=== asac_ is now known as asac === nellery_ is now known as nellery === nellery_ is now known as nellery === nxvl_ is now known as nxvl [13:59] #startmeeting [13:59] Meeting started at 07:59. The chair is mcasadevall. [13:59] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [13:59] Who's here? [14:00] What's the meeting? [14:00] Mobile [14:00] morning StevenK [14:00] * StevenK shores [14:00] * ScottK will be here if it gets qt4-x11 fixed on armel. [14:00] ScottK, DEPWAIT new build hardware [14:01] hey [14:01] hi [14:01] Grumble. [14:01] morning lool plars [14:01] mcasadevall: Is there timeline for this? [14:02] ScottK, soonish I hope [14:02] Have we started the meeting yet, or did I just miss the #startmeeting? [14:02] StevenK, you just missed #startmeeting [14:02] Hi [14:02] * mcasadevall pokes ogra, dyfet [14:02] oh [14:02] paulliu isn't even online [14:05] I guess that's everyone who's going to show [14:05] [link] https://wiki.ubuntu.com/MobileTeam/Meeting/2009/20090811 [14:05] LINK received: https://wiki.ubuntu.com/MobileTeam/Meeting/2009/20090811 [14:05] [topic] action item review [14:05] New Topic: action item review [14:05] [topic] # [14:05] NCommander to investigate https://bugs.launchpad.net/ubuntu/+source/vnc4/+bug/338148 (co) [14:05] New Topic: # [14:05] c/o [14:06] Launchpad bug 338148 in vnc4 "Needs new version from Debian: fails to build with removal of mesa-swx11-source" [High,Triaged] [14:06] [topic] GrueMaster to retest on i386 https://bugs.launchpad.net/ubuntu/+source/linux/+bug/337809 [14:06] New Topic: GrueMaster to retest on i386 https://bugs.launchpad.net/ubuntu/+source/linux/+bug/337809 [14:06] Launchpad bug 337809 in linux "APIC error on CPU 0" [Medium,Triaged] [14:06] GrueMaster, you awake? [14:06] guess not ... [14:07] [topic] Spec Review [14:07] New Topic: Spec Review [14:07] lool, so if I understand our new spec review now, we all update the status, and just give a small update individually in th emeeting, or am I way off? [14:07] Right [14:07] [link] http://paste.ubuntu.com/251356/ [14:07] LINK received: http://paste.ubuntu.com/251356/ [14:07] it's what I have in progress right now [14:07] There are some bugs [14:08] And mainly missing status i nspecs [14:08] Indeed, -wubi is missing [14:09] There are some dups due to the Recently updated blueprints link [14:09] So I'll refine this script output but it's not ready for this week [14:09] Ok [14:09] how about we use http://piware.de/workitems/mobile/karmic/report.html [14:09] [topic] lool's status [14:09] New Topic: lool's status [14:09] tough thats not in order by developer [14:09] ogra: Hmm I'm basing on that script already [14:10] mcasadevall: Nothing much to report on my spec; just discussed stuff about it with the cloud team last week; didn't want to work on it as it's not very team dependent [14:10] So I preferred doing team stuff over the sprint [14:10] [topic] ogra's specs [14:10] New Topic: ogra's specs [14:11] mobile-arm-karmic-offline-installer-gui -> have to fix some issues in qemu this week [14:11] mobile-karmic-freescale-desktop -> waiting for new linux metapackage which should hit the archive this afternoon [14:11] we'll build A4 from universe [14:11] (kernel and meta in universe) [14:12] that gives me some freedom wrt freeze [14:12] thats all [14:12] [topic] NCommander's specs [14:12] New Topic: NCommander's specs [14:12] Not a whole lot to report, Installer changes are still dep-wait mostly on the kernel [14:12] which spec is that ? [14:13] ogra, mobile-karmic-marvell-desktop [14:13] :) [14:13] how about the uboot setup and initramfs issues ? [14:13] ogra, depwait email from Marvell [14:13] any progress on the debian-cd scripts ? [14:14] ogra, no, I had no machine most of the weekend and on Monda [14:14] ok [14:15] I'm currently back home so I'm in much better shape w.r.t. to this [14:15] anyway [14:15] [topic] dyfet's specs [14:15] New Topic: dyfet's specs [14:15] I created a working seed for a lxde desktop (lubuntu) and I am just finishing a meta package for it. [14:15] I also am investigating several arm/banshee issues, and yesterday I had some difficulties getting my env back running here which are now resolved. [14:16] Good [14:16] I think thats all :) [14:16] [topic] plars's specs [14:16] New Topic: plars's specs [14:17] the status I have on there is pretty much up to date [14:17] [topic] StevenK's specs [14:17] New Topic: StevenK's specs [14:17] -wubi is my only one [14:18] ok [14:18] Beside GrueMaster and paulliu did I miss anyone? [14:18] Selecting UNR as the environment and then clicking Install does nothing, I'm going to talk to Evan and/or Agostino about it. [14:20] [topic] Any Other Business [14:20] New Topic: Any Other Business [14:20] * plars upgraded to karmic after the sprint, yay! [14:20] Alpha 4 is this week, I want testers! [14:21] Ok [14:21] going once [14:21] twice [14:21] #endmeeting [14:21] Meeting finished at 08:21. [14:25] very efficent meeting [14:25] czajkowski: I appreciate that as well :) [14:26] It was lool begging for this format [14:26] Well it's not truly there yet since we didn't get all status in a web page [14:59] afternoon [tech board]] [14:59] I poked Mark; Scott is on holiday [14:59] mdz: here? [15:00] cjwatson: yep [15:00] cjwatson: clan has reminded him [15:01] #startmeeting [15:01] Meeting started at 09:01. The chair is cjwatson. [15:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [15:01] Keybuk: ah, thanks for showing up, I know you're on holiday [15:02] [TOPIC] Outstanding actions [15:02] New Topic: Outstanding actions [15:02] (from https://wiki.ubuntu.com/TeamReports/July2009) [15:02] Colin to set up interviews with developers regarding upload privilege reorganisation (for ArchiveReorganisation) [15:02] done yesterday; currently engaging in several dozen separate little conversations, although mostly just gathering data right now [15:02] cjwatson: holiday basically means playing Tales of Monkey Island while catching up with e-mail, blogs and podcasts ;) [15:03] Keybuk: one of those outdoor activity holidays then [15:03] Action: Matt to contact TB nominees to confirm they wish to stand [15:03] mdz: did they all say yes? [15:04] * cjwatson parallelises slightly [15:04] Action: Scott to implement Developer Membership Board proposal (LP, mailing list, documentation, etc.) [15:04] Keybuk: where does this stand? [15:04] cjwatson: sorry, distracted [15:04] cjwatson: sprint, etc. not even picked up ball yet, let along dropped it [15:04] cjwatson: I emailed technical-board@ with their responses, everyone agreed to stand [15:05] I know the feeling. Will carry over [15:05] [ACTION] Scott to implement Developer Membership Board proposal (LP, mailing list, documentation, etc.) [15:05] ACTION received: Scott to implement Developer Membership Board proposal (LP, mailing list, documentation, etc.) [15:05] Action: Jono to see that documentation is updated to reflect the Developer Membership Board [15:05] this is presumably blocked on the implementation [15:05] [ACTION] Action: Jono to see that documentation is updated to reflect the Developer Membership Board [15:05] ACTION received: Action: Jono to see that documentation is updated to reflect the Developer Membership Board [15:05] correct [15:05] Action: Matt to communicate the new patent policy [15:05] done, right? [15:06] Action: Colin to update http://www.ubuntu.com/community/processes/techboard [15:07] unfortunately I only noticed today that I haven't actually got login details yet; apparently newz2000 got derailed by the wiki subthread [15:07] I think there are still things that ought to stay on the website that need to be updated, so I've asked again on that [15:07] [ACTION] Colin to update http://www.ubuntu.com/community/processes/techboard [15:07] ACTION received: Colin to update http://www.ubuntu.com/community/processes/techboard [15:07] cjwatson: yes [15:07] Action: Jono to draft text to communicate the TB's willingness to review key topics submitted by the community [15:08] jono: did this happen? I think I missed it [15:08] cjwatson, sorry, not had a chance to yet, same reason as Keybuk [15:08] I can certainly have it ready this week [15:08] [ACTION] Jono to draft text to communicate the TB's willingness to review key topics submitted by the community [15:08] ACTION received: Jono to draft text to communicate the TB's willingness to review key topics submitted by the community [15:08] possibly eventoday [15:08] Action: Colin to discuss with Bdale (Debian technical committee chair) about having a Debian representative on the Ubuntu TB [15:08] even today [15:09] started, but Bdale would like to engage with us to figure out a bit more of what the role would entail [15:09] ... but nobody's answered :) [15:10] so personally I'd like to see that all or most significant technical discussions (as opposed to more local governance kinds of things) have a Debian seat at the table, for purposes of advice, history, liaison, etc. [15:10] cjwatson: I don't think we can establish the scope of the role without Debian involvement [15:10] so that's sort of part of the role [15:11] right. but the above is a sane kind of general goal, subject to people's availability ...? [15:11] [TOPIC] Debian TC liaison [15:11] New Topic: Debian TC liaison [15:11] cjwatson: yes [15:12] ok, I'll take that back to the thread I started [15:12] [TOPIC] Technical Board nominations [15:12] New Topic: Technical Board nominations [15:12] someone to provide a Debian point of view and liaison on technical issues which would benefit from coordination between the projects [15:12] mdz: is this now ready for the vote to start? [15:12] cjwatson: yes, has been for nearly two weeks [15:13] whose action is this? sabdfl's, as TB admin? [15:13] i can setup votes [15:13] yes, sabdfl has in his mailbox the list of nominees who have agreed to stand for a vote [15:13] [ACTION] sabdfl to create and announce TB vote [15:13] ACTION received: sabdfl to create and announce TB vote [15:13] sent 30 July, "Technical Board nominations" [15:13] I'll skip DAB, covered earlier [15:13] we have a vote process problem, in that LP is not setup to pick m of n [15:14] i don't have a good solution [15:14] also covered current archive reorganisation earlier (interviews in progress; seems that the bulk of people want to go for generalist access) [15:14] sabdfl: set up devotee? O:-) [15:14] surveymonkey? :-) [15:15] mdz, :-) [15:15] openid enabled django based voting site :) [15:15] lets keep it simple, surveymonkey seems ideal [15:16] jono: how to restrict access? [15:16] I was kidding... [15:16] sabdfl: we would email each dev a link with a token [15:16] mdz: fine by me, easy to script [15:16] mdz, really? seems like it could be suitable to recieve votes [15:16] ok, I withdraw the joke and put it forward as a proposal :-) [15:17] sabdfl, you can invite specific email addresses [15:17] jono: even better [15:17] all we would need is a list of email addresses out of Launchpad to copy/paste into it [15:17] easy [15:17] * jono is victorious [15:17] :P [15:17] ok, let's do that [15:17] err, not to turn into an election methods nerd, but m-of-n votes are actually sort of non-trivial in terms of how the votes are resolved. I wasn't aware LP hadn't already settled this [15:17] what voting method do we use? [15:18] cjwatson: the simplest possible, I suppose...everyone chooses M out of N, and we tally the totals [15:18] unless there's a voting widget [15:18] so no transfers? [15:19] I raised in the previous meeting that, it was my memory that LP Polls only allowed a single selection of candidate.. Question, followed by X options.. with one vote. [15:19] first past the post seems simple enough for our needs here [15:19] there's no voting widget [15:20] there are multiple choice (many options) but no way to constrain the number of choices made, afaics [15:20] I guess that's OK, though I haven't thought about it deeply; if you guys think FPTP is OK then let's do that. (In fact it's really "not-last-past-the-post" since this is a 5-of-6 election isn't it?) [15:20] first past the post would be horrible at the tail end, largely random [15:21] sabdfl: if you're speaking of surveymonkey, you can limit the number of choices [15:21] ok, i propose surveymonkey unless a better option jumps out of a quick google [15:21] cjwatson: I've lost all grip on number of nominations and/or seats ;-) [15:21] for 5-of-6 it means the loser will simply be the person whom most people don't select [15:21] I may be getting confused, but surely we just have a surveymonkey survey with one question with the candidates, open for a set time period and we invite all devs to choose one person [15:21] is it much more complex than that? [15:22] jono: that would be terrible at the tail end [15:22] oh, you want just *one* person per developer? that will be crap [15:22] sabdfl, what do you mean? [15:22] jono: yes, we're trying to elect 5 people [15:22] not just one [15:22] * Keybuk thought we each voted for 5 people [15:22] and thus the people/person who appeared in the least number of votes groups-of-5 would lose [15:22] http://www.cs.cornell.edu/andru/civs.html [15:22] Keybuk: yes, that's more what I was thinking [15:22] LINK received: http://www.cs.cornell.edu/andru/civs.html [15:22] so the person with the lowest votes doenst get elected [15:23] IOW the person whom most people don't select. That seems OK as long as each developer has five votes each [15:23] jono: that would mean that each dev could only express an opinion about one candidate [15:23] ooh, civs looks good [15:23] oh I see what you mean [15:23] civs is ideal for this kind of stuff [15:23] * ScottK has used it in other projects. [15:24] can we vote on that? [15:24] civs looks good to me [15:24] I wouldn't have wanted us to implement condorcet in a rush, but given an existing implementation it would be great [15:24] I approve of civs [15:24] [VOTE] use http://www.cs.cornell.edu/andru/civs.html for TB election [15:24] Please vote on: use http://www.cs.cornell.edu/andru/civs.html for TB election. [15:24] Public votes can be registered by saying +1/-1/+0 in the channel, private votes by messaging the channel followed by +1/-1/+0 to MootBot [15:24] E.g. /msg MootBot +1 #ubuntu-meeting [15:24] +1 [15:24] +1 received from Keybuk. 1 for, 0 against. 0 have abstained. Count is now 1 [15:24] it's described as not being super secure but i think it's fine for this purpose [15:24] +1 [15:24] +1 received from sabdfl. 2 for, 0 against. 0 have abstained. Count is now 2 [15:24] +1 [15:24] +1 received from cjwatson. 3 for, 0 against. 0 have abstained. Count is now 3 [15:25] +1 [15:25] +1 received from mdz. 4 for, 0 against. 0 have abstained. Count is now 4 [15:25] mm, it's "not super secure" as in "we have done a detailed security analysis and found the following holes" which is much better than any ad-hoc thing we might do [15:25] ftr, we have 7 nominates for 5 seats [15:25] yes, i think it's fine [15:25] Keybuk: ah ok [15:25] #endvote [15:25] [ENDVOTE] [15:25] Final result is 4 for, 0 against. 0 abstained. Total: 4 [15:25] %!"£$ENDVOTE$£"!% [15:26] /exec -o figlet [ENDVOTE] [15:26] ok, neeeext [15:26] [TOPIC] Ubuntu security policies [15:26] New Topic: Ubuntu security policies [15:26] [LINK] https://wiki.ubuntu.com/SecurityTeam/Policies [15:26] LINK received: https://wiki.ubuntu.com/SecurityTeam/Policies [15:27] kees showed this to me last week [15:27] but I don't think he could make this meeting [15:27] jdstrand: are you here? [15:27] I just read this through and everything other than the executable-file stuff being discussed by mail seems fairly non-controversial, basically a codification of what we're already doing [15:27] I've been talking with Kees and Marc via e-mail about the draft policy for the execute bit, which I feel is not as fully thought through as it should be [15:27] the non-draft bits all look fine to me [15:27] I am here [15:27] I liked the thing you said at the end about the point being to stop MIME handlers subverting the execute bit [15:27] basically, he wanted to document Ubuntu security policies which were not documented yet, and get that blessed by the TB [15:27] and also to propose some new stuff [15:28] * jdstrand reads backscroll [15:28] i.e. the execute permission bit [15:28] I remember we codified a security policy about mDNS, and which services were allowed to be published by default, etc. but I don't see that there [15:28] jdstrand: just starting from New Topic: Ubuntu security policies [15:28] can we vote on the non-draft bits and get that out of the way? [15:28] https://wiki.ubuntu.com/ZeroConfPolicySpec [15:28] Keybuk: ^^ [15:29] any objections to a simple up/down on whether they're ok in bulk? [15:29] mdz: that's the one, that should be linked from there too [15:29] Keybuk: agreed [15:29] cjwatson: don't see why not [15:29] has anyone not read the document yet? [15:29] I probably should have emailed it out in advance, but at least it's short [15:29] Kees e-mailed it [15:29] well, the URL [15:29] oh, good [15:30] [VOTE] Approve non-draft portions of https://wiki.ubuntu.com/SecurityTeam/Policies [15:30] Please vote on: Approve non-draft portions of https://wiki.ubuntu.com/SecurityTeam/Policies. [15:30] Public votes can be registered by saying +1/-1/+0 in the channel, private votes by messaging the channel followed by +1/-1/+0 to MootBot [15:30] E.g. /msg MootBot +1 #ubuntu-meeting [15:30] +1 [15:30] +1 received from mdz. 1 for, 0 against. 0 have abstained. Count is now 1 [15:30] +1 [15:30] +1 received from sabdfl. 2 for, 0 against. 0 have abstained. Count is now 2 [15:31] just for the record, this is https://wiki.ubuntu.com/SecurityTeam/Policies?action=show we're voting on [15:31] +1 (we might want to add to the mdns bits, but what's there is an OK start) [15:31] +1 received from cjwatson. 3 for, 0 against. 0 have abstained. Count is now 3 [15:31] er, https://wiki.ubuntu.com/SecurityTeam/Policies?action=recall&rev=15 [15:31] Keybuk: vote? [15:31] +1 [15:31] +1 received from Keybuk. 4 for, 0 against. 0 have abstained. Count is now 4 [15:31] err [15:32] I just managed to do /+1 ... bloody irssi [15:32] cjwatson: the mdns bits are an earlier TB decision, so I think we can grandfather it in [15:32] [ENDVOTE] [15:32] Final result is 4 for, 0 against. 0 abstained. Total: 4 [15:32] mdz: oh, before my time [15:32] mdz: agree [15:32] someone will need to dig up the reference [15:32] I'd like an [ACTION] for somebody to add that to the document though, jdstrand perhaps as the person whose here? :p [15:32] ok, so shall we continue the execute bit discussion by mail rather than here? it seems rather in-depth [15:33] I'm here also, btw [15:33] [ACTION] jdstrand to add zeroconf reference from previous TB decision to security policies document [15:33] ACTION received: jdstrand to add zeroconf reference from previous TB decision to security policies document [15:33] Keybuk: I can add it [15:34] well, I see I was already assigned it... :) [15:34] I'm told that the contentious bit (no pun intended) of the execute permission policy is WINE [15:34] you can delegate if you like, I just don't like assigning multiple people :) [15:34] mdz: wine and java [15:34] mdz: and desktop files ... [15:34] WINE and Java need the execute bit anyway when using the binfmt support [15:34] cjwatson: who has an issue with .desktop files? [15:35] the only reason they might not is if somebody's mucking around with MIME handlers and subverting the need for the +x [15:35] mdz: desktop files are text files [15:35] Keybuk: that is currently the case [15:35] if you make them executable, then doing ./foo.desktop will mean it's parsed by the shell [15:35] Keybuk: then don't do that :-P [15:35] a file manager or app that doesn't know *.desktop files are special will then start executing them *as shell scripts* [15:35] on the TB election, any objections to running the vote from August 17-31st? [15:35] (which they're clearly not) [15:36] and the desktop file format doesn't permit a leading #! line [15:36] Keybuk: (actually, that is overridable by binfmt_misc; whether that's a good idea is a different question) [15:36] and even if it did, then there wouldn't be a generic way of handling them desktop-agnostically [15:36] fundamentally this is abusing the +x bit ;-) [15:36] also if you drag a *.desktop file out of Evolution and onto the Panel, that will make a Panel launcher for it [15:37] it's not clear whether that requires +x [15:37] Keybuk: it doesn't :( [15:37] sabdfl: ok by me [15:37] (ie. the security or not of your system depends on your aim, and/or understanding of the difference between the panel and desktop) [15:37] sabdfl: fine with me [15:37] I agree with cjwatson that we should probably take this to email, we don't have time to cover it in depth here [15:37] and it's then, on top of that, not clear what the UI of a non-+x launcher on the panel would be, and how you'd make it executable [15:37] mdz: there's already a thread ;-) [15:38] Keybuk: which I haven't read yet, because I was on holiday yesterday [15:38] mdz: pah ;) I managed to _cause_ the thread while being on holiday [15:38] this is a discussion that's been going on literally for years anyway; I'm pretty sure I remember it coming up in 2005 [15:39] so indeed I doubt we're going to solve it right here right now; let's continue at more leisure [15:39] [TOPIC] Handling community problems [15:39] New Topic: Handling community problems [15:39] [LINK] https://wiki.ubuntu.com/ReportingCommunityProblems [15:39] LINK received: https://wiki.ubuntu.com/ReportingCommunityProblems [15:40] this I regret to say I've only skimmed, but it seems basically reasonable. jono, do you want us to have a regular check on issues filed there (presumably we'll get them by mail anyway)? [15:40] cjwatson, I would prefer if the TB had a process to regularly check in with the assigned bugs list [15:40] much of this could be performed over email I assume [15:40] and then issues that require discussion could be moved to meetings [15:41] my only comment, which I passed on to Jono already, is that we'll want to route more issues directly to delegate councils beyond the TB and CC as time goes on [15:41] so let's add a link to the standing agenda, if it's empty it'll only take 15 seconds [15:41] mdz, yes indeed [15:41] but I think it is reasonable to start with TB/CC/community team [15:41] I wanted to trial this with the CC and TB initially and then move to team councils [15:41] cjwatson: agreed [15:41] cjwatson, sounds good [15:41] I am planning on doing an announcement soon [15:42] of the new process [15:42] yes, as I said the other day about something else, I don't think the TB should necessarily be the body of first resort; we should be active, but it's corrosive to foster a culture where everybody just sits back and lets the TB decide [15:42] cjwatson, agreed [15:42] we should only pass on issues to the TB that fall within the mandate of the TB [15:42] (which I don't think this is doing at the moment, for avoidance of doubt) [15:43] jono: the links should probably go to launchpad.net, not edge. [15:43] does anyone have an objection to Jono proceeding with this, or can we move on? [15:43] sladen, I will fix that [15:44] no objection [15:44] no objection [15:44] [TOPIC] Investigate alternative to Google CSE (305905, 402767) [15:44] New Topic: Investigate alternative to Google CSE (305905, 402767) [15:45] sladen requested that we look at this by mail, but nobody has yet taken it up [15:45] what's Google CSE? [15:45] Keybuk: custom search [15:45] oh, didn't that just get dropped anyway? [15:45] thanks all [15:45] I saw a -changes mail fly past [15:45] I believe he's referring primarily to the multisearch widget in firefox [15:46] Keybuk: the time-limited "experiment" was completed, but we still need to decide and implement the final experience for 9.10 [15:46] https://lists.ubuntu.com/archives/karmic-changes/2009-August/006000.html [15:46] Keybuk: default browser page is http://start.ubuntu.com/9.10/ which points to a Google CSE results, which isn't translated and doesn't have most of the additional features people associate with Google [15:46] it doesn't look like the Ubuntu Desktop team are wedded to this implementation, and it looks like they're aware of the issues [15:47] I don't think the TB need to intervene just yet [15:47] sladen: the desktop team is aware of the issues with it and is responsible for making sure the experience is good [15:47] we should trust them to make the right calls, based on the feedback their "experiment" has given them [15:47] sladen: have you spoken directly (pref. not via a bug) with the desktop team about this yet? [15:49] cjwatson: regarding 402767, very much so. Not sure 305905 is a desktop issue though, it's something that requires high-level Google<->Canonical interaction [15:49] (point of order: 10 minutes left) [15:50] I'm not in a position to know what's going on there at the moment, but when I was managing Alexander this was something I would have been involved in, so I'd be surprised if Rick Spencer weren't an appropriate contact for lots of this [15:50] there was a discussion in each of the past two UDS's about search options, I expect there will be another discussion in November [15:50] sladen: having spoken with the team, what made you feel that the TB needed to step in to resolve the issue? [15:50] sladen: if you have new information for that, or suggestions, please contribute them [15:50] sladen: that one is a desktop team issue as well [15:50] it's not clear to me from the visible traffic on the ML and Bug that the desktop team haven't been taking concerns on board [15:50] indeed, my reading in the last few minutes suggests they have [15:51] I'm going to move on based on time, but we can continue this by mail as necessary [15:51] i don't think the TB should be asked to go and explore options, since the desktop team are doing that now [15:51] [TOPIC] Statistics gathering policy [15:51] New Topic: Statistics gathering policy [15:51] cjwatson: I think the next issue is the important one [15:51]