Joseph Kowalski wrote: [snip] > I was prepaired to "go with the flow", until I saw Richard Lowe's mail. > It appears that defaults of "gmacs", "vi" and "none" are equally prevalent > in other systems. If we want to help the newbee, being consistant across > systems is the best way to accomplish that. Since "gmacs", "vi" and "none" > are about equally prevelent, I guess we can't do that. The Pro's I see > for each choice are: > > none: Pushes no style and matches 33% of the data we have. > > vi: Matches the "momentum leader" (Ubuntu), matches historical > Solaris and matches 33% of the data we have. > > gmacs: Matches 33% of the data we have. > > Humm, this is nearly a three way tie, but upon closer examination, gmacs > is in last place. Much to my surprize, "vi" appears to be the winner. > > Its amazing what real data can show. Thanks Richard!
Plese read his email again. The majority of items were ksh88 or ksh88-clones (like pdksh) which completly predates inventions like /etc/ksh.kshrc or "usuability" ... and quoting Ubuntu's ksh93 behaviour was likely the worst item of the whole list (please compare this to SuSE Linux who support ksh93 as part of their product and actively do bugfixing (like Werner Fink's contributions to get all the bugs in japanese environments fixed)). After filtering the ksh88 items it looks like this: none: 1 (MacOSX) vi: 0 emacs/gmacs: 1 (SuSE Linux) ---- Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 7950090 (;O/ \/ \O;)
