Re: [orca-list] Alternatives to Mozilla based software

[Michael Whapples <mwhapples aim com>]

Hello,
Regarding what goes in what update, yes I understand the need for change 
control, it was part of my course I was doing about 6 months ago. I 
suppose it just depends what sort of agility a project wants to have. As 
an example, it seems to be that orca developers are more responsive 
about getting fixes in stable branches. May be that is an unfair 
comparison as I probably fit better with the "normal" user for orca than 
for firefox (IE. accessibility in firefox is more specialist).

On the issue of Mozilla saying only report bugs for builds less than two 
weeks old, may be I have taken it too strictly, but that was how it read 
to me and so they may do better to strongly encourage people to try with 
latest versions and also specify what date the bug was introduced rather 
than just saying "only report bugs for".

It also appears to me that bugs get reported to orca but passed on to 
Mozilla. Sometimes end users don't know whether its orca or mozilla at 
fault, therefore orca may be the place a user will report it. May be 
work needs to go into passing such bugs on (also for other software as 
well).

I do use development versions of some software, firefox and thunderbird 
just aren't ones I want to do that with for main usage (I rely on them 
too much for being productive).

Michael Whapples

On -10/01/37 20:59, Trevor Saunders wrote:
> Hi,
>
> > A bit more detail: In thunderbird 3.1.x the XUL version used gives
> > additional window activated events, which makes orca loose focus if
> > reading messages in a new window (I never got into tabs for email).
> > When I filed a bug report about it, it was seen that it would not be
> > possible to get a fix included in thunderbird 3.1.x. Upgrading to
> > thunderbird 3.3a1 and above has resolved that bug but introduced
> > bugs such as the issue with wrong character being echoed when
> > backspacing and orca being unaware or what text is being inserted
> > into edit fields (eg. the input area for writing a message), so as I
> > type orca does no echo of typed words (I have that echo setting on
> > in orca) and Braille is not updated. Despite having filed a bug
> > relating to the backspace issue and a fix being created for firefox,
> > I am unsure whether that patch has been made to thunderbird and/or
> > which version of thunderbird it will appear in.
> So thunderbird is built from a  build system that uses serveral repos
> including mozilla central so presumably it will happen eventually, but
> to be honest I have no idea what the thunderbird developement plans are
> related to what version of gecko different thunderbird versions will
> use, so I doubt I or anyone else who works on core gecko accessibility
> has an immediate idea, you'd have to ask a thunderbird person.
>
> > I think that is probably more what is my dissatisfaction, the
> > ability of Mozilla to respond to bugs relating to accessibility. It
> > seems like for those of us who follow stable versions (may be trying
> > beta versions and at times alpha builds) bugs are reported too late
> > for Mozilla to get anything done for that stable branch of the
> I understand your problem, and I think it applies to all software to
> some extent.  The trick is that making fixes to stable releases mean
> making changes which means risk of breaking things which would be ...
> unstable.  So you have to limit what you change in stable branches to be
> minor fixes that clearly can't break anything or fixes to major problems
> like security issues or crashes, which accessibility fixes very often
> aren't.  I've thought about this issue before, but never come up with
> any solution.
>
> > product. Also the Mozilla requirement that when filing a bug you
> > should only do it against a build less than two weeks old is very
>   I think your entirpreting this a bit too strictly plenty of people file
>   bugs against branches and everyone is fine with that, just set the
>   version in the report to be 4.x or whatever, and don't be terribly
>   suprised if the bug can only be fixed in trunk or alredy works in
>   trunk.
>
> > restrictive. I am sorry, if the stable builds have sufficient bugs
> > to make me feel this way, why would I want to use something as my
> > main browser or email client which Mozilla accept may not be
> > anywhere near the quality of stable builds? As an example, in
> > checking the firefox issue I mentioned yesterday, I tried a version
> > of aurora but that has messed up a firefox settings file which now
> > prevents firefox 4 loading, so cutting off my fallback option.
> I'm not sure what to say here other than as I said applying fixes runs
> the risk of brekaing stuff...  Other than that I'm using nightly builds
> and they genrally seem to treat me fairly well, and I haven't see any
> issues starting the iceweasel 3.6 that debian ships.  But I'm willing to
> believe that there are bugs in unstable builds that write out bad
> settings or something.
>
> > Working the other way (IE. if I notice a bug in a release and then
> > try a nightly build), this still leaves us in a position of catch up
> yes, I think its pretty obvious that some people need to be testing
> nightly builds etc or issues can't be found before a branch is stable
> which is almost definitionally too late.  I think there are a few people
> in the accessibility community testing nightly builds and the like, but
> not nearly enough.  For example the bug you reported was introduced last
> june and was noticed in a few orca bugs in augest / sept iirc but then
> nobody did much of anything with those reports so nobody realized there
> was an issue until early this year.
>
> > and we hit the issue I mentioned that normally once a release is
> > made Mozilla don't seem to add accessibility fixes until the next
> > major update.
> >
> > That is probably enough of me going on, but I hope may be it gives
> > you an idea why I am so sick of Mozilla stuff and want some
> > alternatives. I will say, I do think its a shame as certain ways in
> > which thunderbird works I do like, its just the state of the
> > accessibility which has got me this way.
> well, fwiw I'm pretty happy with mutt.
>
> Trev