Hi Gabriel,

On Fri, Apr 01, 2005 at 01:38:12PM -0300, Gabriel B. wrote:
> > http://daltonlp.com/daltonlp.cgi?item_type=0&item_id=424 :
> 
> I think that Plan9 would be better suited for that google OS idea. but
> it's getting way of topic :) Back to coda:

:)

> Well, if you mentioned that openAFS already has the funding and the
> stability and the user base, why are you investing time in coda
> instead of implementing what openAFS lacks to be as good as coda?

These are two different systems built to solve different problems.

Coda has some inherent limitations (full file caching) which gives it
the ability to maintain consistency while disconnected.

OpenAFS has other inherent limitations (partial file caching) which
makes it less portable and does not allow for disconnnected mode -
but gives it other features like fast open()+seek() on big files.

So while you can solve artificial limitations (say, max number of files
per directory) or bugs, you can not remove limitations placed
by design, which enable some features implemented by design.

You can not abandon trains and implement their features in ships,
because they go different ways.

Regards,
--
Ivan