| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Currently no events are active in this room - http://summit.ubuntu.com/uds-1311/community-1/ - http://irclogs.ubuntu.com/2013/11/21/%23ubuntu-uds-community-1.html | ||
| === dholbach_ is now known as dholbach | ||
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Track: Community | Automated ubiquity testing with autopilot | Url: http://summit.ubuntu.com/uds-1311/meeting/22063/community-1311-quality-autopilot-image-testing/ | ||
| balloons | hangout -> https://plus.google.com/hangouts/_/76cpjpsnolhdibv2d9as0s29k4?authuser=0&hl=en | 13:52 |
|---|---|---|
| DanChapman | wow is that the time already .... balloons be there in 2 | 14:00 |
| balloons | DanChapman, :-) | 14:00 |
| balloons | DanChapman, https://plus.google.com/hangouts/_/76cpjpsnolhdibv2d9as0s29k4?authuser=0&hl=en | 14:00 |
| balloons | we'll begin in a moment, waiting to round up everyone | 14:01 |
| balloons | :-) | 14:01 |
| elfy | hi - not easy to hear Jean -Baptiste | 14:31 |
| balloons | hey elfy :-0 | 14:31 |
| elfy | in fact I only know he's talking because I can see his mouth moving :( | 14:34 |
| elfy | balloons: I wouldn't ;) | 14:36 |
| balloons | you wouldn't what? | 14:37 |
| elfy | balloons: assume all was ok if an autotest said it was without checking | 14:37 |
| elfy | especially given we've not had much luck with autotests with our stuff | 14:37 |
| elfy | at the moment if we didn't do manual tests then we'd be doing none at all ;) | 14:38 |
| elfy | jibel: if you are Jean - Baptiste - I can't hear you :) | 14:39 |
| jibel | elfy, I amplified the input volume a bit | 14:40 |
| elfy | cheers :) | 14:40 |
| DanChapman | https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bugs?field.searchtext=&orderby=-importance&field.status%3Alist=NEW&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_commenter=&field.subscriber=&field.ta | 14:49 |
| DanChapman | g=autopilot&field.tags_combinator=ANY&field.status_upstream-empty-marker=1&field.upstream_target=&field.has_cve.used=&field.omit_dupes.used=&field.omit_dupes=on&field.affects_me.used=&field.has_patch.used=&field.has_branches.used=&field.has_branches=on&field.has_no_branches.used=&field.has_no_branches=on&field.has_blueprints.used=&field.has_blueprints=on&field.has_no_blueprints.used=&field.has_no_blueprints=on&search=Search | 14:49 |
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Track: Community | One hundred papercuts for trusty | Url: http://summit.ubuntu.com/uds-1311/meeting/22070/community-1311-quality-papercuts/ | ||
| druellan | Anyone watching from summit.ubuntu. Inspect <section class="span-5"> and add width="100%" 8) | 15:02 |
| balloons | hangout link for anyone wishing to join the fishbowl :-) http://summit.ubuntu.com/uds-1311/meeting/22070/community-1311-quality-papercuts/ | 15:03 |
| balloons | AlbertoSN you around? | 15:03 |
| balloons | or anyone else from the papercuts team? | 15:03 |
| balloons | anyone wish to join the fishbowl? http://summit.ubuntu.com/uds-1311/meeting/22070/community-1311-quality-papercuts/ | 15:06 |
| balloons | bah, my bad. wrong link | 15:06 |
| balloons | https://plus.google.com/hangouts/_/76cpimcdcppk2ccl1q020phsgo?authuser=0&hl=en | 15:06 |
| balloons | AlbertoSN, hello :-) | 15:07 |
| AlbertoSN | Hi! | 15:07 |
| balloons | AlbertoSN, https://plus.google.com/hangouts/_/76cpimcdcppk2ccl1q020phsgo?authuser=0&hl=en | 15:07 |
| balloons | https://launchpad.net/hundredpapercuts | 15:09 |
| la_juyis | AlbertoSN, QUESTION: I didn't know the project. It says they're "easy bugs" how easy? | 15:12 |
| la_juyis | AlbertoSN, what kind of person should be able to fix them? | 15:12 |
| la_juyis | AlbertoSN, do you organize some kind of bug squash party? | 15:13 |
| la_juyis | AlbertoSN, are they categorized in any way so you can pick from a list? | 15:16 |
| la_juyis | balloons, thanks for the tip, I watched to a recorded session yesterday and it was hard to follow the answers without the questions first :D | 15:16 |
| balloons | la_juyis, :-) | 15:17 |
| la_juyis | AlbertoSN, do you involve the reporter of the bug in solving it, or try at least? | 15:20 |
| la_juyis | (I'm sorry I'm asking this much, but it looks like an interesting project and I'm curious :)) | 15:20 |
| balloons | ask away la_juyis! | 15:21 |
| balloons | wiki page: https://wiki.ubuntu.com/One%20Hundred%20Papercuts/ | 15:21 |
| la_juyis | AlbertoSN, maybe it's a good idea to have them "ready to pick" | 15:21 |
| la_juyis | AlbertoSN, so that the project can use little chunks of people's time | 15:21 |
| * la_juyis writing a clearer question | 15:22 | |
| la_juyis | I mean, if they're "easy to solve", maybe it's a good idea to involve the reporter - you could gain a new collaborator | 15:23 |
| la_juyis | maybe helping them to solve the bug | 15:23 |
| la_juyis | themselves | 15:24 |
| la_juyis | instead of you solving it | 15:24 |
| la_juyis | just an idea :) | 15:24 |
| qengho | la_juyis: the audience for the bug is not "you" or someone in particular. It could be reporter that fixes. | 15:24 |
| la_juyis | i mean, not just say "DYI", but "how can i help you to solve it" or show the path that needs to be taken, or etc | 15:25 |
| la_juyis | sorry, s/dyi/diy | 15:26 |
| qengho | la_juyis: that's an americanism, I think. "Do It Yourself". | 15:27 |
| la_juyis | thank you qengho :) | 15:30 |
| qengho | Let's get off the topic of "100 Papercuts" philosophy, and let's have a to-do list of what will make it easy to fix 100 small, important bugs in Trusty. | 15:30 |
| AlbertoSN | https://wiki.ubuntu.com/One%20Hundred%20Papercuts/One%20Hundred%20Papercuts%20will%20make%20Ubuntu%20shine/Remaining%20actions | 15:31 |
| qengho | Is there anything in our way now? | 15:31 |
| qengho | I think we should make a list of experts in various subjects to consult when it's not obvious why or how to fix a reported bug. Those can fail the "easy" test of 100PC bugs, but not always. | 15:34 |
| balloons | qengho, what is the current escalation process if you get stuck? | 15:35 |
| linuxtech | balloons: I just sent you an email on this issue. The DPL talked about a how-can-i-help package at http://lists.debian.org/debian-devel-announce/2013/11/msg00000.html and it uses a Debian bug tag gift, and stores the data in a little json file, to ask for help with particular packages. Can we make a how-can-i-help package? | 15:36 |
| qengho | balloons: The right way, usually, for 100PC bugs, is to ask around and then unassign yourself from the bug and expect someone else to know. Many eyeballs, et c. | 15:36 |
| balloons | qengho, oO. really, just unassign yourself.. intereting | 15:37 |
| balloons | I would say a better process would be useful | 15:37 |
| balloons | linuxtech, interesting.. so could this allow folks to "get" help in qengho's situation | 15:38 |
| qengho | balloons: Let's not worship processes. If you can't fix it, move to something you can. The big win of 100PC is raising awareness of problems that dont' fall into anyone's purview but should be fixable. | 15:39 |
| qengho | So, that awareness will bring someone who *can* fix what had you stuck. | 15:40 |
| balloons | qengho, I can appreciate that | 15:41 |
| balloons | there is beauty in that design | 15:41 |
| linuxtech | I can make the how-can-i-help-ubuntu package. I just spent an hour or two trying to figure out how to get the data out of launchpad and I haven't succeeded. | 15:41 |
| linuxtech | My launchpad id is cp. | 15:42 |
| balloons | qengho,what do you think of linuxtech's suggestion? | 15:42 |
| la_juyis | what about organizing little papercuts sessions in f2f meetings like FS conferences that are already being held, with ubuntu supporters attending anyway? | 15:44 |
| la_juyis | and have some scheduled virtual sessions once every two months or so | 15:44 |
| linuxtech | I'll write a launchpad list or two and ask for help, after all the json data shouldn't be on my personal server anyway. | 15:44 |
| la_juyis | AlbertoSN, can we sort them out into useful categories (the list)? | 15:45 |
| qengho | ? | 15:46 |
| AlbertoSN | https://bugs.launchpad.net/hundredpapercuts | 15:46 |
| qengho | https://bugs.launchpad.net/hundredpapercuts/+bugs?orderby=-heat&start=0 | 15:46 |
| la_juyis | and maybe getting reporters more involve could help: if they reported the bug, it's possible that they are annoyed by it and interested in seeing it fixed, and maybe with some help, they can | 15:48 |
| AlbertoSN | papercuts-ninja@lists.launchpad.net | 15:51 |
| linuxtech | It would diplay the new papercut bugs after apt finishes its run. | 15:52 |
| balloons | https://wiki.ubuntu.com/One%20Hundred%20Papercuts/One%20Hundred%20Papercuts%20will%20make%20Ubuntu%20shine/Remaining%20actions | 15:53 |
| balloons | qengho1, any other issues that should be brought up or addressed? | 15:53 |
| balloons | linuxtech, I think perhaps seeing the tool in action will help | 15:54 |
| qengho1 | I don't think we need a how-can-I-help package. 100PC bugs should churn so much that it takes longer to update the package than fix a few PC bugs. Let's not make burdens. The bugs should be our big burden. | 15:55 |
| balloons | qengho1, if you have the knowledge and the bugs are "easy", I agree :-) | 15:55 |
| linuxtech | We don't need to install it on everyone's machine, just people who elect to install it. | 15:56 |
| qengho1 | balloons:There are no formulas for getting rid of bugs. If there were, we would not have those bugs. The Launchpad URL list of many dozens of varying difficulties of bugs is all we should need to give to people. | 15:57 |
| linuxtech | I need to loin it I assume. | 15:58 |
| linuxtech | join the list... | 15:58 |
| balloons | linuxtech, yes I believe so | 15:58 |
| balloons | go for it :-) | 15:58 |
| qengho | My final suggestion is: | 15:59 |
| balloons | ty everyone | 15:59 |
| AlbertoSN | Bye bye | 15:59 |
| qengho | Ruthlessly go through 100PC bugs on LP and remove ones that are stale or not easy. | 15:59 |
| balloons | qengho, feel free to add on the pad as well | 15:59 |
| linuxtech | Thank you and have a good day! | 15:59 |
| balloons | qengho, so your thoughts are not lost | 15:59 |
| balloons | qengho, that sounds like a good idea as a work item | 16:00 |
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Track: Community | Touch Core Apps testing needs | Url: http://summit.ubuntu.com/uds-1311/meeting/22062/community-1311-coreapps-testing/ | ||
| AlbertoSN | Yes, write this down | 16:00 |
| AlbertoSN | And have a good day | 16:00 |
| AlbertoSN | :D | 16:00 |
| AlbertoSN | We'll see what will happen... | 16:01 |
| balloons | hangout for touch core apps testing -> https://plus.google.com/hangouts/_/76cpj9mr34bq9l8h7722ebh2uc?authuser=0&hl=en | 16:02 |
| balloons | come on into the fishbowl, the water's fine :-) https://plus.google.com/hangouts/_/76cpj9mr34bq9l8h7722ebh2uc?authuser=0&hl=en | 16:05 |
| balloons | anyone interested in joining the hangout? if not you get just me ;-) | 16:07 |
| vthompson | balloons, I'll join ya | 16:10 |
| balloons | Dashboard: http://ci.ubuntu.com/smokeng/trusty/ | 16:11 |
| balloons | http://ci.ubuntu.com/smokeng/trusty/touch/maguro/26:20131120.1:20131120.1/5059/ | 16:11 |
| mzanetti | but what about API changes in the service? | 16:13 |
| mzanetti | you want to catch those | 16:13 |
| mzanetti | ideally 2 steps. mock the service, test app with qmltestrunner. have autopilot tests walking through the api | 16:14 |
| cjohnston | using an external serivce for testing can provide you with false failures | 16:16 |
| mzanetti | 2-step testing... 1) mocks + unit tests, 2) integration tests that that it the api still works | 16:16 |
| mzanetti | yeah... you don't need too many AP tests then. Only a few that interact with the server. | 16:18 |
| mzanetti | and if the unit tests fail, you know its your app, if AP fails, you know the server's api has changed | 16:19 |
| cjohnston | you want AP tests for your app too | 16:19 |
| mzanetti | yeah... the test suite that interacts with remote server probably can't block CI | 16:19 |
| mzanetti | depends on the tests | 16:20 |
| mzanetti | yeah. split the test suites | 16:20 |
| cjohnston | if you run the tests but it doesn't block things when it fails, what's the point of running the tests | 16:20 |
| cjohnston | why not just check it locally | 16:21 |
| mzanetti | cjohnston: how? | 16:21 |
| cjohnston | open the app? see if you get data? | 16:21 |
| balloons | mzanetti, cjohnston et la, if you wish: https://plus.google.com/hangouts/_/76cpj9mr34bq9l8h7722ebh2uc?authuser=0&hl=en | 16:21 |
| cjohnston | +1 balloons | 16:22 |
| cjohnston | there needs to be a quick way to get a fix/release out (and SRU'ed or whatever the process will be) so that people don't stay broken | 16:22 |
| plars | ...or revert | 16:23 |
| cjohnston | plars: I think we are talking about if the 3rd party API changes and breaks the app | 16:23 |
| cjohnston | so reverting wouldn't do anything | 16:23 |
| plars | cjohnston: ah, ok I was thinking about the more general case | 16:23 |
| cjohnston | you don't want to have tests that fail randomly | 16:24 |
| cjohnston | having a test that fails due to a 404 isn't good practice | 16:25 |
| plars | how many things other than weather app do we have that depend on remote services? RSS reader does but it shouldn't have 3rd party api problems | 16:26 |
| cjohnston | These 'external tests' more than likely wouldn't be run at a set interval, they would be run on MP or something, so it could potentially be days before the tests tell you there is a problem | 16:26 |
| balloons | plars, not sure offhand, but you can look through the list | 16:27 |
| balloons | clock does also for instance | 16:27 |
| balloons | stock ticker | 16:27 |
| cjohnston | mzanetti: ^ | 16:27 |
| cjohnston | (my comment) | 16:27 |
| mzanetti | cjohnston: no... I'd run them with fixed intervals | 16:28 |
| cjohnston | mzanetti: where? | 16:28 |
| plars | ok, so there's lots more coming | 16:28 |
| mzanetti | cjohnston: not sure I understand... | 16:29 |
| mzanetti | cjohnston: on jenkins | 16:29 |
| cjohnston | mzanetti: I don't think that CI is planning to have fixed interval testing abilities | 16:29 |
| cjohnston | its going to be event based | 16:29 |
| mzanetti | cjohnston: unless it changes from what it is now, that wouldn't be a problem | 16:29 |
| cjohnston | its presently event based | 16:30 |
| cjohnston | afaik | 16:30 |
| plars | maybe on the assert for the testcases that depend on remote services, it could note that it is a remote test, or even in the test case name? That way someone looking at the results might know to take it with a grain of salt | 16:30 |
| mzanetti | cjohnston: note that "on MP" is actually a 15 minutes event based job that just checks if some commit has happened before doing anything | 16:30 |
| cjohnston | mzanetti: right.. but if there is no MP, it doesn't do anything | 16:30 |
| mzanetti | cjohnston: sure. but I don't see a problem with oh. it's 15 pm. run that one test suite then | 16:31 |
| cjohnston | I don't think that we have that in the plans for CI | 16:31 |
| mzanetti | cjohnston: as I said... unless it changes from what it is now, it just works | 16:32 |
| cjohnston | I still don't see how? Jenkins doesn't say "its a specific time, do this" jenkins has a job that runs every X minutes and looks for MPs. if there is no MP it doesn't do anything. you would have to add some sort of logic to give it the ability to say its X time, run tests... | 16:33 |
| mzanetti | cjohnston: all the jobs are scheduled like cron jobs. the ci jobs run on */15 * * * *. when they run the first thing they do is to run a custom python script which checks for a changed MP. So if you set the config to * * 0 * * and don't use that python script. it runs the job every midnight | 16:35 |
| plars | that sounds much better | 16:37 |
| ahayzen | balloons, we use a custom toolbar :) | 16:40 |
| balloons | ahayzen, I ALWAYS forget | 16:40 |
| dkessel | hey community :) | 16:42 |
| balloons | hey dkessel | 16:42 |
| balloons | dkessel, want to write some music tests? | 16:43 |
| balloons | :-) | 16:43 |
| dkessel | hmm :) will the new nexus 7 be supported ? ;) | 16:44 |
| dkessel | balloons, ^ | 16:44 |
| balloons | dkessel, hmm.. | 16:46 |
| dkessel | balloons, tbh my motivation kind of depends on having a supported device to really run stuff on. | 16:46 |
| balloons | dkessel, definitely | 16:46 |
| cjohnston | balloons: whats the problem? need more info | 16:49 |
| plars | sorry, was responding to something else | 16:49 |
| balloons | cjohnston, they exist in jenkins, but not http://ci.ubuntu.com/smokeng/trusty/ | 16:49 |
| plars | what was the question? | 16:49 |
| cjohnston | balloons: are they being run the the daily image testing? | 16:50 |
| plars | no, they are not | 16:50 |
| plars | they need to be added to run in jenkins | 16:50 |
| plars | can you open a bug for them at https://bugs.launchpad.net/ubuntu-ci-services-itself | 16:51 |
| plars | and we will look at adding them | 16:51 |
| cjohnston | balloons: ^ | 16:51 |
| balloons | cjohnston, plars ty | 16:51 |
| mzanetti | fine with me... was just a suggestion | 16:53 |
| balloons | mzanetti, :-) | 16:53 |
| ahayzen | balloons, thanks | 16:53 |
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Currently no events are active in this room - http://summit.ubuntu.com/uds-1311/community-1/ - http://irclogs.ubuntu.com/2013/11/21/%23ubuntu-uds-community-1.html | ||
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Track: Community | Limiting Surveillance | Url: http://summit.ubuntu.com/uds-1311/meeting/22016/limiting-surveillance/ | ||
| dholbach | gQuigs and everyone else: https://plus.google.com/hangouts/_/7acpj8a1gvucftcan7cn47bbm4 | 17:59 |
| dholbach | anyone else who's interested in joining? | 18:03 |
| dholbach | for everyone who's interested in joining: https://plus.google.com/hangouts/_/7acpj8a1gvucftcan7cn47bbm4 | 18:06 |
| dholbach | here are the notes for the discussion: http://pad.ubuntu.com/uds-1311-limiting-surveillance | 18:06 |
| mdeslaur | the issue is the hundreds of mirrors, and most of them wouldn't be willing to turn on https | 18:07 |
| rbasak | https stops proxy caching from working. That's a feature I value. But sure, I can understand that it stops third parties from seeing packages you're installing, and people should be able to do that. | 18:08 |
| mdeslaur | but yeah, having a subset who are willing to do it would be nice | 18:10 |
| rbasak | I had no further comment. Having the option for users to have https would be great. | 18:10 |
| rbasak | Even if initially it's just one mirror! | 18:10 |
| mdeslaur | we can't turn on browser apparmor profiles by default | 18:14 |
| rbasak | Are flash and hangouts constrained right now? | 18:14 |
| mdeslaur | people are using too many insane browser plugins for that to work | 18:14 |
| rbasak | Or are they tied to the browser? | 18:14 |
| rbasak | Perhaps make it really easy for people to constrain their browser if they choose to? | 18:15 |
| mdeslaur | for example, some people are using gpg plugins, so we couldn't even block access to your gpg keys | 18:15 |
| mdeslaur | rbasak: yes, we have a trivial script to enable it | 18:15 |
| mdeslaur | it's shipped, but not enabled by default | 18:15 |
| rbasak | mdeslaur: I wasn't aware of this! I knew it shipped though. | 18:15 |
| rbasak | Perhaps we can point to these things on a single "Ubuntu Privacy" wiki page? | 18:16 |
| alexlist | Hm. I think in the long run these issues with plugins will have to be solved upstream. | 18:16 |
| mdeslaur | https://wiki.ubuntu.com/SecurityTeam/FAQ#Firefox_AppArmor_profile | 18:16 |
| rbasak | Lag is around one minute I think. | 18:17 |
| mdeslaur | ok, perhaps I can join the session | 18:17 |
| mdeslaur | one sec | 18:17 |
| dholbach | maybe you guys should all join :) | 18:17 |
| dholbach | woohoo | 18:17 |
| YoBoY | hi | 18:19 |
| dholbach | there's https://help.ubuntu.com/community/AppArmor which mentions the firefox profile too | 18:19 |
| sbeattie | you could use firefox profiles for that, though I think firefox may have discontinuing that feature on their roadmap | 18:24 |
| rbasak | mdeslaur: OOI, is there an executable tool that runs aa_change_onexec and then execs "$@"? I want to do this for mutt to use lynx to convert my HTML, to contrain it further than what a general lynx profile would usually do. | 18:28 |
| rbasak | (sorry for the aside) | 18:28 |
| mdeslaur | rbasak: aa-exec | 18:28 |
| rbasak | Ah. Awesome! | 18:29 |
| rbasak | I failed to find that the last time I looked. THanks :) | 18:29 |
| rbasak | As a first step, how about stopping the user from using any plugins at all except for the default safe set? | 18:31 |
| rbasak | Then we wouldn't have to worry so much about what arbitrary plugins could do. | 18:31 |
| asomething | somewhere on my list of things to do when I find the time is to package up a Cryptocat webapp | 18:35 |
| asomething | not much to really app right now, just something people might want to check out | 18:35 |
| asomething | s/app/add/ | 18:36 |
| linuxtech | Check on cert generation for various server software so that they have PFS, http://en.wikipedia.org/wiki/Perfect_Forward_Secrecy. | 18:39 |
| asomething | https://wiki.debian.org/FreedomBox | 18:39 |
| pstolowski | dholbach, bless you! | 18:40 |
| dholbach | pstolowski, thanks :) | 18:40 |
| linuxtech | It would also be nice to provide some testing tools so admins can check the existing certs. | 18:41 |
| dholbach | maybe it'd be good to agree on a place where future discussions about this kind of stuff could happen? | 18:45 |
| dholbach | the security team's list? | 18:45 |
| dholbach | just so people who watch the video later on, know where to go | 18:45 |
| mdeslaur | perhaps a privacy specialized wiki page and list would be appropriate | 18:46 |
| YoBoY | désabling lot of scopes is just super hard and takes time | 18:47 |
| YoBoY | (for exemple if you want to disable scopes not with results not localised) | 18:48 |
| asomething | just noticed that when you turn off online search the dash still read "Search your computer and online sources" | 18:51 |
| asomething | just for people's piece of mind that should probably change | 18:51 |
| mdeslaur | asomething: that's reasonable, could you file a bug on that, please? | 18:52 |
| linuxtech | Setup DNSSEC for ubuntu.com. Note Debian.org and fedoraproject.org are using DNSSEC. | 18:52 |
| asomething | mdeslaur, looks like there is already a bug for the tooltip on the launcher | 18:53 |
| asomething | https://bugs.launchpad.net/unity-scope-home/+bug/1202160 | 18:53 |
| udsbotu | Launchpad bug 1202160 in Unity Home Scope "Dash title "Search your computer and online sources" should update according to privacy setting" [Medium,Confirmed] | 18:53 |
| YoBoY | (an old version of the Dash Privacy Interface : http://iloveubuntu.net/dash-privacy-interface-adds-app-icon-focused-mode-and-legal-notice-attention ) | 18:54 |
| mdeslaur | asomething: good | 18:54 |
| dholbach | we're running out of time | 18:57 |
| cm-t_desktop | scrrenshots : http://pix.tdct.org/?action=search&method=tag&tag=dash-privacy-interface | 18:59 |
| rbasak | cm-t: I don't understand why this is a privacy problem in the first place. I understand that users want to have a choice. But in what way is a user's privacy violated when he types something into a box which searches the Internet? | 19:00 |
| dholbach | thanks a lot everyone! | 19:00 |
| rbasak | How is this different from a Google search box? | 19:00 |
| dholbach | track summaries up next: http://summit.ubuntu.com/uds-1311/meeting/22026/track-summaries/ | 19:00 |
| alexlist | time to sleep... it's 3am over here. thanks everyone! | 19:00 |
| === ChanServ changed the topic of #ubuntu-uds-community-1 to: Currently no events are active in this room - http://summit.ubuntu.com/uds-1311/community-1/ - http://irclogs.ubuntu.com/2013/11/21/%23ubuntu-uds-community-1.html | ||
| gQuigs | rbasak: he isn't expecting to search the internet | 19:01 |
| rbasak | gQuigs: so why did he type into a box which says "Search online sources"? | 19:01 |
| YoBoY | rbasak, when someone type something on the dash it's not obvious to know he is sending data on the network. He is not using a webbrowser with a google search box | 19:01 |
| gQuigs | rbasak: we aren't giving him a choice | 19:01 |
| cm-t | exactly | 19:02 |
| gQuigs | rbasak: see: https://www.eff.org/deeplinks/2012/10/privacy-ubuntu-1210-amazon-ads-and-data-leaks | 19:02 |
| gQuigs | rbasak: see What EFF wants from ubuntu | 19:02 |
| cm-t | even if the user want to switch off the onlince sources, if he doesnt know ubuntu, he will probably find into the dash, so will perform an onine search | 19:02 |
| cm-t | it is a privacy concern since the scale of his search is outside the scale he want (localy/online) | 19:03 |
| cm-t | rbasak: ↑ | 19:03 |
| gQuigs | rbasak: it's actually quite similar to what was wanted by the dash-privacy-interface | 19:03 |
| gQuigs | err... provided by | 19:03 |
| cm-t | even if the network can't hack and read his search, it can know this IP has a think using the computer at this time | 19:04 |
| rbasak | I still think it's a stretch to say that there's a privacy leak in the steps it takes the user to find the settings box. | 19:05 |
| gQuigs | rbasak: that's true | 19:05 |
| cm-t | there is a privacy concern, but its coming from the design of the dash | 19:05 |
| gQuigs | but the EFF is there to protect users who may not understand the privacy implications at all; they are looking out for the "best privacy interests" of all | 19:06 |
| cm-t | dash-privacy-interface act to notify the user that ubuntu has great feature to search online, so the user can swith on or off | 19:06 |
| rbasak | I think the concern is that most users don't want to be asked a ton of questions. They just want their computer to work. | 19:06 |
| sparkiegeek | and if they were to search in Firefox to figure out how they turn off the online source for the Dash? is that just as invasive? | 19:06 |
| sparkiegeek | (note I'm +1 on saying it's a stretch - playing devil's advocate here) | 19:07 |
| YoBoY | we only ask for better information of the users on first use. Searching on internet from the dash is not the problem (using or not the canonical server as a proxy of the searches) | 19:07 |
| cm-t | rbasak: excatly, user doesnt want to be asked a lot, so that is why ubuntu is the first to think DPI (dash-privacy-interface) should be removed from the french iso | 19:07 |
| cm-t | but to be removed, there should be a native solution built into the dash | 19:07 |
| gQuigs | I agree that, the "search for privacy" issue is a stretch | 19:08 |
| cm-t | so it is a design issue imho | 19:08 |
| cm-t | ubuntu → ubuntu-fr | 19:08 |
| gQuigs | but the defaulting on, IMO is also unnecessarily reckless with user privacy | 19:08 |
| rbasak | gQuigs: so you think that the Dash deceives users? | 19:09 |
| cm-t | deserve the user's privacy | 19:09 |
| gQuigs | rbasak: based on the bug that we don't change "search online sources" when users turn it off, then yes, we actually now know it does.... | 19:10 |
| gQuigs | it's likely part of the reasons users aren't trusting us when we say we turned it off | 19:10 |
| gQuigs | hence fixubuntu.com | 19:10 |
| rbasak | gQuigs: where's the bug? | 19:10 |
| cm-t | that is why there are rms/eff/mass bad buzz about ubuntu's dash | 19:10 |
| gQuigs | rbasak: https://bugs.launchpad.net/unity-scope-home/+bug/1202160 | 19:11 |
| udsbotu | Launchpad bug 1202160 in Unity Home Scope "Dash title "Search your computer and online sources" should update according to privacy setting" [Medium,Confirmed] | 19:11 |
| === dholbach_ is now known as dholbach | ||
| rbasak | gQuigs: that looks reasonable, thanks. It doesn't leak privacy though. It's the other way round! | 19:12 |
| gQuigs | rbasak: right, but it's likely part of the reason why people aren't trusting us when we tell them, "See we tell users it will search the internet" and then here turn it off in Privacy settings | 19:13 |
| gQuigs | and we still tell them it will search online... ... | 19:13 |
| rbasak | It's a shame that this came about when we made an effort to notify users. | 19:13 |
| gQuigs | rbasak: what do you mean? | 19:14 |
| rbasak | The bug is valid, of course. But it's ironic that we have to respond to critics because we're erring on the side of caution when warning users. | 19:17 |
| rbasak | Since that is exactly what critics are complaining that we're not doing. | 19:18 |
| cm-t | user's privacy is a topic when the user want to share something at a scale he wants, and this scale isn't the want he wanted | 19:18 |
| cm-t | so there is this notifying issue | 19:18 |
| cm-t | to be sure the user know the scale of the search | 19:18 |
| cm-t | but there is another issue: | 19:18 |
| cm-t | how to change the scale | 19:18 |
| gQuigs | rbasak: that's not on the side of caution; it indicates that we aren't sure what's going to happen with the users search | 19:19 |
| cm-t | you allready said the user doesnt want question, may i traduce that by, he does'nt want open a setting tab every time he wants to switch the scale ? | 19:19 |
| rbasak | Well, the point of the interface is that it'll figure out what you want, rather than you having to tell it by adjusting the scale all the time. | 19:19 |
| cm-t | so i think the dash should provide an in-situ way to turn on/off online search | 19:20 |
| rbasak | If you don't want that, then perhaps you're using the wrong interface. | 19:20 |
| rbasak | I think this interface caters for "human beings", which is what Ubuntu sets out to do. | 19:20 |
| rbasak | But Xubuntu and Lubuntu are still available and promoted as alternatives. | 19:20 |
| cm-t | hum | 19:21 |
| YoBoY | rbasak, are you saying we have to stop to support and promote ubuntu because we think there is something wrong from the point of view of a majority of our users ? | 19:21 |
| cm-t | let image possibility | 19:21 |
| rbasak | To be clear, this is My Humble Opinion. | 19:21 |
| cm-t | you work at office | 19:21 |
| cm-t | it is Ubuntu | 19:21 |
| rbasak | Xubuntu and Lubuntu are still part of Ubuntu. | 19:21 |
| cm-t | you can't install your OS | 19:21 |
| rbasak | I certainly don't want to tune my settings for every search. | 19:21 |
| cm-t | user'privacy is the user's privacy, not the host (company for example) privacy in first place | 19:22 |
| rbasak | YoBoY: IMHO, you don't speak for the majority of users (and neither do I) | 19:22 |
| cm-t | so xubuntu, lubuntu is not the awnser to this problem | 19:22 |
| YoBoY | I speak for the users of my locoteam | 19:22 |
| YoBoY | (ubuntu-fr) | 19:22 |
| cm-t | rbasak: ↑ | 19:23 |
| rbasak | Well, I support your ability to patch software as you wish. | 19:23 |
| rbasak | And make that available to users who want it. | 19:23 |
| rbasak | That's the great thing about open source/free software. | 19:23 |
| rbasak | What Unity upstream do is not up to me, though. | 19:23 |
| YoBoY | but if we want to follow the ubuntu brand policy, if we do that, we have to change the name of our live DVD | 19:24 |
| YoBoY | (the ubuntu french edition which is a localised live DVD) | 19:25 |
| rbasak | I understand that problem. | 19:25 |
| rbasak | And that you want something different. | 19:25 |
| rbasak | So the question is: what should Ubuntu do? | 19:25 |
| cm-t | and our goal is not to support dash-privacy-interface, it is suppose to be a temporary fix waiting for a dash design real fix | 19:25 |
| rbasak | That's probably something between the Unity design team and the community council· | 19:25 |
| YoBoY | inform more the users on how the dash work on first use. See it like a tutorial on first launch pointing to the right informations | 19:26 |
| cm-t | 20:20 < rbasak> I think this interface caters for "human beings", which is what Ubuntu sets out to do. | 19:27 |
| cm-t | not sure if i make sense by awsering you after that in many lines | 19:28 |
| cm-t | but | 19:28 |
| cm-t_desktop | cm-t: test | 19:29 |
| rbasak | Don't get me wrong. I think it's great that you've actually written the interface that you're proposing. | 19:29 |
| cm-t_desktop | well, i am the main developper of that | 19:30 |
| cm-t_desktop | and i am maybe the first who want to see it dead | 19:30 |
| gQuigs | And Ubuntu is going against the EFF here... I'm not sure if anyone has more experience protecting user privacy then them | 19:30 |
| cm-t_desktop | because i want ubuntu have a native awser to that problem | 19:30 |
| cm-t_desktop | that is why i wrote in the pad that design/ergonomy team should be here for that point | 19:31 |
| cm-t_desktop | because dash-privacy-interface should be just temporary here until dash has an in-situ switch online/offline higltlighted | 19:32 |
| rbasak | Well the EFF is basically saying that they want Unity made less useful or to bombard the user with questions. | 19:32 |
| cm-t_desktop | exactly | 19:33 |
| cm-t_desktop | so that' is why it is an ergonomy/design issue | 19:33 |
| YoBoY | it's the price to have a all-in-one app_launcher/file_search/dictionnary/calculator/store/… | 19:33 |
| cm-t_desktop | not a dev first level, but ergonomy/design | 19:33 |
| cm-t_desktop | the user story should has the less as possible the user | 19:34 |
| cm-t_desktop | but the user should access as fast and as easy as possible to the swith online/offline option | 19:34 |
| cm-t_desktop | i'm not designer, but an example: [SEARCH STRING here the button] ← unity search bar | 19:35 |
| cm-t_desktop | with a notification overlat when searching for the first time if it is opt-in | 19:36 |
| cm-t_desktop | overlay* | 19:36 |
| cm-t_desktop | like I said I am not designer/ergonomist, but here just some idea, i saw many of them in the pad that was openened 1 years ago | 19:36 |
| cm-t_desktop | them == other idea | 19:37 |
| cm-t_desktop | I want the day team design/ergonomy add this concern of their todo list (not flaming at all they do nothing, i know they have lot of work) | 19:38 |
| cm-t_desktop | since ubuntu ask user using a LTS for no cutting edge feature, ubuntu-fr added dash-privacy-interface in the hope something will come with 14.04 LTS | 19:39 |
| cm-t_desktop | but it seems not (for the moment) | 19:40 |
| rbasak | I appreciate your flame-free approach here. Are you connected to the right team? Right here the discussion has happened but I fear that it will go no further. | 19:40 |
| gQuigs | cm-t_desktop: yea, I was just going to add, please do contact the design team directly | 19:40 |
| rbasak | Perhaps dholbach can help with connecting you to the design team? | 19:41 |
| dholbach | you can ask in #ubuntu-design | 19:41 |
| dholbach | or check out http://community.ubuntu.com/contribute/design/ for a couple more links | 19:42 |
| YoBoY | (last time I contacted them for help on something 7 months ago,… still waiting an answer ^^") | 19:42 |
| cm-t_desktop | let's try anyway | 19:42 |
| YoBoY | +1 | 19:43 |
| YoBoY | we have to try, and if it's not working, we have to try harder ;) | 19:43 |
| cm-t_desktop | it is my message, hey we need that , we need that, but I don't remember asking them direclty | 19:43 |
| rbasak | I am in favour of a "privacy mode" which is activated all the way from lightdm and logs you in as an anonymous user with all settings turned to maximum privacy (at the cost of minimum utility). | 19:46 |
| cm-t_desktop | if this prodcue something, i will be proud (to make something better in ubuntu) but ashamed (waiting 1 years to get in touch with the good people) | 19:46 |
| rbasak | If that were based on Unity, then I figure online searches would be turned off by default. And Tor turned on in Firefox, with Firefox constrained and all plugins turned off. Etc. | 19:46 |
| rbasak | But only for that login. | 19:47 |
| rbasak | Anyway, time to go. | 19:47 |
| cm-t_desktop | rbasak: thanks for sharing your point of view | 19:47 |
| cm-t_desktop | good night | 19:47 |
| rbasak | I appreciate your thoughts on this. I think I disagree with you, but that's not to say that I think you should not be heard. | 19:47 |
| cm-t_desktop | thank | 19:48 |
| rbasak | And I really appreciate that you're being constructive. | 19:48 |
| cm-t_desktop | i am tryng my best | 19:49 |
| rbasak | I think you deserve a straight answer. | 19:49 |
| cm-t_desktop | i am not a unity hater, i really love the feature to search online, but also i am aware of the privacy issues, so i try to find something | 19:50 |
| cm-t_desktop | we have a french speaking uds pleanry session (#ubuntu-uds-plenary) so i have to go | 19:52 |
| === Pricey is now known as pricechild | ||
| === pricechild is now known as Pricey | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!