New Development Proposal

[Carl Nagle]

Phil et. al.,

I am the Project Owner and Architect for SAFSDEV: http://safsdev.sf.net

I am currently tasked with providing a workable remote-controlled
automation "engine" for Android testing which may (or may not) include
a remote-controlled Robotium.

While working on my remote-controlled proof-of-concept I have run into
several scenarios for which autoandroid appears to be the intended
solution. Primarily, a Java API for using the SDK tools (adb,
emulator, etc..) at any arbitrary test runtime--not necessarily during
an ant build.

I'm writing to find out if there is any reason besides resources this
project's activity has waned. Is there any reason why I should not
pursue using and possibly enhancing the Java APIs provided by
autoandroid?

Any insight or advice (or alternatives) will be greatly aprreciated.

Carl Nagle
carl....@sas.com
SAS Institute, Inc
www.sas.com
Planning, Operations, and Strategy
Software Quality Tools and Infrastructure
Project Architect, SAFSDEV
safsdev.sourceforge.net

[Phil Smith]

Hi, Carl

My personal involvement with the project dropped off some time ago
when I changed employment, taking me out of the mobile and java
spaces. That aside, as I was going it became clear that the tools I
was making were in direct competition with those of platform vendors
(i.e. Google.) For example, I understand they have something
functionally equivalent to a story runner now. Of the two of us, if
Google wanted to build the things that fitted the need, I would rather
have them do it. I haven't been following developments since, so I
don't know how applicable their tool chain is to these particular
problems.

That's the only reason that comes to mind. After several major
versions they've probably settled down to the point where it again
makes sense to consider where 3rd party development tools fit into the
ecosystem. I don't have any personal interest in doing this, but if
you have similar efforts, by all means loot whatever parts you want.

I should mention that some time ago the engineers at http://www.kualitatem.com/
reached out to me and asked for stewardship of the auto-android and
positron projects, and I added them a committers. Looking again now,
it seems they haven't particularly done anything with it yet.
Discussion of the tools' future should probably involve them as well.

On Dec 16, 8:57 pm, Carl Nagle <sascan...@gmail.com> wrote:
> Phil et. al.,
>
> I am the Project Owner and Architect for SAFSDEV:http://safsdev.sf.net
>
> I am currently tasked with providing a workable remote-controlled
> automation "engine" for Android testing which may (or may not) include
> a remote-controlled Robotium.
>
> While working on my remote-controlled proof-of-concept I have run into
> several scenarios for which autoandroid appears to be the intended
> solution.  Primarily, a Java API for using the SDK tools (adb,
> emulator, etc..) at any arbitrary test runtime--not necessarily during
> an ant build.
>
> I'm writing to find out if there is any reason besides resources this
> project's activity has waned.  Is there any reason why I should not
> pursue using and possibly enhancing the Java APIs provided by
> autoandroid?
>
> Any insight or advice (or alternatives) will be greatly aprreciated.
>
> Carl Nagle

> carl.na...@sas.com
> SAS Institute, Incwww.sas.com