Layers of Abstraction

Chris mentioned the tendency for some API developers to design their technology in a way where everyone things[sic] they are defining the abstraction layer onto which everything else can be plugged in. I happen to fall squarely in this camp. From my perspective this is the vision of XML Everywhere.

Shortly most of the data application developers on Microsoft platforms process will be available as XML. The next version of Office is all about XML file formats, SQL Server can look like an XML data source and store XML, and with XML Web Services even the messages coming over the network from customers and partners will be XML. Similar things are happening on other platforms; Open Office has XML file formats, both Oracle and DB2 can store XML, and XML Web Services are used across various development and operating system platforms. XML will be everywhere and there will have to be abstraction layers to make all these XML data sources look or behave similarly.

Also with the rise of interacting with XML being the primary way to interact with data there probably should be layers of abstraction that enable other data sources to look like XML so as to leverage the wide support for processing XML that already exists in industry and across various toolsets. This is where things like the XML infoset become even more important and why people like Don get excited by the XML it

XML Everywhere. Ready or not, here it comes.


Issues With the W3C Website

One problem I have with the W3C website is that anyone with an email can post their opinions and rants on the website in a manner that imply that they have official imprimatur when in some cases this is far from the truth. An example of this is which at first blush looks like an official description of the XML data model.

Closer observation leads to the disclaimer "Note: this document is an essay, not a specification." and further observation leads one to the top level page which implies that there are a whole bunch of unmaintained and unofficial ramblings on XML hosted on official looking URIs on the W3C's website. Not all of them have a disclaimer such as



Why Use XHTML?

After reading Mark's most recent blogging about XHTML 2.0 and based on some conversations I had yesterday I'm still hard pressed to see why anyone would use XHTML 1.0 let alone move to something like XHTML 2.0. The only reason to use XHTML I've seen so far has been that XHTML makes it easier for others to process your content (e.g. creating Tantek's feed was easy to do using XSLT since he used XHTML) but this is a weird argument because it implies that the only major benefit of using XHTML is to make it easier for people who want to screen scrape your website.

I am very interested in alternate justifications for moving to XHTML that don't boil down to this. So far I have seen none.


Blogging in the Mainstream

Blogging is so mainstream now it is ridiculous and cool at the same time. Who can't say they don't enjoy reading the daily/weekly postings to the InterWeb buy their favorite authors such as Neil Gaiman, William Gibson, and Dave Berry. Unfortunately none of the aformentioned authors seems to have an available RSS feed. Aaaargh.


RSS Bandit Update

OK, I'm impressed with Visual Studio.NET. I just created a fancy shamncy installer for RSS Bandit by following the instructions here without having to write a line of code or do any scripting. I'd have released it by now to folks if not that I seem to be having some sort of multithreading issue that makes the list view control disappear if I scroll through to many feeds at once. Annoying.


Get yourself a News Aggregator and subscribe to my RSSfeed

Disclaimer: The above comments do not represent the thoughts, intentions, plans or strategies of my employer. They are solely my opinion.



Comments are closed.