=== rog is now known as Guest45233 === Guest45233 is now known as rogpeppe === frankban_ is now known as frankban [12:16] morning [12:37] Hi all. Quick q. [12:37] According to the readme I neet to access https://localhost:8081/ws [12:38] But I have nothing running on that port [12:38] afer doing a "make server" [12:42] hi sepi [12:43] hi I think I know my problem [12:43] I needed to use the rapi-delta branch? [12:43] sepi: yes, either it or rapi-rollup [12:46] sepi: also, the url you pasted is only for the web socket to use. to actually access the juju gui you would use http://localhost:8888 [12:47] sepi: reading the README i see we don't mention that explicitly. i'll fix that now. [12:48] bac: thanks :) [12:48] np. have fun, file bugs! === gary_poster|away is now known as gary_poster [13:26] hiya (was upgrading quantal on the desktop) [13:30] morning [13:30] mattuk1972, hi. I got your question about the service detail page. The box has the two-radius rounded corners, right? I'm pretty sure I see that. Because of that, I want to confer with Makyo and see if we want to do this with svg. I'll do that and get back to you [13:31] * bac tries to remember which days we had dinner paid. [13:31] heh [13:31] good question [13:31] many of them [13:31] esp. second week [13:31] gary_poster: ok ty [13:32] bcsaller, hi. are you in nyc area, or just awake *really* early on west coast? :-) [13:33] gary_poster: also kapil mentioned that theres a requirement to simplify the svg's overall? [13:33] gary_poster: in NYC [13:33] :) [13:33] as of last night [13:33] bcsaller, cool :-) hope everything is ok for you there [13:33] bac: these are my notes, maybe they can help: http://pastebin.ubuntu.com/1334735/ [13:34] Seems mostly back together here but there is still alot to be done in other spots [13:34] mattuk1972, yes. Chrome and IE10 support the service box svgs on the environment mage well. iPad 3 is pretty good. FF is very bad. Apparently the FF problem is largely because of the service assets [13:34] s/mage/page [13:35] mattuk1972, Makyo investigated and reported that FF team is not being very responsive about SVG rendering issues [13:35] gary_poster: weird i did a lot of testing in all the browsers to make sure they rendered well and firefox was one of the best... [13:36] So we need to do something to make that better. I believe I heard hazmat (who is out this week) talked to Ivo about it so he may know details. [13:36] mattuk1972, it's not a matter of rendering accuracy. The problem is that once you draw the boxes, when you try to drag them around it is extremely slow [13:36] ahh [13:36] gary_poster: I can still try to reach out to some of the people I know at Mozilla and see if they can make it more of a priority [13:37] mattuk1972, easy comparison: try http://uistage.jujucharms.com:8080/ in chrome and in ff to see diff [13:37] at least in Ubuntu [13:38] but simplifying the svgs will fix it? removing any transparency? [13:38] haven't tried in os x [13:38] bcsaller, that's worth a try [13:38] thank you [13:38] hmm seems to be ok in osx [13:38] mattuk1972: removing the filters and layers will help alot, yes [13:39] i don't know how much of that i can do before it just becomes a plain grey box [13:39] mattuk1972, that's my understanding. It's probably worth some experimentation coordinating with us to both verify (I think that's already been done, and bcsaller is confirming) and to see how far we can go [13:39] frankban: your records matched my recollection. thanks. [13:40] we can replace all the fancy shading with a png that just says "Firefox is slow." [13:40] ack mattuk1972. Maybe there's a simple experiment we can run to see if it this avenue is even workable? [13:41] to take it right down with a really basic shape just to see if it helps? [13:41] my DSL connection is flaky today, dropping and reconnecting every now and then. [13:41] lol benji [13:41] I.e., you quickly simplify it to a point that you think might be representative "the least detail that would work visually/professionally" [13:41] :-) [13:42] mattuk1972, I think we have already established that a basic shape helps, right bcsaller? The question is, is FF good enough that we can still have something professional looking, or not [13:42] I think [13:42] Since FF is the default browser in Ubuntu, it's kind of important [13:44] gary_poster, mattuk1972: Yeah, simplfying the service boxes helps, but it comes at visual cost. We have options to help make FF faster, make the svgs simpler, add a drag mode that on FF turns them into outlines, or switch FF to use larger static image assests that don't scale as well and try shrinking those down as needed. [13:45] I think an outline on drag could fix our issue at low cost [13:45] like an old school WM [13:45] heh [13:45] yeah, not a bad idea [13:45] interesting that it is ok on os x [13:45] in ff [13:46] yea it really smooth - can't see any dif between it and chrome [13:46] FF uses different platform specific paths for accel [13:46] I *think* its cairo on linux for this, but not sure [13:46] pretty sure it was at one point, but don't know now [13:47] so windows and osx can do very different things [13:47] k [13:47] right [13:47] this is just a border cases that impacts very few sites on a smaller cross section of their browsers [13:48] chrome wrote their own svg accel layer and use it everywhere iirc [13:57] bcsaller: can we have a g+? I need to check with you the solution I have for the aggregation task I am working on. [13:57] tveronezi: yeah, I need a few minutes though [13:58] ok... ping me when you are ready. tkx! [14:14] If anybody said anything to me in the past 15 min or so I probably missed it. Had to switch computers for a few minutes [14:14] So if I haven't replied to something or other, please try again :-) [14:22] gary_poster: standup is still at 12 EST i assume [14:23] bac, yeah, for now that was what I was thinking, though I was going to clarify with everyone. AFAIK we have all now gone through DST transition. We might be able to move the call one hour earlier with Ben on the East Coast, but that's arguably a separate discussion. Lemme ping everyone... [14:24] tveronezi: ok, I can chat now, joining the hangout [14:25] ok [14:32] bac bcsaller benji frankban (hazmat not here) Makyo mattuk1972 teknico tveronezi I think we have all moved from DST now except bac, who will not. bcsaller is now on East Coast USA. What do you say to moving the daily call an hour earlier, adjusted for DST changes, to 1600 UTC (11AM EST, 9AM MST, 12PM AST, 1700 CET)? [14:33] (If I left anyone [14:33] sounds good to me [14:33] gary_poster: I'm cool with that [14:33] 's timezone out I apologize :-) I tried) [14:33] gary_poster: ok [14:33] mm, I left benji's out because I still think he is Eastern sorry :-P [14:34] * benji is livin' la vida UTC [14:34] :-) [14:35] gary_poster: sounds good. no change for me [14:35] ok thx [14:35] I am the man without a timezone. Europe to SF for 1 day to NYC, my body is sure I hate it. [14:35] heh [14:37] gary_poster, Sounds fine. [14:37] gary_poster, Starting today? [14:38] gary_poster, ok for me [14:42] Makyo, that's the idea, but doesn't have to be [14:43] gary_poster, Just making sure :) [14:43] cool :-) [14:55] Makyo: got a sec to catch up on g+? [15:03] frankban, mattuk1972, tveronezi let me/us know asap if the new mtg time I proposed above is OK for you. I'm hoping it is ok for jovan too, when he is back [15:04] gary_poster: sounds good to me too. [15:04] jovan is in the office [15:04] is that 4 my time? [15:04] gary_poster: sounds good [15:05] mattuk1972, yes [15:05] frankban, tveronezi cool thx [15:05] ok sounds good [15:05] cool thanks [15:06] * gary_poster is changing calendar [15:06] jovan says 4 is good too [15:08] bcsaller, Sure, was just replying to email, whenever's good. [15:08] Makyo: I'll hop in the g+ [15:11] g+ is being strange [15:12] hmm, going to restart the router here, it might be acting up [15:12] Alright. [15:16] mattuk1972, fwiw and to be clear I invited you to all the meetings, but don't expect you to actually want to attend all of them. You are invited to attend as makes sense for you. :-) [15:16] gary_poster: cool will do - ty [15:16] cool [15:24] bac, hey. I just found the css spec for the two-radius border syntax. It looks like it is supported. Did you ad Huw try it and reject it, or is it broken for some reason that is not clear on this page? http://muddledramblings.com/table-of-css3-border-radius-compliance/ [15:25] gary_poster: we did not try it [15:27] bac ok. maybe worth playing with. we have more of those two-radii curves coming. [15:27] gary_poster: we didn't know about it. [15:27] gary_poster: i'll try it now [15:27] cool [15:27] thx [15:41] hi gary_poster, mattuk1972 says that cimi tried that two radii method and rejected it since the curve is a more complex bezier curve and cannot be replicated using that CSS approach. [15:42] i mean it might be worth just seeing what it looks like? but i think cimi spet a long time trying to emulate it before [15:42] bac, but we can get a lot closer with it, yeah? for buttons and smaller shapes in particular [15:43] gary_poster: we can try but if they have investigated and rejected it then i'm not hopeful we'll be able to get them to accept a non-image approach [15:44] mattuk1972: does that sound accurate? [15:44] bac, I thought we were hoping to get the buttons in all css, even with the simple rounded curves, and this would be a lot closer. Did I misunderstand? [15:44] from what i can understand. would it be useful to talk to cimi? [15:44] gary_poster: yes, that was the goal [15:45] gary_poster: but if the designers won't allow any deviation it seems like the experiment is not a good use of time [15:45] if you use tat method then it won't be following the visual style of siri [15:45] i.e., i don't know if there is anything "close enough" [15:45] suru even lol [15:45] :-) [15:45] siri, make me a button [15:47] hehe [15:47] if someone can show me how close we can get then it might be ok? [15:47] bac, ^^ [15:47] bac, I can try or you [15:47] just warning that it may get some noise from this end if we don't have the curve right enough [15:48] ack [15:48] gary_poster: if that is what you want i'm willing to try [15:48] bac, I don't see how the attempt should take more than 15 minutes or so, yeah? [15:48] from what i can work out a nice multi radius to try would be 12 and 8 [15:48] cool [15:49] thx [15:49] as thats where the curve start on the horizontal and then the virt [15:54] Makyo, thanks for reply. I can imagine pretty easily what we'd want for the image slice css border approach. If we did the svg approach, what resources/assets would we want? We couldn't use an svg of the whole box; we'd want svgs of the four corners, I guess? or what would you recommend? [15:58] bac bcsaller benji frankban jovan2 Makyo mattuk1972 (if you want) teknico tveronezi call in 2 [15:58] in https://plus.google.com/hangouts/_/409819056ea1551523432ac63f4df4a6feaa5922?hl=en-US [15:59] * gary_poster has to install google talk plugin and didn't realize... === mattuk1972_ is now known as mattuk1972 === mattuk1972_ is now known as mattuk1972 [16:36] reminder this is the last day to the the company survey [16:57] lunch time... brb [17:03] Noticeable improvement in speed in FF with the new asset. Still not great, but definitely better. [17:03] gary_poster, so, I had a look at the pdf for #1074298, any pre-imp hints? [17:03] <_mup_> Bug #1074298: charm panel search results page should match visual design < https://launchpad.net/bugs/1074298 > [17:04] teknico, first, leave out the drop-down, if that wasn't clear [17:04] gary_poster, well, that part indeed was clear, yep ;-) [17:04] cool :-) [17:05] teknico, otherwise, for the buttons, instead of the image, I'd prefer it if you could use the button function that bac made and is tweaking now [17:06] teknico, also the search part should already be done, if that's not clear--the searchbox_icon.png and so on [17:06] I *think* this should just be CSS teknico [17:07] teknico, is all of that clear or would you like to discuss? [17:08] gary_poster, yeah, I thought it would be CSS [17:09] gary_poster, a few specific pointers in code and CSS would speed things up [17:09] teknico, cool. Gimme a sec and then we can have hangout? Maybe set up a hangout and then I'll join? [17:10] gary_poster, great, thanks. I'm sure frankban won't be offended if we employ the usual https://tinyurl.com/see-emily-code :-) [17:10] :-) k [17:14] :-) [18:08] Very, very simple branch up for review, lp:~makyo/juju-gui/replace-service-module Just replacing assets. [18:09] Rietveld: https://codereview.appspot.com/6826057 [18:42] Reviewing... [18:43] you've got one approval, Makyo [18:43] gary_poster, Fanks :) Easy, but hopefully helpful. [18:44] cool [18:46] hi gary_poster -- i've messed with the border-radius and something like "border-radius 4px / 10px" seems close. take a look? [18:47] bac, love to. [18:47] what should I do? [18:47] just change css probably :-) [18:47] gary, just mess with it in your browser [18:47] ack, on it [18:47] gary_poster: i discovered in LESS it has to be written: border-radius: ~"4 px / 10px" [18:47] or else it tries to do division. :( [18:47] huh [18:48] heh [18:48] ~"..." tells it to not interpret [18:48] that's a good weekly hint bac; or at least an email [18:56] bac, cool. my eye likes 4px/7px best but go withe what your eye tells you and show Matt. You could offer to change it live over a hangout tomorrow even [18:57] mm, I could be convinced bt 4px/10px bac [18:57] by [18:57] thanks for exploring bac [18:57] mm? [18:57] bac maybe just land it? [18:57] the "mm" was a sound of thought as I compared the two, side by side [18:58] gary_poster: ok, i'll land it. [18:58] cool [19:42] Out to the vet. [19:58] Makyo: what is the difference between the buildingRelation and the clickAddRelation flags? [20:00] lbox hates me [20:10] bac, gave you an approval despite lbox's displeasure with you [20:15] gary_poster: cool. just on the MP i presume [20:15] bac, y [20:15] it blows up trying to access google [20:15] boom! [20:32] so 'lbox submit' ignores the reviewers from LP. i should've guessed that. no rv = no reviewers in the submit [20:37] * bac -> dog walk [20:59] benji, buildingRelation is true if you did the click to add or the long click, clickAddRelation only in the former case. [21:03] brb... [22:28] tveronezi: there is a conflict in your filter-buttons branch. no big deal but please push a new branch when you get a chance [22:28] bac: ok... tkx. [22:29] tveronezi: i'll have a full look at it tomorrow before you start. [23:12] * Makyo dogwalks.