|
XML and Web Services In The News - 18 January 2007
Provided by OASIS |
Edited by Robin Cover
This issue of XML Daily Newslink is sponsored by SAP AG
HEADLINES:
Open Geospatial Consortium Joins W3C Geospatial Incubator Activity
Staff, OGC Announcement
The Open Geospatial Consortium (OGC) announced that it has taken a role
in the World Wide Web Consortium, "a standards organization that
develops interoperable technologies (specifications, guidelines,
software, and tools) to lead the Web to its full potential." The OGC is
participating in a W3C incubator activity focusing on semantic
geospatial issues. W3C Incubator Activities facilitate rapid development,
on a time scale of a year or less, of new Web-related concepts. The
semantic geospatial activity or Geospatial XG is sponsored by W3C members
OGC, SRI International, University of Southern California Information
Sciences Institute (USC ISI), Stanford University and Oracle and is
chaired by Traverse Technology's Joshua Lieberman. As an initial goal
the Geospatial XG is working to develop a W3C "Note" based on GeoRSS
version 1. This will result in a W3C Web page describing GeoRSS in the
context of both W3C standards such as XML, HTML, and OWL; and OGC's
relevant work, such as the OGC Abstract Specifications and Geography
Markup Language (GML). The OGC and the W3C seek to collaboratively add
geospatial functionality to the emerging Semantic Web in a manner that
is consistent with existing and future OGC standards.. OGC standards are
the product of a successful 12-year open, international, and consensus-
driven effort to overcome obstacles to geospatial interoperability.
See also: Geography Markup Language (GML)
NetBeans Adds Ties to Devices, ALM
Paul Krill, InfoWorld
Sun Microsystems announced that it will advance the NetBeans open-source
tools platform on devices and the application lifecycle management (ALM).
Sun and the NetBeans community are announcing the availability of
NetBeans Mobility Pack for Connected Device Configuration (CDC) 5.5 for
developing applications for devices such as full-featured smart phones.
NetBeans also is being fitted with a plug-in for ALM, which could be a
precursor to a full-scale NetBeans ALM initiative. Featuring code
contributions from Ricoh, the CDC pack provides for visual design and
development of Java ME (Java Platform Micro Edition) applications for
devices using CDC. The CDC technology features an embedded Java solution
for smart phones, set-top boxes, and multifunction peripherals. Sun's
Dan Roberts: "CDC is designed to run on systems with a more powerful
processor and additional memory" than standard phones used for voice
calls. A smart phone can feature applications such as e-mail and
calendaring." The NetBeans Mobility Pack for CDC 5.5 will be available
for download next week. Intland Software, meanwhile, has developed a
plug-in for its Java-based CodeBeamer ALM platform. The open-source
plug-in links to the NetBeans IDE. The company is making available an
open-source version of CodeBeamer for 15 users. Intland is providing
ALM capabilities, including enterprise wiki and project management
functions as well as trackers for bugs, tasks, changes, and requirements.
The United States Department of Agriculture Natural Resources
Conservation Service uses CodeBeamer and NetBeans for a software
development operation that includes more than 675 users and 200 projects
in diverse locations.
See also: the announcement
Logically SOA: A Logical Architecture for SOA
Srikanth Seshadri, JavaWorld Magazine
As the next step in the architectural evolution, service-oriented
architecture (SOA) captures many best practices of the architectures
that preceded it. Taking into account the number of existing systems
and investments made by the industry in them, any new architecture
should leverage and improve upon existing infrastructures and shouldn't
discard any part of these infrastructures. This is really the essence
of SOA; it aligns the existing technical infrastructure closer to
business. This story tries to conceptualize a logical architecture
independent of various implementation methodologies and technologies
for SOA. Such a logical architecture is useful in design and for
organizing systems in enterprises moving towards SOA. The architecture
described is independent of vendor product/terminologies and can be
easily mapped to them. As depicted [in the diagram 'Logical architecture
of an SOA system'], the SOA-specific layers are additional layers
sandwiched between the existing tiers of the current architectures. The
layers in the diagram are stacked in standard manner, based on usage.
The top layers use the services of the bottom layers to satisfy
requirements and the functional expectations from the layers above. This
architecture is intended for use during the creation of a technical
SOA road map for an organization. It helps in identifying the gaps in
the current environment and the target SOA-based infrastructure. The
architecture serves as a good reference for the projects that require
development of reusable SOA frameworks. Mapping the logical
architectural layers to a product matrix will assist in determining
the SOA-related products required for use in projects.
Internet Calendaring and Scheduling Core Object Specification (iCalendar)
Bernard Desruisseaux (ed.), IETF Internet Draft
IETF's Calendaring and Scheduling Standards Simplification (Calsify)
Working Group has announced an updated version of the "Internet
Calendaring and Scheduling Core Object Specification. iCalendar
defines a MIME media type for representing and exchanging calendaring
and scheduling information such as events, to-dos, journal entries and
free/busy information. The definition of the text/calendar media type,
known as iCalendar, is independent of any particular calendar service
or protocol. This memo is intended to progress the level of
interoperability possible between dissimilar calendaring and scheduling
applications. This memo defines a MIME content type for exchanging
electronic calendaring and scheduling information. The Internet
Calendaring and Scheduling Core Object Specification, or iCalendar,
allows for the capture and exchange of information normally stored
within a calendaring and scheduling application; such as a Personal
Information Manager (PIM) or a Group Scheduling product. The iCalendar
format is suitable as an exchange format between applications or systems.
The format is defined in terms of a MIME content type. This will enable
the object to be exchanged using several transports, including but not
limited to SMTP, HTTP, a file system, desktop interactive protocols such
as the use of a memory- based clipboard or drag/drop interactions,
point-to-point asynchronous communication, wired-network transport, or
some form of unwired transport such as infrared might also be used. The
memo also provides for the definition of iCalendar object methods that
will map this content type to a set of messages for supporting
calendaring and scheduling operations such as requesting, replying to,
modifying, and canceling meetings or appointments, to-dos and journal
entries.
See also: the WG Charter
REST Eye for the SOA Guy
Steve Vinoski, IEEE Internet Computing
I'm torn in the Representational State Transfer (REST) and SOA debate:
REST is extremely appealing, but my technical background is firmly
rooted in the SOA camp. In this column, I try to explain REST from the
viewpoint of someone steeped in SOA, with the intention of helping SOA
people understand the value the REST camp so rightfully touts. SOA
encourages several critical development practices, but the most
important ones are establishing and adhering to service contracts and
splitting interface from implementation. SOA proponents regard
interfaces and contracts as being critical to service definitions:
different services have different interfaces — a normal and desirable
characteristic of software systems, whether they're distributed or not.
REST proponents, on the other hand, stand by the uniform interface
constraint. The data variability part of the scalability equation
(that different services expect and deliver different data formats)
remains within REST, even if interface variability is eliminated.
Although data variability is indeed a factor in both SOA and REST
systems, REST has an advantage here as well. REST's foremost concern,
unlike SOA, has always been distribution: it focuses primarily on
ensuring that distributed hypermedia systems can scale and perform
well, by explicitly constraining important aspects of their
architectures to handle issues related to distribution and by
separating critical orthogonal concerns.
WS-Context Moving Through OASIS Standards Process
Eric Newcomer, Blog
"I am reminded to mention that we are finally moving the WS-Context
spec toward a vote for possible acceptance as an OASIS standard. One
of the things that I have always liked about WS-Context (and this by
the way is something I picked up during the Web Services Architecture
work from Mark Baker and others) is the opportunity to model the
context as a Web reference. It can also be passed by value directly
in a SOAP header. WS-Context defines an XML structure in which to
store data elements related to the service execution environment, such
as transaction context (from which the idea was generalized), database
connection IDs, user IDs, security tokens, and so on. These are values
typically generated by the service infrasturcture rather than created
by the application, but need to be associated with the application.
Some of the committee members used to suggest a better name for the
spec would be 'WS-Cookie' and that is a good way to think about its
purpose. The spec underwent somewhat significant revision during the
WS-CAF TC process, but I am glad to say that the RESFful behavior has
been preserved. It is perfectly reasonable to implement a 'context
service' (as defined in the spec) using HTTP GET and PUT to obtain a
representation of, and update the state of, a Web resource in order
to share context across multiple service invocations. I highlight
this because the expected behavior of a WS-* related spec is to pass
information by value as a SOAP header, which WS-Context certainly
does. However I believe WS-Context is a little bit different in also
supporting a RESTful behavior."
SOA Moves Toward Event Handling
By Rich Seeley, SearchWebServices.com
Business event management (BEM) and complex event processing (CEP)
are the next step in the fusing of SOA and BPM in business process
management suites from major vendors, according to Ken Vollmer, analyst
at Forrester Research Inc. BEM is a cousin to workflow systems, but
where the latter focus on human interactions within an organization,
BEM focuses on the business processing and business rules and then
alerts humans when something goes wrong. The goal is to speed processes
up by minimizing time lost because of an exception. As an example,
Vollmer points to a system where a credit exception needs to be granted
to a customer to complete an order. If a sales manager is alerted, the
credit exception for a reliable customer could be granted immediately
and the transaction could then be completed. Without the BEM alert,
the customer might become frustrated, turn to another supplier and
major business might be lost. Vollmer defines CEP as processing
software that "automatically correlates events into patterns that may
represent a threat or opportunity and orchestrates an appropriate
response." As he points out, the concept is not new, having originated
in 1998 in a Stanford University research paper authored by David
Luckham and Brian Frasca. CEP has been slow coming to market, although
Tibco Software Inc., webMethods Inc. and Sun Microsystems, include it
in their integration-centric business process management suite
(IC-BPMS) product offerings. Despite the promise BEM and CEP to take
SOA and BPM technology to the next level, adoption remains slow and
Vollmer said he has not seen much progress in adoption in the six
months since he wrote his first report.
Autonomic Computing
Daniel A. Menasce and Jeffrey O. Kephart, IEEE Internet Computing
The Janury/February 2007 issue of IEEE Internet Computing contains
articles that illustrate the general flavor of the research that's
needed for us to move more quickly toward the ultimate autonomic
computing vision. In October 2001, Paul Horn, IBM's senior vice
president of research, coined the term autonomic computing to describe
a solution to the ever-growing complexity crisis that threatens to
thwart IT's future growth. In this vision, systems manage themselves
in accordance with high-level behavioral specifications from
administrators — much as our autonomic nervous system automatically
increases our heart and respiratory rates when we exercise. In five
years, autonomic computing has evolved into a new subdiscipline of
computer science. Academics and industry professionals attend dozens
of conferences and workshops and write hundreds of papers each year;
universities around the world are offering dozens of courses on the
subject; government agencies in Europe and the US are supporting a
variety of autonomic computing research projects; and several industry
labs have substantial research and development efforts in autonomic
computing. Spurred by this momentum, the worldwide community has
progressed on many fronts. Dozens of products from both well-established
vendors and startups offer hundreds of autonomic features that ease
the administrator's burden. Systems-management standards efforts such
as OASIS's Web Services Distributed Management (WSDM) are under way,
aiming to facilitate the creation of multivendor autonomic computing
systems. However, we've by no means reached our destination. Although
many autonomic components have been developed and are proving useful
in their own right, no one has yet built a large-scale, fully
autonomic computing system — comprising multiple components that work
together to satisfy high-level business goals — that exhibits the
ability to configure, heal, optimize, and protect itself.
Docvert Version 3.0 Supports Word Processor Format Conversion to HTML
Matthew Cruickshank, Developer Announcement
Version 3.0 of Docvert has been announced. The software is described
as a web service software takes multiple word processor files
(typically .doc) and converts them to OASIS OpenDocument and HTML.
The Web Service receives .doc file and converts it to a OASIS
OpenDocument 1.0 which can then be converted to HTML, RSS, or any XML
format. The resulting OpenDocument is then optionally converted to
HTML or any XML. This is done with XML Pipelines, an approach that
supports XSLT, breaking up content over headings or sections, and
saving those results to multiple files (e.g., chapter1.html,
chapter2.html); the result is returned in a .zip file. Docvert is easy
to integrate as it uses a simple REST-style interface, and it's
released under the LGPL so although it's open source there's no legal
problems developing proprietary software ontop of it. The XML produced
is easier to understand and more structured than the WordML or .DOC
formats. New in version 3.0 (1) Post-conversion editing. Upload an
word processing file, preview, and edit it on the browser. Remove
chapters, correct typos, and rebuild the document to download the
resulting HTML. (2) FTP/WebDAV upload. Send the results of a conversion
to your website. (3) Control every tag and attribute with XSLT, PHP,
and XML Pipelines. (4) Image format conversion. Convert WMF files to
PNG and GIF. (5) Microsoft Word Plugin, which installs a toolbar and
allows one-way conversion of the word document to OpenDocument and HTML
(multiple pages, based on document structure, controlling any tag or
attribute). Successful use requires disciplined application of Word
styles: "For all but trivial examples you'll need to use Word Styles
for any conversion software so that it knows how to section your
document and format everything correctly. Like most conversion software,
Docvert ignores font sizes and background colours and instead makes
decisions based on structural Word Styles that describe paragraphs,
headings, lists, tables, etc."
See also: the FAQ document
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. |
