> Will postgresql be a viable database in 20 years? Will SQL be used
> anywhere in 20 years? Are you sure 20 years is your ideal backup
> duration?
>
> Very few media even last 5 years. The good thing about open source and
> open standards is that regardless of the answers to those questions,
> there is no proprietary element to prevent you from accessing that
> data- simply decide what it will be and update your backups along the
> way. Whether such data will be relevant/ useful to anyone in 20 years
> is a question you have to answer yourself. Good luck.
I am not to sure of the relevance, but I periodically worked as a
sub-contractor for an
Oil-producing Company in California. They were carrying 35 years of data on an
Alpha Server
running Ca-Ingres. The really bad part is that hundreds and hundreds of
reporting tables were
created on top of the functioning system for reporting over the years. Now
nobody know which
tables are relevant and with are redundant and or deprecated.
Also year after year, new custom text file reports were created with procedural
scrips. The load
on the server was such that the daily reporting was taking near taking 23 hours
to complete. And
the requests for new reports was getting the IT department very worried.
Worst of all know one there really know the ins and outs of Ingres to do
anything about it.
Well, I take part of that back. They recently upgrade to a newer alpha to
reduce the time daily
reporting was taking. :-)
Regards,
Richard Broersma Jr.
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
