Unless I am missing the point, it seems to me that this problem was solved
long ago by distributed system middleware such as CORBA and DCOM. They use
so called GUID's which contain a locally generated time stamp appended to
the network card identifier which is guaranteed to be a unique number for
each machine. Any windows system is able to generate a GUID on demand
although not from Java 1.1 applets. I don't know about Java 2 but I suspect
that there must be a security problem with accessing the network card
number. Anyway, if the id's are generated on the server (be it a local
server for p2p), it's not a problem. There are several Web applications that
use GUID's generated on client systems to insure that id's stored in the
server database are unique.
I googled the stuff a bit and found a dead IETF draft concerning GUID's and
the explanation of why it died and a hint that in the future we won't be
embedding the network card address in the GUID:
http://www.ics.uci.edu/~ejw/authoring/uuid-guid/draft-leach-uuids-guids-01.t
xt
http://dav.lyra.org/pipermail/dav-dev/1999-May/000245.html
Gil
P.S. Eric, did you check WebDAV for hints on versioning?
-----Original Message-----
From: Eric Armstrong [mailto:eric.armstrong@eng.sun.com]
Sent: jeudi, 26. octobre 2000 22:31
To: unrev-II@egroups.com
Subject: Re: [unrev-II] Jack's Use Case: Context-sensitive representation
Jack Park wrote:
>
> The topic maps folks do this with a *public subject* which is being
> defined as a registered URN. Registering URNs doesn't appear to be
> all that easy.
>
> In a closed system, you just start with the number 0, grab a couple of
> numbers for some reserved things, then start giving every concept a
> new, unique number. Pointrel does this. I suppose you'll need to use
long
> integers to pull that off, but, what the heck. Problem is, you cannot
> share with others; that requires grabbing numbers from some registry.
> Unique URLs
>
I'm delighted to say that I have solved this problem, at least in
principle. The combination of a globally unique userID (say
country+ID#),
plus a globally unique systemID (say, IP address) that includes a CPU
identifier, in a multiple CPU system, plus a synchronized,
miliisecond-level time stamp, produces a globally unique node ID.
That mechanisms allows peer-to-peer node sharing.
It seems like a lot of information, but except for the time-stamp,
nodes only need pointers to shared IDs, so the resource requirements
aren't too extreme.
[Note: If I can *just* get the rest of the versioning process worked
out, I'll have an announcement soon.]
eGroups Sponsor
Community email addresses:
Post message: unrev-II@onelist.com
Subscribe: unrev-II-subscribe@onelist.com
Unsubscribe: unrev-II-unsubscribe@onelist.com
List owner: unrev-II-owner@onelist.com
Shortcut URL to this page:
http://www.onelist.com/community/unrev-II
This archive was generated by hypermail 2b29 : Thu Oct 26 2000 - 14:39:30 PDT