• To: "'Jeff Lowery'" <Jeff.Lowery@c...>,"'Uche Ogbuji'" <uche.ogbuji@f...>,"'Jonathan Robie'" <jonathan.robie@d...>
• Subject: RE: Design as, one hopes, not premature optimization
• From: "Michael Kay" <michael.h.kay@n...>
• Date: Thu, 3 Apr 2003 16:30:21 +0100
• Cc: "'Chiusano Joseph'" <chiusano_joseph@b...>,<xml-dev@l...>
• Importance: Normal
• In-reply-to: <4827EC373E03DB44AC2AB3900430896B02667A@c...>
• Reply-to: <michael.h.kay@n...>

> > the accepted wisdom that premature optimization is evil?

Don't fall into the trap of reading this as "early optimization is
evil", let alone "you can ignore performance during the design stage". 

There is a right time for everything. The right time to select a good
algorithm is early, the right time to add a fast path for a special case
is at the end.

Michael Kay

• Follow-Ups:
  • On the aparent importance of emoticons (was "Design as, one hopes, not premature optimization")
    • From: Uche Ogbuji <uche.ogbuji@f...>
  • RE: Design as, one hopes, not premature optimization
    • From: Jonathan Robie <jonathan.robie@d...>

• References:
  • RE: Design as, one hopes, not premature optimization
    • From: Jeff Lowery <Jeff.Lowery@c...>

• Prev by Date: Re: 3 possible approaches for representing concepts
• Next by Date: Re: Re: ID/IDREF makes XML generation NP-hard
• Previous by thread: Re: Design as, one hopes, not premature optimization
• Next by thread: RE: Design as, one hopes, not premature optimization
• Index(es):
  • Date
  • Thread