Re: Could the Python swallow the Fox?

Tech-Archive recommends: Repair Windows Errors & Optimize Windows Performance

I do understand where you are coming from but I don't think it applies
equally among all development shops.

So, what you're saying is all the people that have been developing apps in
Java, C#, VB.Net, etc for the past few years are either not doing any
business apps or just tricking about what they're really doing?
The issue of this thread, as I understand it, is what moves a small shop
should make. For a small shop, yes, we do need a dynamic language for
writing business apps. I cannot afford to spend my time worrying about
typecasts. That was one of the reasons why no small software shop (in their
right mind) writes business applications in C++.
Now, somebody could validly argue that strong typing may cause things to
execute quicker and that editing your projects is more efficient. On the
execute side, if you want really fast code because you are running some
really complex process or whatever, I tell my clients to buy a faster
machine! Or, I don't use Fox for the task, I will write the routine in
another language.
On the project editing side yes, it is easier to catch errors up front, but
a lot of those errors would not occur if we weren't strong typing to begin
with. OK, employ a heap of programmers and run the project for twice as
long.

You can do things similar to these in .Net. I'm not saying it's easy, but
can be done.
Exactly, I want 'easy'. I am being paid to understand what a business needs
and translate that into software. I want a development platform that makes
my life easier, not more difficult. I want to leave the complexities to the
'technology' gurus.

... but the thorn in the side for me was that it apparently took 2 weeks
for an outsider to port Python onto the DLR but they couldn't do it with
Fox?
It's not "couldn't" do it. It's "wouldn't" do it.
Yes, and that is why it is a thorn in my side. When it was entirely feasible
to do the port they made the decision to dump the work that represents
thousands of manhours by thousands of developers.

I don't know what the definitive answer is. I spent nearly 12 years
developing a framework that lets me build business applications in a
entirely new manner that I don't believe anyone has done. I presented it
once at a Fox User group, someone came up to me afterwards and said it was
the best development environment they had ever seen (maybe it was the only
framework they had ever seen). Now it is finished, I am faced with some hard
choices.

The fact is that VFP continues to work, but external pressures are forcing
us to change (a change that would not have been required if a port to the
DLR had been done). 64 bit machines without 64bit VFP OLEDB drivers, a lack
of VFP developers and a perception by corporations that new products must be
written in .Net. These are forcing the changes on us.

"Craig Berntson" <craig@xxxxxxxxxxxxxxxxx> wrote in message
news:B1889E9E-BE59-45FA-8762-3C94ED0AA967@xxxxxxxxxxxxxxxx
See my replies inline below....

--

Craig Berntson
Microsoft MVP

-------------
"Paul Hemans" <darwin@xxxxxxxxxxx> wrote in message
news:umij7IFGJHA.6052@xxxxxxxxxxxxxxxxxxxxxxx
Craig,
The problem as I see it is that we lose more than we gain by going to c#.
For business programming we need a dynamic language, I

So, what you're saying is all the people that have been developing apps in
Java, C#, VB.Net, etc for the past few years are either not doing any
business apps or just tricking about what they're really doing?

believe that is one of the features that made VFP so powerful, weak
typing of variables, execscript, eval() and macros.

You can do things similar to these in .Net. I'm not saying it's easy, but
can be done.

All the things that a purist would call sloppy are running thousands of
businesses around the globe and have been doing so for decades. Now we
have the DLR in .Net but the thorn in the side for me was that it
apparently took 2 weeks for an outsider to port Python onto the DLR but
they couldn't do it with Fox?

It's not "couldn't" do it. It's "wouldn't" do it.

I know C# and it would cost a lot to transfer my business code into it,
much more than I can afford.

I've never said that VFP apps should be rewritten in VFP. If it works, can
be maintained, and meets your needs, VFP is a great product.

It is well suited to the corporate environment where you can afford a
team of developers, but that is not what this thread was about. If you
are a 1 man band then developing complex applications in C# is a recipe
for disaster as it is not a RAD environment.

C# was never about RAD. It was, and is, targeted to the language purist.
If you want RAD in .Net, look at VB.

It is aimed at the large corporate environments with teams of developers.

I agree with you here and I've been quoted in the press as saying pretty
much the same thing. But it looks like MS is going to go after the small
and mid-size market in VS10. See http://blogs.zdnet.com/microsoft/?p=1595

If you are a contractor (hired gun) then I would definitely recommend C#
but if your an independent shop then productivity is the most important
factor.
I want a language that looks familiar - because the language worked. So
far the main contender is Python because if anything it is possibly a
little more efficient code than even VFP was.

I've talked to lots of people trying to transition from VB 6.0 to VB.Net
and they said it was easier to go to C# because things were familiar, but
just enough different to screw them up. MS has said several times that
their research shows that about 50% of VFP devs that have gone to .Net
have chosen C# and the other 50% chose VB. You're the only person that can
choose what's best for you and your clients. If that's Python, that's
great. If it's PHP, that's great. If it's Java, that's great. If it's
something else, that's great too. But, I certainly wouldn't rule out Mono.

The other project that is interesting is http://guineu.foxpert.com/ but I
think it might take a while for that to be ready for production, I don't
know, (any comments Christof Wollenhaupt?)

Christof is doing some great stuff there. The first I ever saw it was at
German DevCon where he did his session in German. I didn't understand what
he said, but I certainly understood the demos. Very, very cool stuff.



.

Relevant Pages

  • Re: IBM file status 97 (was: Prodcuing an output file onlyonFriday?
    ... love to have a language that has all of the strengths of COBOL but also ... We choose financial applications and take ... outsource their IT solutions and concentrate on their real business. ... The traditional approach of a central shop where it makes sense to have ...
    (comp.lang.cobol)
  • Re: Could the Python swallow the Fox?
    ... For business programming we need a dynamic language, ... I want a language that looks familiar - because the language worked. ... more efficient code than even VFP was. ... ACCEPT Command ...
    (microsoft.public.fox.programmer.exchange)
  • Re: Spring
    ... All companies need finance, and it really makes no difference ... Too many people have a dream about a business of their own and borrow ... It sound like saying "Ladies and Gentlemen, for my next act of masochism ... And when I did have an amp or two in the shop here for 2 years, ...
    (rec.audio.tubes)
  • Re: Spring
    ... All companies need finance, and it really makes no difference ... Too many people have a dream about a business of their own and borrow ... sound is due to how they have done it but i don't think so. ... And when I did have an amp or two in the shop here for 2 years, ...
    (rec.audio.tubes)
  • Re: To all you shop owners out there...
    ... business or going to be out of business soon. ... Pol's ect came over & shop owners took advantage of their fine ... I think I want to be a toolmaker. ... Just try to sell an American product in China. ...
    (alt.machines.cnc)