[OT] Business Requirements Analysis... Sort Of

In article <12s4u31plmj6csf7fgbji575cq3b3nee9t@4ax.com>,
Robert  <no@e.mail> wrote:
>On Thu, 20 Mar 2008 09:17:47 +0000 (UTC), docdwarf@panix.com () wrote:
> 
>
>My parents were married.

Punchline #862: Your mother's husband is buried in Penobscot, Main.  Your
father just landed a seven-pound trout.

> 
>
>Users SHOULD ask for a general purpose query screen that THEY can run
>without running the
>bureaucratic gauntlet every time.

In my experience, Mr Wagner, people don't always do what they 'SHOULD'
(caps original).

>If I were working there, I'd sell them
>the idea, then
>write it for them.

When I started on this contract there was a fellow who expressed a similar
desire; he was told it would be better to address this after the main
project went live.

After Go Live a bunch of consultant/contractors/hired guns were given
their walking-papers... he was one of them.

DD

On Thu, 20 Mar 2008 14:36:56 +0000 (UTC), docdwarf@panix.com () wrote:

>In article <12s4u31plmj6csf7fgbji575cq3b3nee9t@4ax.com>,
>Robert  <no@e.mail> wrote:
 
>
>In my experience, Mr Wagner, people don't always do what they 'SHOULD'
>(caps original).
> 
>
>When I started on this contract there was a fellow who expressed a similar
>desire; he was told it would be better to address this after the main
>project went live.
>
>After Go Live a bunch of consultant/contractors/hired guns were given
>their walking-papers... he was one of them.

This illustrates why software technology in some mainframe shops has not cha
nged in 20
years. It's not because workers are lazy or uninformed, nor because opportun
ities are
absent. It's because management blocks change. They do it to retain hegemony
 over users.

Cracks in the wall appear when users learn their counterparts can run a quer
y in five
minutes. They learn the other company is using Agile, Java and LINQ, then te
ll top
management all their problems will be solved by switching to new technology.
 In truth,
they could get the same with old technology. The difference wasn't technolog
y, it was IT
management's refusal to change.

Some managers welcome change. After my system went live last month, they ext
ended me six
months to improve it. I'm now working on improvements like the above.

On Thu, 20 Mar 2008 08:23:20 -0600, Robert <no@e.mail> wrote:
 
>
>I did that in the Good Old Days. My query program turned into DYL240, which
 became DYL280,
>now called Quick Job.

DYL260 was expanded to DYL280 by DYLAKOR which bought out by ? which
was bought out by CA.  I believe it is now Vision Results.  Quickjob
was another predecessor of DYL280.  If the COBOL standards committee
wanted to see a good way of doing report writing they should have
looked at DYL280 which made Report Writer look dumb.

Clark Morris

In article <t995u3lods5e970rbvqrgnuqr0bl8ucu1i@4ax.com>,
Robert  <no@e.mail> wrote:
>On Thu, 20 Mar 2008 14:36:56 +0000 (UTC), docdwarf@panix.com () wrote:
> 
> 
>
>This illustrates why software technology in some mainframe shops has not
>changed in 20
>years. It's not because workers are lazy or uninformed, nor because
>opportunities are
>absent. It's because management blocks change. They do it to retain
>hegemony over users.

I might agree with your statements, Mr Norris... but I'd quibble with
words like 'sotfware technology', 'mainframe' and 'retain hegemony'.
Let's see... from
<[url]http://groups.google.com/group/comp.lang.cobol/msg/12e0303b6b75284f?dmode=source[
/url]>

--begin quoted text:

I am not sure if this is the cause or it is the difference between 'if it
ain't broke, don't fix it' and 'let's take it apart to see how it works!'

A buddy o' mine used to be Materials/Inventory Manager for a jewelery
manufacturer... he could never understand folks who considered inertia to
be valuable, in and of it'sself.  After a meeting where things got a
bit... heated a VP took him aside and asked, seriously, what the problem
was... after all, in his (the VP's) department they'd been doing things
exactly the same way for the past twenty years.

'Where's your passion for work?', asked my buddy, 'Where do you strive to
do something more, where to you work to make things better?'

'You don't understand', said the VP, 'we've been doing the exact same
thing for the past twenty years.'

The VP could not understand why anyone would not see this as a Very Good
Thing.

--end quoted text

[snip]

>The difference wasn't
>technology, it was IT
>management's refusal to change.

It might have something to do with a given organisation's attitude towards
risk and reward, Mr Norris.  If errors are not tolerated - remember the
buzz-word phrase a decade or so back of 'We can't afford not to get it
right the first time'? - then the surest way to keep a job is to not waste
money or time on things that might not work.

>Some managers welcome change. After my system went live last month, they
>extended me six
>months to improve it. I'm now working on improvements like the above.

I was contracted at my present site in November '03 for a project that
went live in May '05... I think I posted my Musings After Go-Live here.
There was, of course, a massive dismissing of consultants... and since
then there's been a steady leakage of personnel, folks retiring and not
getting replaced, other folks getting fed up with an eternal heaping-on of
More Responsibilities and nobody (as far as I can see) getting hired on.
Meanwhile, the customer list continues to get larger (up about 25%, we
started with about 60,000, we're now at a hair under 75,000... and just
starting go-live on a new phase to bring in another 13,000 or so).

DD

On Thu, 20 Mar 2008 23:57:39 +0000 (UTC), docdwarf@panix.com () wrote:

>In article <t995u3lods5e970rbvqrgnuqr0bl8ucu1i@4ax.com>,
>Robert  <no@e.mail> wrote: 
>
>I might agree with your statements, Mr Norris... but I'd quibble with
>words like 'sotfware technology', 'mainframe' and 'retain hegemony'.
>Let's see... from
><http://groups.google.com/group/comp...=source
>
>
>--begin quoted text:
>
>I am not sure if this is the cause or it is the difference between 'if it
>ain't broke, don't fix it' and 'let's take it apart to see how it works!'
>
>A buddy o' mine used to be Materials/Inventory Manager for a jewelery
>manufacturer... he could never understand folks who considered inertia to
>be valuable, in and of it'sself.  After a meeting where things got a
>bit... heated a VP took him aside and asked, seriously, what the problem
>was... after all, in his (the VP's) department they'd been doing things
>exactly the same way for the past twenty years.
>
>'Where's your passion for work?', asked my buddy, 'Where do you strive to
>do something more, where to you work to make things better?'
>
>'You don't understand', said the VP, 'we've been doing the exact same
>thing for the past twenty years.'
>
>The VP could not understand why anyone would not see this as a Very Good
>Thing.
>
>--end quoted text

I was taught as a lad growing up on the options floor that there is a somewh
at constant
ratio between risk and reward. If a hypothetical stock were flat-lined, its 
beta would be
zero and an option on it would return no more than a T-Bill.

In real life, I've noticed people are poor at risk assessment .. in both dir
ections. Some
like Enron are reckless, causing catastrophic failure. Many more are overly 
risk-averse,
and thereby miss opportunities. They look for guarantees and investments tha
t are a sure
thing. The insurance industry, in large part, is fueled by its customers' po
or risk
assessment.

Mr. Norris lets them compensate by taking vicarious risks.
 
>
>It might have something to do with a given organisation's attitude towards
>risk and reward, Mr Norris.  If errors are not tolerated - remember the
>buzz-word phrase a decade or so back of 'We can't afford not to get it
>right the first time'? - then the surest way to keep a job is to not waste
>money or time on things that might not work.

The surest way to lose an IT job is to use obsolete technology. It may not b
ite you this
decade, but it will eventually.
 
>
>I was contracted at my present site in November '03 for a project that
>went live in May '05... I think I posted my Musings After Go-Live here.
>There was, of course, a massive dismissing of consultants... and since
>then there's been a steady leakage of personnel, folks retiring and not
>getting replaced, other folks getting fed up with an eternal heaping-on of
>More Responsibilities and nobody (as far as I can see) getting hired on.
>Meanwhile, the customer list continues to get larger (up about 25%, we
>started with about 60,000, we're now at a hair under 75,000... and just
>starting go-live on a new phase to bring in another 13,000 or so).

I would think the risk is lack of scalability. 25% isn't much of an increase
. My phase of
the current project added fourteen million active customers. My previous pro
ject for
another company added a similar number. In both cases, the company added a B
ig Unix box to
handle the increase.

On Thu, 20 Mar 2008 20:25:43 -0300, Clark F Morris <cfmpublic@ns.sympatico.c
a> wrote:

>On Thu, 20 Mar 2008 08:23:20 -0600, Robert <no@e.mail> wrote:
> 
>
>DYL260 was expanded to DYL280 by DYLAKOR which bought out by ? which
>was bought out by CA.  I believe it is now Vision Results.

DYLAKOR was founded in the late 60s by Wild Bill Newcomer, who 'borrowed' my
 program from
Tabulating Consultants Inc (TCI). Something happened to Bill and the company
 was taken
over by his daughter, Carole, who sold it to Sterling in 1983. Sterling was 
acquired by CA
in the early 2000s.

DYL240 was a JIT compiler. It read the input script and spit machine languag
e into memory,
then executed it. It ran very fast, usually faster than an equivalent Cobol 
program.
DYL260 was a complete rewrite in which they improved the language but also t
urned it into
an interpreter. It ran as slowly as interpreted Basic. They should have writ
ten a
compiler, as EZYTRIEVE did.

> Quickjob
>was another predecessor of DYL280.

I see CA renamed Quickjob to Vision Sixty. There is also Vision Inquiry for 
database
queries.

>  If the COBOL standards committee
>wanted to see a good way of doing report writing they should have
>looked at DYL280 which made Report Writer look dumb.

In article <seg6u3heplqhmesj0bo9lrjade00te571l@4ax.com>,
Robert  <no@e.mail> wrote:
>On Thu, 20 Mar 2008 23:57:39 +0000 (UTC), docdwarf@panix.com () wrote:
> 

[snip - attributions appear to have gotten mangled]
 
>
>The surest way to lose an IT job is to use obsolete technology. It may
>not bite you this decade, but it will eventually.

It has been suggested, Mr Wagner, that a Fairly Sure Way to lose a job -
IT or otherwise - is to recommend solutions which may, in fact, work but
with which one's Corporate Superiors are unfamiliar or uncomfortable.
 
>
>I would think the risk is lack of scalability. 25% isn't much of an increase.[/colo
r]

One would not know that by the Management Milestones newsletters that get
circulated... but I think it was Twain who spoke of a chicken cackling as
though she'd laid an asteroid.

>My phase of
>the current project added fourteen million active customers. My previous pr
oject for
>another company added a similar number. In both cases, the company added a 
Big Unix box to
>handle the increase.

I offered my skills to the Oracle side of the house at one point... and
they were accepted, too... and then it was discovered that this would be a
violation of a contract they held with the Anderoids.... errrr,
Accenture... and so, back to the Big Iron for me.

DD

On Thu, 20 Mar 2008 18:25:23 -0600, Robert <no@e.mail> wrote:

>This illustrates why software technology in some mainframe shops has not ch
anged in 20
>years. It's not because workers are lazy or uninformed, nor because opportu
nities are
>absent. It's because management blocks change. They do it to retain hegemony over u
sers.

I suspect some do it for that reason.

But others do it because their job isn't to make sure technology is
current.   Their job is to provide support for the business without
costing too much.

The CEO sees IS as a huge expense.   He grudgingly allows increases in
non-revenue producing technology to provide security and to meet
government needs.   But what he really cares about is stuff that gives
an obvious benefit to his bottom line.

That might be "giving salesmen an accurate inventory while in a
customer's office".    But it's not "having the most current
compiler".

And the CIO needs to make the CEO happy.

On Thu, 20 Mar 2008 23:52:28 -0600, Robert <no@e.mail> wrote:

>The surest way to lose an IT job is to use obsolete technology. It may not 
bite you this
>decade, but it will eventually.

I disagree.   The surest way to lose an IT job is to fail to give your
employer a good return on his IT investment.

Of course one can define "obsolete technology" to fit in to this. But
that's not what most techies mean when we use this term.

On Fri, 21 Mar 2008 09:22:20 +0000 (UTC), docdwarf@panix.com () wrote:

>In article <seg6u3heplqhmesj0bo9lrjade00te571l@4ax.com>,
>Robert  <no@e.mail> wrote: 
>
>[snip - attributions appear to have gotten mangled]
> 
>
>It has been suggested, Mr Wagner, that a Fairly Sure Way to lose a job -
>IT or otherwise - is to recommend solutions which may, in fact, work but
>with which one's Corporate Superiors are unfamiliar or uncomfortable.

True, when suggestions come from an employee or contractor. Suggestions carr
y more weight
when they come from an Outside Expert such as the author of an airline magaz
ine article.
That's where most CEOs get ideas about computer technology.
 
>
>One would not know that by the Management Milestones newsletters that get
>circulated... but I think it was Twain who spoke of a chicken cackling as
>though she'd laid an asteroid.

Here we have charts proudly boasting of achievements such as five minute ave
rage query
response and monthly crashes reduced from 100 to 90.
 
>
>I offered my skills to the Oracle side of the house at one point... and
>they were accepted, too... and then it was discovered that this would be a
>violation of a contract they held with the Anderoids.... errrr,
>Accenture... and so, back to the Big Iron for me.

Reminds me of F-100 companies where executive decisions were made like a wor
d association
game. I'll say a word or phrase, you tell me the first brand name that comes
 to mind.
Ready?

Computer: IBM
Business computer: mainframe
Database: Oracle
Network: ummm, either IBM or phone company
Secure business network: IBM
Trusted quality expert: Arthur Anderson (now Accenture)
Computer project management: Deloitte & Touche
Payroll system: Peoplesoft

They not only leased the product but also paid the maker big bucks for on-si
te experts.

They didn't need high-priced executives; they could have gotten the same dec
isions from an
average man-on-the-street.

At Sears, they wanted to go with Oracle. A VP scheduled a meeting with Larry
 Ellison to
sign papers. When he arrived, Larry had something more important to do. The 
BSD (big
swinging dick) retaliated for the snub by switching to Informix. Forget man-
on-the-street,
they could have gotten the same decisions from a kindergarten kid.