David,
Thank you.
I am a bit pressured with deadlines this week, but will sit down
closer to the weekend and answer the second part of your email.
/Anne
On 17. aug.. 2009, at 18.12, David Pollak wrote:
On Mon, Aug 17, 2009 at 1:46 AM, Anne Kathrine Petterøe
<[email protected]>wrote:
I was more just wondering if David would need some help from the
rest of
us?
What I've done so far is to create a project at GitHub (
http://github.com/dpp/esme_g2/tree/master ) I find Git's branching
much
easier to allow for playing and exploring... which will be very
important in
the early days. I will add all the ESME committers to the GitHub
project.
Once the codebase stabilizes, we can move it into the Apache repo.
(If the
Apache powers that be have an issue with this, let's all talk
through it
now.)
I think we do need to get to a release of ESME which includes (1) a
stable
UI and (2) a flexible authentication system (OpenID, User/pwd,
etc.) We
need a stake in the ground that says, "we can ship."
Right now, I've been down in Goat <http://goatrodeo.org>
Rodeo<http://blog.lostlake.org/index.php?/archives/94-Lift,-Goat-Rodeo-and-Such.html
>land
(which, along with Lift, will provide the foundation for G2). My goal
is to allow the description of G2 with something like (not pseudo
code, but
actual Scala code):
ReceiveUpdate[(User, JSON), (User, Update)] %>
ApplyInterpretations[(User,
Update), (User, Update)] %> (DistributeTo[(User, Update), (User, User,
Update)] * PostToInbox[(User, User, Update), Unit])
This will allow for a high level description of the message flow
process.
The message flow (inline, Actor-based, distributed) and logic lookup
(think
Erlang-style run-time changes to business logic) is handed by the
runtime
and does not impact the definition of the business logic.
I've got some good stuff running on my local machine (there's one
set of
code that I can't check into the Goat Rodeo repository until I get
sign-off
from the author, so this stuff is local to me.)
So, what would be great from the ESME team are:
- A list of external sources and sinks of ESME updates. The list
includes: Twitter, iCal, generic RSS. I have no idea of where we
want to
send (sink) ESME updates. In the above update flow, the sources
would hook
to ReceiveUpdate. PostToInbox will result in something other than
Unit...
and that would be an attachment point for sinks.
- A list of apps you'd like to build on top of ESME. My list
includes:
To Do (on a pool-by pool basis, so it could be a work-group to-do
app),
hours tracking on a project by project basis. Once you have the
app in
mind, think about example input (updates) that would drive the
app, think
about a tabular data structure for representing the data required
for the
app, and how the updates mutate the data table(s) (can this be
expressed
with a simple spreadsheet-like formula?)
I hope to formalize the above into (1) coding tasks to write the
simple
bridges between the sources/sinks and the core system and (2)
integration
tests for the application/example code/documentation for end users.
Sound reasonable?
Thanks,
David
PS -- I'm off ESME today, but back on for most of the rest of the
week.
/Anne
On 17. aug.. 2009, at 10.37, Richard Hirsch wrote:
As I said in a previous email, I'm assuming that a branch would be
easiest way for @dpp to get started.
D.
On Sun, Aug 16, 2009 at 10:20 PM, Anne Kathrine
Petterøe<[email protected]> wrote:
David,
I would say we have all agreed that G2 is the way forward.
How you thought about how we proceed from here? Who does what?
/Anne
On 12. aug.. 2009, at 09.15, Vassil Dichev wrote:
Actually, I think this is an extension to microblogging.
If I microblog:
tcw: dickh re: #ESME standup time: 30min
todo: add different authentication to #ESME
working on: #ESME authentication
done: #esme authentication
over the course of a day, those updates are meaningful to
humans, but
can
also be parsed and used to update structured data.
In fact, the idea is not new. Take a look at twitterdata.org :
"...simple, open, semi-structured format for embedding
machine-readable, yet human-friendly, data in Twitter
messages...".
That totally defines what we want to do with actions.
Granted, you won't see as much interest in processing messages on
Twitter as in a corporate environment, but @rtm, @timer and
@iwantsandy are still quite popular.
I also don't see this as something that replaces what ESME
currently
is, but as a way to extend ESME. Most likely the change wouldn't
even
be noticeable, like when Google introduced Calculator- there's no
change in the interface, but if you typed "2+2" or "123 in hex" or
"sin(pi/4) ^ 2", the result would get evaluated.
Vassil
--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Git some: http://github.com/dpp
