RE: <xsl:stylesheet xmlns...

["Paulo Gaspar" <paulo.gaspar@xxxxxxxxxxxx>]

> -----Original Message-----
> From: owner-xsl-list@xxxxxxxxxxxxxxxx
> [mailto:owner-xsl-list@xxxxxxxxxxxxxxxx]On Behalf Of Chris Bayes
> Sent: Sunday, August 06, 2000 04:57
>
> >> >Besides, that often requires to transfer more data.
> >>
> >> Depends on the application
> >
> >That is why I use "often".
>
> I dispute that. It depends on what you send down the line. Obviously you
> wouldn't send a whole SQL database to a client to do a query. You send the
> results. If the results consist of the minimum amount of data then the
> transform will *always* produce more output than input if you are
> transforming to html (transforming to text less). Add cacheing into the
> equation and client-side transformation will *always* produce less line
> traffic.

You do not know my but I can assure I am not SO dumb. I made my measurements
by only sending the necessary XML data.

There is also the overhead associated with fetching 2 documents (XML, XSL)
instead of one (HTML). In most cases I focused on it only saves traffic if
the browser caches the XSLT.

Now, that being an advantage depends on the nature of the site. It only
works better if the template caching is used often enough.


There are NO absolute rules. One has to analyse case by case.


And there is still the issue of having to work with browsers like Netscape,
Opera, older Internet Explorer versions and so on.

> >
> >> >IMHO, server side transformations rule.
> >>
> >> Are you selling hardware too?
> >
> >I agree that it might be different in an Intranet scenario.
> >
> Slightly but not much.

Much. In an Intranet there is more control over the used browser and there
is not so much compatibility concerns. You also tend to know the use cases
better.

>
> >(Is hardware that expensive these days that I would get a profit form
> >preaching this???
> >=:o)

This was an irony. Noticed the smiley?

> Possibly if you are running a data center. The cost of hardware
> is miniscule
> compared to the cost of support staff.

And of development staff.

> Throw more hardware at it is just not an answer to badly designed
> applications as many people are aware.

Correct, but some people are also aware that throwing more hardware in a
concious and controlled way can be a valid strategy to save development
effort and have faster results.

> I won't even tell you what I think of preachers.

Again, noticed the irony?


Have (more) fun,
Paulo


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list