|
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
Having been a Delphi (ala Object Pascal) developer for Windows since
Delphi's introduction, I appreciate the ability to essentially "recompile"
many Delphi applications to run under Linux. As a general rule, I have
found there is little that I can't do with Delphi (with the exception of
device drivers, maybe). The single inheritance model is fine when one
considers that the concept of "interfaces" has been extended to not require
COM to function. Through implementing "interfaces", once achieves much of
what multiple inheritence provides. And, its available in Kylix as well. I
used to be an avid C++ developer. But, have found that Delphi/Object Pascal
meet 98% of my needs.
I do a lot of communications work using sockets. This is where Kylix fails
(in my book). The socket library (DxSock) I use relies very heavily on
threading. Kylix 1 (and I suspect 2), still haven't gotten pthreads down
solid. DxSock was to be included in Kylix. The developer, however, refused
to allow it to because the thread problems would make his code look like
crap. On the Windows platform, his code easily outperforms all other socket
libraries. And, his webserver has achieved higher throughput than
Microsofts IIS. Yes, the library uses blocking sockets rather than the
event driven model. He's in the process of rewriting the threading layer
properly to bring his library to Kylix. Borland, in their typical fashion,
included an Open Source library, INDY, with Kylix. This library certainly
was not ready for production due to the same problems (and more).
But, Kylix 3 is not being targetted at writing servers, per se. Instead, it
being touted as a way to create Web Services with improved support for SOAP
and database access. In these areas, it seems to perform quite well.
Unfortunately, K3 is not yet released and I'm left having to work with K2
for a little while longer. The need to use CLX over the VCL to obtain cross
platform support is also detrimental as not all 3rd party vendors are
offering CLX support (but, that changes weekly).
Tools such as Kylix and JBuilder (and whatever the C++Builder version for
Linux is called) are the key to bringing more "business" oriented apps to
Linux. And, in doing so, it will mean a migration from Windows to
alternative OSes such as Linux. Microsoft seems to be helping us along
these days by changing the way they want to do business. The gut reaction
of many IT shops is to begin searching for alternatives. The OS with the
largest selection of business and developer tools will win. This means
Linux is poised to become more than a blip on the IT radar screen.
Kyle Burton demonstrated to me how Linux could be useful in a business
environment several years ago. I now run it exclusively for my internet
servers. My systems don't crash (I don't remember when I had to reboot the
web server last). And, I'm now making headway to move some of our core
products to the Linux platform. When we succeed, we have the ability to
"black box" our systems. This is possible because one does not need to
write a visual application using Kylix. It's still possible to write
"console" apps or even daemons. But, we all seem to concentrate on the GUI
aspects. But, since the compiler (and libs) are not GCC compatible, many of
the tools and libraries we depend upon are not currently available to us.
Kylix is not intended to be a mere Delphi clone for Linux. C++Builder is
also being ported. And, the JBuilder product was the first to make the
port. JBuilder has won many awards. But, word on the street is that the
products are not selling as well as expected. The end result is that these
tools may be pulled and all effort concentrated on .Net technologies.
Borland may have shot themselves in the foot early on by screwing up the
license. Who knows for sure?
The misconceptions that many have regarding Pascal (and Object Pascal) are
just that; misconceptions. I can debate for hours on the subject. But,
until we can mix and match libs between GCC and Kylix environments and the
price of the real Kylix tools (not the Open Source one) drops substantially
(how many personal developers can affort a $2K price tag) the crowd will
favor GCC. And, I can't totally blame them.
Charles
-----Original Message-----
From: plug-admin@lists.phillylinux.org
[mailto:plug-admin@lists.phillylinux.org]On Behalf Of Noah Silva
Sent: Wednesday, July 03, 2002 10:37 PM
To: plug list
Subject: [PLUG] Pascal?
Hi,
>From the response I saw at the meeting, I assume the two or three people
that asked me about Kylix were the only people who were interested.
Since I heard the groans around the room, I have to ask: Why?
It has been my experience that many techie types (or hardcore C or ASM)
are prejudiced against pascal just as much as Basic, but usually with no
basis, or a very outdated basis. There seems to be a lot of
misconceptions floating around pascal (it's slow, it's not powerful,
it's interpreted...), and I have to wonder why?
Also I should point out that there is no other tool which even comes
close to being able to build and debug GUI applications on linux as
conveniently as Kylix.
I'm not trying to start a flame war, but I am interested on opinions
about this. I have programmed in Pascal and C (and modula and rpn, 68k
asm, etc.) for many years, including designing some compilers in
college, and so far, pascal is still my favorite ;)
thanks,
noah silva
______________________________________________________________________
Philadelphia Linux Users Group - http://www.phillylinux.org
Announcements-http://lists.phillylinux.org/mail/listinfo/plug-announce
General Discussion - http://lists.phillylinux.org/mail/listinfo/plug
______________________________________________________________________
Philadelphia Linux Users Group - http://www.phillylinux.org
Announcements-http://lists.phillylinux.org/mail/listinfo/plug-announce
General Discussion - http://lists.phillylinux.org/mail/listinfo/plug
|
|