Re: Why Emacs needs a modern bug tracker

[David Kastrup <dak@gnu.org>]

Óscar Fuentes <ofv@wanadoo.es> writes:

> Eli Zaretskii <eliz@gnu.org> writes:
>
>>> > Not just evidence: you should volunteer to do actual work.
>>> [snip]
>>> 
>>> ESR volunteered for implementing the DVCS and the bug-tracker. I
>>> volunteered for spam-filtering and monitoring the bug-tracker. I can do
>>> other misc things, such as pre-testing the system and later assist other
>>> users on its use.
>>
>> Yes, I know.  But -- and please forgive me for being so blunt -- I
>> have enough grey hair to remember a few instances in the past where
>> people with good intentions said things like that.  Sadly, intentions
>> are all we are left with to this day.
>>
>> I want to see some action, in addition to good intentions and more
>> talking.
>
> You have experience dealing with ESR. Don't you trust him?

Well, the most recent experience is the VC reorganisation which was
checked into the trunk in a non-finished state and then left in the
lurch for others to sort out themselves for about half a year.

However this may have come about, it is not a workflow that really
inspires confidence in the procedures and communication needed for
people in charge of important subsystems, in particular if one calls for
faster releases.

I can't blame Eli for being sceptical about the wisdom of introducing
more single points of failure hinging on a one-person effort.

So the question remains: where do we all (not just ESR) want to go, and
will others be able to pick up the slack in case that ESR has to drop
the ball for some reason?

> I think that the reimplementation of VC that ESR just accomplished
> involves an amount of work of the same order than implementing the
> systems he proposed. I never was involved with ESR, but his offer
> seems pretty credible to me.

The VC reorganization still appears to be an ongoing effort right now.

>>> The advantages of a living database over a static text file are too
>>> large to enumerate.
>>
>> But the advantage of the flat text file is that it exists.
>
> Seriously, I don't understand your attitude. Is like somebody were
> asking you to take some risk.

Introducing a new dependency for working is a risk.  If people stop
reporting bugs in the current way, one has to change the workflow of the
potential bugfixers.  Not all of them will do this.  If stuff gets
stuck, one can't easily return to the old way of doing things since a
different report-emacs-bug would already be out in the wild.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum