W3C home > Mailing lists > Public > www-xkms@w3.org > January 2005

DRAFT XKMS CR Interop Summary

From: Jose Kahan <jose.kahan@w3.org>
Date: Wed, 12 Jan 2005 17:50:19 +0100
To: www-xkms@w3.org
Message-ID: <20050112165019.GA11063@inrialpes.fr>
Hello,

Following my AI from yesterday's meeting, I compiled a draft summary
of results[1] and linked it to the XKMS home page. X-KISS testing 
is almost done, both X-KRSS and Compound testing needs more work. 
Optional tests are done (unless we add the stringprep one).

Our exit criteria required having two client implementations and two
server implementations for the non-optional attributes and only one
client and one server implementation for the optional ones. In
yesterday's meeting, we decided that to give further interoperability
proof we should have at least two clients being succesful against
each server and, when possible, have a same client contact two servers. 

(I'll check with the QA people to see if it's ok if we don't always 
have the same client contacting two servers if we give a valid reason 
(like lack of time)).

I added some comments and notes. In particular:

- Yunhao, I didn't take into account the results of your client against
  your own server, as we thought it would be more fair to only
  include tests done against other servers when people had both a 
  client and server. Could you update your report and choose "not
  applicable" or something similar, like Tommy did in the column
  referring to his server?

- I considered that a server passed a test when a client reported
  success against it. When this wasn't the case, I couldn't tell 
  if the server supported that feature.

- In some cases a developer reported success and another one failure
  when testing against the same server (e.g., XKRSS-T1). You should
  get in touch to see if the problem came from a misunderstanding of
  the spec or of the test.

- We apparently only have one server implementing XKISS-T17, 
  XKISS-T18, and XKRSS-T9. Yunhao, is there a possibility that could
  support those tests and if yes, could you give some estimation of
  when it could get done?

Thanks everybody for their hard work so far. We're almost done.

-jose

[1] http://www.w3.org/2001/XKMS/Drafts/test-suite/CR-XKMS-Summary.html

Received on Wednesday, 12 January 2005 16:50:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:31:43 UTC