm doing the screencast? Maybe with release duplicates so I can talk about that page in a non-vague way?
2016-01-11 01154, 2016
LordSputnik
Leftmost: I dislike having multiple revisions for what is one change
2016-01-11 01127, 2016
Leftmost
LordSputnik, when would we need multiple revisions?
2016-01-11 01114, 2016
LordSputnik
To add the new relationship set to the data for both entities
2016-01-11 01132, 2016
LordSputnik
Or are you talking about with the current plan?
2016-01-11 01126, 2016
Bookzombie has quit
2016-01-11 01139, 2016
Bookzombie joined the channel
2016-01-11 01150, 2016
Leftmost
With the current plan. I think adding relationship sets that way would have the exact same problem of integrity and just becomes an additional redirect.
2016-01-11 01113, 2016
Freso
stanislas: What mention?
2016-01-11 01115, 2016
Leftmost
My thinking is to stick with the current schema and add constraints that will ensure integrity.
2016-01-11 01126, 2016
Freso
Oh, meeting.
2016-01-11 01143, 2016
Freso
stanislas opatel99 gcilou, talk to you in ~1½ hour. ;)
2016-01-11 01154, 2016
LordSputnik
Leftmost: I hate the current schema for relationships
2016-01-11 01104, 2016
LordSputnik
it's too complicated
2016-01-11 01134, 2016
Leftmost
Unless we can come up with something completely new, it's going to be complex. The alternatives we've discussed just shift where that complexity is, and I think the tradeoffs end up being worse.
2016-01-11 01100, 2016
Leftmost
Ultimately, I'm guessing we're going to end up with complexity, because it's a complex problem.
2016-01-11 01105, 2016
LordSputnik
What are your objections to the EntityLink -> EntityLinkRevision -> RelationshipSet -> Relationships system?
2016-01-11 01130, 2016
LordSputnik
I don't think that the fact it's a separate versioning system is a problem
2016-01-11 01134, 2016
Leftmost
I do.
2016-01-11 01106, 2016
LordSputnik
Why?
2016-01-11 01119, 2016
Leftmost
For one thing, as I mentioned previously, we lose the holistic view of an entity at a given revision. For another thing, it reintroduces the problem of creating revisions for merging entities. I think it also makes it harder to understand revision graphs, it treats what the user sees as a single act of editing as multiple revisions (which I don't think is appropriate with our schema), etc.
2016-01-11 01131, 2016
Leftmost
I feel like it just complicates the overall picture instead of keeping the complexity local.
2016-01-11 01151, 2016
Freso
"we lose the holistic view of an entity at a given revision." ← I think this is a pretty big one, IMHO.
2016-01-11 01116, 2016
LordSputnik
OK, I've just been looking around, and it looks like my issue with (revision_id, bbid) PKs isn't such an issue after all - PG will allow us to have a separate index on the revision_id column, so that's fine
2016-01-11 01141, 2016
LordSputnik
So we could do one revision and update the entity data in a single revision
2016-01-11 01159, 2016
Leftmost
Hmm. What was the problem with (revision_id, bbid) without a separate index on revision_id? I thought postgres would be able to use the tuple when searching by revision_id alone.
2016-01-11 01146, 2016
LordSputnik
Well, it's presumably faster if we have an index on revision_id as well as (revision_id, bbid) when we want to query by revision_id alone
2016-01-11 01102, 2016
chirlu`
reosarevok: It wasn’t fixed or got unfixed later, but having the CORS headers won’t help anyway because the CAA is all about redirects.
2016-01-11 01120, 2016
LordSputnik
Leftmost: OK, so that solves how we do the revision stuff, but I'd like to still have 2 entities per relationship set though
