IMHO..(rant)

tldr; Old bloke yearns for a simpler past

Of course I agree wholeheartedly, but on the other hand there was an 
elegant simplicity that enabled an app to be generated very quickly. A 
well-designed relational model in pdox is little different in a modern 
database. The ground-breaker for me was Codd and I recall selecting ANSA 
Software's then new Paradox Database precisely because it was the 
product that most closely supported Codd's criteria. Similarly Netware 
gave us super fast and reliable networking and servers (in contrast to 
NT), and later switched to linux which went on to great things even if 
it destroyed Netware's raison d'etre. If only we'd known that Borland & 
co would fail to provide that sort of continuity.

Back in the day we had promises of 4GL and 5GL languages. I think in 
many ways with the advent of Windoze the whole concept went backwards 
and app development became much more complicated again. But Micro$oft 
got us by the short and curlies. Sometimes options are a curse, enforced 
simplicity a blessing. I suspect the correct answer lies somewhere in 
between. As a result Delphi did ok.

But again on another hand, languages, databases and communications have 
come a long way, computer speeds are comparatively unbelieveable, 
development tools are amazing, open source tackled the stranglehold of 
proprietary code, and the web promises run-anywhere front ends even if 
it is built on different philosophies that introduce new challenges.

The trend is to use inter-operable components using advanced 
communication techniques built around sockets. No more behemoth apps. 
Good stuff, but personally I still also like the idea of the ability to 
easily build a data-centric app in an environment that looks after me. 
Frameworks have emerged as the answer, but they have their own issues too!

Phew!

- Bernie
On 18/10/17 3:20 am, Larry DiGiovanni wrote:
> Paradox applications, especially DOS PAL apps, tend to be written in 
> certain ways to accommodate the way Paradox works.  Data is modeled to 
> work in Paradox forms, which does not always promote good modeling.  
> What I am saying here is that retaining the original design is not 
> always a good thing.  ;-)
> -- 
> Larry DiGiovanni