> We're missing each other. The initial point of discussion was "what > are the barriers for the adoption of plan 9?" not the development > model or what is the "right" way to do things. My post was not meant > to advocate making rio "prettier". I hope I made that clear. > I'm sure you're right, and I ought to have pointed out at the outset that the adoption of Plan 9 by lots of users is not the good thing that it seems to be. In fact, in my opinion, it is not even likely as it would be, as discussed here, by imitation, which is hardly the role Plan 9 should be playing.
> What I would like to reiterate, though, is that the expectations of a > typical user, even a very smart one, differ from Plan 9 community. If > we could better understand that difference and couch Plan 9 advocacy > in terms a regular user can more readily understand Plan 9 advocacy > would be much more effective. But we'd waste enormous resources in exactly the wrong direction. Plan 9 has an appeal of itself. It is hardly sensible to try to cast Bette Midler in Claudia Schiffer's role. Teaching Claudia Schiffer to act (or sing) like Bette Midler may be more successful, but the result isn't a certainty. I'm sorry we (I, in particular) lost the thread of this conversation, as you point out. I think it should be made clear as soon as the subject is raised that Plan 9 is not anywhere near ready for broad acceptance, largely because it would entail straying very far from its fundamentals, but also because the active community, meaning the contributors, are much more likely to focus on Plan 9's unique properties than on imitating Windows. Perhaps we could (re)assemble an FAQ in which this feature of Plan 9 would be made much clearer, with corroborating evidence from the realm of those assets that the mainstream OSes are still struggling to acquire. ++L PS: I think Plan 9's biggest "mistake" was to drop Alef where it should have become the _only_ development language. I know this is absolute pie-in-the-sky and I accept without qualification the motivation for dropping Alef. But a lot of discussion would have been avoided if there wasn't so much C code out there waiting to be adopted, poorly, into the Plan 9 fold. Andrey's "libssh" is very much a case in point, but anything APE could be used as an example. I do not intend any disrespect by this, I'm merely pointing out that C raises expectations of Plan 9 that are not realistic. Plan 9-with-Alef became Inferno and no-one is suggesting that _it_ should be more widely adopted by adding Linux or Windows features to it. In summary, I think the subject needs resolving. The derided "road map" for Plan 9 should be a topic for study and contribution, as much as are P9P and APE ports, although it is understandable that we all only have the resources (or motivation) to contribute when we're stung personally :-)
