Another Pick Software Company Attempting to Convert To ORACLE - Voice Your Opinion
webmaster d-o-t c-o-m
I'm not familiar with Oracle, so I'm stating my opinions based
on conversions I've done with other environments. Anytime you have to
re-write something for another platform or another environment, you
can expect alot of redesigning. It's my personal opinion that
'trans-fitting' code from one system to another is not worth the
effort if you can develop the applications using flow charts
or design templates written from the original code. Why hammer
a square into a circle when you can just cut a circle out of new
material. It'll look and work a hell of alot better than the hammered
piece.
As far as the market, I have no clue!
I don't own a software company or work for one. I do; however, feel
that the company making the software has the final say in what they
will and will not do. Customer opinions and wants should be respected.
I think that you should also weigh the good and bad points of such a
transform. Time required, money required, support for the new software
and environment. Wouldn't it seem obvious that a programming crew be
experts on the environment before they start writing critical code?
How can a programmer streamline code if they don't know what makes the
OS tick? Custom coding is something that I've dealt with for a while.
You either write a canned package or a modulised package. If the
customer wants custom stuff, you'll have to keep up with:
1) you own versions, upgrades, and fixes. Not everyone will have the
same release.
Plus
2) your customer's individual requests which may or may not be
affected by your code changes - OR may or may not affect
your future code changes/implementations. If one customer can't
have your new 'option' because of their previous custom code,
will you give it to everyone else? what if that guy still
wants it and demands it?? Are you looking at a complete rewrite
to accomdate that one customer?
I've seen some people sell the software outright for customization.
They are not allowed any upgrades or bug fixes once they
customize their site. That's not very ethical to me, but it's a way
to minimize code conflicts on the design, debug, and upgrade side. If
they want the new release, they pay dearly for code merging and
debugging in the end.
That's my 2 cents,
Glen aka Ryengoth
http://members.xoom.com/Ryengoth/pick
Bill Hanna
A few rules of modern business politics:
1. it is easier to convert to another product than to fix your own
2. it is easier to write a new system from scratch than to fix your own
3. technology seldom influences decisions on which platform to move to --
the platform that the VP used to have in his prior company is his best
choice (after all, he knows this system)
4. nobody ever got fired for suggesting Oracle (or was that IBM...)
As far as rewriting in Oracle, I'm sure someone in this NG could tell the
story of Oxford's (Oxford Health Care in CT) attempt to convert to Oracle.
The last I heard (about a year ago), they were loosing money (after being
very profitable) because bills had not been produced over a period of
months, and they were unable to manage workflow because the claims
forecasting reports were not in place. These functions may not have been
perfect in the Pick system, but the move to another platform, at least
initially, cost them dearly.
One of my co-workers likes to say that one reason that businesses move from
Pick to Oracle is because the business will not pay for top Pick talent,
therefore there is a high programmer turnover, for which the software
suffers. Eventually, the Pick product gets the blame, and the move to
another DB is on. Now, these companies _expect_ to pay good wages to Oracle
people, because the market demands it.
The bottom line is that I believe that software should continue to evolve as
time goes on (not just be fixed as features are demanded). As new features
of the DB come out, and customer's business needs change, modules should be
rewritten to take advantage of this and the knowledge that the staff had
gained over time.
Maybe many companies are on this track, but many are not, and they just
don't understand the cost of not keeping pace with what is available.
On the other hand, a complete rewrite from scratch is intriguing. If one
can start with a "correct" database design, then use the features now
available to most Pick-like platforms (GUI, web enabled, transaction
logging, commit/rollback, etc.) the results could be impressive. But for
most fast-paced businesses, by the time the initial product is rolled out,
it will already be obsolete. Resolve the obsolescence issue, and get
long-term commitment from top management, and the results could be great.
Only $5 million... I know of companies that have spent tens of millions in a
conversion effort, then abandoned it!
- r - a - m - a @interacsdotc-o-m Drama
>I'm so SICK of e-mail SPAMGlen Batchelor <webmaster @ allspec d-o-t c-o-m>
wrote in article
David Montgomery
Drama > wrote in message <01be7093$9782fcc0$a9d9b926@ramette>...
Your company should also look at jBase as it has support for Oracle and
other RDBMS's
With jBase you could persue a "gradual" rewrite of the application to
support Oracle and add
GUI features.
David Fisher
We have a very interesting Oracle story, one of our longtime users is in the
garment distribution business and were in an association with the countries
biggest retailer. They decided to move the distribution center to a facility
of the retailer who are big Oracle users. The existing distribution center
ran on MOE (Mentor Operating Environment) on an NCR 3000.
The new facility ran on MOE on an NCR 4300 with two processors and 512MB
ram. It was able to process on a busy working day twice the volume of
transactions that an 8 processor machine with Oracle could. We had to admit
that we could not offer the degree of data integrity at the database level
that Oracle offers but we have never needed it because the fault tolerance
at a hardware level has kept the database up 100%.
At a pure business level the capital cost of the system was a lot less than
an equivalent Oracle system, the cost of running and maintaining the system
is minimal and - most importantly - the functionality and ease of change is
very high.
Here is a real live example of our MultiValue Database performing better
than one of the so-called mainstream products.
Dave Fisher
Drama > wrote in message <01be7093$9782fcc0$a9d9b926@ramette>...
>I recently worked for a "premier Pick software company" that has provided a
>vertical application to a lucrative market (publishing industry) for about
>20 years. It is written in traditional Pick BASIC (no 4GL or Applications
>Generator in use), and has its share of problems, having been piece-mealed
>together over the years by dozens of programmers.
>
>Still, it is pretty good, and they have about 100 satisfied customers who
>have paid big bucks for this software.
>
>Now, although the President of the company is a very down-to-earth
>individual, he does associate with the big-wigs of major corporations, and
>though sometimes wears jeans to the office, other times is dressed in his
>finest.
>
>In a recent meeting he pleaded with everyone to do anything they could to
>help improve programmer and customer service productivity.
>
>But he also announced a plan to potentially re-write the software
>application in ORACLE - an estimated 5-year, 5 million dollar project, all
>because the big-wigs (stuffed shirts) in England at the big publishing
>companies say "if it's not in ORACLE AND GUI, we're not interested".
>
>What is Your Opinion:
>
>1. It is one thing for an end-user to demand something from their in-house
>programmers, but doesn't it make sense for a Software Company like this to
>be making the Technical Decisions, and simply tell the big-wigs "sorry, we
>are the experts here, we know what's best, we cannot have the power and
>flexibility desired with ORACLE, you let us make the technical
>decisions..."?
>
>2. What is the likelihood of such a re-write being successful (knowing the
>extensive "business rules" and "special custom software" that most
>customers have in their customized versions, which they also paid big bucks
>for)?
>
>3. Furthermore, aren't there a LOT of other countries participating in this
>marketplace? Have I heard wrong, for instance, that Pick is big in
>Australia? What other countries is Pick big in?
>
>
>
>(Please no plugs for the MV-ON (Pick on ORACLE) product, they know about
>this product and will check it out if they desire).
>
>--
>d - r - a - m - a @ interacs dot c - o - m (no dashes)
>* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
>Helping Cure Corporate Illnesses Since 1980
>* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
>
Roger Nicholls
>DAVE RAMETTE wrote:
Snippety-Snip...
>
>But he also announced a plan to potentially re-write the software
>application in ORACLE - an estimated 5-year, 5 million dollar project, all
>because the big-wigs (stuffed shirts) in England at the big publishing
>companies say "if it's not in ORACLE AND GUI, we're not interested".
>
>What is Your Opinion:Funny that…, seems a familiar story to me
Dave, That sounds familiar.
When working for a UK Hardware/software provider, I went to a large
hospital after responding to their SOR and being invited to
demonstrate the system, after the demo, which went very well and all
the potential end users were fired-up and said that it was the first
system that they had seen which really addressed the needs of the
staff 'on the shop floor', fast data entry, quick response etc. etc.
The new IT manager then walked in, (you know the sort, I call them
estate-agent types, young, loud, flash, bleeper going off every minute
and causing him to interrupt our meeting while he conversed loudly
with someone on the 'phone) and announced that "If it isn't GUI and
Isn't Oracle, frankly chaps you are wasting your time being here,
thanks very much, goodbye!"
Nice of him to tell us, after asking us to demo and letting us drag a
mini plus all the terminals etc. to the site.
Sometimes the end-user's opinions count for nothing in the corporate
scheme of things.
After three years, this company re-wrote the application in Oracle
with VB front end due to pressure from other customers who said that
this was the way to go.
This was the last Pick application to bite the dust for them. They saw
it as bigger revenue, with sales of bigger Iron, more PC's, more
profit from the Oracle licence and inevitable add-on charges from
Oracle, more support charges for the hardware, more costs to upgrade
all the PC's when they don't perform when they are all running
Business Objects….. Everybody was happy :-)
Fashion seem to dictate corporate policies, with some IT managers who
are valuing their jobs more that the user's needs , many years ago it
was said that 'No-one ever got fired for buying Blue (IBM)', next,
no-one ever got fired for buying Unix, then it was Oracle, then NT
what next? Answers on a postcard please….....
Yes we must move forward, but with caution, it is a minefield for
anyone who wants to develop a new application in knowing which
platform to put it on, this the benefit of 'Pick-like' suppliers
putting their product on NT as well as Linux/Unix.
The problem is that we, as software developers, will need to have a
broader knowledge of the underlying O/S in order to support all these
platforms properly. Just follow the threads on 'why does my network
card.....', 'How do I create a terminal driver for.......', 'Backups
fail with version xxx of xyz with release abc using SCSI card 1234..'
!!
I know that we used to have similar problems with Mc Duck, Mentor, GA,
Ultimate, R83 etc, but they seemed simple compared to today
(or perhaps I'm getting old !)
Whatever happens, I feel that a Pick-like product will survive, be it
Jbase, D3, Uni.. and long live CDP
----------------------------------
ro...@diadem.u-net.com
Diadem Computing Services Ltd (UK)
Tel:01344 774266 Fax:01344 778069
----------------------------------
Tom Packert
>Fashion seem to dictate corporate policies, with some IT managers who
>are valuing their jobs more that the user's needs , many years ago it
>was said that 'No-one ever got fired for buying Blue (IBM)', next,
>no-one ever got fired for buying Unix, then it was Oracle, then NT
>what next? Answers on a postcard please….....
>
Well, some people may be getting fired soon over NT, judging from the
preliminary reports of what it takes to port applications from NT 4 to NT 5.
Many budgets may be in for a rude awakening when they have to spend gobs of
money porting apps developed on NT 4 to the new version. Software that was
already 'done'. M$ will probably solve a lot of this before it becomes too
big of a mess but it is amazing that backwards compatibility is an
after-thought not a design requirement.
As for Oracle, it is no magic bullet for all that ails software. If
management thinks it can change the database and fix the issues, then they
are seriously deluded, the problems lie elsewhere. Oracle IS an exceptional
product, with its own foibles. Sure it is more expensive, but it is just a
cost of doing business. But if buying into Oracle gets you more sales and
bigger sales, and you can afford the entry price, it makes sense.
It may offend our rather, lean, Pick way of looking at things, but, lots of
companies deliver high quality, consistent information services using
Oracle. One problem with Pick is that the companies are so lean, that one
or two people can manage a large application. Not so with Oracle shops. If
one experienced person leaves a Pick shop, WOW, that impacts the company
pretty severely. But with more people, with more specialized skills, the
loss of one person, is less dramatic. Consistency over time, is a valuable
attribute.
I read an article, forget where, that said Oracle needed more people to
develop in Oracle 8 with multivalues. Apparently, so many people are
skilled in 1st normal form and columns/rows, they are having trouble
thinking in nested tables. The apps are not taking enough advantage of the
new Oracle 8 features.
Tom Packert
>On the other hand, a complete rewrite from scratch is intriguing. If one
>can start with a "correct" database design, then use the features now
>available to most Pick-like platforms (GUI, web enabled, transaction
>logging, commit/rollback, etc.) the results could be impressive. But for
>most fast-paced businesses, by the time the initial product is rolled out,
>it will already be obsolete. Resolve the obsolescence issue, and get
>long-term commitment from top management, and the results could be great.
Actually, if they are rewriting from scratch, and already picking the DB
they are locking themselves in. Seems like they should avoid getting locked
into Oracle and build the app as database independent, with Oracle being one
of the supported DBs.
Enter your name here
I am looking to also change the 20 year-old user interface 'paradigm' in a
very successful package. We've decided to rewrite in Clarion from TOpspeed
(the old Turbo Pascal People ) because the 'browse' and 'tree explosion'
paradigms are native to theri template development.
Plus one set of source for 'access','oracle',AS400, DB2 - just let the
development environment know what you want to do and let it do the work
check out www.topspeed.com
P.S. I really hate having to do it because of the productivity with PICK
Basic, put the world must go on
PPS a COmpany in our client list tried it - 50M later they are the highest
expense ratio in the trade and might be forced to merge to survive - by the
way - they're still running their cobol 85 over db2 cics system just to
survive.
Nick
Tom Packert wrote in message <7crc9e$bam$1...@camel19.mindspring.com>...
John Bend
>
>When working for a UK Hardware/software provider, I went to a large
>hospital after responding to their SOR and being invited to
>demonstrate the system, after the demo, which went very well and all
>the potential end users were fired-up and said that it was the first
>system that they had seen which really addressed the needs of the
>staff 'on the shop floor', fast data entry, quick response etc. etc.
>
>The new IT manager then walked in, (you know the sort, I call them
>estate-agent types, young, loud, flash, bleeper going off every minute
>and causing him to interrupt our meeting while he conversed loudly
>with someone on the 'phone) and announced that "If it isn't GUI and
>Isn't Oracle, frankly chaps you are wasting your time being here,
>thanks very much, goodbye!"
I think this same person now works for the Council we have as a customer.
Anyone know a *cure* for these types?
John Bend.
