On Sun, Apr 10, 2005 at 01:40:45PM -0700, Andrew Lentvorski wrote:
> 
> On Apr 10, 2005, at 11:28 AM, Lan Barnes wrote:
> 
> >The universality of web interfaces is a myth.
> 
> Yes and no.  There are ways to do set up code to do rendering  
> correctly.  The problem is that some graphic designer decides that  
> something isn't "pretty" and makes it nonportable.  The number of  
> stories from Apple about how the usability folks had to fend off the  
> graphic arts folks about the original Geneva font are legion.  For  
> those who don't know, the original Mac Geneva font is this really  
> bold, blocky, (might even by monospaced) font that is just absolutely  
> legible on low-resolution pixel displays, but ugly as sin.
> 
> >- state is kludged
> 
> There is no state.  That's the problem.  If you need state, you  
> should not be using http.  The sooner the developers get that through  
> their heads, the better.  Use Java or Flash/Director.
> 

I say again, state is kludged. Amazon/Netflix/etc most definitely keep a
light burning for you once you open a transaction. I can remember
aborting an Amazon buy, coming back a week later, and finding the
abortus in my shopping cart.

> >- performance is for-shit
> 
> Latency can be worked around, most times.
> 

So we agree?

> >- simplicity is enforced by lack of tools
> 
> Actually the Macromedia stuff seems to be pretty damn good.  We just  
> never see it because we are open-source bigots. ;)
> 
> >- the bulk of developers and all PHBs think 100% of the world uses IE
> 
> The developers are the bigger problem than the PHB's.  They use  
> Macromedia on Windows rendering into IE and ship.  Since the bugs  
> never cost the developer anything, they have no incentive to do any  
> more.
> 
> Surprisingly, the PHB's are actually starting to get it.  In fact,  
> the most important ones always did.  I'll note that my financial  
> services websites all managed to work across a really big spectrum of  
> browsers.  Sometimes, they forced me to enable Java (grumble ...  
> mumble), but it was generally the correct solution.
> 

That's good to hear.

-- 
Lan Barnes                    [EMAIL PROTECTED]
Linux Guy, SCM Specialist     858-354-0616
-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to