RE: Joel Spolsky on IE 8 standards mode

I wonder if there is an interesting architectural conclusion to be drawn
for many software systems.  Proposed Good Practice Note for Language
Design: It is good practice to identify the version of the originating
software ( when there is a high likelihood of interoperability issues |
| the specifications are complicated  the specifications implemented are
> x(50) pages ).  

I've seen this issue of needing the exact version come up in
interoperability testing of other complex specs, like the CORBA and Web
Services.  HTTP seems to have gotten by without it, perhaps because
there seems to be well-known subset that is deemed interoperable.  

Cheers,
Dave

> -----Original Message-----
> From: John Cowan [mailto:cowan@ccil.org] 
> Sent: Wednesday, March 26, 2008 10:28 AM
> To: David Orchard
> Cc: www-tag@w3.org
> Subject: Re: Joel Spolsky on IE 8 standards mode
> 
> David Orchard scripsit:
> 
> > Interesting and worthy read @
> > http://www.joelonsoftware.com/items/2008/03/17.html
> 
> Many have pointed out the obvious point that Joel misses:  if 
> IE 8 didn't self-identify as IE, it won't trigger all the 
> IE-specific hacks.
> 
> -- 
> Using RELAX NG compact syntax to        John Cowan <cowan@ccil.org>
> develop schemas is one of the simple    http://www.ccil.org/~cowan
> pleasures in life....
>         --Jeni Tennison                 <cowan@ccil.org>
>

Received on Wednesday, 26 March 2008 17:39:13 UTC