S
seasoned_geek
Why do you believe Qt's implementation of the observer pattern, which they
refer as "signals and slots", is "a major factor in portability"?
Spend 20 years in IT working with all of these products claiming to be
"cross platform".
Go back to the early days of C++ cross platform libraries (or even
later days). The ones which did not have signals and slots. Take a
damned good look at how the compiler and the language specification
were abused with "friends" and preprocessor macros. Signals and Slots
was an intervention by the hand of God. If you don't believe this you
either have zero experience in IT or have never worked in the cross
platform world.
This is simply wrong. To start off, by "platform-independent", in this
context, it can only mean a program that may be successfully compiled for
more than one platform, which hardly means that there is any independence..
This is simply correct. You can and I have compiled a single set of
source full business application which ran on Windows 98, Windows XP,
Windows Vista, Windows 7, 32-bit OpenSuSE, 64-bit OpenSuSE, 32-bit
Ubuntu, 64-bit Ubuntu, and Mac. Not just some tiny little example but
a full ERP product containing customer maintenance, order entry,
invoicing, and inventory management to identify some of its portions.
The type of application most businesses have to have.
and Qt simply does not cover every aspect of computing.
I don't mean to offend, but that statement makes it sound like you've
never worked a day in your life. You certainly have never supported
production applications.
What SPECIFIC business functionality does Qt not provide? It has the
GUI, it has report creation, it has universal file handling, it has
classes to access relational databases and classes to interface via
IP. If you are developing in an MQ Series type of environment you
would need to call the IBM provided MQCONNECT MQOPEN MQGET MQPUT
MQCOMMIT MQCLOSE MQDISCONNECT, but MQ is available on every business
platform and not all businesses are currently utilizing message
delivery. Many are using raw IP and socket services to deal with
Internet communications.
And to claim that Java has somehow more portability issues than C++ software
relying on Qt, you surely don't know what Java is, nor do you know what
developing software with C++ using Qt.
Java has OCEANS of portability issues. I have written 3 books on Java
and more JNI code than I care to remember trying to work around the
pathetic excuse for the Java specification. You need to try running
Java on something other than a PC, then you will realize what a train
wreck that specification really is. Try running Java on an IBM
mainframe, or an OpenVMS midrange, or a Tandem computer. Good (^)(&*)
(*ing luck being able to do anything useful with it. Try getting the
same non-trivial Java applet to run on FireFox, Opera, Konqueror,
Arora, and Chrome. If you do magically get it to run on 32-bit
systems you will definitely be shot down by the 64-bit versions. Not
to mention each and every one of those browsers have different default
security settings so even if there is no syntax error what-so-ever,
the applet couldn't run on a bet. Oh you are soooooo correct, Java
has no portability issues what-so-ever....<cough><cough><hack><hack>
I'm even going to leave off getting it to run on a Windows machine
which, since the lawsuits, doesn't install Java by default, at least
many flavors don't.
It appears you are trying to pull the wool over this newsgroup's eyes, and
the way you are trying to achieve that is based on the idea that everyone
here is nieve, gullible and never used C++, Qt or Java at all. I don'tknow
if you are clueless or a marketing shill, but either way your shoving a
heavy load of nonsense.
It appears you haven't ever had a real job in IT, or you wouldn't make
such uneducated statements.
Hopefully you will know more about programming in general once you
attend college.