[Standards-JIG] Unique room names for JEP-45
Robert B Quattlebaum, Jr.
darco at deepdarc.com
Sat Jun 17 01:10:26 UTC 2006
On Jun 16, 2006, at 2:25 PM, Kevin Smith wrote:
> On 16 Jun 2006, at 21:25, Robert B Quattlebaum, Jr. wrote:
>> In other words, when there is a 1 out of 2^128 chance that there
>> will be a collision, I don't think that it is "correct" to check
>> for collisions. It is redundant.
> Million to one chances turn up nine times out of ten.
I'm sorry, but that doesn't make sense. Also, we aren't talking about
million (1x10^6) to one chances. We are talking about something much,
2^128 = 340282366920938463463374607431768211456 ~= 3.4x10^38
Due to the birthday paradox, we can approximate the chances of a
collision between values to be approximately:
1 - (e^(-(n^2) / (2 * (2^128))))
n = number of samples.
Assuming every person on the face of the planet (~6 billion) created
a 'nameless' conference room on a single MUC server, the chances of a
collision are (approximately):
1 out of 18,904,575,940,052,136,859 (1.89x10^19)
For all practical application, I think we can assume collisions to
not be a problem.
That being said, I'm not against having the server choose a name for
us. I just don't want you to think that doing collision checks when
dealing with probabilities on this magnitude is a matter of
correctness. It is effectively a redundant operation.
Jabber: darco at deepdarc.com
eMail: darco at deepdarc.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Standards