A
Andrew Morton
Ben said:
Err... isn't that from the framework, rather than from the language? Not
that any of this really matters anyway.
Andrew
Err... isn't that from the framework, rather than from the language? Not
that any of this really matters anyway.
Andrew
F
Fernando A. Gómez F.
Abubakar said:ummm Ladies and Gentlemen, this is not a newsgroup for c# vs vb
discussions
But is a newsgroup for C# discussions, is it not? If not, I beg you a
pardon...
F
Fernando A. Gómez F.
Jeff said:
Yep, but wasn't it the only MS language at the time?
Easy man, we're here to talk. Why did you have to go aggressive?
F
Fernando A. Gómez F.
dgk said:
So did I. I guess they still run, they did their purpose quite well. VB
had its market share, and it was designed for a specific purpose. In
particular, we used to create the front end in VB and back end
components in C++ with COM.
Perhaps it was a bad design. At any rate, I agree that OO is not always
the better. Each paradigm has to be chosen according to the design goals
of the application involved.
Please, re-read my post. I never said it wasn't a real language, that
was your assumption. It was designed for specific tasks and under
specific conditions, as happens with any other language. The thing is,
it had all these hidden features and hacks that made the language
unclean and error-prone. Its paradigm was some kind of hybrid between
structured and OO, yet it didn't allowed truly OO design.
You can write apps with any language, and certainly each project must
choose the proper language and platform according to its design goals,
etc. I did that a long time ago, that's why I wrote VB6 code. And that's
why, when needed, I write VB.NET code (although I don't like its syntax,
which I consider too verbose). But that has nothing to do with language
characteristics, rather more with the framework and the programmer's
skills.
And as for that, from an OO perspective, VB6 lacked several features,
despite being marketed as an OO language. That is why I think VB6 sucked.
Best regards.
F
Fernando A. Gómez F.
Herfried said:
Nobody said there should be one... IIUC, Neb said that VB was too
verbose (which I agree, but I guess that it is a matter of getting used
to). Rossum's reply implied that C# was also too verbose, just by adding
whitespace. The point of my post was to state that C# does not have to
be verbose, but that VB had to be verbose.
If only because that was one of the original primary design goals for
the language.
Best regards.
H
Herfried K. Wagner [MVP]
vanderghast said:
Well, I never said that it's the signature only. I just mentioned that I
prefer the long version because it's easier to extend. The long version's
code skeleton is typically generated by a code snippet, including the
backing field.
N
Nobody
"Fernando A. Gómez F." said:
Nope, sure wasn't.
F
Fernando A. Gómez F.
Nobody said:
Sorry I missed that. Which other languages had MS at the time?
Certainly, C++ is not a MS language... Perhaps you refer to J++?
Best regards.
T
Tigraine
I've programmed in both languages and it mostly comes down to personal
style and taste. As of VS10 VB will become again "equal" to C#. The
current 3.5 implementations aren't. C# is way ahead in lambdas and
some other (syntactic sugar).
The only thing that drives me nuts regularily is that VB doesn't allow
line breaks except through that awkward _ operator.
Sometimes there are long method calls I'd like to break down (or a
long LinQ method chain) and VB won't let me do that easily.
VB tends to be more expressive to start with. But once you learn to
avoiding nesting in C# it boils down to the same thing. I try to avoid
methods that are longer than 30 LoC and have more than 2 levels of
nesting {} .. so usually it's no big deal to identify what closing
brace } belongs to what thing. Having to write }} //if only means that
I can't see where I started the block, and that's usually very very
bad.
greetings,
Daniel
B
Branco
Timothy said:
Yes, but, then, not exactly (IMHO). It's true that the lack of
security in the OS allowed for a really fun ride for hackers, script
kiddies and their neighbour's dog, but the political development of
the Windows API, the different models it had (has) to support to
favour legacy compatibility, etc, well, it seems to me that that
allowed for a lot of damage also, due to friendly fire. I seem to
remember a nasty bug in the GetWindowText API caused by some clever
programmers using the window caption to pass (binary) data around
between running instances of their apps. Programmers (from big
companies like Adobe) went through hoops and jumps and twists to force
Windows behave according to their plans and then there were the
counter hoops and anti-jumps placed by MS to revert Windows back to
its (uh...) "correct" behavior (sorry for not providing factual
references, these are stories I remember reading in my long gone --
stolen, mind you -- MSDN Magazine collection *before* it was branded
"MSDN Magazine").
That doesn't surprise me a bit, I had my fair share of non-working
commercial applications or the ones that subverted the system in such
a way as to drive it (and me) nuts.
<snip>
"Fully half of it actually works" is some statement! =)) It seems we
actually agree. Don't you think that having those bad programmers
working in a more "secure", "rich", "easier to learn" and "accessible"
environment (.Net) may produce better applications? In the meanwhile,
nothing would stop the real bad *ass* programmers from giving us
Proccess Explorer and similar (unmanaged, native, API based, etc)
shining apps.
Here in Brazil we have the expresion "Boi de piranha". It's how you
call the cow that is sent alone as bait to cross the river downstream
so it attracts all the piranhas (really, really bad fish) and allows
for the rest of the cattle to cross safelly upstream. To me .Net is a
boi de piranha for the general programmer. I consider myself a
reasonable to good programmer and I notice that I write better .Net
apps than I write "API based" apps (damn, I just realized I'm a
piranha!)
Best regards,
Branco.
B
Branco
mayayana wrote (responding to a post of mine):
Me:
It's certainly a coincidence that this is repeated over and over and
also a cliché, besides being my educated opinion =)))
Well, yes, nicelly put. Of course, nothing is really restraining me
from actually using tools that allowed me access to the API. So, yes,
its a choice I make and I hope others make too, so the guys who "don't
know what they're doing" (API-wise) can come to this bandwagon and
know a little better =D
I don't see the problem in Windows turning (evolving?) from platform
to service. That's actually what Windows -- and it's API -- allways
longed to be (IMHO). This role of providing a thin abstraction layer
between application and hardware was never played very well by Windows
since its inception, and I doubt it defines this OS as we know it.
Yeah, I see your point, but don't get me wrong: I'm not bad mouthing
COM. My opinion is that COM was a fantastic breakthrough in
application interoperability, a really visionary and radical approach
(yes, there was CORBA, but COM actually *existed*, contrary to CORBA).
Trying to learn COM made me a better programmer, really =))
But COM for the masses had an intrinsic complexity that no tool
(besides VB) would easily circumvent (C++ smart pointers were
"discovered" too late in the game when there was a lot of damage
already done by programmers forgetting to call AddRef/Release
appropriately). Thus there were lots of badly written COM application
(or so the legend says). I remember that .Net exploded on our faces
just when COM was preparing to turn into COM+. For a few months MS' PR
was "COM+ this" and "COM+ that", "COM+ is the wave of the future", and
then *boom* there was .Net. It really seemed to me that a political
tour de force had ensued inside Microsoft and this new platform (lol
sorry, "service") got the upper hand, maybe because it provided
answers to COM "faillures" that outshone COM's benefits (DLL hell,
registry bloat, referene counting, anyone?)
The merit of .Net, in my opinion, is to provide this really rich and
*likable* environment (for a programmer such as myself) while it
allows the OS to continue its regular existence: as you notice,
Windows is vastly COM-based still.
I'd install Vista, which would change the contribution of disk usage
from the .Net libraries to mere 2% of the OS =)))
I'm just kidding of course. Or maybe not. Vista takes up 10GB of space
in a regular install, if I'm not mistaken. What drives us to install
such a monster? Even the slim and fat-free (yeah, right) XP seems huge
if we compare its 1GB (I'm trusting your numbers, here) to, say, a
functional Linux install that fits what? 100 MB? And then came World
of Warcraft. Damn, that thing alone is 15GB!!
I'm using VB.Net to write corporate apps that are used in a daily
base. It replaced both VB6 and Delphi as the default development
platforms here, and suffocated to death (I really like this image,
considering the people involved) a movevement that wanted to migrate
all our development to Java. =))
What I see is: .Net is kind of fresh still, and it allowed a freedom
in development that really shines if compared to the alternatives. Is
there bloat? yeps, but this is good fat, not bad fat, and comes mostly
from new services and tools mayde available to you. Is there
instability? uh, I really hardly see it (there was a 3rd party app
once that failed to install with no elucidative message. I installed
the 3.0 redist -- the machine had only 2.0 -- and everything worked
without a glitch). Lack of security? In C# maybe. I suppose it's
easier to write a faulty app (security-wise) in C, unmanaged C++ or
unmanaged C# than to write a faulty managed .Net app (I may be wrong,
of course, we VB programmers once had a history of circumventing the
system).
Best regards,
Branco.
Me:
It's certainly a coincidence that this is repeated over and over and
also a cliché, besides being my educated opinion =)))
Well, yes, nicelly put. Of course, nothing is really restraining me
from actually using tools that allowed me access to the API. So, yes,
its a choice I make and I hope others make too, so the guys who "don't
know what they're doing" (API-wise) can come to this bandwagon and
know a little better =D
I don't see the problem in Windows turning (evolving?) from platform
to service. That's actually what Windows -- and it's API -- allways
longed to be (IMHO). This role of providing a thin abstraction layer
between application and hardware was never played very well by Windows
since its inception, and I doubt it defines this OS as we know it.
Yeah, I see your point, but don't get me wrong: I'm not bad mouthing
COM. My opinion is that COM was a fantastic breakthrough in
application interoperability, a really visionary and radical approach
(yes, there was CORBA, but COM actually *existed*, contrary to CORBA).
Trying to learn COM made me a better programmer, really =))
But COM for the masses had an intrinsic complexity that no tool
(besides VB) would easily circumvent (C++ smart pointers were
"discovered" too late in the game when there was a lot of damage
already done by programmers forgetting to call AddRef/Release
appropriately). Thus there were lots of badly written COM application
(or so the legend says). I remember that .Net exploded on our faces
just when COM was preparing to turn into COM+. For a few months MS' PR
was "COM+ this" and "COM+ that", "COM+ is the wave of the future", and
then *boom* there was .Net. It really seemed to me that a political
tour de force had ensued inside Microsoft and this new platform (lol
sorry, "service") got the upper hand, maybe because it provided
answers to COM "faillures" that outshone COM's benefits (DLL hell,
registry bloat, referene counting, anyone?)
The merit of .Net, in my opinion, is to provide this really rich and
*likable* environment (for a programmer such as myself) while it
allows the OS to continue its regular existence: as you notice,
Windows is vastly COM-based still.
I'd install Vista, which would change the contribution of disk usage
from the .Net libraries to mere 2% of the OS =)))
I'm just kidding of course. Or maybe not. Vista takes up 10GB of space
in a regular install, if I'm not mistaken. What drives us to install
such a monster? Even the slim and fat-free (yeah, right) XP seems huge
if we compare its 1GB (I'm trusting your numbers, here) to, say, a
functional Linux install that fits what? 100 MB? And then came World
of Warcraft. Damn, that thing alone is 15GB!!
I'm using VB.Net to write corporate apps that are used in a daily
base. It replaced both VB6 and Delphi as the default development
platforms here, and suffocated to death (I really like this image,
considering the people involved) a movevement that wanted to migrate
all our development to Java. =))
What I see is: .Net is kind of fresh still, and it allowed a freedom
in development that really shines if compared to the alternatives. Is
there bloat? yeps, but this is good fat, not bad fat, and comes mostly
from new services and tools mayde available to you. Is there
instability? uh, I really hardly see it (there was a 3rd party app
once that failed to install with no elucidative message. I installed
the 3.0 redist -- the machine had only 2.0 -- and everything worked
without a glitch). Lack of security? In C# maybe. I suppose it's
easier to write a faulty app (security-wise) in C, unmanaged C++ or
unmanaged C# than to write a faulty managed .Net app (I may be wrong,
of course, we VB programmers once had a history of circumventing the
system).
Best regards,
Branco.
T
Tim Roberts
rossum said:
It's amazing you were able to recover from that.
Algol 68 was a fascinating language -- it almost literally tried to be all
things to all people. PL/I programmers could write PL/I in Algol 68.
Jovial programmers could write Jovial in Algol 68. Fortran programmers
could write Fortran in Algol 68.
Ada owes a lot to Algol 68.
T
Tim Roberts
Fernando A. Gómez F. said:
By that definition, Basic is definitely not a Microsoft language. It
predates Microsoft by a decade.
The only truly Microsoft language is C#.
C
Cor Ligthert[MVP]
The only thing that drives me nuts regularily is that VB doesn't allow
l>ine breaks except through that awkward _ operator.
Done in VB 10, although not everywhere like in a string, but that is the
same in C# (although there are more places it is simply impossible).
But surely in all Linq methods and much more where on those places you
always wanted it.
However, C#4 needs that senseless semicolon and method () still almost
everywhere
Cor
l>ine breaks except through that awkward _ operator.
Done in VB 10, although not everywhere like in a string, but that is the
same in C# (although there are more places it is simply impossible).
But surely in all Linq methods and much more where on those places you
always wanted it.
However, C#4 needs that senseless semicolon and method () still almost
everywhere
Cor
M
Mike
Tim said:
Good point, but isn't the analogy the same?
VB.NET is to BASIC, what
C# is to C++
C++ and BASIC aren't Microsoft invented but the mutations C# and
VB.NET are. No?
Maybe that is why SUN nips on potential mutation of JAVA by Microsoft
and sued them before it became their own. That is the reason
Microsoft is abandoning Java.NET - they can't control it.
--
B
Bill Yanaire
Does that mean my VB applications will stop working?
HA HA HA HA
HA HA HA HA
J
Jai Singh
Per the decades old flame war whose smoke and flame still linger eternal in
our hearts and minds...
The reality is that no matter what the "reality" is or is not... VB, by its
very name, has always and will always, apparently, suffer from SEMANTIC
Discrimination, based solely upon its lowly, debased name... which no longer
adheres to its type descriptor.
Yet, to interject, C# is nothing but VB (in spirit and form (pun intended)).
It took the spirit and purpose of VB and wrapped it in {}. That's all. C#
is not a powerful, dangerous, low level language (save for unmanaged code
segments) - it's a nice, highly abstract language that is powerful because of
its abstraction (like VB 6 was) and that is a million miles away from message
pumps, unmanaged pointers, and hwnd of C based windows development.
The .NET Framework was a MAJOR advancement for VB and for VC++ programmers
who really wanted a simpler, VB like world to develop in... and wanted the
incredible real world gains that that provides.
VIVA LA GUERRA!
our hearts and minds...
The reality is that no matter what the "reality" is or is not... VB, by its
very name, has always and will always, apparently, suffer from SEMANTIC
Discrimination, based solely upon its lowly, debased name... which no longer
adheres to its type descriptor.
Yet, to interject, C# is nothing but VB (in spirit and form (pun intended)).
It took the spirit and purpose of VB and wrapped it in {}. That's all. C#
is not a powerful, dangerous, low level language (save for unmanaged code
segments) - it's a nice, highly abstract language that is powerful because of
its abstraction (like VB 6 was) and that is a million miles away from message
pumps, unmanaged pointers, and hwnd of C based windows development.
The .NET Framework was a MAJOR advancement for VB and for VC++ programmers
who really wanted a simpler, VB like world to develop in... and wanted the
incredible real world gains that that provides.
VIVA LA GUERRA!
B
Ben Voigt [C++ MVP]
Jai said:
Unless you want to do something useful. In which case the HWNDs, message
dispatch loops, and everything else are right there waiting to bite the
unsuspecting C# programmer.
M
Mike
You have to admit, VB/NET people can certainly hold their own. They
know their place, enjoy it and don't let anyone tell them otherwise.
I was once in a $40M project with 36 or so people during a recession
where many engineers were transferred into the C/C++ OS/2 software
project to give them work. They struggled big time with C/C++ and
OS/2 GUI slowing the project with lots of newbie C/C++ bad coding and
just a terrible time in writing GUI programs. Then something
incredible happen - Microsoft came out with a killer app called Visual
Basic, along with Windows 3.0, MS split with IBM, dumped OS/2 and
because of VB, the project was reduce $15M with 20 people, many
overpaid C/C++ OS/2 consultant were let go, the engineers prospered,
RAD, component engineering and the project was completed. Never
forget that one (I helped decide the switch).
--
know their place, enjoy it and don't let anyone tell them otherwise.
I was once in a $40M project with 36 or so people during a recession
where many engineers were transferred into the C/C++ OS/2 software
project to give them work. They struggled big time with C/C++ and
OS/2 GUI slowing the project with lots of newbie C/C++ bad coding and
just a terrible time in writing GUI programs. Then something
incredible happen - Microsoft came out with a killer app called Visual
Basic, along with Windows 3.0, MS split with IBM, dumped OS/2 and
because of VB, the project was reduce $15M with 20 people, many
overpaid C/C++ OS/2 consultant were let go, the engineers prospered,
RAD, component engineering and the project was completed. Never
forget that one (I helped decide the switch).
--
T
Timothy Casey
[SNIP]Tim Roberts said:
I doubt C# is a language unto itself. C# is simply another variation or
dialect of C, along with C+, C++ etc. (hence the affectation of 'C' with
modifier in the name).
The only language I can trace back to Microsoft is RTF (1987). RTF can be
structured two ways. One uses group hierarchies to define non-overlapping
format zones and results in a structure identical to that of the style of
SGML that uses the generic closure tag; </>. In RTF this is replaced by the
closing curly brace: }. Interestingly, SGML also has another structure
identical to the inline (potentially overlapping) flag system used in RTF
when the group hierarchy is not used. RTF's equivalent of block elements are
deployed with the same structure as that used in IBM's GML, which is the
root for SGML. RTF may well use different control characters, but in terms
of structure it is closer to a SGML and GML than HTML. Nevertheless, RTF,
having its own vocabulary, is much more a language unto itself than is the #
variant of C.
Much of what we consider "new" on PCs has already been done on mainframes
and minicomputers. Honeywell had DDR memory in minicomputers decades before
it was "new" in PCs and some of the decades old 1Gb SSDs used in older ICBMs
are *still* much faster than the brand "new" SSDs made for the PC market.
Macros were used to describe classes before the word 'macro' was
systematically misused to promote MS Office's reinvention of what was once
known as "master mode". Master mode was the first attempt at laying a
foundation for SaaS, once again back in the mainframe days, and once again
master mode was discredited spectacularly decades before malware authors
used the exact same techniques to exploit Office "macro" functionality. In
fact, the same techniques for information theft are still used via
JavaScript, VBScript, ActiveX, XAML, & .NET components served up to the
client as applications disguised as documents.
Relational databases have not changed since the 1970's - only the
programming interface has changed. Hierarchical databases are still plagued
by the same fragmentation issues now as they were when writers like Kroenke
(1977) outlined the methods and conundrums of data storage. Database
normalisation dates back to Codd (1970), and although Goveran & Date (1994)
are credited with the Orthogonal Design principle, the concept of functional
dependence discussed by authors like Kroenke (1977) is not dissimilar.
Basic was first designed and put to use at Dartmouth College by John Kemeny
and Thomas Kurtz in 1964. However as most of the original vocabulary of
Basic continues in Visual Basic, then VB is likewise an evolved dialect of
Basic (in much the same way that Standard Italian is an evolved dialect of
Latin - probably with overlap of variations of local dialect thrown in for
broader accessibility). The world's first IDE came with Dartmouth Basic, as
Basic was not intended to be simply a beginner's programming language but to
serve as an example of the cost savings offered by a little versatility and
simplified UI. This focus on simplicity is probably why Basic remained the
most versatile programming language. In time, console IDE evolved into
keyboard-based IDEs, such as Maestro I of the 1970's and then the graphic
IDE or CASE emerged in 1982 with GraphiText, followed by DesignAid; the
first to support a wide range of structured analysis and design
methodologies.
Then there is the broken paradigm (not the hippy symbol for establishment
perspective, but the real linguistic animal). Human language structures are
verb oriented, whereas modifications like that of Basic into VB have shifted
the fundamental structural orientation from verb oriented to noun oriented.
This makes separation and identification of subject and object vastly
simpler and more consistent than the more variable verb oriented
expressions. OOP was known decades ago as State Driven Programming and has
quite a long history. It was MIT that originally smashed the paradigm with
PDP-1, shortly followed by Daal & Nygaard's Simula 67 in the 1960's, which
interestingly also had the first 'garbage collection' functionality.
I think that VB will probably evolve into the programming equivalent of
"English" because it is used more often in reliable communications regarding
programming methods. VB already has demonstrable roots in GML, SGML,
Dartmouth Basic, PDP-1, Simula 67, & DesignAid, with influences hailing from
C (multilevel programming), Pascal (string assembly systems), etc. Only dead
languages like Latin and Esperanto are pretty. Popular languages are always
'bastardised' and anything but eloquent because it is that flexibility,
which makes a language most accessible to the broadest possible demographic;
and accessibility is the lifeblood of language.