tag:blogger.com,1999:blog-10770855.post1761002580330164497..comments2024-03-28T03:20:57.393-04:00Comments on The Little Calculist: When to use point-free styleDave Hermanhttp://www.blogger.com/profile/00405190527081772997noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-10770855.post-26294356109487995172008-03-22T11:18:00.000-04:002008-03-22T11:18:00.000-04:00Google does better than Wikipedia on this one. Fir...Google does better than Wikipedia on this one. First hit is from the Haskell wiki:<BR/><BR/>http://www.haskell.org/haskellwiki/PointfreeDave Hermanhttps://www.blogger.com/profile/00405190527081772997noreply@blogger.comtag:blogger.com,1999:blog-10770855.post-10654801250323394212008-03-21T23:12:00.000-04:002008-03-21T23:12:00.000-04:00So .. uh .. what _is_ this here "point-free style"...So .. uh .. what _is_ this here "point-free style", anyway? 0.01 seconds of searching on Wikipedia yielded nothing ...offby1https://www.blogger.com/profile/01257566357465364171noreply@blogger.comtag:blogger.com,1999:blog-10770855.post-22890428080720706972008-03-21T19:31:00.000-04:002008-03-21T19:31:00.000-04:00Ye have heard it said in the past: "Magic numbers"...Ye have heard it said in the past: "Magic numbers" must be given names commensurate with their semantic intent.<BR/><BR/>Verily I say unto you: Intermediate values with no semantic intent do not need names.<BR/><BR/>I was talking about this with Tom Conway the other day. He thinks this is the #1 difference between functional and logic programming. Logic programming is centred around the concept of the "logic variable", which means that even the most useless values need to be given variable names.Pseudonymhttps://www.blogger.com/profile/04272326070593532463noreply@blogger.comtag:blogger.com,1999:blog-10770855.post-36684860094883413462008-03-21T15:24:00.000-04:002008-03-21T15:24:00.000-04:00When taking university examinations, I always prog...When taking university examinations, I always programmed in a point-free style. That is, I rarely got any points for my botched answers!<BR/><BR/>-- PaulPaul Stecklerhttps://www.blogger.com/profile/13416750891822431224noreply@blogger.comtag:blogger.com,1999:blog-10770855.post-44643305005158454952008-03-21T13:21:00.000-04:002008-03-21T13:21:00.000-04:00Thanks Dave, this is a clear perspective which res...Thanks Dave, this is a clear perspective which resonates with me immediately. It leaves open differences of personal interpretation of which functions <I>should</I> be viewed as a single data point, which I've never had the pleasure of debating since it was never put in those terms. Of course, for algebraic manipulations you want to push symbols around without thinking about the meaning of intermediate steps, and point-free seems helpful for this.Unknownhttps://www.blogger.com/profile/04205298776211790050noreply@blogger.com