On 06/08/2010 01:04 AM, Dianne Hackborn wrote:
> On Mon, Jun 7, 2010 at 3:21 PM, Olivier Guilyardi <l...@samalyse.com
> <mailto:l...@samalyse.com>> wrote:
> 
>     This openness is your strength, and I think that it will lead to
>     success if you
>     manage to build an active community of contributors. I'm not
>     monitoring this
>     closely, but it looks like it is not the case (yet).
> 
> 
> Development has increasingly moved to the open.  Unlike many open-source
> projects, we have a lot of complications that make it very difficult to
> just throw everything into the open: from the fact that in many
> countries you can't file patents after any information about them is
> publicly available, to needing to work with hardware manufacturers on
> major new features that are important for them to control the PR around
> for their devices.

Okay, I'm not a free software fanatic, but I'm going to tell you right away
(something that you may already know): a big part of the free software community
is likely to consider these reasons as unacceptable.

I do understand these reasons though, but I also know the community, and it's
not without reason that one of the most widely used tool is named 
"Subversion"...

> Development tools, Dalvik, and other things are now done in the open.
>  There is increasing desire to have the framework done in the open, but
> this is challenging because it will by necessity mean we need to be
> working in two branches, to continue to keep sensitive work private
> until it can be released.  Given that we also are often working on two
> more more active internal branches (for the current release under test,
> the next release in development, and often the next release after that),
> the development process for having yet another active branch (with all
> the corresponding merge issues and correctly deciding which branch X
> work should go in) is...  not pleasant to contemplate.

I think that your problem is that the whole of Android is too monolithic. You
are mentioning branches where you could think in terms of standalone modules.

If you were to make it more modular then you could get a boost in contributions.

For example, let's take Dalvik, it's a great VM. So why isn't that included in
Debian/Ubuntu for example, why can't I: aptitude install dalvik?

Whereas WebKit *is*. There even are python bindings. Do you get my point?

With modularity, you could maybe safely turn non-sensitive modules into
standalone projects, which could be used outside of Android. And that could
result in wider interest.

> Also, we have been accepting an increasing number of patches from the
> community.  I have regularly been reviewing and accepting multiple
> patches every week, as have other framework engineers.  This is slacking
> off a bit as we wait for the froyo drop, but I am sure it will pick up
> again.  Most patches don't need to be done on the very most recent head.

That sounds good indeed.

--
  Olivier

-- 
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en

Reply via email to