17:03:07 <isabela> #startmeeting ux biweekly meeting
17:03:07 <MeetBot> Meeting started Wed Oct 26 17:03:07 2016 UTC.  The chair is isabela. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:03:07 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
17:03:11 <isabela> alright
17:03:17 <isabela> this is what I had for the agenda
17:03:22 <isabela> 1. Update on next steps we discussed last meeting for guidelines
17:03:22 <isabela> 2. Update on the state of the user experience for security slider on orfox
17:03:23 <isabela> 3. underexposed event in Berlin \o/
17:03:27 <isabela> anything else?
17:03:33 * Samdney lurks
17:03:49 * linda shakes head no
17:03:50 <isabela> oi Samdney o/
17:04:03 <Samdney> hi
17:04:14 <isabela> ok
17:04:26 <isabela> to point 1 - scouttle i know you are super hyper busy
17:04:40 <scouttle> yes, but happy to talk guidelines!
17:04:54 <isabela> do you think we should still aim for that blog post?
17:04:58 <scouttle> yes
17:05:15 <scouttle> I am still in the process of revising the guidelines based on the Seattle feedback
17:05:29 <isabela> ah
17:05:31 <scouttle> and want to run them by Elio as well, plus convert them to an open-source format
17:05:32 <isabela> that is great to hear
17:05:44 <scouttle> and then I think we should be ready to do the blog post
17:06:03 <isabela> sounds great
17:06:07 <scouttle> realistically, given how busy I am prepping for a conference next week and Underexposed the week after, I think we should table work on the blog post until mid-November
17:06:07 <isabela> do you need help with anything?
17:06:14 <scouttle> I don't think so
17:06:23 <scouttle> I am eager to get it done and out before the end of the year
17:06:32 <isabela> yeah i was about to suggest that we aim for after the underexposed
17:06:36 <scouttle> and to start talking about scoping a proposal to OTF to implement them
17:06:37 <isabela> because i know y'all super busy
17:06:41 <scouttle> :D
17:06:43 <PhilipL> Hi!
17:06:56 <linda> scouttle: I'm coming to underexposed, btw~
17:07:01 <isabela> PhilipL: o/
17:07:04 <isabela> we just started
17:07:12 <scouttle> yes, excited to have you join us Linda!
17:07:16 <isabela> we are on #1 of the agenda - following up with the guidelines etc
17:07:46 <scouttle> one of our sponsors is not being super responsive re: travel funding, so still not sure about how much we'll be able to reimburse, but I'm optimistic
17:08:00 <linda> okie doke (~˘▾˘)~
17:08:00 <scouttle> so I'm glad you already have your ticket secured and aren't waiting on that answer :)
17:08:03 <isabela> summary is that scouttle is incorporating the feedback from seattle, will run it by elio and make it opensource format
17:08:11 <linda> scouttle: ٩(^‿^)۶
17:08:15 <isabela> once that is done they will work on the blog post
17:08:18 <scouttle> by Elio and also the rest of the group of course!
17:08:21 <isabela> a realistic estimation is for mid nov
17:08:35 <linda> After the blog post, will we then try to get applications to adopt the guide?
17:08:54 <isabela> linda: plan is to start with tor browser and our site
17:09:13 <linda> isabela: but does this happen after the blog post?
17:09:24 <isabela> yes
17:09:35 <linda> isabela: ah, okay. Just curious about the timeline.
17:09:53 <isabela> we are hoping to use the blog post to get more cash money for simply secure to have resource to continue helping us with it
17:09:53 <scouttle> I mean, I think work can start before the blog post if folks want, doesn't have to be strict
17:10:02 <scouttle> but ^ yes exactly
17:10:20 <isabela> is ok, i think we have enough going on btw now and mid nov
17:10:26 * linda agrees (•₃•)
17:11:01 <scouttle> indeed!
17:11:06 <isabela> hehe
17:11:07 <isabela> ok
17:11:24 <isabela> anything else on #1? questions, commments?
17:11:40 <linda> not on my end.
17:12:08 <dcf1> Yawning had a question in #tor-dev yesterday: "(do we have art assets from the new style guide thing, ideally in a vector format?)"
17:12:19 <dcf1> So there's your application developer desire already.
17:12:25 <isabela> :) nice
17:12:50 <isabela> me and linda were talking about having a repo for ux team
17:12:51 <scouttle> w00t
17:13:03 <isabela> we can get these there for folks to use
17:13:05 <scouttle> short answer is not yet but we can generate them pretty easily
17:13:23 <scouttle> once we know which assets they wanted specifically
17:13:31 <linda> ☝ (^‿^)☝
17:13:43 <scouttle> (the guide isn't particularly rich in vector assets, but the ones that are there are easy to pull out)
17:14:19 <isabela> i can follow up with yawning - i suspect he is looking for it because he is working on UI for the sandboxing work he is doing for tor browser
17:14:37 <scouttle> cool
17:14:54 <scouttle> one thing that came up in Seattle is what process people should use when they have a question outside of the style guide
17:15:05 <scouttle> that's something I think it would be good for this group to talk about
17:15:20 <scouttle> and, as a consequence, the community process for expanding the style guide over time
17:15:29 * linda nods
17:15:30 <isabela> i am in favor of us using trac tickets with the ux component
17:15:31 <PhilipL> Isn't contacting the mailing list the preferred step?
17:15:37 <isabela> this way is easier for devs and others to follow up
17:16:01 <scouttle> that sounds right to me, PhilipL
17:16:17 <isabela> PhilipL: true - i guess it depends on the request .. if is just a question that should be it
17:16:25 <scouttle> I also wonder if it makes sense to identify a point person to respond to queries, even if just to say "I'm not sure but let me get back to you"
17:16:26 <isabela> PhilipL: if is a task request i would vote for the ticket
17:16:35 <scouttle> that way queries to the UX team don't go into a vacuum
17:17:18 <PhilipL> ok, maybe a ticket would be better. but in case anyone has a question before a ticket, they should write to the mailing list. or should we offer a personal mail adress?
17:17:35 <isabela> i think for ux in general linda is the point person for sure / but guidelines specific she might route it to the folks working on it
17:17:48 * isabela talking for linda :P hehehe
17:17:50 <isabela> sorry
17:17:56 * linda is okay with this
17:18:09 <linda> I would like to know who to redirect question to, if they come.
17:18:22 <isabela> PhilipL: main list /me thinks
17:18:39 * linda is also in favor of this main list idea.
17:18:49 <PhilipL> Yes.
17:18:51 <scouttle> ok, so for now the guidelines say to go to the main list
17:19:06 <scouttle> but when/if we get a ticket component we can suggest that people can do that as an alternative
17:19:19 <scouttle> and Linda will be responsible for making sure all queries get a response
17:19:27 <isabela> hehe
17:19:28 <scouttle> even if it's a "I'm not sure let me figure that out for you"
17:20:11 <PhilipL> In other cases people will find Linda's mail address somewhere else.
17:20:21 <isabela> heheh
17:20:38 <isabela> #item for isabela to review components at trac with linda
17:20:46 <isabela> just a thing i want to make sure to remember :)
17:20:52 <linda> ヽ(•◡•)ﾉ
17:21:07 <isabela> ok
17:21:12 <isabela> should we move to #2?
17:21:36 <isabela> update on the work we are doing with the security slider + orfox / as we gave ourselves the deadline for it to be eom
17:21:43 <isabela> linda: do you want to give an update?
17:22:05 <linda> sure!
17:22:38 <linda> We have finished finalizing the wording for the security slider.
17:22:40 <linda> https://docs.google.com/document/d/1U_tbrC1lpuKHZg_QKD_TUVd6ImjTJ3jWZkSxjXgceZI/edit#
17:23:23 <linda> A bunch of tor people pitched in on the document (15-20?), and we're pretty happy with the wording.
17:23:27 <linda> The main changes were that:
17:23:56 <linda> -we have 3 security settings rather than four (there was no reason for having four and three makes it less confusing for users because there are less options)
17:24:37 <linda> -we re-named our settings from "low, med, high" to "standard, safer, and safest" (people were concerned that the default settings were low, and this more accurately communicates that the extra settings are optional precautions)
17:24:52 <scouttle> nice
17:25:16 <scouttle> (both of those sound like really solid changes)
17:25:17 <linda> -we added an introductory sentence to each setting explaining why changes are made ("disables featuers that are often dangerous, "only allow features required to visit static sites and basic services")
17:25:37 <linda> -we shortened the list of changes (without losing information, we think).
17:25:45 <linda> ヽ(•◡•)ﾉ
17:26:03 <PhilipL> Yes, this is way much better!
17:26:07 <linda> Oh, and we're also thinking of providing a link to the tor maunual for more information.
17:26:19 <linda> But we haven't found where specifically to link to, etc.
17:26:33 <isabela> #item find the read more link for security slider
17:26:34 <isabela> :)
17:26:47 <PhilipL> Although, one thing, isn't the new wording misleading: Standard has "all Tor Browser features" ans Safer "disables features"?
17:27:07 <isabela> maybe disable websites features
17:27:15 <isabela> so they dont think they are tor browser features
17:27:29 <linda> okay. fixed.
17:27:40 <PhilipL> hope that doesn't make the wording too complex. it was just a thought
17:27:48 <isabela> boom!
17:27:52 <linda> No..
17:27:57 <linda> I considered doing that but didn't..
17:28:01 <PhilipL> :)
17:28:02 <linda> But you're right.
17:28:23 <dcf1> isabela: your security slider manual link might be https://tb-manual.torproject.org/linux/en-US/security-slider.html
17:28:33 <isabela> nice!
17:28:36 <isabela> thansk dcf1 !
17:28:36 <dcf1> Also /macosx, /windows.
17:29:34 <isabela> ok so these changes will be applied to orfox first but tbb team will later apply them to desktop as well
17:29:40 <isabela> so experience is consistent
17:29:56 <linda> I also plan to make a wiki page summarizing our process and changes (for posterity).
17:30:00 <linda> ٩(^‿^)۶
17:30:31 <isabela> another thing on this is that once we have a working build for android we will give it to a couple trainers in different countries to try it out with users to see if the ui makes sense and if there anything other modification that might be necessary
17:30:40 <isabela> that might happen around january or february
17:31:31 <isabela> +1k for documenting
17:31:39 <isabela> alright
17:31:53 <isabela> any comments or questions or .... for this point?
17:32:10 <PhilipL> Now the Standard text sounds like a regular browser. The wording fits better to the new Safer text but lost its tor benefits
17:32:35 <linda> >_>
17:33:00 * linda makes slight tweak
17:33:19 <PhilipL> :) nice
17:33:45 <linda> ᕦ(^‿^)ᕤ
17:33:56 <isabela> hehe
17:34:20 <arthuredelstein> FWIW I think this new security slider text looks really good. Great work!
17:34:30 <linda> arthuredelstein: (≧◡≦)
17:34:54 <linda> arthuredelstein: you were a big help! Thanks!
17:35:09 <isabela> arthuredelstein: !
17:35:13 <isabela> :)
17:35:49 <arthuredelstein> Thanks for doing this project! Definitely a major improvement.
17:36:18 <isabela> i am really happy with this work - i think it was a great experience for the ux team working with a dev team project
17:36:32 <isabela> just the fact we had a wireframe and time to work on wording before coding
17:36:35 <isabela> is a big plus!
17:36:37 <linda> ٩(^‿^)۶
17:37:07 <isabela> alright people!
17:37:11 * linda likes to fork stuff and instrument functional prototypes
17:37:17 <linda> maybe on bigger projects.. haha
17:37:34 * isabela thinks we could move to point #3 of the agenda
17:37:46 <isabela> underexposed!
17:37:46 <PhilipL> *** PhilipL agrees
17:37:50 <linda> ☝ +1
17:38:21 <isabela> scouttle: do you want to chat about it a little?
17:38:24 <isabela> we are shipping linda there!
17:38:25 <isabela> hehehe
17:38:36 <scouttle> yay, we're excited to have her!
17:38:49 <scouttle> it's a small invitation-only gathering for UX designers and researchers
17:38:58 <scouttle> on Nov 11th
17:39:07 <scouttle> this year the theme is "Designing for trust in the era of algorithms"
17:39:09 <linda> invitation only.
17:39:15 * linda feels special.
17:39:18 <scouttle> :)
17:39:47 <scouttle> we'll talk about designing for trust in terms of encryption and privacy-preserving software, but also in other edgier things like IoT and machine learning
17:39:54 <scouttle> it should be a good time!
17:40:06 <linda> How can I get involved at Underexposed?
17:40:20 <linda> I'd be happy to talk about Tor in general, or any topics you suggest.
17:40:33 <scouttle> cool, thanks for offering!
17:40:37 <isabela> scouttle: we were wondering if we should bring print-outs like people did last year
17:40:45 <linda> If I can somehow sneak in a UX-specific presentation, or a recruitment pitch.. hehe I'd like that too.
17:40:55 <scouttle> That's a good question. Ame and I are working on all those things now.
17:41:18 <linda> When things are figured out, let me know what works for the event.
17:41:27 <scouttle> It's a small group (definitely <50, probably 25-30), so you should be able to find time to recruit at least during the breaks!
17:41:28 <scouttle> will do
17:41:35 <linda> Sure.
17:41:44 <scouttle> the format is unusual, short talks mixed with break-out sessions
17:42:04 <linda> are all the talks set?
17:42:09 <scouttle> no, we're figuring them out now
17:42:16 <linda> (★‿★)
17:42:20 <isabela> heheh
17:42:21 * linda tries to pounce on a slot
17:42:24 <scouttle> heh
17:42:41 <scouttle> so, as I said the focus is UX for "algorithmic" issues
17:42:50 <linda> Yes.
17:42:58 <scouttle> so, how do we surface the big-data happenings to users, and allow them meaningful interaction?
17:43:02 <scouttle> hmm
17:43:10 <scouttle> now that I think about it, something on the recent slider project might make sense
17:43:19 <isabela> interesting
17:43:20 <scouttle> because you're trying to communicate highly-technical decisions to end users
17:43:31 <linda> We're also considering doing something with the tor metrics team.
17:43:34 <scouttle> and give them access to a meaningful choice without overwhelming them
17:44:07 <linda> To give users an overview of what's going on, for researchers to answer questions, etc. It's not done though.
17:44:27 <scouttle> so, we're careful to not let "algorithmic" mean just "data visualization"... but with that caveat I'd be curious to know more about your plans there and whether you've gathered any data yet
17:44:29 <linda> I can present about the slider though, if that sounds interesting.
17:44:46 <scouttle> it could also just be, "here are some of the challenges Tor faces in communicating about highly technical things"
17:44:56 <isabela> scouttle: that was a ux research GeKo passed to me that is around this area
17:45:01 <scouttle> the talks are usually just 10 minutes or so
17:45:11 <linda> If I could just choose a thing to present on, I was planning to present on my research on the Tor Launcher Interface.
17:45:29 <scouttle> what would the algorithmic angle be?
17:45:34 <linda> I think that might be more algorithmic--I do an evaluation of how users connect to Tor.
17:45:50 <isabela> not sure if there is still time to invte the person - but she did a machinelearning prototype that includes displaying different graphs according to user habits to let them know something is dangerous or not
17:45:54 <linda> Currently, it's manual, but there's talk of automating it.
17:46:23 <scouttle> so, your research method is algorithmic, but I don't think that the UX in question is for an algorithmic system, though, right?
17:46:26 <linda> The evaluation finds that people take a really long time to connect, have trouble with it, and that it might be better if the bootstrapping process was automated.
17:46:48 <linda> And we explicitly discuss alternative ways to safely connect people.
17:46:56 <isabela> scouttle: i am not a big fan of machine learning to help build right behavior / not for our area though / but is an interesting study
17:47:27 <linda> Hmm.
17:47:36 <linda> i.e. rather than asking users to configure a brige or proxy, we ask
17:47:46 <scouttle> Yeah, part of the idea of the focus is that machine learning, IoT etc. is coming whether we like it or not, so what are the security and privacy design/research considerations to make sure users are protected?
17:48:24 <linda> I don't think that there is a machine learning element.
17:48:35 <linda> It's more to keep users safe.
17:48:43 <isabela> yeah
17:48:48 <linda> We could automate the process, but it would reveal to a network eavesdropper that someone is connecting to TOr.
17:49:04 <linda> So we give users full control, but configuring tor-specific components is hard.
17:49:13 <scouttle> yeah
17:49:19 <linda> And many people end up trying a connection that doesnt work, so they reveal that anyway.
17:49:31 <linda> so we were thinking of alternative approaches..
17:49:38 <linda> like automating after the first failure.
17:49:44 <linda> asking users what country they are in.
17:49:53 <linda> asking them if there is risk if the process was automated.
17:49:56 <isabela> i think our thing like the tor launcher or security slider is about communication technical settings for users so they can use things right and be safe against the world of big data
17:50:05 <linda> and discussion of how the automation could happen.
17:50:12 <scouttle> I think that one potential talk framing could be "Like ML systems, Tor is a complex thing that is hard to explain to users. Here are three mini cases where we've been working to make it more accessible."
17:50:37 <linda> Mmmm...
17:50:43 <scouttle> And then spend literally no more than 3 mins each giving a taste of your research, the slider, and your goals for metrics
17:50:46 <scouttle> the key would be to make it brief
17:51:00 <linda> Oh goodness.
17:51:01 <scouttle> the goals of the talks are to be thought-starters more than eg educational
17:51:02 <isabela> scouttle: i am curious if y'all have any case of ML where is good for the user (giving our area /privacy and security)
17:51:04 <linda> Very very brief.
17:51:13 <scouttle> yup! it would be a challenge linda
17:51:15 <scouttle> we can talk more
17:51:19 <linda> okie doke.
17:51:22 <scouttle> I need to run talk more to Ame
17:51:35 <isabela> ok
17:51:55 <scouttle> isabela I think there are cases where people can be helped by ML
17:52:05 <isabela> i agree
17:52:10 <isabela> specially for e-learning
17:52:16 <scouttle> the question is whether the system gathering and using the data respects users' right to privacy
17:52:27 <scouttle> and whether users can get a meaningful understanding of what's going on
17:52:35 <scouttle> and whether they have a meaningful opportunity to opt out
17:52:48 <scouttle> but, that's sort of what I'm hoping we'll get out of the workshop
17:52:56 <isabela> gotcha
17:53:06 <scouttle> in the "era of algorithms" that we're in today, what does it mean for algorithm-based systems to be ethical?
17:53:14 <isabela> i will fwd to you the paper i am talking about / is very related to this topic
17:53:25 <scouttle> beyond just slapping up a privacy policy on your website and calling it done
17:53:29 <scouttle> cool, thank you
17:54:01 <isabela> or the sneaky things like people do
17:54:13 <linda> ヽ(•◡•)ﾉ
17:54:22 <isabela> for instance the settings on android for you to add a phone # to your twitter account to receive notifications has a small text that says
17:54:27 <scouttle> also, PhilipL I have no idea if this is remotely related to the stuff you do or how far you are from Berlin, but if you're more on the experience side than the visual/branding side, perhaps we should talk about whether you might be available :)
17:54:29 <isabela> 'people will be able to find you by your phone number'
17:54:42 <scouttle> yes, indeed
17:54:43 <isabela> even tho you are not checking any boxes allowing that or uploading your contacts
17:54:45 <scouttle> that's a dark pattern
17:54:55 * isabela was part of that :(
17:55:02 <isabela> is all growth tactics pattern :(
17:55:23 <isabela> ok
17:55:32 <isabela> last question on this!
17:55:41 <linda> (⨀_⨀)
17:55:42 <isabela> who from this ux gang is going beyond linda ?
17:55:54 <isabela> PhilipL: ?
17:55:55 <isabela> or elio?
17:56:07 <linda> (╯°□°）╯ come with meeeee
17:56:12 <isabela> hehehe
17:56:21 <PhilipL> All that is really interesting, but I think it is ok for me to not participate. Until the end of the year I am really busy
17:56:31 <linda> got it.
17:56:34 <scouttle> We're hoping to get Elio there.
17:56:39 <isabela> okdoki
17:56:45 <isabela> cool
17:56:55 <scouttle> plus lots of other good, friendly people that will be good for Linda to get to know :)
17:57:01 <isabela> !
17:57:01 <isabela> nice
17:57:05 <isabela> thanks for organizing it scouttle o/
17:57:12 * isabela doesn't have any more questions hehe
17:57:14 <scouttle> of course! :D
17:57:24 <isabela> anything else folks?
17:57:30 * linda shakes head no
17:57:49 <scouttle> not on my end!
17:57:50 <isabela> if not i can end the bot
17:58:08 <isabela> ok!
17:58:11 <isabela> #endmeeting