Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Proposed 2013 Platform Goals

197 views
Skip to first unread message

JP Rosevear

unread,
Dec 21, 2012, 11:17:33 AM12/21/12
to dev-pl...@lists.mozilla.org
Hello all,

As we prepare to enter 2013, I'd like to propose a set of goals for the
platform next year. These goals have had initial review by many
platform leaders and have an initial cut at owners. Owners are
responsible for producing a plan of attack for their area by Jan 15 and
converting those to quarterly goals.

The goals are of course not a complete list of everything we will do in 2013, but they are the key focus points. So if there is a decision to be made between work in support of these goals and other work, please think hard about the trade off and make the best decision you can.

Feedback welcome.

The below is also posted at:
https://wiki.mozilla.org/Platform/2013-Goals

** Goals

1) Enable three application types across the platform: Games (vlad),
Multimedia (roc), Social/Location (TBD) - (overall jst)
+ Have a high profile games engine company port to HTML5 publicly (vlad)
+ Deliver a carrier network interoperable multimedia chat app (jesup)
+ Drive deep integration of a tier one location platform (dougt)

Background: We must close the gap on native apps, but we won't be able to boil the
ocean in 2013. The idea is to focus on three. All three are important
app types for growing our marketplace. Carrier network
inter-operability is a potential revenue growth area as well and a
feature that will generate interest among carriers in shipping FF
Android on their devices. Need to dig a little deeper on how the
social/location focus should manifest itself.

2) Become a market leader in performance (dmandelin)
+ Define and deliver against metrics for startup speed, responsiveness,
and application feature speed (taras)
+ Create and promote our own compelling benchmarking and compliance
story

Background: We want to take our performance work such as snappy much
further in 2013. Its key to giving users a great experience. We are
also unsatisfied with most benchmarks and would like to build and
promote our own in addition to making key improvements on existing
benchmarks that are useful.

3) Increase product line development velocity & efficiency (jpr)
• Ensure web platform technologies are unified across all our products
(overholt)
• Reduce core platform fragmentation/technical debt (jet)

Background: We need to ensure that all our platform features are
available in all our products (ie WebAPIs). We also need to pay down
some excessive technical debt (ie OMTC for desktop, Azure for content)
as we've raced to put out two 1.0 products (fennec native and ffos) over
the last year.

** Product Support

In addition to targeted platform goals, platform must support our user
product deliverables.

1) FFOS (overholt, milan)
+ Support FFOS as we scale

Background: We must continue to build out platform support for FFOS in
key areas such as WebAPI and rendering.

2) FF Desktop (dmandelin)
+ This is primarily about performance in the platform goals

3) FF Android (blassey)
+ OEM/Carrier Support
+ Mobile web compatibility

Background: We want OEMs and carriers to ship FF Android on devices. We
will need to move ahead on performance and marketplace support as well
as explore the carrier and OEM needs more deeply. We also need to
improve through technical means and evangelism compatibility with mobile
websites.

** Overall Engineering Quality

Generally we should always strive to build a secure, stable, performant,
accessible and tested product as a baseline. We should also stick to
our development principles which means open development and process as
much as possible and interaction with all contributors.

1) Software Quality (jpr):
+ security (davidb)

Background: We must make quality software!

Thanks,

-JP





Justin Dolske

unread,
Dec 22, 2012, 6:11:37 PM12/22/12
to
On 12/21/12 8:17 AM, JP Rosevear wrote:

> ** Goals
>
> 1) [...] Social/Location (TBD) - (overall jst)
> [...]
> + Drive deep integration of a tier one location platform (dougt)

Hmm. I don't understand what this is. Could you say a bit more about it?

Does "location" involve geolocation? Is there overlap with SocialAPI work?


> • Reduce core platform fragmentation/technical debt (jet)

+1. I think it's likely something along the same lines will show up on
front-end's radar. :)


> ** Overall Engineering Quality
>
> Generally we should always strive to build a secure, stable, performant,
> accessible and tested product as a baseline. We should also stick to
> our development principles which means open development and process as
> much as possible and interaction with all contributors.

Good to see alignment with front-end here, too. Conner used the term
"table stakes" here, which I really like. It's hard to build the next
new thing unless you've got a solid foundation, and it's wise to
remember that doesn't just happen accidentally (or for free).

Justin

JP Rosevear

unread,
Dec 28, 2012, 8:16:19 PM12/28/12
to Justin Dolske, dev-pl...@lists.mozilla.org
On Sat, 2012-12-22 at 15:11 -0800, Justin Dolske wrote:
> On 12/21/12 8:17 AM, JP Rosevear wrote:
>
> > ** Goals
> >
> > 1) [...] Social/Location (TBD) - (overall jst)
> > [...]
> > + Drive deep integration of a tier one location platform (dougt)
>
> Hmm. I don't understand what this is. Could you say a bit more about it?

Not at the moment I can't, Doug is digging in a bit more on the
requirements here. This derives from the Apps team top 3 app types and
some business development work about integrating a mapping platform.
TBD how much needs to be done in the platform.

> Does "location" involve geolocation? Is there overlap with SocialAPI work?

Probably the former but not the latter, but as above TBD.

>
> > • Reduce core platform fragmentation/technical debt (jet)
>
> +1. I think it's likely something along the same lines will show up on
> front-end's radar. :)

Its on the goal list particularly this year because we have built up an
abnormal amount pushing for the release of two "1.0" type products
(fennec native and b2g).

Thanks,

-JP

0 new messages