[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Request for feedback on SRFI-126
From: |
Marko Rauhamaa |
Subject: |
Re: Request for feedback on SRFI-126 |
Date: |
Mon, 28 Sep 2015 02:56:59 +0300 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux) |
Panicz Maciej Godek <address@hidden>:
> The best thing that Scheme does for programming is that it promotes
> writing software that can be read, and not only executed.
Really? I think the best thing Scheme does for programming is it brings
you closer to God. Seriously, Lisp in general and Scheme in particular
is an intellectual eye-opener — but I wouldn't call it inherently
readable.
Personally, it seems that too many people get too excited with defining
new syntax. It is understandable because it's just so cool, but in
practice, macros should be defined sparingly and locally. Weird macros
can make Scheme code very difficult to read. Instead, one should get
comfortable with the lambda and go wild with it.
> Hash tables, weak references and threading are just implementation details
> that a programmer usually shouldn't care about. Delimited continuations are
> an academic curiosity, and sockets and regexps are just a specific domain
> (I'm sure you could easily find plenty of others; anyway, they are by no
> means fundamental)
Assoc lists are not abstract enough to serve as a generic map. Note that
Python and Java pretend to provide abstract maps but require you to
specify hash functions anyway.
Nothing wrong with hash tables. Throw in balanced trees as well.
> Not every system provides sockets (my pen&pencil certainly doesn't,
> and although it can easily use multiple threads to evaluate
> s-expressions, it surely doesn't need threads as first-class objects)
All Linux system calls should be available to Scheme on Linux. Don't
fall in the Java trap. Perl and Python give access to system facilities.
So does Guile (for the most part).
> Then create them with extended Scheme's feature set of your liking. Or
> is there anything that stops you from doing that? Like the fact that
> the name of the language that you use isn't Scheme, but Racket, Bigloo
> or Guile? Or the fact that there is no pdf out there on the web whose
> naming scheme is R*RS, that would describe the language that you use?
True, I develop exclusively with Guile. So far, I've generally been very
happy with their choices.
Marko
Re: Request for feedback on SRFI-126, Christopher Allan Webber, 2015/09/29