James Carlson wrote: > Michael Bender writes: >> Huh, that's an interesting statement, James. What does Sun Ray have to >> do with this specific case and with how TX handles device allocation? > > To be honest, I cannot tell. I know that two of the three separate > issues presented are related to Sun Ray operation, and assumptions it > makes about device allocation. I can't tell if the middle one is also > related or if it's just a bonus issue.
Sun Ray doesn't make assumptions about device allocation, I asked the TX team what the DA requirements were and we worked together to come up with how to get Sun Ray to work in a TX environment. >> I'm also not so sure what you mean by "baroque" - are you referring to >> the HAL and GNOME bits? Could you clarify your statement please? > > I'm referring (again) to the use of multiple success codes. I think > that's grotesque. Well, how about if we make '0' the error code, and all the non-zero codes success codes ;-)? Anyway, does it really matter as long as it is all documented so that both the supplier and the consumer agree on what code means what? We're not talking about an interface that will be used by a lot of people, so expending a lot of effort on getting it "just right" doesn't seem to be a very good use of time. > As I said before (though Scott Rotondo disagreed), I'm willing to give > up if it makes sense to the consumers. I'm not willing to redesign it > here, especially given the resistance. OK, thanks. mike
