On Sat, Jan 25, 2003 at 08:31:25PM +0000, James Troup wrote: > > Having twenty packages named using a standard aspell-<ll>, and one > > package named 'aspell-<cc>' seems more confusing to me, particularly > > if you're a user who *needs* the package in question and > > consequently are likely to know what the language code is. > Again it's not about <ll> vs <cc>; I don't care what is used in place > of "uk" as long as it's not "uk". What exactly is wrong with "ukr" > for instance? Selectively using the three-letter language code is less consistent, but this seems a reasonable concession to clarity. I do strongly object to using 'ua', which would inappropriately associate the package with a location instead of a language. > > Are you really suggesting that people are going to look at 'aspell-uk' > > and think "Oh, good, just what I needed, a dictionary of words specific > > to Britain"? > Err, yes. I certainly would. Why on earth wouldn't they? There's > precedent for British dictionaries after all (ibritish). I would think a user might at least look at the package description when considering its installation, and there see that it's Ukrainian. I suspect the package name itself will be much more important to people who *are* familiar with the ISO codes than to those who aren't, as the former will likely be expecting the package to be there by name, and the latter will be more likely to search for a keyword (hopefully a keyword other than 'uk' :). -- Steve Langasek postmodern programmer
Attachment:
pgp1RGC9InsLM.pgp
Description: PGP signature