Re: Reminder, blatant ad

x wrote:
> "JOG" <jog_at_cs.nott.ac.uk> wrote in message
> news:1139452676.340724.51300_at_f14g2000cwb.googlegroups.com...
> > XML is fashionable. As such there is a greater pot of research funding
> > for projects related to it, and gradually but inevitably impressive
> > people become involved. They establish entrenched research interests,
> > and the cycle self-perpetuates until something more fashionable
> > arrives. Someone should coin the definition of IT evolution as
> > "survival of the trendiest".
>
> It is like the emperor's new clothes.

I've used that phrase for 1NF before, but, of course, it is just spin. My take on XML is that it got at least two important things right. 1) It has a data model that is more useful in that it can be used to model data for any interface, not just a database interface and 2) it doesn't have any unvalue values, such as the SQL NULL.

There is plenty I don't like including 1) Performance can still kill a project and it seems a dog of a format for exchanging data. 2) It isn't particularly easy to work with in a programming language. It is much easier it is to manipulate the data with JSON, for example. 3) the XML schema drives me nuts and I don't even know how to put that into words. I think Marshall got at one side where you can think of it as untyped without enforcing a schema, but if you wanted to enforce a schema it just looks too painful to do so. I suspect some folks have given up and put constraints and validations in code instead.

But as long as I think of it as an upgrade to csv (comma-quote) format for data exchange, or a way to model data for a query language that is not as restrictive as SQL, it works for me. Cheers! --dawn

P.S. This is still in my blatant ad (for www.tincat-group.com/mewsings in case I didn't mention that ;-) thread and perhaps should move? Received on Thu Feb 09 2006 - 16:13:07 CET