Re: the distinction between data and intelligence
From: Kenneth Downs <knode.wants.this_at_see.sigblock>
Date: Mon, 27 Jun 2005 07:08:01 -0400
Message-Id: <le35p2-b89.ln1_at_pluto.downsfam.net>
>
>
> Yes, such logs are handy for many things. I have seen
> situations however where uncontrolled development had
> occurred and there were many different program objects
> being used for update and enquiry. In this instance, a log
> on the code useage helped rationalise the code, since the
> data logs did not contain an identity for the code used.
Date: Mon, 27 Jun 2005 07:08:01 -0400
Message-Id: <le35p2-b89.ln1_at_pluto.downsfam.net>
mountain man wrote:
> "Kenneth Downs" <knode.wants.this_at_see.sigblock> wrote in message
> news:38u3p2-clp.ln1_at_pluto.downsfam.net...
>> mountain man wrote: >> >>> "Kenneth Downs" <knode.wants.this_at_see.sigblock> wrote in message >>> news:7a30p2-ak7.ln1_at_pluto.downsfam.net... >>>> mountain man wrote: >>> >>>>> Here's a question for you, and anyone into developing >>>>> code within a project that involves change management >>>>> of this code suite over some years .... >>>>> >>>>> Do you build a simple stats history data table for each >>>>> of the code modules? ie: whenever code is invoked >>>>> by a user, this is logged to a table (procid, userid, >>>>> datetime). >>>>> >>>> >>>> Nope. >>> >>> >>> This information can sometimes be handy in order >>> to map and quantify the demographics of use of the >>> elements of the suite. >>> >>> >>> >> >> Not in my case. To map use we look at change logs on tables and columns. >> Since the code is tied directly to the definitions of columns and table, >> measuring one gives a measure of the other. >>
>
>
> Yes, such logs are handy for many things. I have seen
> situations however where uncontrolled development had
> occurred and there were many different program objects
> being used for update and enquiry. In this instance, a log
> on the code useage helped rationalise the code, since the
> data logs did not contain an identity for the code used.
That's why you want to eliminate use of manual coding for data processing, using only the spec to generate all code.
>
>
> Have you ever put in place a rowcounter history
> table for the database, containing table name, table
> row count and datetimestamp as at some routine
> time each day? See link below for surther info.
>
>
Yes. They usually tell us what we already know, such as what tables are getting hit the hardest.
-- Kenneth Downs Secure Data Software, Inc. (Ken)nneth_at_(Sec)ure(Dat)a(.com)Received on Mon Jun 27 2005 - 13:08:01 CEST