harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Geir Magnusson Jr <g...@pobox.com>
Subject Re: jira messages redirected to commits mailing list
Date Tue, 07 Mar 2006 19:10:47 GMT

Tim Ellison wrote:
> Geir Magnusson Jr wrote:
>> Tim Ellison wrote:
>>> Is there some way to teach JIRA not to send so much mail?
>> Stop using it as a chat room. :)
> So what is the right way to use JIRA?
>  - people open an issue,
>  - maybe comment with a test case
>  - maybe attach a patch or two

So far, I think these are important.

>  - I may comment on the issue, with comments that are relevant to that
> specific issue

Also good to know, since it's equivalent to technical discussion on the 
dev list.

>  - when I work on it I assign it to me, and say progress started

That's good to know too, since people can get an event driven picture of 
what's going on.  However, this may be something we could reduce distro 

>  - when I'm done I resolve it
>  - when the reporter has verified it they comment to say so
>  - I close it as verified

ALl of these are important too :/

> What steps should I stop doing?

It was a joke, referring to my continual plea to get some of the 
conversations out and into -dev@

>>> Every state change produces mail to the world - even though it is likely
>>> only of interest to the reporter, assignee, and watchers.  i.e. any way
>>> to solve the problem rather than move it ;-)
>> Every change should be visible to everyone for maximum transparency, or
>> so I believe.  It would be a pain in the rear if one had to explicitly
>> sign up for each jira one was interested in.
> Some people say every JIRA state change / comment is too much 'spam' --
> you want to see them all ...

Yes, because mail is easy to delete, filter, ignore.

More importantly,  I think a model that requires the community to 
individually add themselves to each JIRA as a watcher is one prone to 
failure of oversight.  I know that I'd forget, and I believe that full 
flows like this sometimes catch the attention of a new person to 
participate.  There also is the issue of archiving that mail stream... 
It could be that isn't as important because JIRA has the info, but OTOH 
someone might want to prove something was done with full exposure to the 

I guess my answer right now is that given how we are currently using 
JIRA, I can't think of anything to cut out...

Now that it doesn't go to -dev@, it's easier for those that want to 
participate less?

Maybe we have a separate list for the jira flow that we ask every 
committer to sub to, but then people can just watch -dev@ and/or 
-commit@ and not have to deal with it?

>> That said, once the VM activity gets really honking, we'll probably need
>> a second stream for those...
> Not sure why the VM is special here.

Not special, but different. It will be a separate group of people 
working on different things, so we may want to start segmenting the mail 
streams.  People may really not care about VM stuff if the work on 
classlib stuff, or classlib stuff if they are focused on the JIT or 

We'll have to see.


> Regards,
> Tim
>>> Leo Simons wrote:
>>>> Taking care of this now...
>>>> I will note that this makes it even more important for committers and
>>>> active contributors to subscribe to the commits mailing list - a lot of
>>>> important information is in those jira messages.
>>>> I will also note that it *also* makes it even more important that Jira
>>>> is not used for discussion - that really needs to happen here on the
>>>> mailing list where  everyone can track it. The ASF has had some bad
>>>> experience in the past with too much communication going via the issue
>>>> tracker; this isn't so much a guideline as it is a pretty hard
>>>> requirement.
>>>> - Leo
>>>> On Tue, Mar 07, 2006 at 08:17:45AM -0600, Archie Cobbs wrote:
>>>>> Mark Hindess wrote:
>>>>>> Geir,  There are quite a lot of JIRA messages these days, perhaps
>>>>>> is time to split the JIRA traffic to a separate list with a reply-to
>>>>>> set to harmony-dev.  Or perhaps just have them sent to the commit
>>>>>> list?
>>>>> Yes, please... +1e6
>>>>> -Archie

View raw message