Re: PB 4.0 vs Oracle Forms

From: Tony Rothwell <Tony_at_santen.demon.co.uk>
Date: Fri, 24 Mar 1995 02:01:44 +0000
Message-ID: <850517052wnr_at_santen.demon.co.uk>


In article: <3kq66sINNfj7_at_marvin.wdi.disney.com> Steve Yan <steve_yan_at_cc.wdi.disney.com> writes:
> On the other hand, if you are not an Oracle shop and have
> no experience with their tools, then PowerBuilder is better of the two.

A colleague of mine has extensive experience with PB (incl PB4) and the Oracle tools. His choice is the opposite. He has a lot of praise for PB4, but for developing departmental or enterprise apps on Oracle7 he would choose Oracle's Developer/2000 (Dev/2000). His reasons are:

  1. Development is faster and easier using Dev/2000 because much functionality that requires coding in PB4 is provided automatically by Dev/2000.
  2. Development is faster and easier in Dev/2000 because it integrates more closely and more easily with Oracle7. For example, having PL/SQL as the common language speeds learning, speeds development, and simplifies partitioning.
  3. Dev/2000 is more powerful than PB4. For example, Reports 2.5 is the most powerful reporting tool available, bar none.
  4. The scalability, database performance, and network performance of Dev/2000 far exceeds PB4. Dev/2000 talks to Oracle7 in *the* most efficient manner possible.
  5. Having an integrated environment means one-stop support, and none of this bickering between the tool and database vendor, each blaming each other for the bugs you find.

> 1. PowerBuilder can run comfortably on a 486 machine with 4-6 Meg
> of memory. Oracle will run slow on the same machine and
> most likely require more memory.

Yes, the downside of Dev/2000 is that it requires a lot more memory than PB4. Also, it's not so object-based as PB.

At the moment, PB is too complex to develop personal/small-workgroup apps (1-5 users), but does not scale well enough (without using TP monitors) to build enterprise apps. Whichever direction Sybase shift it in, it's likely to be to the detriment of the other. As they'll probably shift it upwards, it seems to me they'll be needing a low-end development tool soon. Perhaps they might give Borland a call :)

-Tony Received on Fri Mar 24 1995 - 03:01:44 CET

Original text of this message