<div class="gmail_quote">On Wed, Aug 8, 2012 at 1:31 AM, Zooko Wilcox-O'Hearn <span dir="ltr"><<a href="mailto:zooko@zooko.com" target="_blank">zooko@zooko.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">In fact, it is quite likely that there were no files or</div>
directories to which write access was held by people on both sides of<br>
the partition!<br>
<br>
So, empirically, all this distributed consistency stuff that we're<br>
talking about is technically correct, and could probably be very<br>
useful in some specific cases, but with the Tahoe-LAFS access control<br>
architecture -- in which most things are immutable, and most mutable<br>
things are writable by few or only one writer -- such cases appear to<br>
be very rare.</blockquote><div><br></div><div>I operate a Friendgrid, and we have a centralized "Incoming" directory into which most previously unclassified content is uploaded by many users who share the same writecap prior to being moved to a more appropriate location by our content curators who have writecaps to, shall we say, the more organized directory structure.</div>
<div><br></div><div>I think this is a pretty common use case (at the very least, having one or more directories that users of a friendgrid might upload to simultaneously) and I don't think a network partition is necessarily something they're aware of let alone something that would stop them from uploading to such a directory.</div>
<div><br></div><div>I think it's an important case to consider in a system like Tahoe.</div><div> </div></div>-- <br>Tony Arcieri<br><br>