You don't have to assume anything; the reasoning why this was removed
is stated pretty clearly in the closing of the relevant bugs. You may,
specifically, want to read the final comment in the most recent bug:
http://code.google.com/p/chromium/issues/detail?id=100319
where a future way that this may be doable in an extension is mentioned.
--
Torne (Richard Coles)
to...@google.com
Have you also read all the chromium developer comments in other bugs
on this topic? I didn't mean that single comment explains everything;
I was drawing your attention to the goal of having this doable in an
extension in future.
> I personally manage to live without the vertical tabs in Chrome (as
> also without the PKI signing support, which prevents me from using my
> on-line banking with Chrome), but what upsets me and I guess also most
> of the site tabs supporters, is that you (Chrome developers) keep
> insist to have clearly explained the reasons for this decision, while
> you failed to do this.
The reason, as far as I know (I did not work on this feature, but have
followed some of the previous discussions about its removal) is
"wasn't finished, wasn't implemented on all platforms, was an ongoing
maintenance issue, and very few people used it". As pkasting
acknowledges in the comment I referred you to, we are aware that the
few people who did use it were very attached to it, but developer
resources are finite.
I am not a UX designer and have no desire to be, so I will not comment
on the design aspect here.
> But go beyond that absurdity to core principles of software design.
> What decent object oriented programmer (or UX designer) in this modern
> age would say that you should artificially constrain what is
> reasonable for the user to do? (I.e., let users or extensions put a
> side/top/bottom/floating bar anywhere they think makes sense for
> them!)
Pretty much all software has many artificial constraints on what the
user can do, because there are practical limits to how many
configurations it's possible to test, and making it trivial for users
to get into entirely untested states is a recipe for disaster. I
realise you probably won't consider this relevant in this particular
case (since it's "only" a couple of choices) but this logic can
reasonably be applied to a very long list of things, and thanks to
multiplication this produces a lot of configurations very rapidly :)
This is somewhat tangential to the actual side tabs issue in any case;
the removal of the side tab code is *not* just a case of "artifically
constraining" what the user can do, since it was a separate
implementation which had to be kept up to date.
> Just one final point. As you say, the so called pretty clear
> explanation points to a discussion (id-52084) as, "a future way that
> this may be doable in an extension." So not only do we have no clear
> indication if or when it will ever be doable, if doable. But even if
> that development takes place it will still require someone to develop
> the extension.
I am also not a project manager and don't set the direction of the project, but:
Like most open source projects (and most closed source projects, for
that matter), we don't commit to timelines. I understand that this can
be frustrating to users, especially when it looks like a feature is
not just a long way off but may never happen at all, but you have to
understand that the project's priorities are never going to match up
with any specific user's priorities. A very small number of users who
*really want* a feature does not, generally, outweigh a much larger
number of users who would be better served by devoting resources
elsewhere.
> By the way, I separate the reasonableness or not of the decision to
> pull support for the feature, from the clearly unreasonable, cavalier,
> arrogant, thoughtless, unprofessional way it was (and is still being)
> handled.
I really don't know what you expect, here, and at this point I'm done;
I started replying here under the assumption that some or all of the
contributors to the thread were only recently aware of this removal
(when stable updated) and hadn't seen the previous discussions, and
thus would benefit from the explanation; I've now done that, and
continuing to retread the same ground with people who have already
seen the previous discussions is not useful, since I didn't implement,
or make any decisions about, this feature.