From: "Sean E. Russell" Date: 2004-02-13T01:19:18+09:00 Subject: Re: Standard libraries --Boundary-02=_8d6KA0tpjDpO6LV Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline (Dave Thomas: there's a question for you in the second paragraph; if you're= =20 not following this thread, would you please at least scan that paragraph? = =20 Thanks!) On Thursday 12 February 2004 09:59, Gavin Sinclair wrote: =2E.. > > Basically, the code duplication that Hiroshi is complaining about is > > required for good documentation, because there's no other way of =2E.. > Good points, but I don't see the light overall. > > Method documentation is neccesary to help people understand the code - > I agree. Without type information it can be hard to understand a > method at a glance. Sorry, I wasn't being clear. I was trying to say that Hiroshi doesn't like= =20 code duplication, but there is no way to avoid the kind of duplication he's= =20 complaining about because there's no other way to extract that information= =20 automatically. Ergo, if you want good documentation, you'll *have* to have= a=20 certain amount of duplication -- although, I wouldn't call it duplication;= =20 I'd call it documenting constraints. Even so, it *is* difficult to keep=20 those constraints in sync. I remember having a discussion with Dave about the possibility of having rd= oc=20 extract constraints from the documentation and generating or running them i= n=20 the form of unit tests. Dave, I'm a bit out of touch with what you're up t= o=20 these days; have you given this any more thought? Also, since I'm a lazy cuss, does anybody know what the consequenses of hav= ing=20 Ruby source files that consist of enormous amounts of comments is? I'm=20 assuming that the parsing out of comments would add a not inconsiderate=20 amount of time to the loading of files, but since that only happens once pe= r=20 run, can it be ignored? I'm just imagining some literate programming styl= e=20 that's 90% comments, and wondering what the overhead would be considering t= he=20 dynamic nature of Ruby. =2D-=20 ### SER =20 ### Deutsch|Esperanto|Francaise|Linux|XML|Java|Ruby|Aikido ### http://www.germane-software.com/~ser jabber.com:ser ICQ:83578737=20 ### GPG: http://www.germane-software.com/~ser/Security/ser_public.gpg --Boundary-02=_8d6KA0tpjDpO6LV Content-Type: application/pgp-signature Content-Description: signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQBAK6d8P0KxygnleI8RAhXqAKCvUXQZKbvWcciD5Bwhg+5OGNytegCfR/jy Lu493f6YP+lfELIiPeSeiqo= =OunM -----END PGP SIGNATURE----- --Boundary-02=_8d6KA0tpjDpO6LV--