Re: Proposed Design Principles

On Tue, 27 Mar 2007, Maciej Stachowiak wrote:

> On Mar 27, 2007, at 10:55 PM, John Joseph Bachir wrote:

>> How about "optional extensibility is more important than new features".
>> 
>> Instead of introducing new features, introduce a system by which users 
>> may add their own features.
>> 
>> Yes, I dream of syntactic macros in HTML.
>> 
>> But my fantasies aside, I think some note about the value of 
>> extensibility should be included.
>
> I'm not sure I understand this proposal well enough to turn it into a 
> clearly stated and justified design principle. Please help me out. Who 
> is the target of the extensibility? What kind of extensions can they 
> make? What problems does this solve?

One might say that the separation of semantics from presentation several 
years ago was an application of the extensibility principle.

The separation of semantics from presentation not only resulted in a more 
efficient and philosophically sound way to structure web documents; it 
also added a layer of semantic abstraction to the language itself. The 
"class" namespace is a powerful tool that gives web authors arbitrary 
flexibility at a fixed level of abstraction. This was clearly a better 
choice than predefining a fixed number (be it 10 or 100) of classes for 
styling and script reference.

Perhaps a similar abstraction/namespace could solve some present day 
problems, such as (and I am just throwing this out as an example, I 
haven't though of what a great solution would be) the object/plugin mess, 
where every media designer wishes to use their own out-of-spec attributes.

-- 
John Bachir
http://lyceum.ibiblio.org/
http://blog.johnjosephbachir.org/
aim/yim/msn/jabber.org/gtalk: johnjosephbachir
713-494-2704
irc://irc.freenode.net/lyceum

Received on Wednesday, 28 March 2007 07:19:00 UTC