It's been a long journey even though or maybe just because it started out with two specifications. Paul Freemantle wrote about the reason for the last delay of WS-RX but now the Balloting for OASIS Standard - WS-ReliableMessaging 1.1 is on. The critical number of affirmative votes required to approve this ballot is 51, and right now the status page gives a count of 43 positive votes of a potential total of 337, so with more than a week left it finally looks like the last child of the WS-*X family will succeed.
Tuesday, May 22, 2007
Sunday, May 20, 2007
The fight between SAML 2.0 and WS-Federation has been present in Danish eGovernment for some time, and here the choice has fallen on SAML 2.0, being an OASIS standard with a fair product support. There's more background information on that on Denmarks choice of SAML as the federation standard.
Newly the Danish National IT and Telecom Agency (DNITA) has announced an agreement with Microsoft about partial support for SAML 2.0. The central part of the quote from the agreement is:
To support interoperability between WS-Federation and SAML 2.0 based products Microsoft has agreed to support the SAML 2.0 token format in the future release of Active Directory Federation Services code-named Active Directory Federation Services2.
So it's only the Token format and it's not until the next version of ADFS, but god news is always welcome. It also mentions that WS-Federation has recently been submitted to OASIS, and adds:
This step further enables interoperability between federated environments that deploy SAML 2.0-based products and those that deploy WS-Federation-based products.
Well I'll say the jury is still out on that one as I've touch on in Federation power fight in the backyard of OASIS. As for the prior fight between SAML 2.0 and WS-Federation, I've found some articles like Microsoft Backs Web Services-Federation Against SAML 2.0 For Identity Federation and Vendors team on WS-Federation standard
First there was the clash of reliable web services messaging in OASIS. Second the outlook for a potential double ISO standard for document formats and now the next power fight looks to be on federation standards inside Oasis.
About a month ago OASIS let the word out with a Call for Participation on Web Services Federation. So what's the party about? you might ask, well you can start by reading the complete mail - it's long, so at a glance it looks like a lot of work has to be done, I guess it's a draft for the TC charter. The reason I write looks like is that the work kicks off on a version 1.1 of the WS-Federation specification, and I can't tell how close a match there is between the charter and the latest specification. One evil thought is that since the description is so detailed it could look like an attempt to rubber stamp the submitted specification, but a bit more on that later.
The two versions of the specification can be found on the IBM WS-Federation site. The first version of Web Services Federation Language (WS-Federation) from July 8 2003 is 41 pages. The WSDL has 9 wsdl:message's:
and 12 wsdl:operation's (spread over 6 wsd:portType's ):
This initial version was developed by
- BEA Systems, Inc.
- IBM Corporation
- Microsoft Corporation
- RSA Security, Inc.
- Verisign, Inc.
Web The second version (Version 1.1) is from December 2006 and has triple the original size with 124 pages. This version has some new participants and RSA missing (the new ones in bold):
- BEA Systems, Inc.
- BMC Software
- CA, Inc.
- International Business Machines Corporation
- Layer 7 Technologies
- Microsoft Corporation, Inc.
- Novell, Inc.
- VeriSign, Inc.
As for the WSDL for version 1.1 it's a bit strange. There's only one wsdl:message called
SignOut and two wsdl:operation's in seperate wsdl:portType's. I haven't dived into the specification yet, so maybe all the former operations stick, though not likely and in Appendix I - WSDL there's an example with the text
The following illustrates the WSDL for the Web service methods described in this specification and it references 5 specifications whereof only one is a standard:
- WS-Trust (OASIS Standard)
- WS-Eventing (W3C Member Submission 15 March 2006)
- WS-Transfer (W3C Member Submission 27 September 2006)
one conclusion could be that since 2003 there has been a massive convergence (and spread over new WS-Star specs) that the majority of operations are defined in other specifications.
Counting points - the fight is on
In the mix of paranoia, conspiracy and facts it's interesting to read the comments resolution log (pdf). There has already been said and written a lot about this, here are a couple of my picks: Burton Group, Eve Maler and Tim Bray. As usual the coverpages has a great collection of resources.
Tuesday, May 8, 2007
Browsing the news over at W3C I discovered the news about the second edition of SOAP 1.2:
- SOAP Version 1.2 Part 0: Primer (Second Edition)
- SOAP Version 1.2 Part 1: Messaging Framework (Second Edition)
- SOAP Version 1.2 Part 2: Adjuncts (Second Edition)
- SOAP Version 1.2 Specification Assertions and Test Collection (Second Edition)
For one I haven't even got over to SOAP 1.2, though Axis2 makes it possible for me. Secondly what is a second edition of the same version? Is this like going from 1.2.0 to 1.2.1? The answer is given there:
The second edition incorporates corrections for the known errata in the first edition
Uhmm, so it's just a write up of all the errors in the first edition. There are also multiple editions of other W3C standards like the Extensible Markup Language (XML) 1.0 that's in the fourth Edition and XML Inclusions (XInclude) Version 1.0 that's also a second edition. How often do they make these editions and on what grounds? I've search the site and didn't find anything on it, so I guess the groups make it up for them selfs when they feels it's time for a write up.
Posted by Sweetxml at Tuesday, May 08, 2007
I my home we don't have a TV, so when one of my girls wants to watch the children's program from The Danish Broadcasting Corporation (Danmarks Radio - DR) we use the streaming service that they offer. The problem with this service (the chosen format aside) is that it's way to often overloaded with the result that we're thrown off or i hangs for long periods (restart of streaming somehow seems to help, but is really annoying to do ex. every 5 minute).
Some time ago I saw that DR are working together with Joost to give greater availability. Back then I went to the Joost website to find that it's still in beta, and that you can't just try it out, you'll have to be invited, sigh. One evening I saw that the most popular search on Technorati was for joost invite and saw just how fast people commented to get an invite. Last weekend I was lucky enough to meet someone who could invite me. Since i looked last they removed the 5 invite per user limit, but anyway I still needed the invite. Last night I tried it out, and it turns out that only a few programs from DR are featured, so it'll not help getting a better signal for DR children programs. If I had the time I could probably waste some time on the sci-fi channel, but i do not have that much time :-). The stream quality in general seemed good, and hopefully I'll find some more time to check it out. Also it'll be interesting to see just how annoying the adds will be, since I didn't see any when I tried it out.
Posted by Sweetxml at Tuesday, May 08, 2007
The X-members of the shinning WS-star just keeps producing new standards. Now it's WS-Transaction version 1.1 that's become an OASIS standard. It's also one more of these standards to consist of multiple specifications, here the three (from the OASIS standards list):
- Web Services Coordination (WS-Coordination) v1.1
- Web Services Atomic Transaction (WS-AtomicTransaction) v1.1
- Web Services Business Activity (WS-BusinessActivity) v1.1
I'm not sure what happened to version 1.0, if there ever was one cause I can't find it on the WS-TX TC homepage. Judging from the list of companies that has participated in this standard, it looks much more promising than when I looked at WS-Context (maybe they are more related that i know of, but doesn't really change my view on it). Since I'll probably never have time to look into this standard I'll take the version short description of the specifications from the press release:
WS-Coordination enables an application service to create the context necessary for propagating an activity to other services.
WS-AtomicTransaction defines agreement protocols for short-lived activities having the all-or-nothing property.
WS-BusinessActivity defines protocols for long-running transactions that require compensation-based agreement.
Working together, these specifications enable existing transaction processing, workflow, and other systems to hide their proprietary protocols and operate in a heterogeneous environment