David E. Ross wrote:
On 7/28/11 9:43 AM, Robert Kaiser wrote:
David E. Ross schrieb:
That is sometimes called "lemming behavior", which might be appropriate
in this case.  After all, the legend is that lemmings follow each other
off a cliff, into the ocean, and to their deaths.

You mean Firefox will die just as Chrome has and SeaMonkey has before?

Robert Kaiser


 From my long experience (30+ years) in software QA, the long-term cost
of frequently churning out versions with new features and not merely
bug-fixes will eventually either weaken necessary QA through shortcuts
to meet schedules or else consume developer resources to the point of
weakening the organization.  It appears that frequent releases might
already be driving away users who otherwise not only have to update the
product but also find that needed extensions no longer work.

Extensions constitute one of the major features of Mozilla-based
applications.  Today, the problem with extensions results from the fact
that, like SeaMonkey itself, most extensions are developed as hobbies by
volunteers who are not paid.  With frequent new versions of
Mozilla-based applications, extension developers find themselves in the
same situation as the Red Queen in Lewis Carroll's "Through the Looking
Glass", running as fast as they can just to stand still and having to
run faster than they can to make progress -- while still having to earn
a living.

+1

I'm a fan of feature release followed by n (n>=1) bug fix releases, so you get a usable version fairly often, like stability releases for the Linux kernel, that kind of thing.

In other words, having versions the (majority of) users find better than the last stable version. Having people generally feel that your QA has vanished is particularly bad when doing volunteer work, since the only reward you get is people thinking you have produced something great.

The fact that 2.2, with broken address book, was not considered a brown bag release and quickly followed by 2.2.1 indicates that quantity is more important than quality. If you never add names to your address book, 2.2 is fine. If you get 5-10 new addresses a day in "Collected addresses" and have to type them into the production AB by hand because you can't move them with drag and drop, it's worth the pain to fall back to an earlier (working) version, and disable automatic updates.

I know SM uses some shared TBird code, the one person I know who uses TBird tells me address book is still broken in the nightly she tried. Sigh.

I'm grateful for the work people do, but I think the whole Mozilla effort has lost its way. It feels as if Firefox is the only thing which still gets QA resources and fixes in a timely manner. And IIRC the fixes in SM address book were rejected for TBird, so it would have to be maintained in SM long term.

--
Bill Davidsen <[email protected]>
  We are not out of the woods yet, but we know the direction and have
taken the first step. The steps are many, but finite in number, and if
we persevere we will reach our destination.  -me, 2010


_______________________________________________
support-seamonkey mailing list
[email protected]
https://lists.mozilla.org/listinfo/support-seamonkey

Reply via email to