> From: Brett Lymn [mailto:blymn_at_baesystems.com.au]
> To: Steve Wray
> According to Steve Wray:
> >
[snip]
> >Coda, it seems, will never be a production-quality
> >filesystem. It will always be experimental.
> >
> 
> Ummm "never say never".  Unless I misread what Jan said he did not say
> that this problem cannot be addressed, simply that it will be a big
> step to take in the development of Coda.  For a lot of people the
> limit of 7000 files in a directory is not a big issue and hence it has

How many aspects of coda are hardwired in this fashion?
How many numerical limitations are there which are set (artificially?)
low *because* coda is experimental?
Are there any other limitations which, if altered, would create a
'new epoch in the development of coda'?

These are the crucial issues that need to be addressed. Or among them
anyhow -- I can think of another right away;
'Will removal of debugging code & routines be feasable and will this
cause any incompatibilities with prior versions?'

> not reared it's head before.  The Coda team need to look at what can
> be done to address the issue and try to address the scalability.  I
> see it more as a hurdle (albeit a large one) than an insurmountable
> obstacle.  Hopefully it will simply mean the next step in the
> evolution of Coda!

What I was suggesting was taking what has been learned from the experiments
with coda and building on it.. evolution, but a new species, as it were
;)
distinguished by incompatibility much as real species are!