You can also just describe it to the mail list first to see if people
argue about it being a feature/bug etc.

There's a certain level of confidence/embaressment with going public and
being wrong, but provided it's not something someone does a lot I think
it's a fine thing to do.

The perfect situation in my view is best seen on IRC, but can work on mail
lists too:

Jorg pipes up and says he's seen what he thinks is a bug. Stephen listens,
asks a question and Jorg provides an answer. Stephen agrees it's a bug and
either asks Jorg to report it in the tracker, or Jorg volunteers.

It's possible that Stephen may just say he'll fix it there and then, or
take a private patch, though this leads to a poorer audit trail. Still,
bug-fixing is the primary game here, not providing audit trails.

Hen

On Tue, 20 Apr 2004, Stephen Colebourne wrote:

> Yes, write the junit test to prove the problem, then solve it if you can
> (take advice from the mailing list if not), finally post patches to bugzilla
> (use cvs diff -u format).
>
> Stephen
>
> ----- Original Message -----
> From: "J�rg Schaible" <[EMAIL PROTECTED]>
> > Hi,
> >
> > I suspect I met a bug in org.apache.commons.io.FileUtils:
> > what should I do?
> (3) > *- AND should I go to the bugs tracking apps in apache.org?
> (1) > *- AND should I write a junit class showing the point?
> (2) > *- AND else write a patch?
> > *- or instead keep it for myself?
> >
> > Please tell me ASAP.
>
> In the order above.
>
> -- J�rg
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to