222715 - Mozilla 1.5 not stable on OS X 10.1.5, support should be delisted, homepage updated.

Status: VERIFIED FIXED

(mozilla.org :: Miscellaneous, task)

Description

[Paul]

User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.5) Gecko/20031007
Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.5) Gecko/20031007

I Love Mozilla. The borwser has seriously improved my internet experience...
even my tech weary wife uses it it is so wonderful--she has to admit, it is a
better browser. However things went very very wrong when I installed Mozilla
1.5. I lost all my bookmarks, I can not open a pdf (it just won't download) and
when I click on preferences:navigator:helper applications:appliction/pdf, the
browser crashes.

There is something very wrong with the Quality Assurance/Quality Controll that
this happened. Mozilla is mature enough that stuff like this shouldn't happen. I
can no longer recomend/use mozilla in this condition... this makes me very sad.
I might mark as blocker (since i will be taking a long vacation from mozilla),
but since finding a solution is a long term thing, I will mark as normal

Reproducible: Always

Steps to Reproduce:
1.try to install Mozilla 1.5

Actual Results:  
software takes a serious step backwards

Expected Results:  
software takes an incrimental step forwards

Comment 1

[Boris Zbarsky [:bzbarsky]]

Hardly a QA failing... QA found the bug, filed it, developers fixed it... but
after the branch.  And drivers did not take the fix on the branch.

*** This bug has been marked as a duplicate of 213639 ***

Comment 2

[Paul]

Regressions are _prima facie_ proof of a QA/QC failings; I am reopening this bug. There 
needs to be an assurance that regressions do not happen--the QA system is somehow 
broken. 

Definitions:
regression: a functionality loss from one version to the next
_prima facie_: at first sight
_prima facie_ proof : proof that needs no additional verification, once it is seen no 
additional arugmen is necessary.

Comment 3

[Boris Zbarsky [:bzbarsky]]

.

Comment 4

[Daniel Wang]

This bug is probably INVALID or WONTFIX.

INVALID because QAs can't really tell what the programmers should do, and QAs
can't hold release back because of major bugs. And also that 1.4.1, not 1.5, is
supposed to be the stable build.

WONTFIX because, well, since the breaking up of Netscape we are awefully short
on # of people doing QA. Unless bug 60455 is fixed (no, daylight saving doesn't
count), we can do nothing about this.

Comment 5

[Paul]

Look what's at stake here is weather or not Mozilla is a viable program or just a neat toy 
for the developers--I would think developers would be interested in this issue.

Why? If Mozilla doesn't have a functional QA process, I can't update my copy or 
recomend the softare to anybody (but a programer/QA tester). After this stuff hapened, my 
wife went back to IE (makes me cry) and said, "well, that was nice while it lasted." end of 
story.

Comment 6

[Ken Snider]

Paul, use 1.4.1.

If you want a stable build, 1.4.1 is for you.

1.5 is an ongoing development build. If that doesn't sound like the kind of
thing you'd like to be doing, then stick to 1.4.1.

Otherwise, how about a look at http://www.mozilla.org/quality/help/ and see if
there's anything you may be able to contribute?

Marking INVALID.

Comment 7

[Paul]

Good point. I have a new major QA/QC failing. The main page makes it look like 1.5 is 
stable build, but it is not.

Put your self in the unknowing user's shoes... Mozilla 1.5 is apparently a development 
build but...

1) Mozilla 1.5 is promenently featured
2) Mozilla 1.4.1 is not featured or indeed directly linked to.
3) Mozilla 1.5 is not listed as "developer edition" or "beta" or "alpha"
4) mozilla 1.5 is reffered to as "... the browser that PC Wrold calls 'Best of 2003.'" But it 
isn't, its the developers version of that browser.
5) there is a section titled "for testers", Mozilla 1.5 isn't there either--the nightly build is 
there.
6) There is a section titled "for developers" Mozilla 1.5 isn't there
6) You can buy 1.5 on CD!!! WHAT? There is a developers build that you can buy on 
CD?

Who are these developers/testers that want 1.5? Why call a nonstable release anything 
but "beta?" (see MySQL's home page for their naming convention.) Notice that when you 
go there they distinguish between "alpha" and "production."

Perhaps the root bug is to have a time line that requires a release of 1.5 at all. i.e. 1.5 
may have been released because it was time, not because it was ready. I don't know...

I am reopening this bug because this is still a major QA failing, it is just that it was not the 
step I identified. The QA failing was to not list 1.5 as a developer release and make that 
clear on the home page.

This is an open issue, it can be fixed by changing the Mozilla home page.

Comment 8

[Mitchell Baker]

This idea about 1.5 not being stable is eroneous.  1.4.1 is the long-lived
stability branch to be used by those who want stability most -- particularly
those who distribute mozilla as part of their product releases.  I don't know
how the idea got started that 1.5 is not intended to be stable, but it is incorrct.


To the bug reporter:  1.5 clearly has a bug that bothers you a great deal.  You
might file the bug under an appropriate summary, the QA failure heading is
probably inaccurate, as the comments indicate.

Comment 9

[Paul]

Changing Summary on advice from comment #8. Also, suggest that all OS's not used in 
by Developers or QA workers should be delisted. Notice that you can't even build Mozilla 
in 10.1.5

I am bothered. This browser has the potential to be a great browser. But if it starts 
claiming stability it does not have people will get burned and never come back. It was 
very wise for Mozilla to wait a very long time before going 1.0, it would be silly to throw 
away existing good will.

Comment 10

[louis bennett]

mozilla 1.5.1 was released to address issues specific to mac os 10.1.x, so this is fixed. 

if there are specific issues in mozilla 1.5.1 on mac os 10.1.x, please search for corresponding 
bugs, or file a new one if you can't find an appropriate existing bug.

Comment 11

[Daniel Wang]

v