Jira issues against trunk

[Martin Sebor <se...@roguewave.com>]

It occurs to me that our current process of assigning bugs
to a yet unreleased branch (e.g., 4.2) might lead to the
misleading impression after the branch is released that it
was/is full of bugs even though all the bugs may have been
resolved before the release went out the door. I think we
need a separate version entry in Jira for regressions
introduced during development. Any suggestions for what
this entry ought to be? Would labeling them simply "trunk"
work or is there any benefit in being more specific (e.g.,
<version>-dev)?

Martin

Re: Jira issues against trunk

[Martin Sebor <se...@roguewave.com>]

I agree. Unless someone has a better idea (hint -- it would
be nice to get feedback from others as well :) I'll go with
trunk.

Martin

Andrew Black wrote:
> The only benefit I can see in using <version>-dev over trunk, is it
> would provide a view into what release the bug was discovered (and
> fixed) in.  This assistance may be offset by the extra maintenance
> hassle of creating the new name every release, and having to deal with
> people inadvertently reporting bugs as being part of a previous trunk
> version.
> 
> My tendency is to cast a vote for using 'trunk'.
> 
> --Andrew Black
> 
> Martin Sebor wrote:
>> It occurs to me that our current process of assigning bugs
>> to a yet unreleased branch (e.g., 4.2) might lead to the
>> misleading impression after the branch is released that it
>> was/is full of bugs even though all the bugs may have been
>> resolved before the release went out the door. I think we
>> need a separate version entry in Jira for regressions
>> introduced during development. Any suggestions for what
>> this entry ought to be? Would labeling them simply "trunk"
>> work or is there any benefit in being more specific (e.g.,
>> <version>-dev)?
>>
>> Martin

Re: Jira issues against trunk

[Andrew Black <ab...@roguewave.com>]

The only benefit I can see in using <version>-dev over trunk, is it
would provide a view into what release the bug was discovered (and
fixed) in.  This assistance may be offset by the extra maintenance
hassle of creating the new name every release, and having to deal with
people inadvertently reporting bugs as being part of a previous trunk
version.

My tendency is to cast a vote for using 'trunk'.

--Andrew Black

Martin Sebor wrote:
> It occurs to me that our current process of assigning bugs
> to a yet unreleased branch (e.g., 4.2) might lead to the
> misleading impression after the branch is released that it
> was/is full of bugs even though all the bugs may have been
> resolved before the release went out the door. I think we
> need a separate version entry in Jira for regressions
> introduced during development. Any suggestions for what
> this entry ought to be? Would labeling them simply "trunk"
> work or is there any benefit in being more specific (e.g.,
> <version>-dev)?
> 
> Martin