[
Lists Home |
Date Index |
Thread Index
]
actually len i think the trick is to keep building simple things that
can interact. the total system can be amazingly complex, but each
component simple, verifiable, and understood.
unix was built on that premise, djikstra's predicate calculus almost
dictates it, it is the heart of my database/application work and
philosophy of how to build very large systems, but most importantly it
is essential to the existence and success of the web - and by extension
this group.
so, in this matter i disagree. lots of simple things are still simple
things, even if their interaction is so complex it can never be fully
understood.
rick
On Wed, 2004-04-07 at 05:32, Bullard, Claude L (Len) wrote:
> Problem is, one simple thing is simple. Two might not be.
> More than that likely aren't.
>
> No one sets out to deliberately design obscure complex
> products unless they are David Lynch.
>
> len
> -----Original Message-----
> From: Jim Rankin [mailto:jimbokun_lists@mac.com]
> Sent: Tuesday, April 06, 2004 2:32 PM
> To: rjm@zenucom.com
> Cc: xml-dev@lists.xml.org DEV
> Subject: Re: [xml-dev] Competing Specifications - A Good or
> Bad Thing?
>
>
>
>
>
> On Apr 5, 2004, at 5:53 PM, Rick Marshall wrote:
>
>
> simple things are good. they can
>
> be understood, implemented, and used in the lifetime of a
> human being.
>
> complex things are of no use because they can't.
>
>
> I just added this to my signatures list.
>
>
> -jimbo
>
>
> Excelsior! XML Marshaller for Cocoa
>
> http://www.homepage.mac.com/jimbokun/Excelsior.html
>
|
