Thank you very much!
On Thu, Oct 27, 2022, 5:08 PM Clay Jackson (cjackson) <
Clay.Jackson@xxxxxxxxx> wrote:
Wise words from MWF.
The problem I have with “best practices” is that they are typically only
“best” within some rather narrow set of constraints, which often includes
time.
Many moons ago, before such things as Index Organized Tables, and in fact
using a database other than Oracle, a shop I was in had “best practices”
that “all tables must be indexed” and “wherever possible, all access to
tables must be via an index”. Some wunderkind took that so far as to write
“access wrappers” for all tables and mandated that “all access must use the
wrappers”. We had a really small lookup table (under 100 rows), something
around work centers, as I recall, that was the most frequently accessed
table in the shop. Needless to say, performance problems around that
table were rampant. I was a “hero” for 10 minutes when I replaced a
process that called the access routine several thousand times with a
“direct read” of the table. When wunderkind found out, I was the goat. I
was a GS-3, he was a GS-13, and performance was sacrificed on the altar of
conformance.
There’s a term in medicine that I think might be appropriate:
“Standard of care”
Or the corresponding Tort law term
“Duty of care”
Of course, there’s the age old “excuse”, “It seemed like a good idea at
the time”
And ANY “standard”, or “best practice” or “usual practice” or WHATEVER you
want to name it, needs some sort of documented EXCEPTION process, the rigor
of which is commensurate with the possible harm that could be caused by not
complying with said standard.
Clay
*From:* oracle-l-bounce@xxxxxxxxxxxxx <oracle-l-bounce@xxxxxxxxxxxxx> *On
Behalf Of *Mark W. Farnham
*Sent:* Thursday, October 27, 2022 2:22 PM
*To:* christopherdtaylor1994@xxxxxxxxx; oracle-l@xxxxxxxxxxxxx
*Subject:* RE: What's that line again about 'best practices'?
*CAUTION:* This email originated from outside of the organization. Do not
follow guidance, click links, or open attachments unless you recognize the
sender and know the content is safe.
James Morle suggested something along the lines that they should be
renamed Usual Practices (or something like that). I’ve called them Standard
Minimum Starting Points and I pointed out that the only best practice I
know of is to not allow things to be called best practices. Calling
something a “best practice” tends to stifle attempts to do better.
IF you can get something called a best practice into your service delivery
standards and you implement that practice, you have a legal defense whether
or not the users can do anything or not.
Nothing can be proven to be a best practice. Things called best practice
are sometimes really just good enough to be acceptable.
You’ve probably caught the drift I believe “best practice” is a harmful
term. Some things called “best practices” are really quite good initial
starting points or usual practices that are just fine unless you need
something better.
*From:* oracle-l-bounce@xxxxxxxxxxxxx [
mailto:oracle-l-bounce@xxxxxxxxxxxxx ;<oracle-l-bounce@xxxxxxxxxxxxx>] *On
Behalf Of *Chris Taylor
*Sent:* Thursday, October 27, 2022 1:59 PM
*To:* oracle-l@xxxxxxxxxxxxx
*Subject:* OT: What's that line again about 'best practices'?
Mark or someone has an idiom I want to save this time....
Something about best practices being written by people who don't have to
support them or something .....
Chris
