13:58:51 #startmeeting ux team monthly sync august 13:58:51 Meeting started Tue Aug 4 13:58:51 2020 UTC. The chair is antonela. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:58:51 Useful Commands: #action #agreed #help #info #idea #link #topic. 13:58:55 hello! 13:59:20 is anyone around for the ux monthly sync? 14:00:04 im starting sharing the pad here https://pad.riseup.net/p/1D8sK8Zy74b_0qclC97I-ux-team-monthly-2020-keep 14:00:10 o/ 14:00:18 i saw Fabby around a few minutes ago :) 14:00:23 oi thurayya! 14:00:32 Hi everyone! 14:01:43 hi Fabby hi antonela :) 14:02:26 oh thanks for joining us Fabby 14:02:30 thanks for the invitation, David and Nathan from my team will be joining shortly 14:02:36 perfect 14:02:48 we are very exited to share with you the news about the Onion Browser 14:02:53 i shared the pad a few lines before 14:03:26 if you want to share any link or resource, you can paste it here, i'll share with the lists after the meeting 14:04:15 so, as a quick intro, I wanted to talk with Guardian folks about the research, design and development of Onion Browser 14:04:32 Ok, we have a step by step tutorial on or YouTube channel, I'll share that 14:04:38 super 14:05:21 I have some questions in my notebook for you Fabby, you can tell me when i can start :) 14:06:13 you can go ahead 14:07:04 so, you have been working with Orbot before and with Onion browser now, how the change of OS/platforms impact on the design decisions? How Apple users are different from Android users? Do we all want privacy and security anyways? 14:09:46 hi n8fr8 o/ 14:09:51 hi _hc o/ 14:09:54 Both types of users are concern about their privacy, but our decitions when in came to the Onion Browser were deeply affected by Apple restrictions 14:10:03 <_hc> hi! 14:10:15 "Apple requires all web browser apps to use the same core web rendering engine. Due to this limitation, we are unable to compile and include our own web engine, based on Firefox Gecko, as with Tor Browser for desktop and Android" 14:10:55 right, and now we are moving to Phoenix with TBA which has its own constraints too 14:11:04 Fenix not Phoenix 14:11:51 fabby, how you have been collecting feedback from users and how it became an actionable design decision during this iteration? 14:12:04 maybe my team can add to that, here is Antonela's question agaion "so, you have been working with Orbot before and with Onion browser now, how the change of OS/platforms impact on the design decisions? How Apple users are different from Android users? Do we all want privacy and security anyways? " 14:12:20 oh yes :) 14:13:07 There are a different set of design constraints on iOS (relative to Android), and a new set of technical constraints 14:13:59 There are also patterns Apple uses for Safari across platforms, and users may expect to see those same patterns in alternative browsers. 14:14:35 Regarding feedback: we have done user surveys before we did our latest round of work, and afterward. 14:14:40 User feedback is very important to us, we always reach out to users before we start the development of new apps or features , we reach our again during the development phase and after we finish the project, usually we do it through the IFF community, you guys, our partners and our database of testers 14:17:49 We also get a lot of great feedback, bug reports, enhancement requests, etc through Github for both projects. We may all want to host our own infrastructure or move to Gitlab, but something about Github makes it a much more social place for at least some users to connect with us 14:18:22 so we at least leave it up for that, while mirroring our projects to gitlab, etc, for those who want to work there 14:18:51 For example, in our initial user ssurvey, users were very interested in support for multiple tabs and management of open tabs. So, this was a big focus in our OB2 work. 14:19:06 n8fr8: right, github also brings you the entire github community and developers exposure, which is relevant for open source careers 14:19:22 r3ckd_gp: interesting, that was a miss in the previous version 14:19:37 Fabby: yes, having rounds of feedback across the entire development process is crucial so you can iterate on each version to adjust assumptions 14:20:04 folks, how do you separate major milestones in terms of features vs small iterations? how do you plan that? minor releases vs major releases? 14:20:17 antonela: there was a different architecture in the initial release, trying to mimic a pattern of Apple, and it just didn't work. 14:20:35 r3ckd_gp: yep 14:21:47 antonela: we put all our 2019-20 work into a single "major release" bucket - going from v1 to v2. But, we planned interim activities that had sets of updates that "made sense" to release as a unit 14:22:38 i wonder how we can articulate some specific funding for some specific feature but make it successful in their final release by spliting it in smaller milestones that also contemplates room for iterations in minor releases, lets say (1.0 major, 1.1 minor) 14:23:12 the problem right now is: we deliver a feature and we dont have room for iterations. The initial funding should contemplate iterations and/or smaller previous releases 14:23:12 What I am trying to say is that we planned interim releases, in advance, that made sense in terms of addition-of-user-benefit. I think we strayed from that path only once, adding one new interim release that was not planned 14:23:37 right, i've been following your minor releases in test flight and i saw the improvements over the time :) 14:24:16 that is why im asking this, because we should incorporate this regular development releases in the aim to be more (i'll say it) _agile_ 14:24:23 I will admit that we had several BIG interim releases and UX did fall into that category. This was partly because we felt there was no way to subset that work meaningfully 14:24:51 i see, yes 14:25:26 The biggest UX items were tabs, circuit display, and on-boarding. From a project management perspective, and a design perspective, we grouped it like that and released it like that 14:25:55 yes, and we are close to have feature parity 14:26:22 in fact, onion browser has better feature parity with desktop than tba 14:26:46 folks last question, and then you have some minutes of open floor to share anything you would like 14:27:07 the okthanks design tools have been helpful as well in simulating a new UI, for design iteration feedback. i wasn't a fan at first but it has grown on me. 14:27:18 antonela: one item you mentioned in your comments to us released to the management of tabs, specifically that tabs disappear when the app is not in use. That feature is BY DESIGN because it was a v1 feature, but there is a setting for mitigating or changing it 14:27:24 the fact it can run full screen on a phone, and feel like you are using the app is very helpful. 14:27:39 n8fr8: it is! prototypes are powerful 14:28:08 not to mention the paper prototypes that were done for other apps... which are great for in-person workshops, once we can have those again 14:28:09 r3ckd_gp: right, the stress between persistent and ephimerous features is real 14:28:22 agreed about the design tools. Also, I forgot to mention SETTINGS - the design team did a lot of work re-organizing those and making it easier for newer users to understand. 14:28:28 n8fr8: paper prototypes are lovely for early stages, indeed 14:29:18 another interesting aspect is that the lead designer (kaci) who worked on Onion Browser, actually wasn't that familiar with Tor. in some ways, her distance from the inner workings, gave her the necessary space to take on the typical user perspective 14:29:27 Yes, we got great feedback from the security levels 14:29:36 in short, antonela you know way too much about tor! :) 14:29:45 oh yes, sometimes the matrix eats youuu 14:29:47 hahhaa 14:29:56 people really like them, it made it easier for everyday users 14:30:20 but maybe there is an approach for hiring short term outside designers to provide ideas for specific areas 14:30:23 i mean, as a designer i often try to show another projects as an example just to show that we are not the only ones trying to fix _that_ problem 14:30:40 n8fr8 and antonela: yes, it has been a challenge to know how much of Tor-workings to expose, given the range of user maturity on iOS. On Android, at least initially, everyone asked for knobs and dials on all Tor settings. OB2 has a different approach 14:31:20 r3ckd_gp: exactly, that is a pain in open source projects, i dont think is just attached to Tor (not defending folks here haha) 14:31:38 but the old but still relevant bazaar vs cathedral thinking still runs 14:32:12 i feel human-centered design approaches are helping open source projects to move forward from that 14:32:20 by putting users in the center of the question 14:33:17 I honestly think it's more of a problem of the "squeaky wheel gets the grease" and there are a set of Tor users who are very very vocal about demanding the more absurd behavior and default settings. 14:33:33 is true 14:33:54 but is recurrent in open source projects, fans are more willing to demand 14:34:16 and the question is, who is the user who can be vocal? and who cannot but still needs to access internet? 14:35:00 True 14:35:11 so, my last question is in fact related with security settings 14:35:27 i saw an iteration in security settings and i think that is a very nice case for user research, i wonder, did you got any feedback about it? specially how per-site settings are impacting on the general configuration? 14:35:47 (per-site settings are still a wip in desktop, btw) 14:35:53 I'll let Fabby answer that one specifically 14:38:19 It made easier for users to have their customized settings for specific websites and not being concern about safety every time they start a new session, at the same time they had the option to give more flexibility to the websites they use the most and they trust 14:38:41 on the per-site settings, this was a major discussion point. from my personal perspective, there was this one core issue of "why won't my video play on X newsite or YouTube?" and that the desire to have get a video to work which required loosening security restrictions, should not then change the default for all sites 14:38:54 We also got comments like " It finally looks and feels like any other browser" 14:39:11 so backing up Fabby on what she said, but also this major issue of "youtube doesn't work" kept coming up 14:39:25 or "i can't watch a video on CNN" etc 14:39:35 or as a technically challenge user this made it easir for me to understand the risks and make my own choices 14:39:35 right, and is the best case to explain with per-site settings are a real need 14:39:44 the worst part of making that video works in desktop is that users downgrade their security and they never put them up again 14:40:04 and for sure, the frustration of having that dead tab 14:42:00 agree with n8fr8 that this was a major discussion point and a difficult one to solve in-the-general-case due to some of the must-haves being very fine-grained. 14:43:34 antonela, you also mentioned "random crashes" in your comments: we're super-aware of that and did make incremental progress on that problem. But this is one of the key areas where iOS is very different from our well-understood Android environment and we continue to have significant difficulty 14:44:30 yes, the pain is not the random crash per-se (it usually happens when the phone went to sleep mode) 14:44:30 on the next steps front, we are working to make MOAT more usable (faster, more polished) and starting to integrate Snowflake (both as bridge and client) on both Orbot and Onion Browser. We should find ways to share/collab on those to ensure consistency of experience and language. 14:44:45 the worst part is starting it again and lose all your browsing 14:45:06 figuring out how to take users through the bridge setup path is another challenging area of course 14:45:11 n8fr8: right! let me share what we are doing on that front 14:45:22 doing, by actually working on it 14:45:34 and please, feel free to jump in those tickets as well 14:46:44 https://gitlab.torproject.org/tpo/anti-censorship/trac/-/issues/40004 14:47:05 and this is for TBA https://gitlab.torproject.org/tpo/applications/fenix/-/issues/34405 14:47:49 it is still being discussed, I really want to make Tor Browser proactive on offering the best bridges available to users 14:48:16 we discussed in the past magic links, QR's 14:48:59 if we ask the user to give consent and answer a puzzle we can make it painful just the first time, but not everytime 14:49:17 antonela: "the worst part is starting it again and lose all your browsing". That is a FEATURE not a bug. That was the initial design users wanted "tabs go away in background" 14:49:41 r3ckd_gp: i know, but instead, is painful because i did not close the browser! the browser crashed! 14:50:10 by painful i mean, not expected 14:50:40 agreed that crashes are unfortunate. 14:51:02 ye, but also the crash is not random 14:51:28 most of the times it happened when the phone went to sleep 14:51:42 I have to test, though: if one uses the setting of NOT closing one's tabs on background, what happens after a disorderly exit? It might be true that we have enough memory of the tab composition to restart them then. 14:52:03 im not sure, i can try it deeper 14:52:38 crashes when phone goes to sleep: I thought that was partly fixed when we moved to a newer Tor release, but I might be wrong on that. 14:52:58 i bet is a networking issue, but im not sure 14:52:59 yes 14:53:36 and a quick update about the work with bridges, we met a group in HK in the past week and we are using this bridges script to learn more about bridges issues to inform the next iteration (you can sneak peek too) 14:53:37 https://gitlab.torproject.org/tpo/ux/research/-/issues/4 14:54:10 (and commenting for sure) 14:54:21 people we are reaching the hour 14:54:27 FOLKS thanks for joining us today 14:54:35 that was a very insightful conversation and i'm glad to have you all building apps around Tor to still bring users to the network <3 14:54:51 is always a pleasure to collaborate on hacking stuff :) 14:55:39 Thanks for the invitation! 14:56:03 anything you want to share Fabby? is the survey still running? 14:56:11 I'll keep the link for the lists 14:56:32 ahh should we update tpo.org/download to point to the new onionbrowser webpage? 14:56:35 We haven't close it, you guys can still add to it and we will appreciate it greatly 14:56:41 instead the apple store? 14:56:46 Fabby: oki, perfect 14:57:00 https://www.torproject.org/download/ 14:57:00 Yes, please do, I'll add it to the note pad 14:57:04 thanks Antonela 14:57:22 Fabby: super, doing that MergeRequest soonish 14:57:34 r3ckd_gp, n8fr8 thank you both also for joining! 14:57:46 and thanks Fabby for being a bridge, as always <3 14:58:07 agreed, thanks for Fabby for keeping this connection open. Bye now! 14:58:07 have a nice week folks! stay safe! 14:58:40 #endmeeting