gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] Comments on 0.2


From: Karsten Hilbert
Subject: Re: [Gnumed-devel] Comments on 0.2
Date: Fri, 23 Jun 2006 16:15:31 +0200
User-agent: Mutt/1.5.11+cvs20060403

On Fri, Jun 23, 2006 at 12:21:49PM +0800, Syan Tan wrote:

> hashing also begs the question of what do you hash?
Which is IMO one of the two important reasons a hash won't
work well. The other one is performance.

> for different data types, and different for different versions,
Well, the hash would have to be over a concatenation of all
columns casted to text.

> client hashed on a ,b  , y and his hashed on a , b, x because it 's a new
> version made at a different time, in a different place etc... 
Well, because of that the hash generation would have to take
place in the backend by means of a trigger updating a hash
column which would then be matched against just like XMIN.

> sequence numbers are pretty simple in comparison.
Yes, such as XMIN.

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346




reply via email to

[Prev in Thread] Current Thread [Next in Thread]