From: Fisher Mark <FisherM@is3.indy.tce.com>
To: "'URI'" <uri@bunyip.com>
Subject: Re: URN Resolution Paths Considered Harmful
Date: Mon, 10 Jul 95 11:06:00 PDT
Message-Id: <30016ADC@MSMAIL.INDY.TCE.COM>
(Apologies for the late reply, as unfortunately URI group activies cannot be
at the top of my priority list...)
To reiterate, I think that we can learn from what database software vendors
had to suffer through during the 1980s. Pre-relational databases (esp.
hierarchial and network CODASYL) forced programmers to specify up-front how
a particular "name" could be resolved. Much time was spent where I worked
(a non-relational DB company) on developing a tool that allowed you to query
the database arbitrarily, as customers needed to make queries ("resolve
names") in manners not allowed by the hard-coded index paths. The big
reason that relational databases have supplanted CODASYL databases is that
the process of determining access paths ("resolution paths") is off-loaded
to the computer. That increase in flexibility has changed the creation of
small DB programs from a painful task to relatively trivial task, as well as
enabling the production in finite time of large DB program suites
manipulating hundreds of tables. This advance in database technology is
comparable to the advance in programming capability provided by
compiled/interpreted languages vs. assembly language.