Adam Turoff on Fri, 27 Jun 2003 10:23:28 -0400 |
On Fri, Jun 27, 2003 at 09:45:52AM -0400, Kurt Starsinic wrote: > On Jun 27, Adam Turoff wrote: > > FWIW, I used to work on a system mostly written in FORTRAN where there > > was one function/sub definition per file. That part of the model *can* > > work. > > What would you do about, e.g., file-scoped variables to be shared > by several functions? The last time I wrote FORTRAN (far back in the > last millennium), the only level of externally-sharable scoping was > COMMON, so it wouldn't have come up in FORTRAN (or COBOL, I think). > In Perl and C, though, as well as in many other "modern" languages, > it's a real issue. Ah, good point. I think the Java solution is to not have file-scoped globals, but make everything class-scoped. The analog here is to make packages the unit of granularity, much like it is today in practice. But that practical matter means that a wiki where each sub/method has its own entry is slightly more difficult to build, or at least assemble into a full set of sources. Z. - **Majordomo list services provided by PANIX <URL:http://www.panix.com>** **To Unsubscribe, send "unsubscribe phl" to majordomo@lists.pm.org**
|
|