You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@oodt.apache.org by Justin Erenkrantz <ju...@erenkrantz.com> on 2010/12/10 13:49:23 UTC

Discussions in JIRA vs list

I notice that there is a bit of back/forth going on in the JIRA
comments.  In my opinion, there is no easy way to read this via email
as JIRA pooches the formatting of the comments (I can't easily tell
what is being quoted and what isn't).

My hunch is that it may be best to move the discussions from JIRA
other than the most mundane back to this list.  After the discussion
is complete and a resolution is agreed upon, a link to the mail
archives in the JIRA issue should be sufficient to "close the loop".
(This is the strategy we use in Subversion.)

More fundamentally, I have concerns about -1's being cast in a JIRA
comment...they should be casted on list to ensure that everyone can
chime in properly and fully understand what's going on.

Cheers.  -- justin

Re: Discussions in JIRA vs list

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
> I apologize for getting meta and sticking on this point, but I'd like
> to ensure we don't place artificial barriers to contributions by
> having lengthy conversations and votes inside of an issue tracker.

I'd like to ensure the same: that we don't regulate people who like to use JIRA to have such discussions and force them to instead duplicate them information they'd normally put in there in an email as (which JIRA already does).

> 
> I find the OS/IDE analogy rather weak - the fact that I use Mac OS X,
> GMail, or refuse to use Eclipse doesn't impact you in what OS or IDE
> you use.  I have options to contribute constructively to any
> discussions even respecting my choices (and you do yours).  However,
> the fact that you use JIRA for detailed design discussions and
> conversations means that no one else can follow detailed conversations
> via email efficiently - and are forced to use JIRA.

Umm, I'm sorry I find that analogy rather weak. JIRA isn't the Death Star. It's not something that prevents anyone from doing anything. It's simply a user interface around discussions, the same way that an email client is the same. All Apache committers can have JIRA just like they have Apache (or non-Apache) email. All JIRA interactions are sent to the list. There is no difference between those interactions then email -- in fact, on the contrary they are * in fact * email. Arguably email with annotations (just like I just used * to emphasize something).

> 
> The fundamental issue is that you can't follow or contribute to a
> lengthy conversation in JIRA without using JIRA.  If you actually
> tried to follow along via email, first off, you can't reply to issues
> via email.  Furthermore, JIRA's emails don't have proper references or
> threading, so you just get a bazillion individual non-related posts.


Looks good in mod_mbox to me?

http://mail-archives.apache.org/mod_mbox/oodt-commits/201012.mbox/browser

> In Gmail, since it has no other clues, it does try to group all
> comments from one author together - so it means that I see all of
> Brian's comments on an issue in one "conversation" and all of your
> comments in a completely separate "conversation".  It's very very
> painful to follow as I see Brian's responses to you first without
> seeing your comments first.  My head hurts trying to follow along -
> ugh.

Uh-huh: the real problem here is Gmail!

> 
> Email is a great common denominator that lets people use the tools and
> workflow they desire.  JIRA doesn't.  It raises the bar significantly
> for contributing to discussions.  Again, JIRA is acceptable for issue
> tracking, but please let's not make it the place where conversations
> and votes are held.  -- justin


I agree with you mostly but there are a lot of things that we key off JIRA (e.g., our CHANGES.txt file for one) where I really appreciate as a user of the software having nice detailed discussion when I got to follow the changes in a particular release. That's just me though and I maintain that you and I can go back and forth but what it really comes down to is whether you prefer vi or whether I prefer Eclipse. :)

Cheers,
Chris


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


Re: Discussions in JIRA vs list

Posted by David Woollard <wo...@gmail.com>.
I'm just glad I did not have to step in and restrain you two ;)...

-Dave W

On Dec 11, 2010, at 3:19 PM, Mattmann, Chris A (388J) wrote:

> OK Justin and I just had a love fest on Skype. :) Or humorous banter. Or something else.
> 
> Anyways, I'll try and keep my -1s to the dev list (watch out Kale! ^_^) and Justin and others will be understanding when some of us JIRA fanboys have tons of discussion in JIRA. Deal. Deal!!!
> 
> Cheers,
> Chris
> 
> 
> On Dec 11, 2010, at 2:56 PM, Justin Erenkrantz wrote:
> 
>> On Sat, Dec 11, 2010 at 9:35 PM, Mattmann, Chris A (388J)
>> <ch...@jpl.nasa.gov> wrote:
>>> I likened the debate to being which OS is best, which IDE is best, etc., because the truth of the matter is: none of them are, and there's more than one way to skin a cat.
>> 
>> I apologize for getting meta and sticking on this point, but I'd like
>> to ensure we don't place artificial barriers to contributions by
>> having lengthy conversations and votes inside of an issue tracker.
>> 
>> I find the OS/IDE analogy rather weak - the fact that I use Mac OS X,
>> GMail, or refuse to use Eclipse doesn't impact you in what OS or IDE
>> you use.  I have options to contribute constructively to any
>> discussions even respecting my choices (and you do yours).  However,
>> the fact that you use JIRA for detailed design discussions and
>> conversations means that no one else can follow detailed conversations
>> via email efficiently - and are forced to use JIRA.
>> 
>> The fundamental issue is that you can't follow or contribute to a
>> lengthy conversation in JIRA without using JIRA.  If you actually
>> tried to follow along via email, first off, you can't reply to issues
>> via email.  Furthermore, JIRA's emails don't have proper references or
>> threading, so you just get a bazillion individual non-related posts.
>> In Gmail, since it has no other clues, it does try to group all
>> comments from one author together - so it means that I see all of
>> Brian's comments on an issue in one "conversation" and all of your
>> comments in a completely separate "conversation".  It's very very
>> painful to follow as I see Brian's responses to you first without
>> seeing your comments first.  My head hurts trying to follow along -
>> ugh.
>> 
>> Email is a great common denominator that lets people use the tools and
>> workflow they desire.  JIRA doesn't.  It raises the bar significantly
>> for contributing to discussions.  Again, JIRA is acceptable for issue
>> tracking, but please let's not make it the place where conversations
>> and votes are held.  -- justin
> 
> 
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Chris Mattmann, Ph.D.
> Senior Computer Scientist
> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
> Office: 171-266B, Mailstop: 171-246
> Email: chris.a.mattmann@nasa.gov
> WWW:   http://sunset.usc.edu/~mattmann/
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Adjunct Assistant Professor, Computer Science Department
> University of Southern California, Los Angeles, CA 90089 USA
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 


Re: Discussions in JIRA vs list

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
OK Justin and I just had a love fest on Skype. :) Or humorous banter. Or something else.

Anyways, I'll try and keep my -1s to the dev list (watch out Kale! ^_^) and Justin and others will be understanding when some of us JIRA fanboys have tons of discussion in JIRA. Deal. Deal!!!

Cheers,
Chris


On Dec 11, 2010, at 2:56 PM, Justin Erenkrantz wrote:

> On Sat, Dec 11, 2010 at 9:35 PM, Mattmann, Chris A (388J)
> <ch...@jpl.nasa.gov> wrote:
>> I likened the debate to being which OS is best, which IDE is best, etc., because the truth of the matter is: none of them are, and there's more than one way to skin a cat.
> 
> I apologize for getting meta and sticking on this point, but I'd like
> to ensure we don't place artificial barriers to contributions by
> having lengthy conversations and votes inside of an issue tracker.
> 
> I find the OS/IDE analogy rather weak - the fact that I use Mac OS X,
> GMail, or refuse to use Eclipse doesn't impact you in what OS or IDE
> you use.  I have options to contribute constructively to any
> discussions even respecting my choices (and you do yours).  However,
> the fact that you use JIRA for detailed design discussions and
> conversations means that no one else can follow detailed conversations
> via email efficiently - and are forced to use JIRA.
> 
> The fundamental issue is that you can't follow or contribute to a
> lengthy conversation in JIRA without using JIRA.  If you actually
> tried to follow along via email, first off, you can't reply to issues
> via email.  Furthermore, JIRA's emails don't have proper references or
> threading, so you just get a bazillion individual non-related posts.
> In Gmail, since it has no other clues, it does try to group all
> comments from one author together - so it means that I see all of
> Brian's comments on an issue in one "conversation" and all of your
> comments in a completely separate "conversation".  It's very very
> painful to follow as I see Brian's responses to you first without
> seeing your comments first.  My head hurts trying to follow along -
> ugh.
> 
> Email is a great common denominator that lets people use the tools and
> workflow they desire.  JIRA doesn't.  It raises the bar significantly
> for contributing to discussions.  Again, JIRA is acceptable for issue
> tracking, but please let's not make it the place where conversations
> and votes are held.  -- justin


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


Re: Discussions in JIRA vs list

Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On Sat, Dec 11, 2010 at 9:35 PM, Mattmann, Chris A (388J)
<ch...@jpl.nasa.gov> wrote:
> I likened the debate to being which OS is best, which IDE is best, etc., because the truth of the matter is: none of them are, and there's more than one way to skin a cat.

I apologize for getting meta and sticking on this point, but I'd like
to ensure we don't place artificial barriers to contributions by
having lengthy conversations and votes inside of an issue tracker.

I find the OS/IDE analogy rather weak - the fact that I use Mac OS X,
GMail, or refuse to use Eclipse doesn't impact you in what OS or IDE
you use.  I have options to contribute constructively to any
discussions even respecting my choices (and you do yours).  However,
the fact that you use JIRA for detailed design discussions and
conversations means that no one else can follow detailed conversations
via email efficiently - and are forced to use JIRA.

The fundamental issue is that you can't follow or contribute to a
lengthy conversation in JIRA without using JIRA.  If you actually
tried to follow along via email, first off, you can't reply to issues
via email.  Furthermore, JIRA's emails don't have proper references or
threading, so you just get a bazillion individual non-related posts.
In Gmail, since it has no other clues, it does try to group all
comments from one author together - so it means that I see all of
Brian's comments on an issue in one "conversation" and all of your
comments in a completely separate "conversation".  It's very very
painful to follow as I see Brian's responses to you first without
seeing your comments first.  My head hurts trying to follow along -
ugh.

Email is a great common denominator that lets people use the tools and
workflow they desire.  JIRA doesn't.  It raises the bar significantly
for contributing to discussions.  Again, JIRA is acceptable for issue
tracking, but please let's not make it the place where conversations
and votes are held.  -- justin

Re: Discussions in JIRA vs list

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
On Dec 11, 2010, at 1:18 PM, David Kale wrote:

> Sorry, Chris, but overall, I'd say I come down on the side of "have involved
> conversations via email and not on JIRA."  For all the wonderful interfaces,
> apps, etc., out there, I think that email is still the most accessible
> medium across devices, OSes, etc.

Hmmm. I'm not sure that the debate is which is the most accessible. 

I'm more on the side of not trying to legislate where people have conversations, so long as it "gets back to the list." JIRA fits this criteria in spades.

Whether those conversations have the words "bq." before blockquote or "{code}" before a code example (the prior two examples being the case of JIRA) or whether someone sees ">>" to indicate blockquote or "----code---" or some other annotation (e.g., the prior two examples being the case of some mail client) is really irrelevant IMHO.

I likened the debate to being which OS is best, which IDE is best, etc., because the truth of the matter is: none of them are, and there's more than one way to skin a cat.

Cheers,
Chris


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


Re: Discussions in JIRA vs list

Posted by David Kale <da...@cs.stanford.edu>.
Sorry, Chris, but overall, I'd say I come down on the side of "have involved
conversations via email and not on JIRA."  For all the wonderful interfaces,
apps, etc., out there, I think that email is still the most accessible
medium across devices, OSes, etc.


On Fri, Dec 10, 2010 at 3:34 PM, Mattmann, Chris A (388J) <
chris.a.mattmann@jpl.nasa.gov> wrote:

> Hugs back to you bud. I'll point you to the proliferation of operating
> systems, IDEs and other devices. Not everyone can be perfect and use a Mac,
> Eclipse and Maven :)
>
> But I put up with them nonetheless :)
>
> Cheers,
> Chris
>
> Sent from my iPhone
>
> On Dec 10, 2010, at 2:42 PM, "Justin Erenkrantz" <ju...@erenkrantz.com>
> wrote:
>
> > On Fri, Dec 10, 2010 at 2:33 PM, Mattmann, Chris A (388J)
> > <ch...@jpl.nasa.gov> wrote:
> >> Yeah JIRA isn't exactly intended for consuming over email. I usually
> just click the links when the email comes so that I can be taken to the real
> place that the magic happens: JIRA itself.
> >
> > That's the crux of my concern - an issue tracker shouldn't be the
> > place that "magic" happens.
> >
> > I shouldn't be forced to follow JIRA to know that you are vetoing
> something.
> >
> >> Eh, I'm not sure on this one. I think it depends on how you consume the
> information from JIRA or a mailing list. A mailing list discussion may be
> great if you're on the go, or reading it on your blackberry: easy to reply
> to, quickly able to read, that type of thing. However, what if you want to
> search it? Or search across a number of them? Google is great for that too,
> (as are the mail archives) but there's some extra clicks there that you have
> to do to find what's going on. It's also not real time in terms of being
> able to search latest issues. Some mail-archive systems are real time, but
> you have to leave the context of your reader to go use them (open a browser,
> click, click, that type of thing).
> >> ...snip, snip...
> >
> > Perhaps, you might not have understood what I was suggesting.  So,
> > I'll point you at Karl Fogel's section from Producing OSS entitled "No
> > Conversations in the Bug Tracker":
> >
> > http://producingoss.com/en/bug-tracker-usage.html
> >
> > As usual, Karl states it far more eloquently than I can.  And, *of
> > course*, Karl has an entire section related to exactly this.
> >
> > *hugs*  -- justin
>

Re: Discussions in JIRA vs list

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
Hugs back to you bud. I'll point you to the proliferation of operating systems, IDEs and other devices. Not everyone can be perfect and use a Mac, Eclipse and Maven :)

But I put up with them nonetheless :)

Cheers,
Chris 

Sent from my iPhone

On Dec 10, 2010, at 2:42 PM, "Justin Erenkrantz" <ju...@erenkrantz.com> wrote:

> On Fri, Dec 10, 2010 at 2:33 PM, Mattmann, Chris A (388J)
> <ch...@jpl.nasa.gov> wrote:
>> Yeah JIRA isn't exactly intended for consuming over email. I usually just click the links when the email comes so that I can be taken to the real place that the magic happens: JIRA itself.
> 
> That's the crux of my concern - an issue tracker shouldn't be the
> place that "magic" happens.
> 
> I shouldn't be forced to follow JIRA to know that you are vetoing something.
> 
>> Eh, I'm not sure on this one. I think it depends on how you consume the information from JIRA or a mailing list. A mailing list discussion may be great if you're on the go, or reading it on your blackberry: easy to reply to, quickly able to read, that type of thing. However, what if you want to search it? Or search across a number of them? Google is great for that too, (as are the mail archives) but there's some extra clicks there that you have to do to find what's going on. It's also not real time in terms of being able to search latest issues. Some mail-archive systems are real time, but you have to leave the context of your reader to go use them (open a browser, click, click, that type of thing).
>> ...snip, snip...
> 
> Perhaps, you might not have understood what I was suggesting.  So,
> I'll point you at Karl Fogel's section from Producing OSS entitled "No
> Conversations in the Bug Tracker":
> 
> http://producingoss.com/en/bug-tracker-usage.html
> 
> As usual, Karl states it far more eloquently than I can.  And, *of
> course*, Karl has an entire section related to exactly this.
> 
> *hugs*  -- justin

Re: Discussions in JIRA vs list

Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On Fri, Dec 10, 2010 at 2:33 PM, Mattmann, Chris A (388J)
<ch...@jpl.nasa.gov> wrote:
> Yeah JIRA isn't exactly intended for consuming over email. I usually just click the links when the email comes so that I can be taken to the real place that the magic happens: JIRA itself.

That's the crux of my concern - an issue tracker shouldn't be the
place that "magic" happens.

I shouldn't be forced to follow JIRA to know that you are vetoing something.

> Eh, I'm not sure on this one. I think it depends on how you consume the information from JIRA or a mailing list. A mailing list discussion may be great if you're on the go, or reading it on your blackberry: easy to reply to, quickly able to read, that type of thing. However, what if you want to search it? Or search across a number of them? Google is great for that too, (as are the mail archives) but there's some extra clicks there that you have to do to find what's going on. It's also not real time in terms of being able to search latest issues. Some mail-archive systems are real time, but you have to leave the context of your reader to go use them (open a browser, click, click, that type of thing).
> ...snip, snip...

Perhaps, you might not have understood what I was suggesting.  So,
I'll point you at Karl Fogel's section from Producing OSS entitled "No
Conversations in the Bug Tracker":

http://producingoss.com/en/bug-tracker-usage.html

As usual, Karl states it far more eloquently than I can.  And, *of
course*, Karl has an entire section related to exactly this.

*hugs*  -- justin

Re: Discussions in JIRA vs list

Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
Hey Justin,

First off, can you report this discussion as an issue in JIRA first?

(I kid, I kid)

> I notice that there is a bit of back/forth going on in the JIRA
> comments.  In my opinion, there is no easy way to read this via email
> as JIRA pooches the formatting of the comments (I can't easily tell
> what is being quoted and what isn't).

Yeah JIRA isn't exactly intended for consuming over email. I usually just click the links when the email comes so that I can be taken to the real place that the magic happens: JIRA itself.

> 
> My hunch is that it may be best to move the discussions from JIRA
> other than the most mundane back to this list.  After the discussion
> is complete and a resolution is agreed upon, a link to the mail
> archives in the JIRA issue should be sufficient to "close the loop".
> (This is the strategy we use in Subversion.)

Eh, I'm not sure on this one. I think it depends on how you consume the information from JIRA or a mailing list. A mailing list discussion may be great if you're on the go, or reading it on your blackberry: easy to reply to, quickly able to read, that type of thing. However, what if you want to search it? Or search across a number of them? Google is great for that too, (as are the mail archives) but there's some extra clicks there that you have to do to find what's going on. It's also not real time in terms of being able to search latest issues. Some mail-archive systems are real time, but you have to leave the context of your reader to go use them (open a browser, click, click, that type of thing).

OTOH, say you're sitting at your desk, and you want to see what's going on with JIRA development for a project/etc. You've got several nice views (the components classification pane, the upcoming fix versions pane), you've got a search interface that's real time updated with the latest issues, etc, etc. Also on each issue you've got the ability to have these nicely formatted (admittedly weird-looking when represented in emails) issue descriptions and comments discussions, just like you'd get if you were looking at a mail archive view of a prior message. So, it definitely has its advantages too. 

> More fundamentally, I have concerns about -1's being cast in a JIRA
> comment...they should be casted on list to ensure that everyone can
> chime in properly and fully understand what's going on.

Heh, I was the one that was throwing around my opposition to a design that was being proposed, so happy to chime in on this one. I don't think there's a hard and fast rule as to where design displeasure should be cast. Arguably my displeasure was being barfed all over the list by the JIRA issue emails you likely received ^_^ At the same time, my -1 comment on the particular issue (OODT-72 [1]) also came across on the list (see here [2]) too. 

One thing I did just notice though is that JIRA issue notifications go to commits@oodt.a.o instead of dev@oodt.a.o. Anyone opposed to me changing it to dev@oodt.a.o? Then I think folks on dev (that care less about SVN commits) would also be privy to the JIRA email discussions and comments too. Thoughts?

Cheers,
Chris

[1] https://issues.apache.org/jira/browse/OODT-72
[2] http://s.apache.org/Hey

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++