Category: XML

Web Graph Diagram API

For a current project I need to be able to display a graph of objects on a web page, which is essentially a set of nodes and lines between these nodes. After looking around the Internet I came across JSVis which is a JavaScript library that allows you do exactly what I wanted and even has AJAX support to dynamically create the graph.

Links
JSVis Home Page

IBM Releases AJAX Toolkit

IBM has released a toolkit to help developers create web applications that use the AJAX connection technique to allow web pages to open a connection back to the originating server and request more data. AJAX therefore allows developers to create web pages that can provide more dynamic page behaviour with additional data controls without having to continuously reload the page, which takes time and bandwidth. The IBM toolkit is available from the alphaWorks site as part of the Emerging Technologies Toolkit (ETTK) and has been proposed as an Eclipse Incubation Project.

Links
IBM AJAX Toolkit Framework

Semantic Services

A friend from IBM just sent two links through to me for some Ontology and semantic based work IBM has on their Alphaworks site.

Ontology-based Web Services for Business Integration
A semantic Web services proof-of-concept demonstration for the industrial sector that shows service discovery, composition, and business process transformation.

Semantic Tools for Web Services
A set of Eclipse plug-ins for semantic matching and composition of Web services. (This is an ETTK technology.)

Links:
Ontology-based Web Services
Semantic Tools for Web Services
IBM Emerging Technologies Toolkit (ETTK)

FacesClient

One of the big problems with developing web application is creating good interactivity for the user without having to refresh the page. Using multiple technologies like AJAX it is possible to enhance this interaction but the work required for developers to achieve this is great. Therefore using the Model View Controller (MVC) concepts from Java Server Faces the FacesClient has been developed. IBM has an implementation of this JavaScript library and components and is available from their AlphaWorks site in the Emerging Technologies Toolkit (ETTK). If you would like to just have a read then below is a link to the FacesClient developer guide.

Links
Emerging Technologies Toolkit (ETTK)
Faces for Laszlo
FacesClient Components, Part 1 : Portlet Programming with FacesClient Components
FacesClient Components Developer's Guide (second edition)

XMLTV for Adelaide, Australia

I have been using a program called TVxb to download TV Guides and create XMLTV compatible XML files for use in a number of other programs. Until recently this has been working although my source provider changed so I have to recreate the settings file. This new file can be found here and works for all free to air TV channels for Adelaide, Australia.

Links
TVxb.ini - TVxb INI file for Adelaide, Australia
TVxb Home Page

Wiki in a Browser Page

I find wiki's a really cool idea. If you are not familiar with them they are essentially a way of creating web pages using a very simple markup language, so simple that anyone can edit a wiki page, even if you know nothing about HTML. The whole idea behind wiki's is to make it easier for people to create and edit web pages. The other benefit of wikis that creating links between pages is all based on creating a wiki link with the name of the new page, that is it. Well now on the meat of this article, enter GTDTiddlyWiki. GTDTiddlyWiki is an implementation of TiddlyWiki that allows you to create and edit a web page that is stored completely in your browser. Yes that is right none of the information is stored on the server, which allows you to edit a TiddlyWiki page even without a connection to a server. In fact it does not work by connecting to a server at all. TiddlyWiki works by using CSS and Javascript to save and manipulate the page contents and then save them in the file for future reference. The GTDTiddlyWiki implements the “Getting Things Done” concept a personal productivity method by David Allen. Have a go for yourself.

Links
GTD Tiddlywiki
TiddlyWiki
Getting Things Done Book

Javascript and PHP

There has been lots of fuss about using Javascript and the XMLHTTPRequest object to communicate between the server and browser page. This communication method allows web pages to get extra data or communicate with the server quicker than refreshing the whole page just to retrieve a small amount of data. JPSPAN is an api to easily connect JavaScript with PHP scripts, have a look here.

Links
JPSPAN Home Page

Another example of XUL

Conor Dowling has sent me a link to another application called 325 Factlog implemented using Mozilla's XUL interface definition language. This application uses RDF to store its database of facts and shows us another way of creating applications. The goal of the “325 Project” is to comprehensively document the year 325 when the west became Christian.

Have a play at http://www.the325project.org

If you are interested in the technical background or knowledge of some of the issues that have been encountered creating this XUL application then have a look at the technical blog.

Links
The 325 Project home page
Technical Blog
Direct Link to the XUL Application (requires either Mozilla or Firefox)

xNAL: Name and Address Standard

I really do think that the number of XML standards out there is completely overwhelming, but on the other hand extremely relevant and required. Although I would like to see a little more convergence between some of the similar standards. xNAL is a OASIS developed standard that is to define the data requirements for address and naming information.

xNal is broken up into two standards:

• xNL: Name language
• xAL: Address language

Links
xNAL Report on OASIS Cover Pages

XForms Beta Release

Mozilla Foundation has announced the beta release of XForms 1.0 Recommendation. XForms is a new XML standard being developed by the members of the Mozilla Foundation to define web forms that are mode extensible and greater functionality than the current HTML forms.

Links

XForms 1.0 Beta Press Release
Mozilla XForms Home Page
W3C XForms 1.0 Recommendation
W3C XForms Test Suite

JSON: JavaScript Object Notation - Lightweight Data Inter-change Format

I have been given a link to JSON (JavaScript Object Notation), which is a project to create a data interchange format that is as language independent as XML but not as heavy in the bandwidth and parsing requirements. JSON is based on the JavaScript Programming Language and has bindings in the following programming languages:

• C
• C#
• Java
• JavaScript
• Objective CAML
• PHP
• ML
• Ruby

Here is a sample JSON message:

{
"glossary": {
"title": "example glossary",
"GlossDiv": {
"title": "S",
"GlossList": [{
"ID": "SGML",
"SortAs": "SGML",
"GlossTerm": "Standard Generalized Markup Language",
"Acronym": "SGML",
"Abbrev": "ISO 8879:1986",
"GlossDef": "A meta-markup language, used to create markup languages such as DocBook.",
"GlossSeeAlso": ["GML", "XML", "markup"]}]}}}

The same text expressed as XML:

<!DOCTYPE glossary PUBLIC "-//OASIS//DTD DocBook V3.1//EN">
<glossary><title>example glossary</title>
<GlossDiv><title>S</title>
<GlossList>
<GlossEntry ID="SGML" SortAs="SGML">
<GlossTerm>Standard Generalized Markup Language</GlossTerm>
<Acronym>SGML</Acronym>
<Abbrev>ISO 8879:1986</Abbrev>
<GlossDef>
<para>A meta-markup language, used to create markup languages such as DocBook.</para>
<GlossSeeAlso OtherTerm="GML">
<GlossSeeAlso OtherTerm="XML">
</GlossDef>
<GlossSee OtherTerm="markup">
</GlossEntry>
...
</GlossList>
</GlossDiv>
</glossary>

Links
JSON Home Page

BPMI to Start Work on BPXL

BPMI.org is planning a Think Tank meeting in Miami on March 1st 2005 to talk about the future of BPEL. BPMI is planning to start work on a new standard called Business Process eXtension Layers (BPXL) that will help exchange information between process modeling tools and process management engines.

The original BPEL standard was developed through a OASIS Technical Committee. However BPMI see that this process would be too slow to develop the new standards and are therefore endeavoring to manage the new standards development themselves outside the confines of OASIS.

Links
Computer Business Review Article
BPMI.org
BPMI Think Tank Announcement

WS-Reliability 1.1 Standard Approved

OASIS has announced that their WS-Reliability 1.1 standard for reliable messaging using web services has been approved by is members as an OASIS Standard. Here is a description of WS-Reliability 1.1 from the standard document itself:

“WS-Reliability is a SOAP-based ([SOAP 1.1] and [SOAP 1.2 Part 1]) specification that fulfills reliable messaging requirements critical to some applications of Web Services. SOAP over HTTP [RFC2616] is not sufficient when an application-level messaging protocol must also guarantee some level of reliability and security. This specification defines reliability in the context of current Web Services standards. This specification has been designed for use in combination with other complementary protocols (see Section 1.4) and builds on previous experiences (e.g., ebXML Message Service [ebMS].)”

See the links below to read more about WS-Reliability or to download the standard and associated schema files.

Links
WS-Reliability Article
WS-Reliability 1.1 Standard
OASIS Web Services Reliable Messaging Technical Committee (WSRM TM)

Actual Example of a “Rich” Web Client

One of my friends and colleagues (Matthew Phillips) has sent me a link to a page about an application that has been written using Mozilla's XUL. As I have previously written XUL is the XML markup language used by the Mozilla-based Browsers (Mozilla and Firefox) to render the user interface. This site talks about using XUL to develop a “rich” web application to search for products on Amazon.com. You can run the rich client as either an extension or online (like a web page). I believe this type of rich client is the future for all web-based applications that require a centrally managed application repository while retaining highly interactive client-side capabilities.

Have a go for yourself. (You will need to install Firefox or Mozilla)

Links
MAB Rich Client
Download Firefox
Download Mozilla

Metadata: Dublin Core Metadata Initiative

Metadata is poorly used on the Internet and most intranets to describe objects (normally web pages). I have used and accessed metadata for numerous implementations and systems, and have come across elements that contained the name “DC” before an element like “title” or “description” and have never really thought about what it was for or what it stood for. While researching information about a recent article about DITA I found a reference to the Dublin Core Metadata Initiative (DCMI). The DCMI is trying to provide better metadata standards for ALL resources that are interoperable. This means the “DC” actually stands for “Dublin Core” and means that the element conforms to the Dublin Core standard. For a complete Dublin Core Definition see the following example for the Dublin Core home page:

<?xml version=“1.0”?>
<rdf:RDF xmlns:rdf=“http://www.w3.org/1999/02/22-rdf-syntax-ns#” xmlns:dc=“http://purl.org/dc/elements/1.1/”>
<rdf:Description rdf:about=“http://dublincore.org/”>
<dc:title>Dublin Core Metadata Initiative (DCMI) Home Page</dc:title>
<dc:description>The Dublin Core Metadata Initiative is an open forum engaged in the development of interoperable online metadata standards that support a broad range of purposes and business models. DCMI's activities include consensus-driven working groups, global conferences and workshops, standards liaison, and educational efforts to promote widespread acceptance of metadata standards and practices.</dc:description>
<dc:date>2004-10-05</dc:date>
<dc:format>text/html</dc:format>
<dc:language>en</dc:language>
<dc:contributor>Dublin Core Metadata Initiative</dc:contributor>
</rdf:Description>
</rdf:RDF>

This example is an XML/RDF instantiation of the Dublin Core definition. As you can see Dublin Core defines the “Description”, “Date”, “Format”, etc. elements that describe the contents of the document. By using this worldwide accepted standard you can describe a document so that any search engine, Internet spider or content management system can use this metadata.

Defining metadata using Dublin Core is really only one part of the problem, the other problem is that the systems still do not “understand” the meaning of the data defined in the metadata. This understanding is where the concepts behind the semantic web can be applied, although the semantic web concept is still developing therefore is not quite ready for prime time use.

Links
Dublin Core Metadata Initiative Website
DCMI Documentation

Technical Documentation - XML Definitions

I have known about DocBook for quite some time as an XML schema to develop technical (or any) documentation. Recently I have stumbled over the DITA (Darwin Information Typing Architecture Article) standard, which is an alternative technical documentation definition. Unlike DocBook, DITA is not just for defining a single technical documents but more about the definition of topics that are later linked together. This creates for greater perceived flexibility, however I have not yet tried DITA so I can not comment in detail. For a good overview have a read through the Introduction to DITA article. I also found the FAQ found inside the DITA download to also help in my understanding.

Links
DocBook Website
Introduction to the Darwin Information Typing Architecture Article
Download DITA

“Rich” Web Clients

I am a keen advocate for “Rich” internet clients. By “Rich” I do not mean Java Applets, but web pages that provide a more complete user experience without having to request new pages from the server whenever the user clicks a button or link. There are few different approaches that provide this type of functionality including XUL (Mozilla Platform), Macromedia Flex and traditional JavaScript and DOM. Many web developers do not exploit the powerful capabilities provided with JavaScript and DOM. A recent example of this type of “Rich” client is Google's new Gmail capability that uses JavaScript and DOM to create a very quick and snappy user interface. By using JavaScript Google has been able to only send the basic requests and data transfers rather sending whole page changes back to the client. This of course takes quite a lot more client-side JavaScript development but does create a very powerful user interface. If you are interested in more information about the Gmail Agent API have a look here.

The Mozilla Platform is a more developer friendly approach as you define your application's interface in XUL and then link it together with JavaScript. However this approach requires your users to use a Mozilla-based browser (Mozilla or Firefox).

For the most interactive “Rich” client experience that will work on any browser Macromedia Flex is your product. Flex is based around a similar concept of Mozilla by defining your interface using XML documents and then linking it together with a scripting languages (Macromedia ActionScript). The difference is that Flex's client is built inside the Macromedia Flash player (available on almost every platform and browser), which allows for highly dynamic and pretty user interfaces. Flex is deployed using a J2EE server and can communicate back to the server by making Web Service calls or exchanging Java objects.

Gmail's approach works on all browsers but is technically much harder to implement. Mozilla's approach works well but requires a Mozilla-based browser. Macromedia's approach provides the best of both worlds but is not free and requires extensive server support. So if you are a web application developer please consider one of these options and provide a more advanced experience for your users.

Links
Mozilla Web Site
Firefox Site
Gmail Agent API Article
Macromedia Flex Site

Enterprise Agility: ESB, SOA, Web Services

Web Services Pipeline has another great article about web services by Fred Cummins. This article talks about enterprise agility and how Enterprise Services Bus (ESB), Service Oriented Architectures (SOA) and Web Services can help create an agile organisation. The article is a must read for anyone interested in the way web services are heading as it identifies a good list of requirements to support an Agile environment, or really any large web services deployment. The article also goes onto to concisely identify the middleware requirements and the components that create a complete enterprise service environment.

Links
Web Services Pipeline Article

NewsTicker: RSS Feed Ticker

NewsTicker is a great little program that displays you RSS feeds in a ticker at the top of the screen (see the screenshot to the right - click for larger version). It is a pretty simple program, which I like. Some of the great features it has are:

• Change the ticker font.
• Choose the colour for the font and background.
• Change the speed of the ticker's scroll speed.
• Can change the level of display (above/below other windows or as a normal window).

There are a few features that I would like to see in future versions:

• Ability to remove RSS feeds once they are clicked (read) or using a modifier key to mark as read without opening the page.
• Display the contents of the article (from the RSS feed) when you mouse over an entry. This could be displayed in a popup window so that it does not get in the way.
• Ability to customise the colour of the link once read rather than just using a darker colour.
• Drag RSS feed URLs to the ticker to add new feeds.
• Have the ticker disappear when there is no news to display.

So now you must think that I hate this program, not correct. I believe that this is quite a good program and I am surprised that we have not seen other ticker style RSS feeds with the minor additions identified above I believe that this will be a great program to monitor news and other feed-based information.

Links
NewsTicker Home Page
VersionTracker NewsTicker Page

WSRP 1.0 Primer Released

I know I have not blogged much lately about the web service standards but I have been pretty busy at work so have not wanted to do research in this area at home. However one important document has been released by OASIS about Web Services for Remote Portlets (WSRP).

“Web Services for Remote Portlets (WSRP) is a web services protocol for aggregating content and interactive web applications from remote sources.” - WSRP 1.0 Primer

This document is a primer to describing the concepts and implementation behind the WSRP standard for users who want to learn how to use this standard. The document is a public release and the WSRP team is requesting comments by the 14th October 2004.

Links
WSRP Primer 1.0 (PDF)
OASIS WSRP Technical Committee Page

FOAF: Friend Of A Friend Vocabulary

The Friend Of A Friend (FOAF) project is aiming at creating a language that uses both XML and RDF to define a language that can describe the relationships between people and resources in a machine readable format. This works by individuals defining their own FOAF definitions and the exposing the URL of this FOAF file so that others can access and use the data. Exposing the FOAF files allows for a FOAF aggregator to consume multiple FOAF files from multiple individuals and by merging the data can define a complete community of interest.

For a basic description of how FOAF works have a read of Edd Dumbill's article on the IBM DeveloperWorks site. For more detail go to the FOAF home page.

Links:
FOAF home page
DeveloperWorks Introduction to FOAF article

Server-side RSS Feed Aggregator

While going through the logs of this site I noticed that someone is connecting up to the RSS feeds produced by a client called FeedOnFeeds. Not knowing this client I decided to find out what it was, to my surprise I found out that it was a RSS reader that ran as a piece of server software rather than a desktop application. The idea behind having a server-side application is that you can use a single RSS reader even if you use multiple computers or work in different locations. By only having one client means that you know exactly which feeds you have and have not read, an idea that has great appeal for me. I have not yet installed the software so I can not comment on the use of the software and how well it works but once I have it up and running I will put up another blog.

Links
FeedOnFeeds RSS Server-side Aggregator Website

Thinlet: Java GUI toolkit

Thinlet is a Java GUI that replaces the now standard Java Swing graphics library. Thinlet provides a GUI environment that is specified by properties in an XML file separating the business logic layer from the graphical layer of your application. The XML format used is the standardised XUL (XML User interface Language) which is also used to display all graphical elements in Mozilla and Firefox browsers.

Thinlet follows the current trend to define user interfaces with XML. It has been reported that Microsoft will also be following this model in their new operating system (codenamed Longhorn), which is using XAML to define application user interfaces.

Links
Thinlet Home Page

Gluecode and Geromino

Gluecode who produces Portal and Business Process products has announced its support of the Open Source Geronimo project. Geronimo is an Apache Foundation project working towards releasing an open source J2EE container to rival the large J2EE providers like IBM and BEA.

Links
Apache Geronimo
Gluecode
Gluecode Products Page
Article on Gluecodes support to Geronimo

Receive RSS feeds by email

If you would like to receive your RSS information via email instead of using your own RSS reader then you can use ERA (Email RSS Aggregator) available at http://era.indecorous.com/

If you are interest to know what the emails look like here is a sample email for the Slashdot RSS feed:

-- START OF EMAIL --

<http://slashdot.org/>

Plans for International Space Station Cut Back
<http://slashdot.org/article.pl?sid=04/07/23/230240>

Sajma writes “Reuters is reporting: NASA and its space partners on Friday
approved a scaled-down International Space Station with fewer astronauts
and less science so the United States can meet a 2010 deadline for ending
shuttle flights, a top NASA official said. Space agencies in Russia,
Europe, Canada and Japan gave unanimous approval to a NASA plan that
means the orbiting platform, now about half completed, will never become
the beehive of scientific and commercial research once envisaged.”

First seen: 2004-07-24 14:00:00
Last updated: 2004-07-24 14:00:00

---

History Of Doom Movie Debuts
<http://slashdot.org/article.pl?sid=04/07/24/0340252>

Rogerpq3 writes “A G4/Tech TV feature on DOOM 3, offering a history of
the DOOM franchise in the words of the folks at id Software. The clip can
be found on the air on The DOOM Franchise, episode 310 of a series called
'Icons.' The piece offers clips from DOOM 3 and other games, interview
footage, analysis and more, and for those without access to the program.
You can download the movie at: 3DGamers, Doom3HQ, Doom3.de, Doom3maps.de,
and FileShack. It's really worth the download for any Doom and id fans
out there. (Thanks: BluesNews)”

First seen: 2004-07-24 15:00:00
Last updated: 2004-07-24 15:00:00

---

Peter Gabriel: Digital Music Downloading's Future
<http://slashdot.org/article.pl?sid=04/07/24/1336208>

securitas writes “CNN International's Becky Anderson interviews musician
and OD2 online music service co-founder Peter Gabriel about the future of
digital music downloads. The interview covers Gabriel's motivations in
starting OD2, how technology has changed the music industry business
model in the favor of artists and away from the big record labels, and
where the small, independent artist fits in. Gabriel's words have weight
because of his insights as both a musician/artist and a businessman who
guided a digital music on demand distribution (OD2) and download service
to success.”

First seen: 2004-07-24 16:00:00
Last updated: 2004-07-24 16:00:00

---

Marian The Robot Librarian
<http://slashdot.org/article.pl?sid=04/07/24/145208>

nusratt writes “BBC reports on robotics researchers in Spain, who have
developed a prototype which can retrieve books from library shelves while
patrons are present. 'When it receives a request for a book, its voice
recognition software matches the titles with the book's classification
code to identify which bookshelf stack to go to. The robot navigates its
way to the bookshelf, using its infrared and laser guidance system, and
scans books within a four-metre radius. Once the book is located, it has
to grasp it and take it off the bookshelf, which is not a simple as it
might seem. For this, the team had to develop special fingertips like
nails, with one nail longer than the other. 'For me that was the hardest
part. All the other things were current state of the art technology,'
said Professor Pobil.' The article also discusses using robots to assist
in digitizing library materials.”

First seen: 2004-07-24 17:00:00
Last updated: 2004-07-24 17:00:00

---

Are You Annoying?
<http://slashdot.org/article.pl?sid=04/07/23/2129250>

cweditor writes “This Computerworld article looks at some habits of
people in general and IT pros in particular that can drive co-workers
crazy.”

First seen: 2004-07-24 18:00:00
Last updated: 2004-07-24 18:00:00

----------------------------------------------------------------------------
This e-mail was sent by era, the e-mail RSS aggregator, using content from
<http://slashdot.org/slashdot.rss>
The content is subject to any copyright restrictions expressed in the file at the above URL.

Comments, queries, and complaints should be sent to <era.info@indecorous.com>. More information can be found at <http://era.indecorous.com/>.

-- END OF EMAIL --

In using the service you have to first send an email requesting an account and then subscribe to the RSS feeds you would like delivered, it is possible to both of these in the same email. After sending any email with commands a response email is generated to confirm the command which you need to remove a string and respond to, which I found a little annoying as you have to do this ever time you send a command.

The system works well and provides a clean and fairly easy mechanism to receive RSS feeds via email. The makers are also working on a web front end so that you can set up your subscription via an easy to use web page but this has not been released yet.

Links
Email RSS Aggregator

Web Standards coming to a desktop near you.

Web Standards like JavaScript, DOM, HTML, CSS, XML are starting to become more common place in desktop applications. Mozilla already uses XML to define its GUI using their XUL markup language and then use JavaScript as the glue. Apple has also previewed their new Dashboard capability that will allow developers to create widgets that are quickly displayed or hidden by entering a hotkey shortcut. These new capabilities are becoming more and more prolific and signifies a paradigm shift of using these standards in desktop-based application development rather than relegating their use in web based applications using standard client-server architectural designs.

The reason that these standards are becoming more prolific for desktop applications is because of their flexibility, ease of use and platform independence. As more and more applications are becoming network enabled it is a logical step for these web-based applications to be run directly on the desktop. These standards are also very powerful and could be used to develop applications that run both as web-based applications accessed by a standard Internet browser or as fully fledged desktop applications.

The common uses for these standards include:
- Use of XML to define and build GUI components
- Using javascript and other scripting languages as the runtime glue to build the application
- providing lightweight user interfaces that provide a quick access to information that can also be skinned so that users can select the overall look and feel of their application.

I foresee that over the next couple of years you will see a greater number of these localised web applications appearing that support and use web-based standards to provide standalone or network centric applications.

ActiveBPEL becomes opensource

Active Endpoints has announced that its BPEL runtime engine has been released as open source. For more information read the press release or the article on TheServerSide.com site.

Links
ActiveBPEL Engine
Active Endpoints Press Release
TheServerSide.com Article

An Overview of Business Process Management (BPM)

InfoWorld has done it again by providing yet again a very interesting article about Business Process Management (BPM) by Eric Knorr. This article talks mainly about the use of the BPEL standard in supporting BPM but it does do a very good job of defining the overall problem and context of BPM and how BPEL supports business modeling. The article also concludes with a very brief (and I mean short) look at some of the other standards also being developed. What is still required is a good comparison between each of the competing standards, there is some work that I am conducting that should provide this outcome.

Links
Go to InfoWorld Article
Email Eric Knorr

Oracle ventures into Web Services, SOA, BPEL and J2EE

Oracle has been involved in J2EE application servers for quite some time now with their Oracle Application Server but with their newest release they are surfing the web services and Service Oriented Architecture (SOA) wave. The newest version of JDeveloper supports full Model-View-Controller (MVC) implementations using a selection of technologies/standards including Struts, Java Server Faces (JSF), Enterprise Java Beans (EJB), Java Server Pages (JSP) to develop your applications. To support SOA Oracle is supporting full development of Web Services and now includes BPEL support for services process execution modelling and execution via the recently acquired Oracle BPEL Process Manager product. All these technologies can be combined and developed using the graphical JDeveloper environment that also allows the developer to view the underlying XML structures. For more information about Oracle's Web Services support and the components supporting SOA, BPEL and MVC development go to this page.

Links
Oracle Web Services Page
Oracle Application Server
Oracle BPEL Process Manager product
JDeveloper

The speed of Web Services

I have implemented a number of web services projects and many of my critics have always commented that “web services are slow”. Well this is true, however there is a big “but” in this statement. Web services are slow when you compare them to such protocols like RMI (Remote Method Invocation) where you can transfer serialised objects between program components using tightly coupled, well known interfaces but this is not the area where web services should be used.

Web services have been developed for environments where you are unsure about the environment (including software, hardware, operating systems, processes and designs). Web services excel when you require information to be transfered from one system to another but do not own or know the details of the other system. These uncontrolled environments is where web services excel and RMI fails. This interoperability ease is why web services have done so well and with the Basic Profiles being developed by WS-I further improves interoperability between web service implementations.

Back to the original statement of why web services are slow. Web services are a very fat protocol as dealing with XML documents is a very CPU intensive process and sending XML documents is not an optimal solution because of the amount of data required to be transferred, for these reasons web services are slow. It is for these same reasons why web services are not the solution for everything, they do however provide a very useful capability when interconnecting disparate systems or developing a very dynamic and unstructured communication environment.

For a interesting discussion on the speed of web services have a read of Jeff Schneider's Blog on “The Suckline” where he discussed that web services provide the lowest common denominator when it comes to system interoperability.

BPEL - Business Process Execution Language

BPEL is still in the press and news feeds as the next new thing in the Web Services space. BPEL is a definition language that uses XML to define business processes at a level where business analysts can define an enterprises processes but in a structure that can be used to execute this defined process and coordinate web service activation.

A new article at InfoWorld talks about the BPEL standard. But BPEL is not the only fish in the sea with WSCI (Web Services Choreography Interface) and WSCL (Web Services Conversation Language) as competing standards under development by the W3C standards organisation. BPEL is being developed under the OASIS standards organisation and is supported by some of the larger application and middleware vendors like IBM, BEA and Microsoft. To stop the W3C and OASIS fighting about these standards Oracle and a few other vendors are starting work to coordinate between these standards efforts. The idea is for BPEL to focus on the internal processes while WSCL on external processes so that these two languages are complimentary. Time will tell as to which standard becomes dominant.

Service Oriented Architecture Presentation

The Pennsylvania State University has produced a very good powerpoint presentation outlining what Service Oriented Architecture (SOA) is, its relationship with Business Process Management (BPM) and where these concepts have evolved from.

Download Presentation

BPML and BPEL4WS

The Business Process Modeling Initiative (BPMI) is an interesting organisation responsible for the development of three specifications, BPML (Business Processes Modeling Language), BPMN (Business Processing Modeling Notation), BPQL (Business Process Query Language). These specifications are to be used for the design and development of business processes. If you are familiar with BPEL then BPML is a superset of BPEL, for more details on the interelation then read this very short article. Also of interest is that the BPMI site has a very good list of articles about the Business Process modelling space and can be found here.

XML Routers

A company called Sarvega is started to sell XML Routers that allow for XML content to be routed around a network. This allow messages like web services SOAP messages to be routed around the network based on their content. Sarvega also produce an XML Firewall to create a security layer for web services like normal firewall secure IP based networks.

Go to CNET News article for more information.

The next wave for application development

The first applications were only run on large mainframes with a limited number of users normally running batch processing. This was then replaced with the introduction of personal computers and the development of applications to run on these operating on local files. Soon applications were started to be developed that still ran locally but accessed remote databases and networked file repositories. From here we have now a trend in the development of web-based clients (eg. HTML in a browser) to deliver applications to the user, otherwise known as thin clients.

More and more you are seeing applications being developed using thin clients as the user interfaces, however there are many issues with the use of thin clients to provide dynamic or interactive applications for users. Now there is a new trend gathering momentum to use XML-based rich clients to develop applications.

Coach Wei has produced a short article on the Computerworld about XML Rich clients and this new trend.

Go to Article

Business Process Execution Language (BPEL)

Many organisations and vendors are now providing external interfaces to applications via web services. These web services allow external applications to access the services or interact with the internal capabilities of the application. This simple service architecture allows for the basic application interaction and integration between systems, however these simple service framework does not support coordination to support workflows where a series of web services need to be called to support a single business process.

A new standard currently under development is the Business Process Execution Language (BPEL) that defines these business processes. An extension to BPEL is BPEL for Web Service (BPEL4WS) that defines how BPEL can be used to define the required workflows when using web services.

IBM has released an article that describes the use of BPEL4WS in IBM WebSphere J2EE Server. The BPEL standard can be found in a number of locations on the Internet (go here).

For a List of the different web services both IBM and Microsoft provide a list of standards:
Go to IBM Web Services Standards Page
Go to Microsoft's Web Services Standards Page

Why XML has been so successfull

XML was first approved by the W3C as a standard in 1998. Since then the uptake of this standard has been quick and extensive, why is this so?. Peter Abrahams on the IT-Director.com website discusses this topic and describes some of the reasons why XML has had such a great uptake. Peter discusses that the real reason that XML has done well is not because of superior technical capabilities but more about its benefits to business and commercial industry.

Go to full article

New Scientist Magazine RSS Feed

Probably the best science and technology magazines on the market is New Scientist. Well after sending an email suggesting that they should provide their news as an RSS feed (XML Syndication) I received a response saying that they already had one that was not well publicised. For all you science geeks out there hese is the RSS feed:

http://www.newscientist.com/syndication/news.rdf