W3C home > Mailing lists > Public > whatwg@whatwg.org > September 2004

[whatwg] RE: Degrading of web applications

From: Didier PH Martin <martind@netfolder.com>
Date: Fri, 10 Sep 2004 09:37:20 -0400
Message-ID: <000101c4973b$4c2def80$c901a8c0@DIDIERHOME>
Hi Chris,

> I think this mention of fancy behaviour is the key. What I am talking
> about by a web application is a *single* page where all server
> interaction occurs through XMLHttpRequest. In this instance all
> functionality of the particular application is provided through
> javascript - a collection of fancy behavious all combined to make the
> application. Browsers that do not support javascript can not use the
> application. As such it means nothing for that web application (in
> this case just one page) to degrade as there is no content to view and
> no functionality available.
> 

right. You may need to degrade for some device also able to support
javascript. For instance, pocketPC connected to the intranet with 802.11b
connections. The screen is smaller but still support javascript and
xmlHttpRequest.

Two scenarios:

a) you create a single rendering language (xml based) produced by different
servers. That language is interpreted differently in different devices.

b) you create a single domain model language (xml based) and let XSLT
perform different transformations on these different devices.

The good point about (a) is the relative low cost of porting to different
devices. The impact though is the poor quality of the porting. Some layouts
are hard to degrade gracefully for different resolutions and capabilities.

The good point about (b) is that you gained maximum adaptability to the
different devices. Allowed other processes to parse and process the data.
Therefore you killed two birds 1) get a web service, 2)transformation of
this data to different devices without bad degradation (let's say a more
controlled degradation). The bad point is the increase development price for
the different transformation templates.

Cheers
Didier PH Martin
Received on Friday, 10 September 2004 06:37:20 UTC

This archive was generated by hypermail 2.3.1 : Monday, 13 April 2015 23:08:19 UTC