Version number changes for Thunderbird

Thomas Düllmann thomas.duellmann at gmx.de
Mon May 30 22:09:55 UTC 2011


First of all, thanks to Wayne for pointing me to this interesting and 
informative communication channel, and hi to everyone known and unknown.

It's unfortunate I'm starting out here with a critical post, furthermore 
it's somewhat emotional (blame it on the late time of the day, and 
reduce the severity of the message accordingly after reading), but 
anyway, here's my comment on the version number changes planned for TB:


My whole-hearted agreement on Karsten's scepticism against the new 
versioning intentions.

Thunderbird is not known for moving very fast in its development, and 
where it has moved fast (as in the case of the new global search, or 
maybe the quick filter bar), it has left a desert of UI problems and 
desiderata that are unsolved to this day. My suspicion is that the 
merging of MozMessaging with MozLabs with the intention of developing 
new phantastic products to cover the full range of today's communication 
channels will not exactly improve the manpower situation for core 
Thunderbird without bells and whistles, on the contrary. The language in 
that announcement seems way too defensive to be fully trusted...
Fast development trains will only help if there is sufficient manpower 
to actually develop, instead of a dependence on God-sent volunteer 
contributors like Jim who is currently restoring big chunks of one 
largely neglected and deteriorated core functionality of TB: attachments 
UI. While other deteriorated corners will continue to languish, although 
concepts and suggestions are mostly out on the table.

What I am trying to say is that exploding version numbers that do not 
reflect actual feature changes that are relevant to the user (which 
Gecko versions are not) will add insult to injury from a normal users 
perspective:
While version numbers will be rocketing skywards under the new scheme 
(starting with that leap from 3.x to 5), there will be nothing tangible 
in terms of bugfixing and UI-improvement / added features that will 
actually justify those leaps from the traditional user's viewpoint where 
version number changes reflect visible improvments. It is a myth that 
version numbers could be de-emphasised, as many of our users are very 
aware of version numbers: from painful experience widely documented in 
the forums like getsatisfaction, bugzilla etc., they have come to 
associate version number changes in Thunderbird with more complications 
of workflow, continuous erosion of screen real estate, and other dangers 
which overshadow some of the significant improvements that co-occured.

> we felt that matching Firefox would make it clearer
> for developers as to which version of Thunderbird was based on which
> gecko/Firefox version.

That main reason for changing the versioning system starts out all wrong 
because it focuses on the small no. of developers rather than the large 
userbase. Furthermore, I suppose people that are actually capable of 
developing Thunderbird will also be able to find out Tb's gecko version 
even if it's not in the title of the product. If I am not mistaken, 
finding out the version number used to be as simple as going to help > 
about. Why do I care that much? Maybe because it's another one of those 
changes in Thunderbird that the world does not need, and that may turn 
out to cause more harm than good to an already endangered product. 
Imagine Thunderbird 10 and we may still not be able to search all of our 
address books in one go - wouldn't that be weird?

I'm not always as pessimistic, most of the time I just find lots of bugs 
and suggest improvements, and than patiently wait for some years till 
someone picks up on it (or even acknowledges the problem, as in so many 
unbelievable keyboard/focus issues). A good indicator of such hopeful 
occasions used to be an increase in the version numbers of the product. 
With the new suggested versioning system, the version numbers would 
become largely meaningless to that regard. Yes, version numbers do 
matter and they should not be abused for technical reasons that are 
irrelevant for the larger parts of the userbase, or for a false 
understanding of marketing that bloats version numbers without improving 
the product at the same pace.

Otherwise, with respect to matching FF's version no: So far I naively 
assumed that the difference in version numbers between FF and 
Thunderbird actually had some truth value with respect to development 
(due to differences in manpower, resources, you name it). It may not be 
wise to cover up that truth by pretending to be what we are not.

Best wishes and greetings,

Thomas

On 27.05.2011 00:06, Karsten Düsterloh wrote:
> Mark Banner aber hob zu reden an und schrieb:
>> Whilst we could have kept the same numbering system, or adopted a
>> different one, we felt that matching Firefox would make it clearer
>> for developers as to which version of Thunderbird was based on which
>> gecko/Firefox version.
> I outright doubt that this is solving any real-world problem.
>
> Chrome-angst-driven version number frenzy might make (very limited, IMO)
> sense for a competing browser, but do users really care about the Gecko
> version of their _mail_ client?
>
> I do understand, though, the assumed marketing "value" of pairing the
> version number with FF. I just think it's nonsense. ;-)
>
>> We'll also be de-emphasising the version numbers in our releases,
> Well, you may do, but will the users? ;-)
>
>
> Karsten
> _______________________________________________
> tb-planning mailing list
> tb-planning at mozilla.org
> https://mail.mozilla.org/listinfo/tb-planning
>




More information about the tb-planning mailing list