Hi!
Thanks for sharing.
Here are a few comments and questions :)
tl;dr: I'm highly skeptical about the goals as they are currently defined.
2013/12/19 Stormy Peters <
spe...@mozilla.com>
> 2014 Goals
>
> Back in August a group of us met in Toronto to discuss the Developer
> Program, what it meant and how we’d measure its success. We agreed
> that the goal
> of the Mozilla Developer Program is to enable, inspire and collaborate to
> make the Web the primary platform used to create connected experiences
> across devices.
>
I should miss something but what exactly are the goals? I mean,
practically? "enable, inspire and collaborate" are not goals, they are
inspirational purpose.
I would be interesting on how each dev engage sub team are turning those
top level ideas into concrete goals? As a volunteer it would be easier for
me to have such concrete goals to help Mozilla reach them.
>From my experience a good practical goal is define as such:
- It's written short
- It gives a point to reach
- It provide a way to measure its success
Say another way: it must be realistic, actionable and measurable. Otherwise
it's just speech in the wind.
> We also decided on 5 metrics for 2014. Since then Patrick Finch figured out
> some ways we could measure those goals and we’ve iterated on the goal for
> 2014 taking into account organic growth and our impact.
>
I would be curious to know how you are measuring those goals. To be
accurate the measurement need to be very accurate in order to avoid false
impression.
> 24.5 % of mobile web developers using open web technologies. (17%, July
> 2013.)
>
How did you get such numbers? By itself it can mean anything. The way you
get such metrics is a very strong indicator of what it exactly means.
> 30% of web developers are using Firefox tools. (15%, July 2013.)
>
I'm skeptical about the interested of such metrics (and again, would be
interested in knowing how such value is gathered). It's a quantitative
information that mean nothing about the adoption of the web as the
platform. There is nothing wrong by using other developer tools, and
FIrefox (even without other browsers' tools) has a long tradition of third
party tools.
> 24.5% of apps developed in HTML5 across all platforms. (17%, July 2013.)
>
Again, I'm not sure about what it means. Do you mean that a quarter of HTML
Web Apps are running on all browser or does it mean 75% of the Web Apps in
the wild are made for one browser only?
And what does "platforms" mean? Browser? OS? Device? All of those three?
And in that case I'm impressed if you truly are able to check "all"
platforms.
So a metric that is so much debatable is not necessarily a good one. Maybe
it needs to be polished.
> 13,000 apps in Firefox Marketplace. (1,448, July 2013.)
>
Nice metric but it mean nothing about having the web being the platform. It
just mean that people are pushing their web app to the Mozilla Marketplace.
I agree that it's a strong metric for Mozilla but beware, it can be tree
hiding the forest.
> 1,000 developers/month contributing to Mozilla Developer
> Program.(473/month, July 2013.)
>
What means "contributing" in that area? Pushing an app to the marketplace?
Editing MDN? Something else?
So as you can see, once pushed in the wild, those metrics needs some
serious explanation about how they are gathered and what they mean exactly.
Otherwise it's a troll prone topic.
Q1 Goals
>
> Keeping in mind the 2014 goals, we brainstormed as a group what we could
> measure and target for Q1.
>
> For each of the 2014 goals we came up with a quarterly goal for Q1:
>
> - 18% of mobile web developers using open web technologies.
> - xx% of web developers are using Firefox tools.
> - 18% of apps developed in HTML5 across all platforms.
> - 4,200 apps in Firefox Marketplace.
> - 539 developers/month contributing to Mozilla Developer Program.
>
I'm curious about how you get to those values? As I'm especially involved
with the MDN team and I cannot figure how it can be actionable for that
team.
> - Need to add Stack Overflow numbers to this
>
This is something that really puzzled me a lot. Yes Stack Overflow is an
awesome place for Web dev. But it's only for English speaking people. How
to engage more people that do not speak English? Mozilla has an incredible
community beyond the English speakers. There is also so many amazing web
developers who do not speak English at all. How do we plan to reach all
those people? English should not be the alpha and omega of our work. We
need some serious plan to move forward English. That way we will be able to
reach our goals faster and more efficiently in the long run.
> We also brainstormed other metrics that might show if we are making
> progress or not.
See my first point, but I strongly think we should first think in terms of
action and only think about metrics to measure the success of those action.
Absolute metrics are hard to get, difficult to interpret and subject to
outside side effects.
> Here’s some of the ones we came up with. We expect that
> every group will take the bigger goals and figure out how their team can
> impact them. For example, the documentation team might measure how many
> visitors to the MDN Apps page end up submitting an app on Marketplace to
> help understand how their work impacts the apps goal.
>
Okay, let's take that example. First let's assume it's possible to get such
information (this would be in itself an interesting topic but it's not the
place). This typically a bad goals in the long run. If the MDN team is
expected to improve that transformation chain, they will focus on having
the MDN Apps page promoting the Firefox Marketplace on many point from the
obvious to the more subtle. Such a hard quantitative goal will force them
to "forget" about the quality of the documentation in favor of some kind of
marketing message. In the end we will lose developers that will start to
think about MDN as the place to get information about Mozilla's Web Apps
instead of having them remembering of MDN as the best place to know
everything about the Web Platform. In the long run the will turn their back
on us... but the short terms goals would have been reached.
A better goal would be to measure how many web apps are updated to be
available cross platform after a visitor has read the MDN Apps page (but I
seriously doubt it's possible to measure that in the wild but for the app
on the marketplace, it could be possible).
> - 21,000 newsletter subscribers (current number: 19,899)
>
I'm so sorry to be that skeptical... but measuring newsletter subscribers!
in 2014? Seriously? With so many web dev using social network, I'm not sure
it's a relevant metric. I think we should better focus on a social networks
strategy.
> - x # of Firefox OS tagged questions or answers on StackOverflow
> - # of MDN active contributors
>
We need to define what "active" means.
> - # of code base committers
>
Which code base?
> - # of apps from PhoneGap
> - # of apps submitted from MDN referrals to Marketplace
> - # of page hits for HTML5 pages on MDN
>
We should globally improved the visibility of MDN and focus a lot more on
localization for that.
> - # new simulator add-on downloads
> - # of page hits on tools on MDN (doubled with the redesign)
> - # of MDN referrals to simulator
>
What's the point of this?
> - measure the funnel from MDN to submit an app
>
If you wish but it a very short term goal that mean nothing in the regard
of having the web becoming the preferred platform for developing Apps.
> Working Remotely
>
> We have a very geodistributed team that is functioning extremely well. We
> regularly get other people who approach us for best practices. Here are
> some of our current best practices:
>
> - IRC - we are all on irc whenever we are working. This is hugely
> beneficial to our team. An even bigger bonus would be if we could get
> all
> of Mozilla to adopt this.
> - Asynchronous. While we have our fair share of meetings, we work hard
> to
> make sure that communication can happen asynchronously to accommodate as
> many schedules and working styles as possible. We do a lot of our
> communication in mailing lists, wiki pages and bugs. Decisions made in
> meetings are shared out through email, wiki or bugs as well.
>
It's true but it seriously need to be improved. Mailing list are underused
across the dev engage teams.
> - Bugs. We track all of our work in bugzilla. We use needinfo to flag
> individuals. We document decisions made in meetings with a comment in
> the
> bug. We encourage people to file bugs for all issues and make sure we
> capture key information there.
> - Meetings. We use meetings wisely. We have as few of them as possible.
> We
> have many IRC meetings - instead of vidyo - to enable more people to
> participate and to facilitate information sharing. We document decisions
> that are made. We have much of the discussion in email before hand -
> this
> enables people to participate even if they can’t make the meeting time.
>
We should avoid "closed doors" meeting. There were way to much of them in
2013. Volunteers could have the false impression they are not welcome as
they have to dig deep to have the necessary information about schedule, and
decision process.
> - Encourage the “do-ers” to make decisions and work across teams, as
> opposed to working through managers.
>
>
> Things we could do better:
>
> - Instead of saying no to proposals, we should encourage others to
> develop
> their idea and help them figure out how to implement it.
>
It would also be good to explain avery "no". Sometimes, "no" is a valid
answer but it must be explain, espacially when the decision making process
is fuzzy (and sometimes it's more than fuzzy as even the payed staff does
not know where the answer come from).
> - Agenda for all meetings. We should have a clear agenda for all
> meetings
> ahead of time.
>
And a reasonable schedule that allow people to prepare the meeting.
> - Bugs. We communicate well in bugs but we don’t always have bugs for
> all
> work.
> - Devengage Cafe. Create a Vidyo room just for hanging out together.
>
Interesting but IRC already provide this, what's the advantage of having a
video conference room here?
> Today we’ll discuss Developer Program value add - what we can best offer
> developers to help them create open web content - and the process for
> proposing contributions.
>
It's good to remember that "open web content" is something very different
than "Firefox OS content" or "Having Apps in the marketplace". Mozilla's
tools and projects are just a means to promote the web platform. They are
not a goals by themselves, if they fail it's not a problem as long as the
standard web succeed.
Best,
--
Jeremie
.............................
Web :
http://jeremie.patonnier.net
Twitter : @JeremiePat <
http://twitter.com/JeremiePat>