[12:24] <mapreri> are NBS binaries removed regularly/automatically from the archive?   /cc cjwatson
[12:30] <cjwatson> mapreri: semi-automatically - i.e. http://people.canonical.com/~ubuntu-archive/nbs.html reports on them, we process the ones that don't have reverse-deps pretty frequently
[12:30] <cjwatson> mapreri: but the ones that have reverse-deps there need manual attention
[12:30] <cjwatson> mapreri: some of those may be false positives (recommends or alternatives-where-first-is-still-available), which are hard to deal with automatically
[12:32] <mapreri> cjwatson: i was looking because i removed pbuilder-uml yesterday, but I don't see it in the list (and britney complains about it).  does it have to be in the release pocket?
[12:32] <mapreri> (which kinda defeats the purpose of that britney check, maybe)
[12:33] <cjwatson> mapreri: yes.  there's a corner case in what you're thinking of, it's only a problem if a package goes through multiple versions in -proposed without migrating and one of the non-final versions has NBS binaries
[12:33] <cjwatson> that has no automatic report but every so often I go through excuses and work out what to remove ...
[12:34] <mapreri> well, this one hit proposed yesterday for the first time, before there was nothing in proposed.
[12:34] <cjwatson> though in this case ... yeah, not sure what's up there
[12:34] <cjwatson> probably an extra-weird corner case to do with arch: all vs. any
[12:35] <cjwatson> the whole business of -proposed being a partial suite makes things complicated in britney
[12:35] <mapreri> arch:all is so funny
[12:36] <cjwatson> mapreri: removed, anyway
[12:36] <mapreri> cool
[12:38] <mapreri> cjwatson: + do you confirm that what's keeping libpodofo out of release is the need of a transition for it?  update_excuse is silent, but i learn out to read update_output...
[12:38] <mapreri> learnt*
[12:39] <mapreri> even if it seems to write stuff (and behave, maybe?) a bit differently than debian's britney
[12:41] <mapreri> seems like usrmerge might need some poking, btw.
[12:43] <cjwatson> mapreri: needs rebuilds of calibre, krename, and scribus, yes
[12:43] <cjwatson> mapreri: I believe usrmerge needs an initramfs-tools merge
[12:44] <cjwatson> mapreri: also coreutils
[12:44] <cjwatson> Conflicting with the current version of an Essential package isn't going to go well for it :-)
[12:45] <mapreri> oh. eheh :)
[13:14] <flexiondotorg> cyphermox, infinity I have an idea I'd like to discuss.
[13:14] <flexiondotorg> Not sure if it is feasible.
[13:15] <flexiondotorg> Need some wise heads to bounce ideas off.
[13:57] <LocutusOfBorg>  In debian the libpng16 transition is mostly "done" I mean, we patched the sources except for a few build failures, and we are waiting for the release team to proceed
[13:57] <LocutusOfBorg>  I did ~30 NMUs and they are pending, and I'm planning to merge ubuntu whenever possible
[13:57] <LocutusOfBorg> do you think we can arrange a transition for xenial?
[13:58] <cyphermox> flexiondotorg: shoot
[13:58] <flexiondotorg> Would it be possible to make another image for Ubuntu MATE that just use the ubuntu-mate-core meta-package?
[13:59] <flexiondotorg> I'd like to create an Ubuntu MATE Basic edition which is heavily stipped down.
[14:03] <flexiondotorg> cyphermox, ^
[14:32] <apw> cjwatson, initramfs-tools merge, ugg
[14:33]  * apw looks at usrmerge
[14:34] <apw> cjwatson, yes so it does, fun, i've got that merge on my list as soon as i get the previous version uploaded
[18:26] <ginggs> hi, is there a reason we don't have ceres-solver in ubuntu ? https://tracker.debian.org/pkg/ceres-solver I don't see it in the sync blacklist and it was uploaded to unstable at the end of november.
[18:45] <doko> ginggs, no idea, just synced
[18:46] <ginggs> doko, thanks
[18:53] <cjwatson> ginggs: The reason was that it was previously removed from Ubuntu as a consequence of a removal from Debian, and auto-sync defers all packages that have previously been removed for manual attention.
[18:53] <cjwatson> In this case I agree it was fine to reintroduce it.
[18:54] <ginggs> thanks, cjwatson, is there a list of these packages somewhere?
[18:55] <cjwatson> http://people.canonical.com/~ubuntu-archive/auto-sync/current.log
[18:55] <cjwatson> pretty raw form
[18:55] <cjwatson> doesn't divide down by reason, you need to read through it
[21:27] <doko> cjwatson, please could you build gradle using the gradle binaries from unstable? validated that it builds
[21:31] <flexiondotorg> cyphermox, Yo
[21:50] <cyphermox> flexiondotorg: hey
[21:50] <flexiondotorg> Hi
[21:50] <flexiondotorg> Back home now.
[21:51] <flexiondotorg> I was enquiring about the notion of creating an Ubuntu MATE Basic image.
[21:51] <flexiondotorg> Is that something the build system can accommodate?
[22:14] <cyphermox> flexiondotorg: in theory yes
[22:15] <cyphermox> flexiondotorg: what comes to mind is what level of simplification are you looking at? because we already have ubuntu-core / ubuntu-server which can be pretty minimal
[22:16] <flexiondotorg> cyphermox, Basically an Ubuntu MATE that is "built" using just the ubuntu-mate-core meta package.
[22:16] <cyphermox> at the point where you're looking at a command-line setup, I start to have a hard time how you can both do a minimal setup and a branded one
[22:16] <flexiondotorg> So still Ubuntu MATE, with desktop. But with much of the bundled applications removed.
[22:17] <flexiondotorg> There is a demand for this for people to make bespoke setups for stuff like Steam or Kodi or whatever.
[22:18]  * xnox uploaded something into ubuntu, instead of ppa =(
[22:18] <cyphermox> yeah, I can picture the kiosk idea to some degree
[22:18] <xnox> nodejs - 4.2.6~dfsg-1ubuntu3
[22:20] <xnox> i did block-proposed, which is good enough.
[22:21] <doko> mdeslaur, ^^^
[22:21] <xnox> block 1537922
[22:21] <xnox> bug 1537922
[22:28] <flexiondotorg> cyphermox, So how can I progress this?
[22:29] <infinity> flexiondotorg: To be fair, starting from a desktop/live installer ISO isn't how I'd build a bespoke kiosk image, I'm not sure many would.
[22:29] <knome> cyphermox, fwiw, the xubuntu team has a similar effort underway.
[22:29] <infinity> flexiondotorg: Doing a d-i netboot and then adding the packages you want would be a more common approach.
[22:30] <flexiondotorg> infinity, I'm not planning to make a kiosk image.
[22:30] <knome> infinity, since the "core" name seems to be the blocker for some people, would you think "base" would be more approachable for them?
[22:30] <flexiondotorg> What the UBuntu MATE community are asking for a minimal desktop.
[22:30] <infinity> flexiondotorg: Well, whatever "bespoke setups" people might have.
[22:30] <cyphermox> knome: flexiondotorg: I'm not sure how to deal with this. My understanding is that new flavours would have to go through TB approval, but I'm not sure if it qualifies as a new flavor or as a JFDI case.
[22:31] <flexiondotorg> People can then use it as a "base" to make their own thing.
[22:31] <flexiondotorg> cyphermox, Is this not is the same category as the alternate image for Lubuntu?
[22:31] <cyphermox> (I'm leaning towards jfdi but hey)
[22:31] <knome> flexiondotorg, we can help with the technical side
[22:32] <knome> or, the social request side, or whatever
[22:32] <flexiondotorg> knome, Thanks.
[22:32] <cyphermox> flexiondotorg: technically it's not much more than having a seed and making an image with that seed
[22:32] <knome> infinity, what flexiondotorg is proposing is basically the same as the "xubuntu core" image
[22:32] <infinity> knome: I would certainly prefer to see people stop using "core" for this, but I'm more trying to understand the use-cases.
[22:32] <infinity> knome: And yes, I know what you're both been proposing. :P
[22:32] <knome> for us, the usecase is that not all people want firefox, thunderbird and libreoffice
[22:33] <knome> (and a lot more)
[22:33] <flexiondotorg> infinity, OK so if we go with -base (for example) are there any technical reasons blocking this?
[22:33] <knome> so we want to offer a smaller image for them so they can save bandwidth and installation time
[22:33] <flexiondotorg> infinity, What knome said. Same from Ubuntu MATE.
[22:33] <knome> the use cases are endless
[22:33] <infinity> Okay, but in both cases, this is targetted to end users?
[22:33] <flexiondotorg> There is a large user group who what to currate their own system.
[22:33] <knome> for us, yes, totally
[22:34] <knome> advanced end users
[22:34] <flexiondotorg> Sort of "make your own adventure".
[22:34] <knome> we will make sure the "core" image is working though, eg. not just stripping packages like headless chicken
[22:34] <flexiondotorg> Again, what knome said.
[22:34] <infinity> Kay.  I think I misread flexiondotorg originally, feeling like this was targettted at people doing custom images, and a live ISO is a horrible starting point for third parties to customize images.
[22:34] <flexiondotorg> Advanced/experienced users.
[22:34] <knome> for us, there might be some replacements for some packages too
[22:35] <flexiondotorg> For Ubuntu MATE it is simply a build from ubuntu-mate-core.
[22:35] <knome> i don't know if anybody in our team has thought about creating customized images from that image, but i guess it helps with that too
[22:35] <flexiondotorg> He says "simply" assuming this is indeed simple.
[22:36] <knome> flexiondotorg, we have a xubuntu-core task set up already.
[22:36] <knome> flexiondotorg, so even in that regard, the same as you
[22:36] <knome> now we basically just want an image for that task.
[22:36] <infinity> It's relatively simple, other than exploding your QA matrix.
[22:37] <flocculant> infinity: our 'core' lives on the tracker already in a basic state
[22:37] <flocculant> all *I'm* waiting for now is it to be  a daily build - so I can actually ask people to test it
[22:37] <flexiondotorg> I have a small team who have said they will test a "base" image.
[22:38] <flocculant> http://iso.qa.ubuntu.com/qatracker/milestones/351/builds/105604/testcases
[22:38] <knome> exactly what flocculant said. we want it to start running in the dailies "officially"
[22:38] <knome> (everything else is prepared for us, except for the merge...)
[22:38] <infinity> So, before we have an explosion of things named core, could we maybe all agree on consistent branding for "a flavour without all the extra desktop applications"? :P
[22:39] <flocculant> :)
[22:39] <knome> infinity, i've got an "ack" for being okay with "base", if the "core" name REALLY is the blocker
[22:39] <flexiondotorg> infinity, I "think" Xubuntu and UBuntu MATE can agree on "base". I'm certainly OK with that.
[22:39] <knome> (we want this to progress too, but we don't want desktop/dvd)
[22:39] <flocculant> 2 of our release team would ack that - pretty sure the other one would too
[22:39] <infinity> knome: I've said before that I wouldn't block on "core" in your case, but I'm less thrilled about it, the more people follow in your footsteps. :P
[22:39] <knome> (as was previously suggested by a few people)
[22:39] <knome> infinity, i understand and remember you said that
[22:40] <flexiondotorg> infinity, So is "base" agreeable from your point of view?
[22:40] <knome> core is our #1 bet, because we've already advertised it as that on our website and more, but if the cd image doesn't approve that, then we'll live with "base"
[22:40] <infinity> Something like "base" or "basic" would be fine.
[22:40] <knome> *cd image team
[22:40] <infinity> And you could maybe rebrant the other ISO as "Full" or something, to be descriptive.
[22:41] <knome> infinity, as you probably can imagine, we don't want to do another merge proposal with the new "base" name unless it's really ok with the cd image team
[22:41] <infinity> So people get an idea that ISO 1 has a basic setup, ISO 2 is what you want to download for gandpa so he doesn't have to hunt for a web browser or word processor after install.
[22:41] <knome> yes, that's one option
[22:41] <flexiondotorg> infinity, Can do.
[22:42] <knome> the downside of that is we need to rename our main ISO at that point
[22:42] <flocculant> not completely sold on renaming what we currently have tbh
[22:42] <knome> but i guess you don't mind if it's base/desktop?
[22:42] <infinity> Well, not in the machinery or anything, just in pretty charts on websites.
[22:42] <knome> yeah, pretty charts are pretty charts
[22:42] <flocculant> if the new one is -base then it stands to reason the bigger download isn't base imo
[22:42] <flexiondotorg> And already have the -desktop suffix.
[22:43] <knome> infinity, so... how do we go from here?
[22:43] <infinity> I like "basic" over "base", if we're bikeshedding.  We're describing a flavour of an OS, not a chunk of it.
[22:43] <infinity> So, an adjective is better.
[22:43] <infinity> But I'll be happy with either of those, if you two agree.
[22:43] <knome> in my mind, basic refers to being "simple", or not the "premium" version, so i prefer "base"
[22:44] <knome> (besides, i would still consider our desktop image the "basic" (or regular) image)
[22:44] <infinity> Yeah, fair enough on the basic != premium.  I think that's why I suggested "light" or "minimal" last year.
[22:44] <flocculant> infinity: in simple terms for me - our main image is stuff added to the base image iirc now
[22:45] <knome> minimal has the potential to be messed up with the mini ISO again, so not that
[22:45] <infinity> I don't think so, TBH.
[22:45] <knome> (and xubuntu core is not really "minimal")
[22:45] <infinity> We don't promote the mini ISO in any way, only advanced users even know they exist.
[22:45] <knome> and for xubuntu, the word "light" is a red light, because we have been previously targeting low-end machines
[22:46] <knome> (which need a "light" OS)
[22:46] <infinity> But yeah, I could live with base, if you two like that best out of the non-core options.
[22:46] <knome> so i would say we want to avoid that too
[22:46] <knome> flexiondotorg, please ack once more that you are fine with "base"
[22:46] <knome> ^ we are
[22:46] <flexiondotorg> I am fine to "base".
[22:46] <knome> great!
[22:47] <infinity> I would definitely recommend you both do some Microsoft-style charts on your websites to tell people the difference.
[22:47] <knome> infinity, do you need to run this by other cd image members, or are you ready to merge?
[22:47] <knome> infinity, i'll pass on that to our marketing team
[22:47] <infinity> "You get all this great stuff, plus a web browser, office suite, etc preinstalled".
[22:47] <knome> wait, i'm on that team
[22:47] <infinity> Cause, yay confusing.
[22:47]  * knome facepalms
[22:47] <knome> ;)
[22:47] <infinity> Hahaha.
[22:47] <flexiondotorg> infinity, Ubuntu MATE will do that.
[22:47]  * flocculant is off now - night all
[22:47] <knome> infinity, we'll likely do it the other way, but yeah, marketing semantics
[22:48] <infinity> knome: Anyhow, yeah, if we s/core/base/ across the board, I think I'll be happy revisiting your MP.  I haven't read it in a while to remember if it was otherwise okay, but I don't recall it being awful.
[22:48] <knome> i don't think there is anything else
[22:48] <knome> when can we expect to hear from you about the name?
[22:48] <infinity> You two agreed, I'm fine with the name now.
[22:48] <infinity> So, you've heard from me.
[22:49] <infinity> It's more about me finding a bit of time for a final review and merge.
[22:49] <knome> ok,
[22:49] <knome> so should i ask Unit193, who proposed the merge, to add you as another reviewer?
[22:49] <infinity> base has no conflicts with other products, and bonus points if we (in)formally decide that this is what we call flavours with fewer bits installed.
[22:49] <infinity> I'm a fan of consistency where we can get it.
[22:49] <knome> i'll applaud for consistency too.
[22:49] <flexiondotorg> +1
[22:50] <infinity> knome: I'm in the team that's proposed as a reviewer, I'm sure, but a fresh pointer to the MP itself wouldn't hurt.
[22:50] <knome> infinity, slangasek claimed the review, so you aren't... but here it is: https://code.launchpad.net/~unit193/ubuntu-cdimage/xubuntu-core/+merge/268167
[22:52] <infinity> knome: Right.  This'll need some fixing up for name change proposals.
[22:52] <knome> infinity, acknowledge, but we are not willing to do that unless we can be certain that the new name is okay.
[22:52] <knome> infinity, as you can probably imagine...
[22:52] <infinity> knome: And I might still agree with slangasek that this shouldn't be whole new projects, but types (like desktop/dvd), which is a transparent thing to the end user, we can name the ISOs and label the webpages however we want.
[22:53] <knome> tbh, i don't understand most of the technicalities here, i'm mediating between the two parties to get this done :)
[22:53] <infinity> Just from a POV of backend machinery and paperwork, it makes more sense for them to be subtypes of "xubuntu" rather than whole new projects.
[22:53] <infinity> Yeah, that's cool.
[22:53] <Ukikie> Yeah I'm pretty sure that's exactly how it is now, just not using the 'dvd' one.
[22:54] <infinity> I'm hip deep in other work right now, can you set yourself a reminder to yell at me about this near the end of the week?
[22:54] <infinity> Oh, indeed, it just created a new type.
[22:54] <knome> infinity, ACK.
[22:54] <infinity> I didn't read the MP, just the comments. :P
[22:54] <knome> thin the comment was about the earlier revision
[22:54] <infinity> So, we could just make a "base" type instead, doesn't really bug me.
[22:54] <knome> then we updated the MP
[22:55] <knome> now we want to make sure we don't do more work unless we know it will be the last time we change this
[22:56] <infinity> Ahh, no, it's still a different project in there.
[22:56] <infinity> So, it's a different project *and* a new type.
[22:56] <infinity> Which doesn't make much sense.
[22:57] <infinity> But yeah, poke me violently on Thurs/Fri, and we'll get this moving for realz this time.
[22:57] <knome> infinity, i'll poke related people
[22:57] <knome> thanks!
[22:58] <knome> we'll try to help mate get their stuff in line before that too, so you can kill two birds with one stone
[23:05] <flexiondotorg> infinity, cyphermox Thanks for helping with this.