Re^2: Generate a unique ID

The length of the random part will decide about the probability to detect such a problem.

Whilst you're right about the probability, the question still arises, do you try to detect the problem?

Ie. Would you be happy with a sort utility that had the possibility to produce wrong results occasionally?

If so, what probability would you accept?

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

Comment on Re^2: Generate a unique ID

Replies are listed 'Best First'.
Re^3: Generate a unique ID (fault tolerance) by LanX (Saint) on Nov 16, 2010 at 11:01 UTC
You missed the point: Detecting the existence of a problem gives you the chance to improve the algorithm that you expected to be perfect! Think about the recent problems with Rolls Royce jet engines in the Airbus A380. The redundancy of this airplane - it can still safely fly with only 2 of 4 engines working - helped discovering that this totally unexpected problem exists. Now the engineers can either try to improve those engines or Airbus can switch to another supplier. But without the redundancy they would have no clue now, why the airbus might have crashed. Absolute security is an illusion. Cheers Rolf	[reply]
Re^4: Generate a unique ID (fault tolerance) by JavaFan (Canon) on Nov 16, 2010 at 11:11 UTC
But without the redundancy they would have no clue now, why the airbus might have crashed. Without the redundancy, they may not have had a failing engine. ;-)	[reply]
Re^5: Generate a unique ID (fault tolerance) by LanX (Saint) on Nov 16, 2010 at 11:52 UTC
Without redundancy, we may not have any air traffic at all! =) Cheers Rolf	[reply]