I don't think anything which increases testing and development is
necessarily good. It might do so and hurt the primary goals of the
project.
I also think projects with too many features and too many bugs are not
very useful. Wether that happens at 10 minor bugs per feature or at
1000, I don't kno
Well, this shouldn't be taken too literally. I'm not even sure
something like perfect code exists...
My point is more that a patch should be made good enough, in the
maintainer opinion, before landing. Delaying necessary changes after
the patch has been landed is not a good maintenance strategy.
Without reference to the actual patch, adding one major feature and
adding 10 minor bugs is still useful, because it increases testing and
therefore development.
On Wed, Mar 27, 2013 at 07:01:21PM -0300, Gonzalo Odiard wrote:
> It's true, but this comments are in the context of new features,
> and
It's true, but this comments are in the context of new features,
and specifically big features like the proposed Journal changes.
In this context, if we add 1 feature and 10 bugs, is not a good deal.
Gonzalo
On Wed, Mar 27, 2013 at 6:53 PM, James Cameron wrote:
> I disagree. A 90% working pat
I disagree. A 90% working patch should be reviewed or even accepted,
if it improves the situation more than it degrades the situation.
Don't let the good be the enemy of the perfect.
In particular, if the patch fixes a high priority ticket but opens
three low priority tickets, the project has st
Hi Ajay,
>
> I don't know the specific history of this patch but, as a rule,
> patches should be pushed to mainline when they are fully working, or
> at least they are thought to be. If the maintainers point out issues,
> they should be solved before committing. Often the remainning 10% is
> the o
On 27 March 2013 19:01, Ajay Garg wrote:
> Yes, I understand that :)
>
> "My" 10% intended any minor UI-tweaks (string change, text-positioning,
> etc), that are often not workflow-critical, and very easy to fix.
> Obviously, anything that breaks any use-case is a reason enough to reject
> the pat
On Wed, Mar 27, 2013 at 11:21 PM, Daniel Narvaez wrote:
> On 27 March 2013 17:39, Ajay Garg wrote:
> > Just one request ::
> > =
> >
> > please include the patch ASAP if it works 90% of the time (although it is
> > working 100% for me at this point :P ).
> >
> > If it is thought that
On 27 March 2013 17:39, Ajay Garg wrote:
> Just one request ::
> =
>
> please include the patch ASAP if it works 90% of the time (although it is
> working 100% for me at this point :P ).
>
> If it is thought that some minor-change/bug-fix is needed, it can be done
> "in" the mainline.
9 matches
Mail list logo
