On Tuesday 24 September 2002 17:04, Peter A. Cassetta wrote:

[snip]
>
> Another topic that has come up recently has been ease-of-use vs.
> performance.

I think I was the one guilty of that. But I was comparing the performance of 
Tapestry with ASP.NET. All I said was that in comparison, Tapestry 
performance deteriorates badly , but I'm in NO WAY dissatisfied with Tapestry 
performance .. like I said I work mainly for intranets (600-800 users), and 
in this environment Tapestry is more than fast enough.
In fact ... sometimes just for kicks  I show the users the HTML source of 
pages in Tapestry applications and show them the last line .. you know... the 
one that goes like this:
<!-- Render Time: 12ms -->
And they all go "UAU!!" :-)


>
> I think ease-of-use has been Tapestry's main goal from the beginning, and I
> think it remains its real selling point. Sure, tune the performance where
> appropriate, but Tapestry doesn't have to be the fastest framework to be
> successful.


Agreed, and to quote on of my favorite Usenet posts:
"Early optimization generally results in buggier, harder to maintain code that 
isn't any faster because the wrong things have been optimized."


>A few Web sites need serious efficiency (your Googles and
> Amazons), but the vast majority just need to present a good experience to a
> modest number of simultaneous users.

Yes... the problem is that many programmers have delusions of greatness :-)


> Heck, hardware is a whole lot cheaper
> than development man-hours.

I keep earing this everywhere... you guys in the States must make a fortune, I 
must consider emigration :-)


>Nobody is going to adopt Tapestry because it is
> fast (though admittedly a very few might avoid it if it has the reputation
> of being dog-slow). 

Why do you think Microsoft make such a spectacle of the Petshop thing?
People (in general) have this idea that faster == better.


>What is going to spur adoption of Tapestry is the
> development-speed argument. From a development-speed standpoint, the
> framework foundation of Tapestry is unmatched (not to mention the wonderful
> Inspector). 

Actually what drove me to Tapestry wasn't the saved development time, it was 
the sheer beauty of it :-)
I come from a ASP/VB background and I was getting sick and tired of that mess, 
designers constantly asking me "Is it OK to move this part of code to the 
bottom of the page" , "Could you take a look and see if I didn't mess 
anything when I added the picture", "Ohh ,I'm sorry ,that was an include 
file,I thought it was just a comment" ...well you get the picture, and tools 
like dreamweaver only help to some extent
The perfect separation of roles was at let me to Tapestry... the rest of the 
things (and are allot of them) were the iceing on the cake.


>All it really needs now is a set of high-level components to
> clinch the development-speed question and some improved tutorials to help
> newbies get productive with it quickly (I know progress is being made on
> both fronts). If, in the end, it remains slower than some other framework
> at deployment time, then it loses out on a few projects. But if it is also
> the most productive for the developer, then it will win out on many more.


That is true, and I hope that Howard doesn't try to be all things to all 
people, and lose the focus, I must admit that I was a little nervous when I 
'eard that Tapestry 2.3 will have the guts all thorn apart to suport 
integration with "tag based frameworks", but I think Howard is approaching 
this with great caution, making an extensive testing framework to make sure 
that everything stays in place.

Cheers,

Luis Neves



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Tapestry-developer mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/tapestry-developer

Reply via email to