|
Val's Blog
Lots of stuff for Web 2.0 freaks and Javaholics
|
|
|
Frederik Brooks: "Adding manpower to a late software project makes it later." |
[ Login ] |
|
Since its inception in 1998, XML has gained a lot of popularity and a broad acceptance in the industry and academia. I take as a proof the increasing number of standards and vocabularies based on XML as well as the growing number of software houses and independent software development groups that have adopted XML as the base language for the support files of their products (deployment descriptors, configuration and property files, preferences, etc). Undeniably, the members of the W3C XML Activity, such as Microsoft, BEA, Sun, etc, have managed to erect XML as the de facto standard for interoperability. Of course, some will object that XML-based applications are not always (often?) 100% interoperable because of the legacy stuff each party throws into the mix, but this is another debate. Together, the pervasiveness and the ubiquity of XML have triggered another problem that is serious enough to have warranted the creation of a new working group. The ever increasing use of XML for exchanging data and invoking remote services has had an enormous incidence on the network bandwidth and the processing power of the computers and devices that take part such XML interactions. If we just start to feel the embryonic effects of XML's fame today, this will eventually become more perceivable by the time web services will have truly taken off (if ever) and when XML will have become unavoidable because standard-setting industries will have started shipping XML based products and stopped providing support for alternative technologies. The new XML Binary Characterization (XBC) working group that has been created in 2003 as part of W3C's XML Activity is chartered to create a recommendation for a binary format for XML. It is worth noting that the essence of this new recommendation will be different from the XML-binary Optimized Packaging recommendation created under the Web Services Activity umbrella. This new format shall drastically lower the bandwidth and processing power needs for processing XML communications, but as nothing comes for free, this will happen at the cost of human readability. Some companies, such as Microsoft, have already devised some binary format for XML that they use internally mainly for efficiency purposes. The goal of this new XML standard is not to deprecate current XML standards, but to interoperate with them and with other related technologies in the XML Stack. After careful considerations, the XBC working group has concluded that: 1) Binary XML is needed (see the use cases below); 2) Binary XML is feasible; 3) Binary XML must be produced by the W3C in order to preserve XML interoperability; and 4) Binary XML must integrate with XML. Furthermore, the XBC working group has identified 18 use cases where Binary XML will most surely bring huge benefits:
As we can see, these use cases cover pretty much 80% of the areas where XML is already in use today. The question is: can we do better than what we are already doing? The XBC members tend to agree that conventional XML leaves much room for improvements in the context of the above mentioned use cases. According to W3C's track record, I'm pretty much sure that a majority of people will benefit from whatever comes out of the pipe. In a couple days, I'll try to come up with some concrete example in order to delve a little more into this binary mud.
Other interesting resources:
TrackBacks[0]
Comments[0]
Posted by val on May 6, 2005 4:11:42 PM CEST
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Content © Val | Powered by Pebble 1.9.1 |