Considering Easy-To-Implement OpenSLIM

[Tom Snyder]

On today's conference call Chris K. made the astute point that
implementing all of WebDAV can be a hinderance to immeidate adoption
of OpenSAM by document-storing Home applications.

Thanks, Chris, for pointing this out!!

Thus we are now considering spec'ing a "lightweight" version of
OpenSAM that relies on HTTP only, instead of the HTTP extensions
WebDAV. The straw-man name for this lightweight version is OpenSLIM.

Given current industry trends and use patterns, OpenSLIM could satisfy
many (most?) customer requirements for a first stage mashup. And, all
the work put into doing OpenSLIM leads directly to OpenSAM -- no
wasted effort.

The differences are:

OpenSLIM: requires only HTTP authenticated access to the Home server.
Can be done in days with Apache.
--OpenSAM: requires WebDAV authenticated access to the Home server.

OpenSLIM: can be launched SSO to edit a single document stored at the
Home application.
--OpenSAM: can be launched SSO and can browse, create, and edit any
document stored at the Home.

OpenSLIM: has *no* access to other content stored at the Home -
pictures, videos, sounds, text, etc.
--OpenSAM: has complete access to other content stored at the Home.

OpenSLIM: has no access to meta-data stored with a document.
--OpenSAM: has complete access to meta-data stored with a document.

Please post your feedback to this concept as we hammer out an initial
spec. Thanks!

[Chris Kohlhardt]

Thanks for this Tom.

I think this is a great start.

For authentication, has there been much thought into officially
supporting OpenID?

For data storage, what are Home Applications doing to implement WebDAV
on their side? Is WebDAV simple file based storage, or can you have a
database masked behind WebDAV?

Tom, can you explain a little more about how the data storage would
work in OpenSLIM?

thanks!

-chris

[Anton Bar (G.ho.st)]

That's totally bikeshed, but I would rather keep the OpenSAM brand.

How about OpenSAM Base (or even w/o the Base, just OpenSAM) for the
easy spec and OpenSAM Full or Extended for the original spec?

I wouldn't imply that the easy to implement spec is basic or partial.
This is maybe true comparing to the original spec, but otherwise,
on its own, it's a rather comprehensive solution.

Just my 2c.

Anton.

[Tom Snyder]

The WebDAV overview that Chris suggested is now at:
http://www.opensam.org/WebDAV%20Primer.html.

Please let us know what we should add.

Thanks!

(and good points, Anton, brand dilution is a danger).

[Tom Snyder]

The 4.1 Application Launch document that specifies SSO has been
updated to require only HTTP, not WebDAV. We now recommend the HTTP
HEAD request to test our access to the server rather than the WebDAV
PROPFIND request. (note that any type of request was valid to perform
this test, so the change does not break any current mashups).
See http://www.opensam.org/4.1%20Application%20Launch.html

We have a new "Implementation Road Map" document that strives to
clarify the portions of OpenSAM that are needed to accomplish
different levels of functionality.

The document breaks OpenSAM up into 3 levels: Level 100, Level 200,
and Level 300.
See http://www.opensam.org/pg_Implementation%20Roadmap.html

[Anton Bar (G.ho.st)]

BTW, I'm not sure whether "bikeshed" is a common expression.
I meant this: http://en.wiktionary.org/wiki/bikeshed.

On Apr 3, 12:38 pm, Tom Snyder <tomsn.sny...@gmail.com> wrote:
> The 4.1 Application Launch document that specifies SSO has been
> updated to require only HTTP, not WebDAV. We now recommend the HTTP
> HEAD request to test our access to the server rather than the WebDAV
> PROPFIND request. (note that any type of request was valid to perform
> this test, so the change does not break any current mashups).

> Seehttp://www.opensam.org/4.1%20Application%20Launch.html