XML and Web Services In The News - 28 December 2006

Provided by OASIS | Edited by Robin Cover

This issue of XML Daily Newslink is sponsored by BEA Systems, Inc.


 Updated IETF Internet Draft: The Atom Publishing Protocol
 amplee: A Python Library Implementing the Atom Publishing Protocol
 Converter Enables Conversion Between ODF and UOF
 SOA Standards SCA and SDO Homeless for Holidays
 Grid for Particle Physics Project Coming Together
 Amsterdam Tests Open Source Software

Updated IETF Internet Draft: The Atom Publishing Protocol
Joe Gregorio and Bill de hOra (eds), IETF Internet Draft
Members of the IETF atompub Working Group have released an updated version of the "Atom Publishing Protocol" specification, companion to "The Atom Syndication Format", published as an IETF Request for Comments #4287 in December 2005. The Atom Publishing Protocol is an application-level protocol for publishing and editing Web resources using HTTP and XML 1.0. The protocol supports the creation of arbitrary Web resources and provides facilities for: (1) Collections: Sets of resources, which can be retrieved in whole or in part. (2) Service: Discovering and describing Collections. (3) Editing: Creating, updating and deleting resources. Atom Protocol Document formats are specified in terms of the XML Information Set. The Atom Publishing Protocol uses HTTP methods to author Member Resources as follows: [a] GET is used to retrieve a representation of a known resource. [b] POST is used to create a new, dynamically-named, resource. When the client submits non-Atom-Entry representations to a Collection for creation, two resources are always created &mdsh; a Media Entry for the requested resource, and a Media Link Entry for metadata (in Atom Entry format) about the resource. [c] PUT is used to update a known resource. [d] DELETE is used to remove a known resource. The draft document only covers the creation, update and deletion of Entry and Media resources. Other resources can be created, updated, and deleted as the result of manipulating a Collection, but the number of those resources, their mime-types, and effects of Atom Protocol operations on them are outside the scope of this specification. Along with operations on Member Resources, the Atom Protocol defines Collection Resources for managing and organizing Member Resources. Collections are represented by Atom Feed documents and contain the IRIs of, and metadata about, their Member Resources. The Atom Protocol does not make a distinction between Feeds used for Collections and other Atom Feeds. The only mechanism that this specification supplies for distinguishing a Collection Feed is its appearance in a Service Document. [Note: the diff version reports 85 change blocks, 459 lines changed or added.]
See also: Atom references

amplee: A Python Library Implementing the Atom Publishing Protocol
Sylvain Hellegouarch, IETF atom-protocol List Announcement
An announcement was posted for the release of amplee 0.3.6 — called 'alpha', but "... you can be safe and use it already as it stands; it should soon move to beta." amplee is a Python implementation of the Atom Publishing Protocol (APP), as specified in draft 11 or later. amplee's aim is to provide an API close to the APP specification which can be used in any Python code, and HTTP handlers to make it easy to integrate an APP store into a web application. amplee has different levels and can be used in a number of ways: (1) At a minimum it's just a pure Python library implementing the Atom Publishing protocol. In that case the API model is almost a 1-to-1 mapping of the APP spec. Modules in the amplee.atompub package include [a] store: APP does not define the meaning of store. We use that concept to describe the outter envelop that carries related APP entities. A store is the layer between the underlying storage (database, filesystem, etc.) with APP entities; [b] service: amplee attaches a service entity to a store. A service also has a list of workspaces it handles; [c] workspace: a workspace belongs to one service and has a list of collections; [d] collection: amplee collection module allows CRUD operations to be made on members. (2) It has built-in members for common data type such as XHTML, audio formats (MP3, Ogg, Flac, WavPack) or Open Document Text format. (3) You can also use built-in storages to persist members of a collection. Currently supported: filesystem, subversion, zodb, and database backend via dejavu. amplee is distributed under the Creative Commons Attribution-ShareAlike2.5 License.
See also: Atom Protocol Python Group

Converter Enables Conversion Between ODF and Chinese Document Format (UOF)
D.C. Parris, LXer Linux News
Peking University recently released a program to convert office documents between OpenDocument Format and the Specification for the Chinese office file format based on XML (UOF for short). Both standards are XML office document standards, UOF being a "National Standard of the People's Republic of China". The converter, which took nearly a year to complete, enables users to convert text, spreadsheet and presentation documents between ODF and UOF. The project's team members learned from IBM's OpenDocument Format experts, and at times, worked closely with them in resolving challenges. The team also participated in the UOF technical committee, providing insight into the differences between the two formats. Their participation in the project led to the adoption of several proposals by the UOF committee to integrate some of ODF's advantages into the UOF standard. Consequently, conversion between the two formats improved as well. The converter is licensed under the terms of the GNU Lesser General Public License. Ensuring a Free/Open Source solution was one of the goals of the project. Being able to convert documents properly between ODF and UOF should be a boon to millions of computer users, especially those using the Chinese national standard. The converter is known to work with OpenOffice.org and EIOffice 2007.
See also: Rob Weir

SOA Standards SCA and SDO Homeless for Holidays
Rich Seeley, SearchWebServices.com
SCA and SDO, emerging standards designed to provide programming models developers could use in creating Web services, have yet to fully mature and reach a standards body, but 2007 appears to be the year they will. Hopes were raised this past July that SCA/SDO would find a standards home for Christmas. Open SOA (OSOA), the vendor group, including IBM, BEA Systems Inc. and Oracle Corp., which is currently working on the standards, indicated last summer that they would make a decision on a standards body by the end of 2006. Now, it appears it will be sometime in 2007. In response to an e-mail request for an update on the status of SCA/SDO, Graham J Barber, program director, SOA partnerships at IBM and an editor working on the specifications, wrote: "We intend to take the main SCA specifications to a 'Version One' level around the end of the first quarter of 2007. It's our expectation then that these specs, together with the published V2.1 SDO specifications will be handed to a standards body for future stewardship." Whether or not it is fully mature or blessed by a standards body, SCA is already being incorporated into vendor products, and Ted Farrell, chief architect and vice president of tools and middleware at Oracle, says it's usable now. SCA is already included in Oracle WebCenter Suite, designed for developers working on SOA and Web 2.0 projects, he said. Rogue Wave Software, a division of Quovadx, Inc. announced earlier this month that is had incorporated SDO, including adding it to the product name, in its SOA tool suite, HydraSDO. [Ted] Farrell said SCA/SDO are following a similar path to Business Process Execution Langauge (BPEL). It was a combined effort of IBM and Microsoft, then joined by BEA, SAP AG and Siebel Systems, which is now part of Oracle. In 2003, those vendors then submitted BPEL for Web services (BPEL4WS) 1.1 to the OASIS open standards organization, which has advanced the renamed standard to its current form as WS-BPEL 2.0. Developers and architects working on SOA projects need to work with standards as they are today and should not wait for all of them to mature and become official, argues Peter S. Kastner, vice president enterprise integration for the Aberdeen Group, a Harte-Hanks Company. He believes those working on SOA projects will have to evolve with the standards.
See also: the 2005 announcement

Grid for Particle Physics Project Coming Together
Jon Brodkin, IDG News Service
Research data gleaned from beams of protons colliding 40 million times a second inside the largest scientific instrument ever built will soon begin flowing into an international network of computer centers designed for scientists trying to uncover the underlying structure of the universe. Physicists hope the project will find evidence that could lead to the discovery of extra dimensions, and evidence for the Higgs boson, a theoretical particle that has never been observed but which scientists believe endows all objects with mass. In late 2007, the Large Hadron Collider will open at CERN, the European particle physics laboratory in Geneva, Switzerland, and produce proton collisions that will create various types of subatomic particles. Five regional computer centers in the United States are among those being set up to analyze the Hadron data. One operated by the University of Chicago and Indiana University is now up and running. Overall, the project involves 158 institutions in 35 nations. The collaborative center is already analyzing test data in preparation for the launch of the Hadron Collider, which is the centerpiece of an experiment known as ATLAS (A Toroidal Large Hadron Collider Apparatus). Data will be distributed worldwide using grid computing, the use of geographically distributed computing resources. ATLAS data will flow from Switzerland to 11 'Tier-1' centers worldwide, including Brookhaven National Laboratory on Long Island, N.Y. The Tier-1 centers will then distribute data to Tier-2 centers, including the Chicago-Indiana center and four others in America.

Amsterdam Tests Open Source Software
Peter Sayer, InfoWorld
Microsoft Corporation has two years in which to convince Amsterdam city authorities that it can offer better value for money on the desktop than open source alternatives. The city will conduct tests of open source software on desktops in two departments in the first half of 2007. It doesn't intend to phase out the use of proprietary software completely, but expects the tests to lead to a reduction in the scope of its contract with Microsoft, its current desktop software supplier, the city said. That contract expires at the end of 2008. Two departments, the city's housing service, and the local authority for the borough of Zeeburg, will test the Linux operating system on desktop PCs, city authority spokeswoman Marjolijn van Goethem said Thursday. A study conducted for the city council earlier this year showed that the use of open software ensures better exchange and storage of digital information. Amsterdam is not alone in considering the switch from proprietary to open source desktops: The German city of Munich is in the process of migrating desktop PCs to the Linux operating system. Other local and national administrations around Europe have cited safeguarding access to their own data as a reason for preferring software based on open source and open standards over proprietary applications. The Belgian government voted earlier this year to adopt the Open Document Format (ODF) popularized by the open-source desktop productivity suite OpenOffice.org for document storage and exchange, while authorities in France and Denmark are also evaluating the benefits of ODF.
See also: ODF references

XML.org is an OASIS Information Channel sponsored by BEA Systems, Inc., IBM Corporation, Innodata Isogen, SAP AG and Sun Microsystems, Inc.

Use http://www.oasis-open.org/mlmanage to unsubscribe or change an email address. See http://xml.org/xml/news_market.shtml for the list archives.

Bottom Gear Image