On reiserfs and external attributes

On reiserfs and external attributes

> The Reiser4 external attribute API is different from the API used in the 2.6 kernel. Applications, to use this API, will have to be rewritten to use the special-purpose reiser4() system call.

> The solution seems reasonably clear: Reiser4, once it's ready, can be merged with its new ways of doing things.

I don't understand how these two excerpts can be reconciled. Surely reiser4(), a hack specific to a single filesystem, is not "the new way of doing things"?

I respect Mr. Reisers work, but I have the slight feeling he's promoting his
baby at the expense of Linux as a whole in this matter.