[
Lists Home |
Date Index |
Thread Index
]
On Tuesday 15 January 2002 03:07 pm, Paul T wrote:
> 1. XML is a tree.
>
> 2. UNIX filesystem is a tree.
>
> 3. find -name -exec script {} \;
> makes a simple, practical and scalable
> processing model for UNIX filesystem
> ( OK - for some of us ). It supports inserts
> and updates, actually.
>
> Was there some project in the past,
> that has tried processing XML trees
> *this* way?
reiserfs... and there are others. reiserfs is the only filesystem I've
come across recently that allows me to manipulate millions of
documents at once, with reasonable speed.
There are marked differences between filesystems and XML documents.
That said, for *many* cases, this approach would work just fine.
|
