=== Andre_Gondim is now known as Andre_Gondim-afk [00:03] :) [00:03] six minutes of developer time! [00:03] luxury! [00:04] poolie: I blieve we should I mark every high bug that has not been fixed in 3 releases to low (clear it was not high enough for someone to close it) [00:04] sinzui: a quicker way to close off obsolete milestones and series would also be luxurious... [00:05] or, indeed, maybe they should just be automatically removed from the list shown for bug targets [00:05] Yes, I think you are right [00:05] when I was a boy, we had to participate constructively in four meeting in the snow before being allowed to petition for a keyboard -- and then we'd get a belting and we'd like it. [00:06] jml: You can't expect to have more than a few minutes of dev team in a team named like yours... [00:06] Er, dev *time* [00:08] :) [00:08] (although I guess it's now just Launchpad Code, rather than Launchpad Bazaar Integration :)) [00:12] hey all wats up [00:12] havent been in here for a while [00:13] poolie: apis; we could do it yourself [00:13] true [00:13] sinzui: there is a problem with projects with a greater bug-add than bug-close rate [00:13] oh, you said 'high' bugs specifically. [00:14] lifeless: yeah, I like to call that problem "software" [00:14] jml: :P [00:18] lifeless: that's actually pretty normal [00:18] for it to be slightly higher [00:18] it's somewhat proportional to the number of users + size of code base [00:22] SamB: reference? [00:22] exarkun: greater bug-add than bug-close rate [00:23] I mean, the number of open bugs open is going to be somewhat proportional to those things [00:23] hmm. too many opens. [00:23] SamB: I mean cite your sources [00:24] exarkun: sources? [00:24] I'm using a simple randomized model [00:25] where users stumble into bugs at random [00:25] ie, making stuff up :) [00:26] yeah! [00:26] exactly === bryce_ is now known as bryce [00:48] SamB: I'm saying that using a algorithm to close bugs in such situations is a problem [00:48] SamB: not that any given % of projects are in that situation === deadoomik is now known as deadoomik|sleep === matsubara is now known as matsubara-dinner [01:00] i use a complex algorithm to determine heuristically the most opportune time to ask my wife for sex. [01:01] we last made love the night before i started developing my algorithm. [01:15] mneptok: :P [01:35] Hi everybody, I wanted to know if there is way to access the number of downloads for a package in a PPA [01:35] djails, no, not currently [01:35] but it's on our roadmap [01:36] so it will likely happen in the next month or two [01:36] beuno: cool, thanks === poolie1 is now known as poolie === matsubara is now known as matsubara-afk [06:52] what am I doing wrong here? [06:52] After installing, the following source dependencies are still unsatisfied: [06:52] gccxml(inst 0.9.0+cvs20090508-0~jaunty0 ! = wanted 0.9.0+cvs20090508) [07:00] 0.9.0+cvs20090508-0~jaunty0 is before 0.9.0+cvs20090508 [07:00] oh hang on [07:00] its not [07:00] but perhaps you have a equality constraint? [07:01] 0.9.0+cvs20090508-0~jaunty0 is less than 0.9.0+cvs20090508 [07:02] tsimpson: no, its not, the -0 is after, the ~jaunty0 is between 0.9.0+cvs20090508 and 0.9.0+cvs20090508-0 [07:02] I made the same mistake on first reading it [07:02] lifeless: $ dpkg --compare-versions 0.9.0+cvs20090508-0~jaunty0 lt 0.9.0+cvs20090508;echo $? [07:02] 0 [07:02] dpkg thinks it is [07:02] ugh [07:02] dpkg is arguable insane [07:03] true, but we all must obey it [07:03] It does actually make sense, let me explain [07:04] The version comparision is not simply applied to the entire string [07:04] First, the string is split into 'version' and 'release' components at the last hyphen [07:05] Now here's the trick.... if there is no hyphen, that's the same as an empty release component [07:05] Now here's the second trick... the empty component compares identically to 0 [07:05] Hence, 0~jaunty0 is before 0 [07:05] maxb: sure [07:06] maxb: though I wouldn't say 'component' that has a rather specific meaning in debian packaging [07:07] hmm, yes. ok 'parts' then [07:07] in particular the bit under discussion is the 'debian revision' [07:08] http://www.debian.org/doc/debian-policy/ch-controlfields.html#s-f-Version has the actual algorithm [07:08] and its interesting to note that youre explanation is false, even if it fits the behaviour;) [07:08] concretely in this case, the matching digits are stripped [07:09] so this end up comparing '' to '~jaunty' [07:09] actually no [07:09] more complex... [07:09] ah right [07:09] '0' and '' is 0 and 0 [07:09] then '~jaunty' and '' is < [07:12] ...which I what I said :-) [07:13] toss in some extra fun about apt and dpkg not *quite* having the same behaviour in some obscure corner cases just for fun, to complete the technical discussion :-) [07:15] maxb: not quite the same - you missed the nondigits, digits loop in the algorithm :) [07:15] not that I'm pedantic or anything ;) [07:15] maxb: apt behaves differently? thats very odd.. [07:16] * maxb hunts a reference [07:17] believeable, but od [07:17] d [07:20] so how do I make the it accept any debian version which is equal to an upstream version? [07:20] tansell: >= upstream, < (upstream+1) [07:21] IE 0.9.0+cvs20090508-0~jaunty0 and 0.9.0+cvs20090508-1 and 0.9.0+cvs20090508-0~jaunty1 [07:21] lifeless, that makes me sad :( [07:21] tansell: indeed; something is tickling my mind though [07:21] I think it may be that -0~jaunty0 is not equal enough [07:21] and that -1 will work wth =0.9.0+cvs20090508 [07:21] I'd need to test though [07:22] 0.9.0+cvs20090508-0~jaunty0 ! = wanted 0.9.0+cvs20090508 [07:23] indeed; I said though that 0~jaunty0 may be your problem [07:23] try 1~jaunty0 [07:26] tansell@tansell:~/src/google3$ dpkg --compare-versions 0.9.0+cvs20090508-1~jaunty0 eq 0.9.0+cvs20090508;echo $? [07:26] 1 [07:28] eq really does mean *equal* [07:28] dunno :) [07:28] dpkg: --compare-versions bad relation [07:29] IE if you give it something like == it complains with that [07:29] possibly eq=equal [07:30] man page says eq is fine [07:32] man page says eq is fine? [07:32] lt le eq ne ge gt [07:32] man dpkg [07:34] you guys should steal the opensuse dependency analyser [08:49] * cafetiere is commonly seeing 'failed to connect to launchpad' errors the first time i go to a page. the error implies i should only report it if it persists. but as its nearly every time i go to a page i think it warrents a whine [08:50] typically the page is fine even on an instant reload [08:50] cafetiere: that's the bug mentioned in the topic. [08:51] wgrant, ok thanks ... stupid topic is right truncated so i missed it === mkorn is now known as thekorn === mpt_ is now known as mpt [09:26] hi, I have a problem registering a project [09:26] I'm trying to register ubuntu-md project [09:27] but launchpad says its used already by somebody else, but I can't see anything registered at http://launchpad.net/ubuntu-md [09:27] any help will be appreciated [09:30] Am I alone on this channel ? :-/ [09:30] stas: That project has probably been deactivated, so it's not visible any more. I'd recommend asking at https://answers.launchpad.net/launchpad/+addquestion [09:30] You can probably steal that name. [09:30] wgrant: thanks a lot === deadoomik|sleep is now known as deadoomik [09:41] how do multiple ppa names work [09:41] I have a ppa called releases [09:41] putting that in the dput conf -> fails [09:41] or is the first ppa created special? [09:42] Rejected: [09:42] Could not find PPA named 'releases' for 'subunit' [09:42] is the exact error [09:44] lifeless: 'releases' there is the display name. [09:44] The first PPA is always named 'ppa' [09:44] So yes, the first PPA is special. [09:44] Subsequent PPAs have an additional field on the creation page. [09:44] if it has confused you, though, it's a bug. [09:45] thanks [09:46] we could certainly make it clearer providing local instructions for 'installing packages from this ppa' and 'uploading packages to this ppa' [09:46] by 'local' I mean related to the context PPA. [09:47] cprov: And having an uneditable field for the name when there are no existing PPAs. [09:47] stas: it was deactivated. I just renamed the old, so please reapply/register when you're ready. [09:47] spm: thanks :) [09:47] it works [09:48] wgrant: uhm, sorry, I don't get it. [09:48] Hello there [09:49] cprov: At the moment, the first PPA creation page doesn't show the name, but subsequent ones have a field for it. Perhaps the initial page could show the name, but with just the text 'ppa' rather than a textbox. [09:49] That would make things less confusing, as lifeless clearly took the 'Display name' field to be the URL name. [09:49] Could anybody tell me how long it usually takes for an SVN import to kick off? [09:50] LenzGr: It has to be manually reviewed, but I've found it normally happens within a working day or two. [09:51] wgrant: OK, then I guess I need to be patient :) - I submitted one 42 hours ago... [09:51] wgrant: yes, that can be done, but will only be seen by people who can 'edit' the PPA. [09:51] * LenzGr wants to encourage a project to choose bzr instead of git [09:51] cprov: I've filed a bug [09:51] cprov: lifeless' problem was that he assumed he could name the initial PPA. [09:52] wgrant: it asked me for a name [09:52] cprov: Because there was a name (well, 'display name', but 'name') all the same. [09:52] Thus I would like to demonstrate how easy the migration from SVN to bzr/launchpad is... [09:52] I assumed I understood what it was asking :) [09:52] lifeless: thanks [09:52] in fact in this case, I would rather have liked to be able to edit the first ppa's name [09:53] as it would be clearer to people setting up apt [09:53] Right. [09:53] I don't see why it's locked. [09:54] Should the tuple of data within an Unauthorized really be returned as content? The first element is the repr of the object in question, which seems like it could be a security risk. [09:54] (when using the API, this is) [09:55] wgrant: kinda legacy. [09:55] lifeless: why didn't you create a new PPA with the name you wanted ? [09:56] lifeless: I see you are concerned about used who already use it, but renaming would break it in the same way, right ? [09:57] cprov: I'd be happy to rename this; its new [09:57] cprov: I wasn't given the choice to set the real name, only the display name [09:59] lifeless: right, the default PPA name is locked as 'ppa'. That's just the legacy thing, I thought you had a different problem. [09:59] I was confused [10:00] because I was asked for a name, I gave it, but it didn't work in dput etc [10:00] It can't be a legacy thing; the form must have been changed a couple of releases ago to explicitly hide that field when no other PPAs exist. [10:00] lifeless: anyway, in this circumstance you just activate the default one and don't use it, like a misnamed branch in LP. [10:00] cprov: mmm, that seems... hard for people to discover [10:01] I'm now fine; data learnt and on my merry way [10:01] wgrant: the legacy I refer to is the IPerson.archive property, it points to the default PPA. [10:01] I'd like to prevent others failing/make it easier, and thats why the bug report ;) [10:01] cprov: Right, I wondered about that. [10:01] wgrant: yes, that's the underlying problem that needs fixing. [10:02] cprov: But all the code handles IPerson.archive being None fine, so absence of an Archive named 'ppa' would surely not break anything? [10:02] Apart from the redirect, obviously. [10:03] wgrant: redirect, signing_key propagation and API would break atm. [10:03] but they are all fixable. [10:03] cprov: Didn't you fix the signing key propogation bug a couple of days back? [10:03] And the API is buggy. [10:04] wgrant: yes, I did, and it rely on IPerson.archive [10:04] cprov: Wasn't the problem that it *couldn't* rely on that, because it might have not had a key due to being disabled? [10:04] wgrant: but it can easily change from name='ppa' to 'the oldest one' [10:04] That's true. [10:48] Hello folks. [11:17] Any one alive? [11:19] * qball checks the bodies for a pulse [11:19] no [11:20] there's one [11:20] * hyperair points at qball [11:21] * wgrant finishes qball off. [11:21] Not any more! [11:21] * hyperair points at wgrant [11:21] but now there's another! [11:21] This is awkward. [11:22] matti: Have you a question? [11:23] heheh [11:26] Yes. [11:26] Sorry. [11:27] I have sent an e-mail to feedback@ and I was wondering if anyone had a chance to look at it... as I haven't heard anything back :) [11:28] matti: That's going to require an actual Launchpad person to answer, and nobody speaking here recently is one. Maybe one will appear soon. [11:29] wgrant: Oh I see :) [11:32] matti: Unless your question needs to remain private, you might get a better response in more public places (eg. here or Launchpad Answers) [11:35] wgrant: I was asking about account merge. [11:35] matti: Do you have access to both email addresses? [11:36] wgrant: No. [11:36] ;] [11:36] wgrant: I know the procedure, no worries :) [11:36] matti: OK. [11:37] wgrant: I just want to confirm that they actually got my e-mail. [11:37] wgrant: As it is some time right now since I have sent it :) [11:38] matti: How long? [11:38] Since monday :) [11:39] Ah. [11:39] Hi to all... I sent a request a few days back to request for a ppa I created (thinking I could remove it) I have no use for a second ppa attached to my account. How do I or who do I need to talk to? === deadoomik is now known as deadoomik|away [11:42] * tumbleweed assumes someone knows that the *.launchpad.net cert (used for edge.lp.net) expires tomorrow [11:43] I remember last time it expired.. it was a few hours before they replaced it, IIRC. === root is now known as Guest53949 [12:03] Is Rosetta amongst the Launchpad stuff due to be free'd? [12:04] Odd_Bloke: Yes. [12:04] \o/ [12:04] wgrant: Thanks. :) [12:05] wgrant: That's happening RSN, not in a later release? [12:05] Odd_Bloke: Parts of Launchpad (including all of Rosetta, as far as is publicly known...) will be open sourced on July 21st. [12:06] Ah, excellent. [12:06] Elleo: 12:05:55 < wgrant> Odd_Bloke: Parts of Launchpad (including all of Rosetta, as far as is publicly known...) will be open sourced on July 21st. [12:06] ah, cool [12:06] now we just need to convince mattl to be non-free for 2 months... :/ === mrevell is now known as mrevell-lunch [13:01] bug 360846 [13:01] Launchpad bug 360846 in launchpad-foundations "appserver isn't recovering like it should causing too many oopses" [Critical,In progress] https://launchpad.net/bugs/360846 [13:01] I would like to know why malone didn't parsed the command I gave via the email interface. The comment is https://bugs.launchpad.net/ubuntu/+source/amsn/+bug/371103/comments/12 , and I expected it to change the assignee and to add some subscribers. Can anybody tell me what I've done wrong? [13:02] Ubuntu bug 371103 in amsn "aMSN fails to launch when msntranslator or gnotify plugin is used" [Undecided,Fix committed] [13:02] petski, you didn't add a space [13:02] Secondly, "-----BEGIN PGP SIGNED MESSAGE-----\nHash: SHA1" is hidden by default, why isn't my "SIGNATURE" as well? [13:03] beuno: where should I have added some space? [13:03] petski, before the commands [13:03] status confirmed [13:03] ahhh! [13:04] It's in bold ... https://help.launchpad.net/Bugs/EmailInterface ... :S ... sorry [13:06] petski, happens to the best of us :) [13:06] not sure what happened with the GPG sig [13:06] maybe BjornT or gmb can enlighten us [13:06] thanks beuno :) [13:11] wsfgn? [13:11] * gmb reads scrollback [13:12] petski: Good question. I'm not sure. I suspect that's a bug. [13:13] gmb: could you please verify if it's a bug or not. If it's the case, I'll file a report if that's needed. [13:15] petski: Well, it should be hidden and it isn't, so that's a bug. === mrevell-lunch is now known as mrevell [13:15] I mean, it might be that there's some reason for it that I don't know about, but if that's so we'll just mark the bug Invalid. [13:15] petski: So please file it :) [13:16] gmb: I will, thanks :) [13:19] I hate this crack [13:19] kiko, get a new dealer already [13:21] erm.. git tags are like bzr branches? [13:21] savvas: git tags are like bzr tags. git branches are like bzr branches. [13:22] git confusion is also like nothing ever seen before. [13:22] oh ok :) [13:22] I thought bzr doesn't use tags :P [13:22] They're not quite the same, but they're similar. [13:22] bzr can attach a tag to label a revision. [13:22] gmb: looks like a dup of LP #161822 [13:22] Launchpad bug 161822 in malone "new bugs by email requires gpg signature but does not strip it from bug report" [Medium,Triaged] https://launchpad.net/bugs/161822 [13:23] wgrant: are the tags visible somehow in launchpad? [13:23] savvas: Only in loggerhead, I suspect. [13:23] wgrant, we don't show them in loggerhead just yet :/ [13:24] petski: Okay, thanks. [13:24] beuno: Oh. [13:24] beuno: is there an open bug wishlist for it? :) [13:24] savvas, there is [13:24] and there [13:24] is a patch [13:25] which doesn't quite fit [13:25] * beuno looks for the bug # [13:25] bug 246739 [13:25] Launchpad bug 246739 in loggerhead "tags are not available" [Wishlist,Confirmed] https://launchpad.net/bugs/246739 [13:25] cool, thanks! [13:26] savvas, poke me if you put together a patch ;) [13:26] it should be very simple to do [13:27] will do [13:27] I've never seen the loggerhead, but I'll try :P [13:27] *loggerhead code [13:27] savvas, there's some code in a patch on the bug [13:28] to get you started [13:28] bzrlib is super nice [13:29] oki doki [13:43] Is this normal ??? https://launchpad.net/builders [13:44] as most builders are in a manual state? [13:44] VK7HSE: That's to stop Karmic builds, as there's a bit of a disaster being fixed right now. [13:45] Although I'm not sure why they care if PPAs are affected. [13:45] Oh ok!... just push a new release of Me TV .... [13:46] anyway I't will build when its ready then ! :) [13:46] It will, yes. [13:50] mpt: Did you mean to mark the public, triaged datepicker change bug as a duplicate of a private one? That doesn't seem like an entirely good idea (although it did reveal a security hole) [13:51] wgrant, oops, no, I should have made the original public. I've just done that now. [13:51] And please do report the security hole. :-) [13:51] mpt: Already half-way through. Thanks. [13:53] * wgrant kicks the appservers again. === gord_ is now known as gord === rowinggolfer is now known as notRowingGolfer [14:22] hey savvas === abentley1 is now known as abentley [14:22] are you the gent who took a look at openmolar? [14:23] hey :) [14:23] erm.. [14:23] refresh my memory :) [14:23] med student from the baltic state? [14:23] the very same :P [14:24] I was checking your profile. impressive. [14:24] oooh you had a problem with your locks [14:24] ah that's right. [14:24] thank you :) [14:24] trying to get by in both "lifes" hehe [14:24] I was putting 2 and 2 together and making 5 [14:25] I assumed that you had a look at the project because it is loosely a medical database app [14:25] and you'll probably agree the medical profession needs such apps? [14:28] it does, badly if you ask me :) [14:28] does it support sqlite by the way? [14:29] no, but that would be trivial to do. [14:29] the python database api really nicely pulls all that stuff together. [14:30] I think being database agnostic is important nowadays [14:30] with big bad oracle on the march [14:30] it would be great addition if you ask me :) you could have an all-in-one binary and another one split to server/client [14:30] Soyuz people, can anyone tell me how often cron.daily is run these days? And is it the same for the development distribution as the stable versions? [14:31] sladen: Hourly. [14:31] savvas: that's a fair suggestion. I'll look into it. [14:31] sladen: And yes, it's the same for all series. [14:31] sladen: (although I'm not a Soyuz person) [14:31] wgrant: ta. [14:31] but he knows more about soyuz than some soyuz people ^^ ;) [14:31] sladen: PPAs are published three times an hour, though. [14:31] noodles775: pfft === asac_ is now known as asac [14:43] Any LP person present? :) [14:43] matti, there's like 30 of us [14:43] :) === BjornT_ is now known as BjornT === notRowingGolfer is now known as rowinggolfer [15:13] Hello! What happens when I push a project to lp:~jblount/project-name/awesome-branch, then months later push a new unrelated branch to the same place? [15:13] jblount, it will tell you they are unrelated [15:13] unless you push --overwrite [15:15] beuno: Nice, thanks. [15:15] (for what it's worth, I've been giving my branches more and more ridiculous names because I was nervous about what might happen) === sinzui changed the topic of #launchpad to: https://launchpad.net/ | Help contact: sinzui | Join https://launchpad.net/~launchpad-users | Channel logs: http://irclogs.ubuntu.com | We're having some intermittent problems, we're working on them, see bug 360846 [15:16] jblount, bzr has some smart built into it :) [15:17] is there a way to regenerate a diff without doing resubmit on a merge proposal? [15:17] * jblount notes that bzr will *attempt* to protect him :) [15:17] a mergee updated his branch with fixes, and I want to review them but I dont want to lose the review history [15:18] lamalex_2, there's some API to do it [15:18] and I think a script [15:18] rockstar would know [15:18] * rockstar knows nothing [15:19] I mean I know how to do it in my bzr tree [15:19] anyone know how to set-up an SSH master connection to launchpad? [15:19] but I'm at work, not doing my work :) [15:19] lamalex_2, so, which diff do you want to update? [15:19] reviewing merges instead [15:20] rockstar: https://code.launchpad.net/~gotomail/do-plugins/TimerApplet/+merge/5456 [15:20] er [15:20] thats not it [15:20] https://code.edge.launchpad.net/~ulrik-sverdrup/do-plugins/screen/+merge/5680 [15:20] i always forget windows cant select to copy [15:21] lamalex_2, so, when you submit a merge proposal, it creates what we call a "Review Diff" that can't be changed. However, if you look at the API (and you do, right? :) there's also the PreviewDiff that can be changed/updated with an app like MAD. [15:21] lamalex_2, https://edge.launchpad.net/mad [15:22] well if I wanted to do it myself I could just do bzr diff no? [15:23] or does this actually update the LP ui [15:23] lamalex_2, so, I have a plugin that's now part of bzr-launchpadplus that uses the command line to do reviews, and it generates diffs for me. [15:23] * wgrant hasn't worked out why the review diff is separate. [15:24] ah, sftp can maintain a master connection [15:24] lamalex_2, MAD links the preview diff so that you can download it, but that's the only UI change. [15:25] yah, it seems like I could just use bzr diff. I dont see the advantage of MAD [15:27] hey all, how do I join a team in launchpad? [15:28] the review diff not updating is going to be quite a pain I fear [15:29] james_w: It thoroughly confused some members of my team when we started using it. [15:29] I can imagine [15:29] lamalex_2, well, if you run MAD as a cron job like we do in the Launchpad team, we always have it. [15:29] it definitely /should/ update [15:30] removing it when it is out of date would be better than leaving it unchanged [15:30] lamalex_2, it's a fundamental part of a review. [15:30] what's a fundamental part of review [15:30] No, a correct diff is a fundamental part of a review. [15:31] An incorrect one is not. [15:31] ^ truth [15:31] * beuno agrees and thinks we should have an "update this diff" button or an automatic-update-on-page-load [15:32] beuno: One can already update the diff, but only the secondary one. [15:32] is it resource consumption that means that it's not always up to date? [15:32] wgrant, but not from the webui? [15:32] The one that is only linked to in one place, and should be the same as the displayed one, but is in fact separate. [15:32] james_w, I think that's the argument, yes [15:32] cant you check if there are new revs since it was generated? [15:32] and only generate it if there are newer revisions? [15:33] lamalex_2, yes. Still icnredibly resource intensive [15:33] For some reason which I cannot fathom, there are two diffs. One is rendered nicely on the MP page, but the other is just linked. [15:33] seriously, I must be missing something big: where in the lp ui do I go to join a team? This is hte team I'm looking to join: https://launchpad.net/zope2 [15:33] One can update the linked diff through the webservice, but not the displayed one. [15:33] ChrisW, that's not a team :) [15:33] ChrisW, https://edge.launchpad.net/~zope2-dev [15:33] well, how do I get to handle bugs on it? [15:34] I guess that's the team you want to join? [15:34] and where did you find that? [15:34] You can perform most bug actions without being a member of any team. [15:34] ChrisW, "Driver: Zope 2 Developers" [15:34] beuno: keeping preview diffs up to date would be massively more expensive than keeping review diffs up to date I assume [15:34] clicked on the link [15:34] james_w: How? They should be identical. [15:34] wgrant: nope [15:34] james_w, I don't know the details, thumper does [15:35] beuno: what's "Driver" supposed to mean?! [15:35] wgrant: review diff can only change when there are new revisions in the source branch, preview diff will change when there are new revisions in either [15:35] so pushing a new revision to trunk would mean updating every preview diff [15:35] ChrisW, "the guy in charge"? I don't know, I kinda hate that word as well [15:35] james_w: Oh, right, I think somebody told me that ages ago. [15:36] * ChrisW finds launchpad baroque and difficult to use [15:36] james_w: Is it the preview diff or the review diff that's visible? [15:36] you would only update the review diff when someone pushed up new revisions to the source branch, which would presumably be in response to review comments [15:36] certainly wouldn't use it for any of my own projects... [15:36] wgrant: the review diff is the visible one I believe [15:36] james_w: So it's the cheap one, but it's immutable!? [15:36] yeah, that's my understanding [15:37] ChrisW, I'm working on it :) a lot of changes coming down the pipe in the next few months [15:37]