Board index » kylix » Re: What to do with Kylix

2004-02-12 08:52:43 PM
kylix0
In article < XXXX@XXXXX.COM >, "Will Honor" <(NOSPAM) @
(NOSPAM).net>says...
Hi,
www.theregister.co.uk/content/4/35481.html
Phil
 

"Phil Shrimpton" < XXXX@XXXXX.COM >wrote in message
It's the extra 1 percent that matters. It's the boring stuff like adding
serialization capabilities to classes, implementing good security, and
documenting class libraries that makes C# and .NET viable."
The same applies to almost *any* Open Source project. It's that boring last
1 percent that makes all the difference and is often missing.
Amazing how rational, objective and realistic some can be when it comes to
Open Source projects that they don't like. But the objectivity quickly
vanishes when they support the project.
 

Ender wrote:
--
Nick Hodges -- TeamB
Lemanix Corporation
 

{smallsort}

Anders Ohlsson (Borland) wrote:
slip in public, with all us paranoid Kylix fans watching. That's a
feat I'd normally associate with myself <g>.
-Brion
 

"Nick Hodges (TeamB)" < XXXX@XXXXX.COM >wrote in message
support, it's just one tool among many; including many that are free.
IMO, the Kylix experiment has obviously failed and any further effort and
support for Open Source related development would probably be better spent
on CBuilder/X. Speaking strictly from a business perspective.
 

This has discussed million times. Linux community usually does not care
about closed source commercial apps like Oracle, VMWare, Kylix. Being
developer of such tool one should function like best member of linux
community, care about compatibility between distros, quickly fix bugs,
collect and use information about problems, release updates on regular
basic... i.o.w. do all things in that way so users has no neccessity to fix
problems themselves (and use open source technology advantages). With open
sourced app you have choice - to wait while developer fix problem or to fix
it youself. With closed source app (Kylix) you have to wait vendor. If the
vendor does not move a finger to eliminate problems why customer should use
application? It is absolutely not matter how you call that app, or what
position you give to it on the market. If quality of app is low and does not
going to improve, it will fail anyway.
Look on Oracle. They certified very few distros, but actually Oracle server
work nearly flawlessly on other widespread distros. So there is no problem.
Look on VMWare. Selling closed source product they very bothered about
compatibility with new distros and kernels.
 

Will Honor wrote:
nothing warranty that mono will be 100% with ms
--
Borland rulez pages.infinit.net/borland
 

"Ender" < XXXX@XXXXX.COM >wrote in message
- Oracle is good, VMWare is not?
- Server apps are a lot easier to distribute than desktop apps?
- The many Linux variants are a disaster for proprietary desktop software?
 

JQP wrote:
So, what is the point of DOT.NET?
The point has been and always will be MS' attempt to stop the Java momentum
and lure programmers into coding in a "supposed" possible xplatform that
could be taken to other platforms for the sake of keeping people in the MS
only Java realm, aka Dot.NET.
Anyone, and I do mean anyone, who looks OBJECTIVELY at NET and Java can
readily see that they are basically the same thing, with different
keywords. No, NET is not as mature as Java, and yes Java and NET each have
things the other does not. However they are minor differences, not major
differences.
The NET mavens have always claimed that NET is xlanguage. So, big deal. So
is Java, if you want it to be. But many say, MS pushes xlangauge aspect.
Sure, it is their main selling point, they believe, over Java. So, they use
it as a marketing ploy.
However, no company is going to solidify on a multiple language environment,
when there is no need to do so. It doesn't make IT nor business sense.
However, it does give those who are using something othert than the
Microsoft stable of tools, a false sense of security.
Some may actually believe it gives their lesser known and used langauge an
equal footing with the others. Sorry folks, this is naive. Yes, there is
some short term benefits of being able to use SOME of the previous code
from your language. However, for a long term strategy, I am afraid it too,
is a naive notion. The industry will, more or less, solidify behind one of
two languages as they have always done. On Windows, that was Visual Basic
and Visual C++. On Unix, it has been C/C++ and in the past five years,
Java. On the AS400 is was Cobol and RPG. On the OS 390 and MVS Mainframes,
it was Cobol. All of these platforms had other languages available, but who
used them? There have always been other players, but what is their overall
marketshare?
On Windows, neither Visual Basic nor Visual C++ were the best, but they
were the predominent, accounting for roughly 80-85% of the marketshare. And
now, with all using the exact same compilers, same runtimes, same byte code
(okay Intermediate code) structures, you think that the chances of a
predominent player emerging will be less?
Anyone looking at C# and then at Java can see these are two peas out of the
same pod. To deny this means you have looked at the one but not at the
other, or you are coming to the observation and use/testing prejudicially
predisposed to find one different than the other. I constantly here remarks
that one "loves C#", but hates Java. That is simply ridiculous. They are
the same animals, with some keywords and library name changes, basically.
Yes, there are few things many like better in C# (like properties, enums),
while others like things in Java better (like the lack of assemblies, which
do get kinda hairy from a coding perspective).
Who I truly feel for is those who will actually buy into the MS/Mono "lure",
that NET will be truly xplatform (MS will not commit to xplatform, nor deny
it, BTW). Those people are going to commit only to find, it will not be so,
precisely as the author states. I do not believe, personally, that MONO
will ever be 99% compatible, the best that they can hope for is 70-90%,
possibly, before MS changes the standards again, just as they have with
everything else. Even if MS became an entirely different company, with a
entirely new different mindset(people who believe this, I have some ocean
front property in Arizona for sale, want to buy it? :)), the natural growth
of the platform itself, would vote against this possibility. Having two or
more totally separate entities working on something that is supposed to be
completely interchangable, is not going to work, period. Without complete
and total collaborations of what consititutes what and what goes into X and
what does not, such a venture is doomed, from the beginning, to fail. Take
two different sets of programmers and give them the exact same requirements
to build a system, but they do not work on the project together as a team.
What will be the outcome. Will you have the same set of APIs? The same
class/object models? The same calls to internal and external methods? Of
course not. There would be simularities, but the finished product, even
with the same general specs, would be different.
If anyone desires or needs true xplatform today and for the forseeable
future, Java is the ticket. Java is and always has been geared and existed
for truly xplatform work. It works the same whether you are developing and
compiling on Linux, Windows, Mac, etc. Take that SAME finished jar/war
file developed on Linux and run it on MS Windows, Mac, or whatever. It will
run the same.
With NET it will always be hit and miss. The only way to avoid this would be
to use something like Mono exclusively, even on Windows platforms. So, why
not do this? Because services on Windows written with Mono.NET, I would be
willing to wager, will not work correctly, and/or will not be compatible.
MS will indeed seek to keep it this way. Why will MS keep it that way?
Because they are evil?
No, MS will do this because they are a business existing for the sole
purpose of keeping marketshare they presently have and increasing the
marketshare they desire to have. Why? Simple, MS is a for profit business.
That's the way it works in the business world, you try and keep the
customers you have and go after the customers you do not have. That doesn't
make them evil, it makes them a successful business. Sure MS could be nicer
about it, and less underhanded in their business practices, but they are
only doing what a successful business desires to do and even eventually
must do, to stay on top.
This is the way this has been from Day 1. NET is MS' attempt to create the
second iteration of Visual J++, a platform specific Java. They were unable
to co-opt Java from Sun, so they started to invent their own MS Java, first
called Cool and now NET. If anyone would look back to Gates in 95-96 about
Java, and the remarks from Gates and MS after they were prevented from
making a MS only Java, you could see the picture and plan unfold.
 

J>- Oracle is good, VMWare is not?
J>- Server apps are a lot easier to distribute than desktop apps?
Sorry, seems my bad English... Oracle is good, VMWare is also good. Kylix is
not good. The truth is that on Linux almost any reasonably complex product
should be constantly checked, revised, tested on compatibility with new
distros/kernels and so on... i.e. developer should regularly throw man-hours
on that product, or that product will be forgotten...
J>- The many Linux variants are a disaster for proprietary desktop
J>software?
No, not disaster. But they require attention. If attention is not paid, then
they will be disaster.
 

Ender wrote:
part of these standards committees, I would not waste my time on them.
 

Yep, that is all I use kylix for (web,daemons,console apps)
It works just incredible with Synapse for TCP/IP servers
www.ararat.cz/synapse/
The GUI part is just such a pain in the ass with QT etc etc and it looks
like shit.
The should have used wxwindows instead of QT.
"Nick Hodges (TeamB)" < XXXX@XXXXX.COM >wrote in message

Tony Caduto wrote:
problem is, for commercial use, requires a license.
OK Borland covered the license part for us but then bound it to QT 2.0 for
perpetuity. BAD MOVE, BAD!!!
 

i'm not hate it either). However, to work with Linux one should try to
understand it's philosophy and way to do various things. Entrie Linux
development it's a stream. It is very easy to swim with the stream, and
tough to swim across or against the stream.
 

Hi Nick,
I agree with you that Kylix desktop applications are unwieldy. However,
this very week I helped deploy a command-line (XML processing)
application on Linux, which was compiled with Kylix 3.
More often, however, I use the web server or web service capabilities of
Kylix on Linux - using Linux as the server, and Kylix as a development
environment for (web) server applications.
For me, Kylix is already a server development tool. <g>
Bob Swart (aka Dr.Bob - www.DrBob42.com)
--
Bob Swart Training & Consultancy (eBob42) Borland Technology Partner
webmaster UK-BUG / DDG Developers Group - IntraWeb Authorized Trainer