Cart
[Home] [By Thread] [By Date] [Recent Entries]
On 31 Aug 2001 10:40:14 -0700, Fuchs, Matthew wrote: > Perhaps because there is often no single "best-practice" - I think I'd > expect you of all people to realize that. Often what is most important is > there be consistent practice - the principle of least surprise, or something > like that. Consistency is often more important than optimality. I think you must have a rather different understanding of "best practices" from mine. My understand of "best practices" isn't "optimal way to do things" but rather "how best to avoid difficulties within the context of a given specification set". In that understanding, best practice typically values consistency as much as (and frequently more than) optimality. Common XML [1], for instance, is a best practices document focused on consistency rather than making optimal use of XML 1.0's features. I had thought this usage of "best practices" was pretty ordinary in computing (things like IETF Best Current Practice documents), but maybe it's rarer than I'd thought. I'd originally written: > > Simple best-practice solutions are fairly easy to come up with, but > > seemingly just as easy to shoot down, suggesting that there > > may never be > > consensus on these issues. [1] - http://simonstl.com/articles/cxmlspec.txt Simon St.Laurent
|
