W3C home > Mailing lists > Public > www-talk@w3.org > March to April 2004

Re: complexity (was: Re: XHTML and RDF)

From: Orion Adrian <oadrian@hotmail.com>
Date: Thu, 15 Apr 2004 13:50:01 -0400
To: www-talk@w3.org
Message-ID: <BAY1-F45e9F6EBSotwB00041f88@hotmail.com>

> > Can you give an example of how doing it X's way is going to be Z out of
> > business? You mentioned this, but I fail to see how a construct in a
> > language is going to harm to Z?
>Sure. Company Z is writing software for a device with limited features, on
>a system which has built-in JavaScript support, but little else, and
>company X wants the technology to use XPath and XMLSchema, instead of
>JavaScript, for its dependency tracking.
>There is no way to implement XPath and XMLSchema in the device Z is
>targetting. Thus, if X gets its way, Z cannot use the technology.

This seems to be a very limited example. It also doesn't follow that the 
hardware requirements of a single technology or small set of technologies 
should override the needs of authors and implementers.  While I understand 
this is a problem for Z, there are a lot of specs out there that are 
designed for a much larger market.

It just seems wrong to say to implementers everywhere that the next 2 years 
of your life are being controlled by hardware that has the same lifespan.

Limited resources are a problem with many devices, but they cannot be a 
dictating factor in broader technologies. The problem with small devices is 
that their capabilities differ wildly and change regularly with new models.

Perhaps the approach in general is flawed as well. Why does supporting web 
technologies require such a huge burden on devices. People are trying to 
essentially standardize and replicate work that has already be done. I can't 
think of a single technology that has come out of the W3C that hasn't 
existed elsewhere in a more compact form for a longer period of time. The 
W3C's job is to reinvent the wheel, but they're not doing a very good job if 
everytime they do it they have a particular car model in mind. It doesn't 
server the greater community and eventually people will stray.

Perhaps there are better examples of what you're talking about, but I still 
fail to see how non-dependant specs are affected. They too are overly 
complex to write, read and implement. I have suggestions I've been working 
out in my head for several weeks now that covers both the larger problem and 
it's solution. Given some more time to test implementations of it, I'll post 
it somewhere on the W3C list network.

Orion Adrian

FREE pop-up blocking with the new MSN Toolbar  get it now! 
Received on Thursday, 15 April 2004 13:51:04 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:33:05 UTC