Mailing List Archive
tlug.jp Mailing List tlug archive tlug Mailing List Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]Re: [tlug] OT: XML?
- Date: Tue, 7 May 2002 00:23:05 -0600
- From: Matt Gushee <mgushee@example.com>
- Subject: Re: [tlug] OT: XML?
- References: <F22lUJaZEG17SEChHz70000d778@example.com>
- User-agent: Mutt/1.3.27i
On Tue, May 07, 2002 at 01:03:06PM +0900, Jean-Christian Imbeault wrote: > > I've been reading up on XML this weekend (sparsely I admiit it) and was > interested in using it for for my web pages. However I couldn't find any > information on why using XML on plain web pages would be a Good Thing(tm). I'm not so sure it would. I've been working with markup languages (first SGML, then XML) for several years, and have been working as an XML consultant and trainer for the last two. One thing that has become very apparent to me is that most users of XML are large corporations. In fact, to make a gross generalization, there is a tremendous knowledge gap between large and small businesses. For many corporate IT types, XML is already old news ... whereas last month I went to a meeting of small business owners with web sites. I explained what I do in what I thought were very simple terms, but I got a lot of blank stares. Maybe 1 of the ~10 people in the room actually understood what I was talking about. Well, there's a reason for that ... I'm currently trying to find out if it's a good reason or not, because I would really rather be working with small businesses if I could. The trouble is that using XML really effectively takes a lot of design work, or data modeling to be precise. If you're just going to throw something together, there isn't much point. Now, what I would like to know is, how complex do the requirements have to be before the overhead of a customized XML solution becomes worthwhile? As far as I know there hasn't been any serious research on that point ... though there are similar questions you could ask (and that maybe have been asked and answered) about other techniques and methodologies ... e.g., at what point does a formal methodology such as the Rational Unified Process become cost-effective for software development? Or ... is there a way to make XML products and services generic enough that individuals and small businesses can afford to use XML, and still get real benefits out of it? Now, I'm not talking about products like OpenOffice and MS Office, which use XML file formats, because in those cases what gets saved is just formatting instructions, just as word processor files have always been. They aren't attempting to encode semantics and structure, just using XML as an open, easily processable file format (well in the MS case it's obfuscated to the hilt, but still nominally open). What's special about XML (or SGML or LaTeX, but very few people care about them these days) is that it allows you to create self-describing documents/datasets. But there's a small problem with that, which is that it takes a lot of work to ensure that the descriptions you are encoding are (a) appropriate to your content and (b) make sense to whoever you are exchanging information with (or whatever system is processing that information). In the abstract, there should be a continuum between the totally customized solutions that some corporations are spending millions of bucks on, and the invisible, totally generic (and not very interesting) XMLized products that you are probably already using. In the abstract, there should be a way to implement XML solutions with just a little bit of customization, that will help make information just a little more manageable for small businesses, or make small websites just a little easier to build and maintain. To my knowledge, nobody has yet found a business model that would make that a reality. > I could think of a few useult things such as making a web site more easily > searchable but other than that ... That's the theory. Unfortunately, that falls into the large category of largely unexploited uses of XML. That's partly because most of the effort put into XML development by most vendors has focused on exchanging data between applications or databases, and partly because effective searching depends on natural language, and once again the techies have grossly underestimated the complexity of natural language. > Can anyone point me to any ressource that explain *why* and *how* XML can > be put to good use on a rather simple web site? I don't know what you mean by simple, but broadly speaking I would say: * XML is very useful for complex, semi-structured documents or datasets (e.g., technical manuals and knowledge bases, insurance claims, medical or police records) * XML is useful for highly structured datasets such as financial accounts, but relational databases and compact, grid-structured exchanged formats are more useful. * XML is not very useful for documents where content is secondary to presentation, such as many marketing documents. * There are an awful lot of borderline cases. So I would say, if your goal is to learn XML, by all means use it for your website. But if the site is as simple as you say it is, you might do just as well by sticking with plain ol' HTML. -- Matt Gushee Englewood, Colorado, USA mgushee@example.com http://www.havenrock.com/
- References:
- [tlug] OT: XML?
- From: Jean-Christian Imbeault
Home | Main Index | Thread Index
- Prev by Date: [tlug] exec -a
- Next by Date: Re: [tlug] xfs, jfs, reiser
- Previous by thread: [tlug] OT: XML?
- Next by thread: Re: [tlug] OT: XML?
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links