On Monday 03 December 2001 09:30 pm, you wrote:
> On Mon, Dec 03, 2001 at 08:52:57PM -0500, Scott Young wrote:
> > The Binary tree automatically handles change in update frequency.  A user
> > could insert two versions less than a second apart without problem, or
> > leave it for a year without updating.  It can scale down the time
> > interval infinitely small too.
>
> Ok, I wasn't claiming that this scheme is inferior to the one proposed,
> I was just letting you know about some prior work in the area.
>
> The core question with any of these schemes, given that determining that
> a key is not present in the network can be very time-consuming, will
> this be fast enough to be useful?
>
> Ian.

Multiple keys can be tried at once.  Once the first document is hit, it is 
downloaded then displayed in the browser.  Freenet then works behind the 
lines to search for the most recent version and it then downloads it.  The 
user can then hit "Refresh" to get the more current version.
The search tree should run in O(log(n)) time where n is the sum uncommon 
depth of the back and forward searches in the tree.  This can be optimized by 
accurate "guessing" at how deep the first part of the tree is (like if a site 
updates daily then do the tree depth closest to a day) and if it finds 
something there, it starts proceeding down the tree.
With this rampant guessing, the keys that exist should replicate very quickly 
and become easy to access, and when keys are hit other less likely requests 
for keys could be dropped - like a heurestic, but that all depends on 
implementation.  Also, the more the user requests the site the better the 
requesting can become.

I am willing to work on this after I figure out the internals of Freenet 
(although the poor documentation is slowing me down).  After I get the Binary 
Tree DBRs working well, I would then like to work on redundant block 
matching, although I shouldn't get too far ahead of myself.  I still need to 
get over the hurdle of understanding the internals though.

Scott Young

_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://lists.freenetproject.org/mailman/listinfo/devl

Reply via email to