From: | Jurgen Defurne <jurgen(dot)defurne(at)pandora(dot)be> |
---|---|
To: | David Siebert <david(at)eclipsecat(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Foxpro |
Date: | 2002-05-04 05:39:44 |
Message-ID: | 3CD37420.D425225E@pandora.be |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin pgsql-general |
David Siebert wrote:
>
> I have a wannabe programer that works for me and he my not for long, nagging
> me that we should use FoxPro instead of PostgreSQL and Java.
> I have tried and tried to explain this to him. I have tried to show him that
> FoxPro is not mainstream anymore. He is a pain in my butt. Would anyone like
> to point me to a comparison between FoxPro and PostgreSql. I can find
> comparisons between DB-2 or Oracle or MySQL and Postgres but not FoxPro. I
> can guess why. I hate being a dictator about things like and I have tried to
> help him by saying he could do one little internal project in FoxPro but he
> can not even get FoxPro to use ODBC to connect to Postgres. I do not care to
> learn anymore about ODBC than what I need to hook Openoffice calc to my
> database.
>
> By the way this guy does not even know how to do a binary or to decode a BCD
> date. When I asked if it FoxPro had a binary shift he told me it had lots of
> keyboard funtions :(
>
I used FoxPro and Clipper between 1990 and 1994. They where both nice
tools for their time, and FoxPro's file and index format was the most
powerful of all xBase dialects.
Another thing that was very powerful in FoxPro was the fact that you
could embed SQL statements into your programs, like embedded SQL ala
COBOL. It supported INSERT, DELETE, UPDATE and SELECT statements and you
could move your result into a cursor, so as a programmer you did not
have to worry where your results went. These features made powerful
joins possible so I used them a lot, and learnt a whole lot along the
way about SQL data statements. I wrote a kind of defect tracking system
with it for rolling stock (yes, talking about railroads here), which
supported 11 users over a Novell network and this ran rather well
looking at the hardware we then had (all 386sx/33 MHz systems, server
386/33 Mhz w/ SCSI, over a 10 Mbit Ethernet). I know from the
documentation that if you had Btrieve, then you could do real
transactioning on the file server. One did not have to do much about
providing software to the workstations either, if you started programs
from the server drive, it created a local cache where it put the
run-times and the compiled programs.
Between 1995 and 1997 I was in factory automation, and then between 1997
and 2000 I was back in multi-user database systems using more powerful
hardware and software, first WANG PACE and Cobol, later ORACLE on HP/UX.
The most important thing that I hated on FoxPro was that I did not have
ways to nicely keep track of relationships, and that there was not a way
to bind code to tables (what I would find out later where called
triggers). So, FoxPro has nice things, and one can be very proficient
and even efficient with it, but it does really miss some features which
are to be found in really professional databases (and which were used in
WANG PACE and ORACLE) :
- Maintenance and checking of relationship integrity
- Triggers
- Real transactioning
- Scalability
Looking back, one can see that indeed all xBase has descended from
personal computers and that multi-user abilities mostly have been tacked
on based on the underlying OS, while for systems like DB2, Oracle, WANG
PACE and of course postgreSQL, multi-user was in the design from the
beginning. It gives me a different sense when I think about both. What
not many people seem to know however, was that FoxPro's predecessor,
FoxBase was also available on Unix platforms. This development would of
course have been stopped by MS when it bought FoxPro and its mother
company.
Emotional aside : as far as 1994, I found the then chief of FoxPro a
traitor for selling his company to MS. For someone who had shown to be
able to create and market a successful product, I do not think he had
valid reasons to take this step. It did not bring us customers benefits,
because when we upgraded to FoxPro 2.6, I did not see much difference
and all marketing stopped after MS took over.
Jurgen Defurne
LinuxIdee
From | Date | Subject | |
---|---|---|---|
Next Message | tony | 2002-05-04 06:10:30 | Re: Foxpro |
Previous Message | Justin Clift | 2002-05-03 22:26:33 | Re: Foxpro |
From | Date | Subject | |
---|---|---|---|
Next Message | tony | 2002-05-04 06:10:30 | Re: Foxpro |
Previous Message | Mark kirkwood | 2002-05-04 04:49:24 | Re: On Distributions In 7.2.1 |