1928778 - Request for data collection - "% amount of users with setting [open links in tabs instead of new windows] enabled"

Status: RESOLVED WONTFIX

(Firefox :: Tabbed Browser, enhancement)

Description

[Dan]

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:132.0) Gecko/20100101 Firefox/132.0

Steps to reproduce:

search for a probe that measures this

Actual results:

none appeared

Expected results:

A probe to measure this would be useful for the attached data collection request (about to be attached to this report in a moment)

Comment 1

[Dan]

Attachment: request.md

Comment 2

[Dan]

What questions will you answer with this data?
If the setting still is required to be included in about:preferences#general and if it wouldn't suffice to have it as an about:config pref instead

Why does Mozilla need to answer these questions? Are there benefits for users? Do we need this information to address product or business requirements?

Determine whether a product or platform change has an effect on user or browser behavior:
If over 99% of users have this setting enabled, then we could consider to add the easy, short, memorable string newTab as a new HTML <a> target Attribute to the HTML parser to handle it the same as the much less memorable string _blank, so target="_blank" would have the same outcome as `target="newTab" (_blank would still be working of course).

This makes writing HTML a little easier for many people who are about to learn HTML in the future.

Can current instrumentation answer these questions?
Not to my knowledge

<table>
<tr>
<td>Measurement Description</td>
<td>Data Collection Category</td>
<td>Tracking Bug #</td>
</tr>
<tr>
<td>Measuring the amount of users (in %) who have the setting "open links in tabs instead of new windows" activated</td>
<td>1</td>
<td>1928778</td>
</tr>
</table>

description of the ultimate data set in a public, complete, and accurate way
ultimate data set shows a % of users who have the about:preferences#general setting "open links in tabs instead of new windows" enabled.

How long will this data be collected?

• I want this data to be collected for 3 months initially (potentially renewable), to measure a trend if there is any.

What populations will you measure?

• Rapid release channel and ESR
• All countries
• All locales

If this data collection is default on, what is the opt-out mechanism for users?
to untick Allow Firefox to send technical and interaction data to Mozilla in about:preferences#privacy

Please provide a general description of how you will analyze this data.
a visual graph across 3 months with the % of users measured every 24 hours, so about 90 measurement points in the graph (to identify any trends)

Where do you intend to share the results of your analysis?
On bugzilla and maybe on my own tech blog aswell

Is there a third-party tool (i.e. not Glean or Telemetry) that you are proposing to use for this data collection?
No.

Comment 3

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Firefox::Tabbed Browser' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 4

[Chris H-C :chutten]

Comment on attachment 9434953 [details]
request.md

This data looks like it will be implemented in Firefox, in which case it gets to follow Firefox's process for data collection review which happens as a part of code review in Phabricator (this process is new since May). No forms required!

Comment 5

[Dan]

(In reply to Chris H-C :chutten from comment #4)

Comment on attachment 9434953 [details]
request.md

This data looks like it will be implemented in Firefox

What's your source for this statement?

Comment 6

[Dave Townsend [:mossop]]

(In reply to Dan from comment #5)

(In reply to Chris H-C :chutten from comment #4)

Comment on attachment 9434953 [details]
request.md

This data looks like it will be implemented in Firefox

What's your source for this statement?

The bug appears to be about instrumenting the setting "open links in tabs instead of new windows" which is a Firefox setting and so the data collection would have to be implemented in Firefox.

Comment 7

[Dave Townsend [:mossop]]

I think reading back over the bug there has been a misunderstanding here. The data collection request process is used when we already intend to collect some new data and we need confirmation that doing so does not violate our lean data practices.

I think maybe you're at the earlier step of that process and requesting that someone write a probe to collect the data you're interested in. Is that correct? I think that is unlikely to happen unless you're willing to write the probe yourself and we would need to understand more about why you need this data. I don't think the data would really be helpful in guiding a change to the HTML spec which has the potential to break existing websites, though if it is we would want to see the discussion on the HTML standard showing that such a spec change is under consideration and needs this data to move forwards.

Even without implementing a probe I can tell you that the vast majority of users have this setting turned on because it is the default and most users don't change their defaults without being prompted to do so.

Comment 8

[Dan]

By "a website breaking" you solely mean if it were to be accessed by other browsers other than Firefox (when using "target="newTab"" instead of ""target="_blank"" as soon as Firefox would be able to handle "target="newTab"" in the same way as ""target="_blank""), right?

Comment 9

[Dave Townsend [:mossop]]

(In reply to Dan from comment #8)

By "a website breaking" you solely mean if it were to be accessed by other browsers other than Firefox (when using "target="newTab"" instead of ""target="_blank"" as soon as Firefox would be able to handle "target="newTab"" in the same way as ""target="_blank""), right?

Existing websites may already be using newTab (fairly unlikely but possible) in which case any browser making it act in the same way that _blank does would break them. And if only Firefox were to implement this then sites would likely not use it as then they wouldn't work in other browsers. That's why any such proposal should happen at the HTML spec level so all browsers implement it. We wouldn't make any such change in Firefox before it appeared likely to become part of the spec.

Assuming that the answer to my previous question is yes you are requesting we start collecting data about this preference in order to support that change in the HTML spec I think this is a WONTFIX, even if there were meaningful discussion happening at the HTML spec level I don't believe that this data would impact the discussion in any way.

Comment 10

[Dan]

in which case any browser making it act in the same way that _blank does would break them

No I meant that the browsers would be able to interpret both "_blank" and "newTab" (and treat "newTab" the EXACT same way as "_blank".). But yes, websites in other browsers would behave differently in that latter case of only Fx having that. Dave do you realize that it appears to be incredibly hard to see any changes to official W3C spec levels? I'm not gonna waste a massive amount of time to propose another change in the WICG if even my previous proposal (from october last year!) hasn't been answered to anymore by any of the 4 engineers there since the summer... (Please see for yourself!! https://github.com/WICG/proposals/issues/125 )

Comment 11

[Dave Townsend [:mossop]]

(In reply to Dan from comment #10)

in which case any browser making it act in the same way that _blank does would break them

No I meant that the browsers would be able to interpret both "_blank" and "newTab" (and treat "newTab" the EXACT same way as "_blank".).

Which is a breaking change from how the target="newTab" works currently. Again fairly unlikely any sites are using this string (though that might be something it would actually be worth gathering data on, though it might be hard to do in a privacy sensitive fashion), but still not something to ignore.

Anyway this is largely irrelevant when it comes to the question on whether we should add a probe to collect the number of users that change the browser.link.open_newwindow as I don't see how the number of users who toggle that setting has any bearing on whether it makes sense to add a new magic target string that web developers can use to always create a new tab.

But yes, websites in other browsers would behave differently in that latter case of only Fx having that. Dave do you realize that it appears to be incredibly hard to see any changes to official W3C spec levels? I'm not gonna waste a massive amount of time to propose another change in the WICG if even my previous proposal (from october last year!) hasn't been answered to anymore by any of the 4 engineers there since the summer... (Please see for yourself!! https://github.com/WICG/proposals/issues/125 )

It is challenging. In part because it is important that we need to be a sure that all the rendering engines are likely to implement a given proposal because not doing so leads us to an even more fragmented web than we already have today.