[Edu-sig] Ditch "self"?

Paul D. Fernhout pdfernhout at kurtz-fernhout.com
Wed Oct 24 15:19:43 CEST 2007


kirby urner wrote:
> Since you've rejoined, 

To be clear, I never left. Just been listening and learning and focusing my
writing on other stuff (even towards a new PataPata version focused mainly
on Jython).

> I promise to keep quiet until you leave, which
> may be never.

I know you posted similar things in your interactions here with Art, but I
personally hope you continue to post your interesting thoughts on edusig,
whether we agree on everything or not or whether another topic comes of here
of immediate interest to me. As I said, I am focusing mainly on other things
right now, but this issue of "self" in declarations was just one I have
wrestled with in the context of trying (and perhaps failing) to make Python
easier to use and learn (and I appreciated Dethe's help in coming up with an
implementation alternative).

By the way, on rereading my post, I realize I should make clear, while I
feel "self" in function declarations may be superfluous as a matter of
opinion (given the typical indentation of the declaration of such functions
inside the scope of a class definition), I like the explicit use of "self"
within the body of a function (unlike the common standard practice in, say,
Java or the language "Self" where it can be frequently omitted); the two
issues are different ones; I should have been clearer on the distinction,
and as I realize now I think your post was likely more focused on the issue
of "self" in the body of a function as opposed to the declaration, and in
that area I think you, Guido, and I (as well as Smalltalk's designers) all
agree on the value of an "explicit" self.

All the best.

--Paul Fernhout


More information about the Edu-sig mailing list