[sugar] upcoming goals for sugar for april and may

[Christopher Blizzard]

Hi, folks!  Yesterday most of the sugar team and I sat down and tried to
hash out what we thought were reasonable goals for the upcoming month.
In attendance were myself, Eben, Dan, Daf, Robert McQueen, Tomeu and
Marco.  Highlights are included here:

SCHEDULE
========

We talked a bit about schedule.  Everyone thought that sticking to a one
month development cycle seemed to work out pretty well.  That is, spend
about 3 weeks doing development and about a week of stabilizing coming
to a stable release at the end of a calendar month.  So based on that
our "April Goals" are goals that we hope to include in a build available
at the end of April.

HIGHLIGHTS
==========

Here are the list of items that we decided were important for April:

1. Mime types as activity attributes.  Right now there's a hard coded
list in sugar of mime types -> activities.  These include pdf, html,
images, etc.  We need to make sure that that's part of the attributes
for an activity bundle so that we don't have to have that static list.
For example, this should let the helix guys install their app without
having to ask us first.

2. Person to person chat.  This will use an XMPP server as the
transport.  Should be simple enough to hook up to any chat server,
including the school server, google or some other server out there.
Collabora has a server set up for us to test against.

3. Group chat.  Based on the discussion in the room it's pretty clear
that there's a decent amount of design work that needs to go on here.
Might be chat rooms on a server, might be based around activities (as
per the original design) might be something else.  Eben and others will
work on this during the month.  Not sure how far we'll get but it would
be great if we can get something like this working.

4. Working Journal and app integration.  This is the most important
thing that we'll be doing during the month of April.  It's also the
highest risk.  Our data store story is weak and Marco didn't think that
April was going to be enough time to get everything done.  So the
strategy is to get as much working as possible and maybe we'll end up
with some apps in the Journal and some apps out of it.  Eben pointed out
that if we hooked it up in such a way as it would work with the current
save/open buttons/arch it might make things smoother.

In any case, we need to get this going again as it's a huge part of our
story and is required to make the project a success.  So we're going to
be spending a lot of our time on it.

5. Video chat app.  This is a killer app for us.  The Collabora guys
will be handling a lot of the work here.  We have all the pieces that we
need, but we're lacking a design.  Eben will work on a simple design so
we can at least get a point to point connection (i.e. entering an IP
address/hostname) going or using XMPP as the startup mechanism, ala
gtalk.  We hope it will be useful, more than just demo-ware.

6. Simple sample tubes apps.  The Collabora guys have just gotten app
communication over XMPP working which is the first step to setting up
some a very easy mechanism for activity writers to be able to
communicate between activities.  We'll need to do a couple of sample
apps that show how this will work.  The two activities that we
identified as good examples are Erik's camera app and the block party
app.  The camera app is great because it's about sharing things that
kids are creating and the block party app has gone through a few
revisions and will be the basis for a "howto" on how to write a full
fledged mesh-aware activity.

7. Be able to download and install an activity from the web.  Eventually
we'll have a safer mechanism for doing this, but for now it would be
great if we had the ability to unzip an archive, drop it in the right
directory and ask the sugar-shell to reload the list of activities.
Marco says that a lot of this is in place today, but it will still need
some work.

8. Grab scrolling.  That hand key is on the keyboard and we should be
using it.

9. XO in the middle of the mesh view.  This is pretty minor but
completes the zoom experience.

10. Some kind of mesh search.

MAY GOALS
=========

The May goals are much more hazy as they are a couple of months out.
But the highlights include:

1. Getting peer to peer XMPP working well enough to demo and test and
some apps that use it.  This is critical for these days.

2. Web site for developers and end users so that it's easy to find
activities.

3. Online updates (requires resources be brought to bear that we don't
have today.)

4. Documentation for APIs (which should be getting more solid) and a
stronger focus on the larger development community.

5. Sugar extensions and activity extensions as part of the base API.