<warning>Sorry!</warning>
<body>
My personal feeling on this is that it's about perception and honesty. We in the tech industry all know Microsoft is only pushing the XML buzzword because it will sell more copies; it sounds like a feature the under-informed ordinary users are afraid to be caught without, because they might fall behind the competition without it. There's no integrity behind pushing software based on features you make people think they need when they might not. We're not used car salesmen, are we?! ;)
The statement that XML is XML is as true as saying XHTML Transitional === XHTML Strict. It is easier to work with XML than binary for sure, but when the XML doesn't make sense, or if it isn't using XML (a META language) to describe the content, then it's not using XML to solve a problem. It's using it for the sake of saying they're using it. Technology for technology's sake. That's why we have such a usability problem in this industry -- users aren't considered, instead we do whatever we want and whatever we tell them is cool, and we let marketing worry about making users believe it has relevance to them. No wonder software marketing is so heavily buzzword-laden.
Here is some perfectly valid XML:
<body>
<font size="5"><b><i>The end is near!</i></b></font><br />
<font size="2">By Joe Smith</font><br />
<p></p>
<font size="3"><b>W</b><font size="1">INNIPEG, MB - FARMERS HAVE BEEN</font><br />
<font size="1">reporting that hundreds of chickens and buffalo are<br />
turning up dead, due to a mysterious bug that crept<br />
out of nowhere.</font>
<p></p>
</body>
This will parse just fine into a completely meaningless data structure. You can transform it into something else, but it doesn't look like the type of XML you'd want to rely on -- is the format going to stay the same? Is it always changing? What's the meaning of each tag?
When the meaning of each tag is visually driven and does not enhance the document, it detracts from it. If you asked me what I'd recommend to get data out of a structure like that, I'd say screen scraping using Perl is probably your best bet. There are plenty of scraping libraries around, and regexes are familiar and powerful enough for many developers, and they're at least as easy as developing XSLT transformations. XSLT can be simple for simple documents, but crap like what we see above makes XSLT like pulling teeth. Image a document longer than 12 lines that was like that (you don't have to imagine, just click view source on any large web site ;))!
The point is, Microsoft is not pro-developer, and I am very skeptical (without having read that article yet, so I'm not judging from that) that their XML output will be anything but a headache to work with. Of course, somebody out there will task themselves with the creation of a nice MSWordXML2DocBook.xsl, and we'll all benefit from that (duplication of effort is just plain wrong after all).
I just pity the fool who would go through that much trouble for such little gain -- and I wouldn't be surprised if a security update broke the compatibility just enough to screw this guy's work up once in a while.
</body>
</rant>
Cheers,
Lux
On Thursday, December 19, 2002, at 07:23 PM, Chris Harrington wrote:
Hi,
I personally found this article totally off-target. XML is XML. I don't care if they use a Schema or not and I don't care if it is "proprietary" or not. An XML file format for Office is going to revolutionize workflow applications of all sorts - as anyone who has struggled with developing Office in the pre-XML world knows quite well. I seriously doubt that it will take a "very, very skilled XML programmer" to make use of this framework. But so what if it does? That will just make me more valuable. Microsoft is not claiming to be making Office open - they are making it more developer-friendly.
Chris
At 10:56 AM 12/20/2002 +1100, you wrote:
Hi all, There was a discussion a few months back about XML, it's openness and the new Office 11.I was maintaining that XML is just a delivery mechanism and the openness comes in the documentation of the schema. It appears that the news services have now noticed that as well. I'll say it again. Just because product 'Z' uses XML that does not make it open, interoperable or even a smart thing to do. XML is being used as a buzz-word for vendors to push products to unsuspecting customers. XML is _not_ the answer to all questions. Here's an interesting article on the not-so-open-Office 11. http://news.com.com/2100-1001-977880.html?tag=lh Cheers, -- jon -- Jon Eaves <[EMAIL PROTECTED]> http://www.eaves.org/jon/ -- http://cms-list.org/ trim your replies for good karma.Chris Harrington Active Interface, Inc. -- http://cms-list.org/ trim your replies for good karma.
-- John Luxford President and Chief Developer ______________________________ SIMIAN systems Driving Web Content Management ______________________________ web : http://www.simian.ca/ email : [EMAIL PROTECTED] phone : 204.946.5955 -- http://cms-list.org/ trim your replies for good karma.
