Feature #11020
Installation procedure can be intimidating.
0%
Description
The first page in the installation sequence assumes too much. Instead it should inform the user that several options will be presented beyond the initial page. The subsequent page again assumes the user will want to install an amnesic variation of linux while displaying a link to the DVD/USB procedure (most commonly used prior to the overhaul) in relative obscurity. I would instead (with equal visual emphasis) tell them that they have several options. It is at that point that the site should inform the user of the time it will take to finish the procedure.
When I first saw the need for 2 USB sticks, I almost impulsively closed the window with the thought that another great security tool has lost focus of the importance of a smooth unassuming UX. It’s only when I squinted to the bottom right hand corner of the screen that I saw my preferred option in blinding green text. I understand the security implications of using a purely amnesic OS, but I have little problems saving and reintroducing my state file to maintain my guard node.
Please properly represent the entire range of TAILS deployment options. Otherwise I fear very few will take the leap.
Files
Subtasks
History
#1 Updated by intrigeri 2016-01-29 12:48:07
- Target version deleted (
Tails_2.0)
(Just removing the buggy target version, I’ll let frontdesk triage this further.)
#2 Updated by sajolida 2016-01-31 10:27:04
- Category set to Installation
- Status changed from New to Confirmed
- Assignee set to Tonda
- QA Check set to Info Needed
- Type of work changed from Website to End-user documentation
- Affected tool set to Installation Assistant
Thanks for the feedback!
- By “first page” you mean https://tails.boum.org/install/ or something else?
- I’m not sure to understand what you mean by “assumes the user will want to install an amnesic variation of linux” as Tails is nothing but an amnesiac version of Linux. So yeah, we’re assuming that people know what Tails is about when they start the installation procedure. We know that we have some work to do on explaining better the properties of Tails, but this is out of scope here.
- Which link on which page are you referring by “displaying a link to the DVD/USB procedure”?
- What was you “my preferred option” and why?
#3 Updated by Tonda 2016-02-07 03:25:52
By “first page” you mean https://tails.boum.org/install/
Yes this one.
*I’m not sure to understand what you mean by “assumes the user will want to install an amnesic variation of linux” as Tails is nothing but an amnesiac version of Linux. So yeah, we’re assuming that people know what Tails is about when they start.
I was hinting at the fact that the primary use case for tails is to remain untraceable. Installing defeats the amnesic quality. Having the installation of it be the primary option while leaving the iso/usb creation option relatively hidden seems odd considering the projects original focus on amnesic usage. The three options should be equally highlighted. IMO the dvd/usb option should be first (far left).
*Which link on which page are you referring by “displaying a link to the DVD/USB procedure”?
https://tails.boum.org/install/win/index.en.html
It’s almost as if it is encouraging users to install instead of the more common and originally intended usage. If the team is worried about guard node attacks, why not just inform the user on that particular page of the issue and potential mitigations like importing/exporting the state file into tails to retain their guard. Better yet why not give users to option import/export their state file via a simple GUI accessible in the desktop.
Example:
User starts up tails, a message pops up and says “if you have a state file click here to import it”. Once clicked the file browser opens for file selection. Done!
This can even be done in the greeter after the user inputs a password.
When the user goes to power down their system, and option to export their state file could then provided.
What was you “my preferred option” and why?
Who knows how many infections or vectors accumulate over time. The clean slate approach provides users greater confidence that their system isn’t compromised. An attacker would then need to start from scratch. As long as a user’s state file is imported there should be little issue. Security being paramount, this is the most secure option is it not?
#4 Updated by sajolida 2016-02-07 16:18:07
- Status changed from Confirmed to New
- Priority changed from Elevated to Normal
I don’t understand which three different options you are referring to. A vast majority of our users, and that’s probably the recommend setup, install it on a USB stick. So I don’t understand the distinction you make between “installing” and “iso/usb creation”. Also note that the more proheminently displayed scenario are always USB installation. So I don’t understand what you mean by “leaving the iso/usb creation option relatively hidden”.
Did you try to follow https://tails.boum.org/install/win/usb/ until the end?
#5 Updated by Tonda 2016-02-08 19:38:27
The word installation invokes a sense of permanence. Tails boot disks are in most cases ephemeral. IMO the wording should be changed but that’s another issue. The first displayed procedure (furthest to the left) requires two usb drives. This is relatively expensive and intimidating. Why not instead show the last two options first. They are the easiest, least costly, and the most commonly used methods of deployment IMO.
#6 Updated by sajolida 2016-02-09 11:09:05
- Status changed from New to Rejected
- Assignee deleted (
Tonda)
> The word installation invokes a sense of permanence. Tails boot disks are in most cases ephemeral.
This is not true. More than half of the WhisperBack reports we receive are from people using the persistent volume. “Persistent” being an antonym of “ephemeral”.
> The first displayed procedure (furthest to the left) requires two usb drives. This is relatively expensive and intimidating.
That’s a technical limitation that we are explaining to the user.
And 8GB USB stick nowadays cost less than 5€.
> Why not instead show the last two options first.
From our user testing sessions, people are much more interested in installing to USB sticks than burning DVDs. For security reasons, using virtual machines is not recommended.
In the future, please make sure to explicitly name things you are referring to, like “the last two options”, or “the first page”, please give their label and URL. Otherwise I’m playing a constant guessing game and waisting precious time.
> They are the easiest, least costly, and the most commonly used methods of deployment IMO.
This is not true both from our user testing and the media statistics we get from WhisperBack reports.
so I’m rejecting this ticket as it’s based on false claims.
#7 Updated by Tonda 2016-02-15 02:15:27
- File Proposal.jpg added
Intrigeri please weight in on this. I’ll keep it simple. Check the attachment or follow this link: https://imgur.com/78So5nL
#8 Updated by intrigeri 2016-02-18 16:42:32
> Intrigeri please weight in on this.
Not sure why I’m expected to “weight in”. I’m not working on this aspect of Tails myself, and sajolida provided actual facts that back his reasoning, which is quite convincing.
#9 Updated by Tonda 2016-02-22 01:44:06
What facts? Those from current users of Tails who already have the technical knowledge to install the OS? Or the non tech savvy that have perused the website and made their observations?
I don’t care about the former as they need little hand holding. New users on the other hand need things to be crystal clear and as easy as possible to understand. That is the goal of all of this UX redesign isn’t it? To make is simple for even grandma with her poor eyesight to use Tails. So why is this relatively small but useful change being rejected? Take a look at the imgur image in my last post. Does it not make sense? I know this is anecdotal but only people in my local social circle with intermediate computer knowledge felt somewhat comfortable with the two prominent options. They still felt that my suggestion would lower the barrier.
I asked you (intrigeri) to weight so that I might get another perspective on the matter instead you…. Nevermind. Please, forget your technical backgrounds for a moment and put yourself in the shoes of a technical novice. The reason for the UX research is because this team obviously has a hard time doing this.
Please, reconsider the change in the interest of greater adoption of this fine OS.
#10 Updated by sajolida 2016-02-22 08:41:12
As explained in Feature #11020#note-6, we based our choices on the statistics we got from WhisperBack reports and observations we did during user testing sessions. That’s what I was referring to by “facts” about our user base.
#11 Updated by Tonda 2016-02-23 01:15:00
Nice! So is my proposal still rejected? Or was that small sample of testers enough to make the team confident that granny and simple bob can smoothly jump into the OS? Did you check the image I was referring to? This one: https://imgur.com/78So5nL . Does it make sense to make all the options more prominent? I don’t see how it could hurt. Forgive me for taking so much of your precious time.