Notes from Contribute Group meeting on 6/23/11

3 views
Skip to first unread message

davidwboswell

unread,
Jun 24, 2011, 4:24:56 PM6/24/11
to
Notes from Contribute Group meeting on 6/23/11

present: davidwboswell, aakashd, stas, marcia, williamr, janet,
pbiggar, sdaugherty, anthony, brian king, juanb, kairo, vineel

- Case Studies

William Reynolds gave a case study of the Army of Awesome, a light
weight way to help people with support questions. Some of the keys to
its success include sufficient promotion in key places, the tool makes
things easy (there's a way to tell who has already been responded to
and there are sample responses you can use, people can get started in
minutes) and there is a way to give recognition (both in the tool
itself with the leaderboard and one-on-one when William follows up
with people who have been responding to questions.

There was a discussion about being able to use this tool for more than
support requests and William said he was starting to look into this.
Stas said it would be interesting to see if this could help with
localization and it would be worth experimenting. This also fits into
the general theme that we've discussed before about how there is a
need for additional tools to help bring new contributors in.

- Community Numbers

David discussed the updated version of the community infographic that
can be found at

http://wiki.mozilla.org/Community

The infographic only has rough numbers in it (100s, 1000s, etc) since
that is all the information we have to work with. Across the
community there are gaps where we don't know who is involved.

For example, l10n has good visibility on who checks in strings since
those people have commit access and an LDAP account. But it is known
that there are many localizers who translate and then hand the strings
off to others to check in. For these people we have very little
information since they haven't interacted with any of our tools that
would track their activity.

This sort of gap seems to be common in all project areas. Two other
examples: Paul Biggar said he knows there are many people interested
in coding coming to him but there's a gap between expressing interest
and getting into Bugzilla. There's very little visibility there but
there are doubtless ways to help people get plugged into Bugzilla
(people need to figure out how to build Firefox and learn how to
interact with Bugzilla before we start getting the chance to interact
with them now).

For QA it's the same thing. We known there are many people
downloading the nightly because we can see those ADUs but there's no
visibility there until someone learns how to take the next step and
report bugs with the nightly. In this case a good place to reach
those people we don't know yet is the Nightly First Run page -- maybe
a screencast of creating a Bugzilla account and filing a bug so we can
then start having a dialogue with those people in our tools.

Anthony has just started to try to put more visibility into QA Test
Days by creating a Google Doc about who is coming to each event.
He'll report back after he gets a sense of how this visibility is
helping or not.

Stas had another good example of how visibility could help. For
instance, if we could see who has liked Firefox on Facebook and also
submitted a mark on the MarkUp site, contacting that person and
helping them get more involved would be a good idea. We can't really
do that right now though since all of our sites are run as separate
islands (this led to a discussion about how to get single sign-on
working).

In general though, having more visibilty could help us move people
interested in contributing up the ladder of involvement. Williamr had
a good example of how we could act on this now. We could post on
Facebook and say 'If you submitted a mark on MarkUp, you may also
like…'. We don't need to know who did what here since people would
self-select. He also said the email program is a way to do this since
it is a one-to-one relationship already.

In general, with anything we do involving visibility into
contributor's activities there were several comments about needing to
approach this in a way that respected everyone's privacy.

Jeff Hammel

unread,
Jun 27, 2011, 12:08:27 PM6/27/11
to mozil...@lists.mozilla.org
On 06/24/2011 01:24 PM, davidwboswell wrote:
> Notes from Contribute Group meeting on 6/23/11
>
> present: davidwboswell, aakashd, stas, marcia, williamr, janet,
> pbiggar, sdaugherty, anthony, brian king, juanb, kairo, vineel
>
> - Case Studies
>
> William Reynolds gave a case study of the Army of Awesome, a light
> weight way to help people with support questions. Some of the keys to
> its success include sufficient promotion in key places, the tool makes
> things easy (there's a way to tell who has already been responded to
> and there are sample responses you can use, people can get started in
> minutes) and there is a way to give recognition (both in the tool
> itself with the leaderboard and one-on-one when William follows up
> with people who have been responding to questions.
>
> There was a discussion about being able to use this tool for more than
> support requests and William said he was starting to look into this.
> Stas said it would be interesting to see if this could help with
> localization and it would be worth experimenting. This also fits into
> the general theme that we've discussed before about how there is a
> need for additional tools to help bring new contributors in.
>
Is this tool available/previewable anywhere? Is it live or???

Janet Swisher

unread,
Jun 27, 2011, 12:33:34 PM6/27/11
to Jeff Hammel, mozil...@lists.mozilla.org
On 6/27/11 11:08 AM, Jeff Hammel wrote:
> On 06/24/2011 01:24 PM, davidwboswell wrote:
>> Notes from Contribute Group meeting on 6/23/11
>>
>> William Reynolds gave a case study of the Army of Awesome, a light
>> weight way to help people with support questions.
>>
> Is this tool available/previewable anywhere? Is it live or???
http://support.mozilla.com/en-US/army-of-awesome

Axel Hecht

unread,
Jun 27, 2011, 5:03:42 PM6/27/11
to
Am 24.06.11 22:24, schrieb davidwboswell:

> Notes from Contribute Group meeting on 6/23/11
>
> present: davidwboswell, aakashd, stas, marcia, williamr, janet,
> pbiggar, sdaugherty, anthony, brian king, juanb, kairo, vineel
>
> - Case Studies
>
> William Reynolds gave a case study of the Army of Awesome, a light
> weight way to help people with support questions. Some of the keys to
> its success include sufficient promotion in key places, the tool makes
> things easy (there's a way to tell who has already been responded to
> and there are sample responses you can use, people can get started in
> minutes) and there is a way to give recognition (both in the tool
> itself with the leaderboard and one-on-one when William follows up
> with people who have been responding to questions.
>
> There was a discussion about being able to use this tool for more than
> support requests and William said he was starting to look into this.
> Stas said it would be interesting to see if this could help with
> localization and it would be worth experimenting. This also fits into
> the general theme that we've discussed before about how there is a
> need for additional tools to help bring new contributors in.

I disagree with stas that l10n is a good field for experiments here.

The big players in the field of crowd sourcing like facebook don't have
this figured out yet at all. For example, the bosnian community needs to
do "string vote sprints" every other day, to vote out some apparently
indian team that's hijacking their work every now and then. And that's
with fb having bots and filters fighting malicious entries already.

Also, translation isn't an exact science, there's not always a right or
wrong to choose from.

I'm not opposed in principle to have something awesome for l10n. But
anything short of a game changer on how people deal with translation
submissions isn't going to help the community. "game changer" meaning a
better contribution model than github or facebook.

l10n would probably be in the endgame of a crowd contribution effort,
and not the initial playfield.

Axel

Staś Małolepszy

unread,
Jun 27, 2011, 6:23:30 PM6/27/11
to Axel Hecht, mozillians
Axel Hecht, Mon Jun 27 14:03:42 -0700 2011:

> > There was a discussion about being able to use this tool for more
> > than
> > support requests and William said he was starting to look into this.
> > Stas said it would be interesting to see if this could help with
> > localization and it would be worth experimenting. This also fits into
> > the general theme that we've discussed before about how there is a
> > need for additional tools to help bring new contributors in.
>
> I disagree with stas that l10n is a good field for experiments here.

I agree with you disagreeing with that, because it's not what I said :)

What I meant was that it would be interesting to see how we can reuse
the Army of Awesome's tools in the area of l10n. I didn't mean to make
l10n a ginuea pig for experiments.

I don't think we currently use Twitter in any way to support our l10n
efforts. Examples would be: reaching out to new contributors,
chaperoning new localizers or closing the feedback loop between the
localizers and end users reporting translation errors/suggestions.

I don't have any concrete ideas in mind right now, but I'm simply
wondering what could we do to make the most of the communication
channel that is Twitter.

> The big players in the field of crowd sourcing like facebook don't
> have this figured out yet at all.

I'm not convinced this is true, btw. Facebook seems to be doing rather
well with crowd-based reviews of suggestions. I'd like to see if this
could work for us, too.

-stas
--
Staś Małolepszy
+1 650 276 8268

David Boswell

unread,
Jun 28, 2011, 1:51:07 PM6/28/11
to Staś Małolepszy, mozillians, Axel Hecht
> I agree with you disagreeing with that, because it's not what I said
> :)

Apologies if my notes weren't clear.



> I don't think we currently use Twitter in any way to support our l10n
> efforts. Examples would be: reaching out to new contributors,
> chaperoning new localizers or closing the feedback loop between the
> localizers and end users reporting translation errors/suggestions.

Just to follow up with with Stas said, how can we use more channels, like Twitter, to find people who might be interested in getting involved with Mozilla projects?

So for support this is pretty easy -- you can see when someone says 'Why isn't my Firefox working?'. With l10n it might or might not be realistic to see someone saying 'Why isn't my Firefox available in Malay?' -- but if we could see those sorts of comments we could follow up with people interested in joining the l10n community.

David

Reply all
Reply to author
Forward
0 new messages