VB.NET is not VB!
Jonas Dygd
until now, since I'm a C++ programmer (C# nowadays).
The one thing that struck me the most was the forcing of paranthesis to not
only functions with no parameters, but with subs as well. How hard could it
have been to have kept the old syntax?
And non-zero based arrays are gone, with the CLR standard as an excuse. That
is stupid, and could have be worked around by just substracting the lower
index in all references. That is, if the developers had actually wanted to
create a new version of the VB language.
After taking a look at this new language, it looks to me as it was written
by C++ programmers, who's trying to show VB programmers what's best for
their own good. It wouldn't surprise me if the all the VB compiler does is
parse through the VB code, converts it to C# and compiles it using the C#
compiler.
Joe "Nuke Me Xemu" Foster
> I've been using .NET for a while but I really haven't bothered to look at VB
> until now, since I'm a C++ programmer (C# nowadays).
>
> The one thing that struck me the most was the forcing of paranthesis to not
> only functions with no parameters, but with subs as well. How hard could it
> have been to have kept the old syntax?
>
> And non-zero based arrays are gone, with the CLR standard as an excuse. That
> is stupid, and could have be worked around by just substracting the lower
> index in all references. That is, if the developers had actually wanted to
> create a new version of the VB language.
They almost had a new version of VB, until it got axed.NET...
news://news.devx.com/3b4cda9b$1...@news.devx.com
> After taking a look at this new language, it looks to me as it was written
> by C++ programmers, who's trying to show VB programmers what's best for
> their own good. It wouldn't surprise me if the all the VB compiler does is
> parse through the VB code, converts it to C# and compiles it using the C#
> compiler.
"Yeah - we've got that!"
http://groups.google.com/groups?selm=3ACD4562.609C%40cluestick.org
http://groups.google.com/groups?selm=3B2BE988.50%40cluestick.org
VB is dead. Long live VB!
--
Joe Foster <mailto:jlfoster%40znet.com> "Regged" again? <http://www.xenu.net/>
WARNING: I cannot be held responsible for the above They're coming to
because my cats have apparently learned to type. take me away, ha ha!
Domenic Albano
> by C++ programmers, who's trying to show VB programmers what's best for
> their own good. It wouldn't surprise me if the all the VB compiler does is
> parse through the VB code, converts it to C# and compiles it using the C#
> compiler.
Well, VB has always been written in C++. So umm, this is nothing new!
"Jonas Dygd" <jo...@novasoftware.se> wrote in message
news:uYrQk$IUBHA.2220@tkmsftngp05...
Jonas Dygd
news:O4sdQVMUBHA.2020@tkmsftngp07...
> > After taking a look at this new language, it looks to me as it was
written
> > by C++ programmers, who's trying to show VB programmers what's best for
> > their own good. It wouldn't surprise me if the all the VB compiler does
is
> > parse through the VB code, converts it to C# and compiles it using the
C#
> > compiler.
>
> Well, VB has always been written in C++. So umm, this is nothing new!
>
When I used "C++ programmers" in this context I wasn't just talking about
people who are capable of writing programs in C++. I was talking about
programmers who prefer C++, and over VB. What language the VB.NET compiler
was actually written in is irrelevant.
Jonathan Allen
> parse through the VB code, converts it to C# and compiles it using the C#
> compiler.
It would surprise me. C# doesn't have all the constructs that VB.Net does.
For example...
Catch Err as Exception When X > 0
C# doesn't have the When clause, a CLR supported construct. They can
simulate it, but there is a performance hit to do so.
This is just the tip of the ice berg. There are many things that VB.Net can
do that C# can't. Some are just syntactic sugar and automagical constructs,
others are actual CLR features that C# didn't implement.
--
Jonathan Allen
"VB is dead. Long live VB!"
- Joe "Nuke Me Xemu" Foster 10/9/2001
"Jonas Dygd" <jo...@novasoftware.se> wrote in message
news:uYrQk$IUBHA.2220@tkmsftngp05...
Jonathan Allen
I see you finally decided to accept the new VB. Welcome aboard.
--
Jonathan Allen
"Joe "Nuke Me Xemu" Foster" <j...@bftsi0.UUCP> wrote in message
news:#MyL1VJUBHA.2116@tkmsftngp03...
Joe "Nuke Me Xemu" Foster
> > After taking a look at this new language, it looks to me as it was written
> > by C++ programmers, who's trying to show VB programmers what's best for
> > their own good. It wouldn't surprise me if the all the VB compiler does is
> > parse through the VB code, converts it to C# and compiles it using the C#
> > compiler.
>
> Well, VB has always been written in C++. So umm, this is nothing new!
Much of the first few versions was written in assembly, yet VB 1/2/3
didn't have MOV or JZ statements. Why, then, must VB.NOT's arrays
start with 0, just because some C-weenies can't conceive of an array
starting with anything else?
--
Joe Foster <mailto:jlfoster%40znet.com> On the cans? <http://www.xenu.net/>
Zane Thomas
I don't know whether most people have seen enough of your posts to know that
there is *nothing* that can be said which will shake you loose from your
belief that VB6 was the be all and end all of programming languages.
So this post is for the benefit of others ...
> Why, then, must VB.NOT's arrays
> start with 0, just because some C-weenies can't conceive of an array
> starting with anything else?
It has been suggested that all that's needed to implement non-zero based
arrays is to simply subtract the lower bound before, presumably, passing the
array to some method written using some other language. That won't work.
Let's suppose that you had an 1-based array, passed it to some c# method
which held on to it and then later passed some integer to a method which was
then used as an index into the previously-passed array. Off by one error.
There is no general way to address that issue other than by requiring
consistency between languages (well, unless you would resort to the absurd
suggestion that all languages be aware of and track array-indexing issues
presented by all other languages).
IMO if you have to choose between zero-based and one-based arrays, then
zero-based is the best choice because a) it's more efficient in the first
place and b) calculations of indices into the array are generally more
efficient since they don't require +1 all over the place.
Ok, rant on d00d.
Zane
Joe "Nuke Me Xemu" Foster
> I don't know whether most people have seen enough of your posts to know that
> there is *nothing* that can be said which will shake you loose from your
> belief that VB6 was the be all and end all of programming languages.
That's a *lie*, and you know it. If, as you assert, I had some sort of
"belief" that "VB6 was the be all and end all of programming languages",
surely I'd denounce any and all routes to implementation inheritance in
VB instead of talking about the aborted *real* VB7 or about providing
inheritance through bolt-ons and add-ins? Would I have asked about an
"Inherits" keyword to complement "Implements"?
http://groups.google.com/groups?selm=11021%40bftsi0.UUCP
http://groups.google.com/groups?selm=924808591.409.13%40news.remarQ.com
http://groups.google.com/groups?selm=tjci65bv9n67b7%40corp.supernews.com
http://groups.google.com/groups?selm=%23bzKWjkOBHA.1788%40tkmsftngp03
http://groups.google.com/groups?selm=tlgpr9ge699r2a%40corp.supernews.com
"You _are_ wrong. Next time we're in Redmond together, I'll introduce you to someone
who will verify that the VB team was within two months of having full inheritance in
the Classic version before they were pulled to the damn .net project."
-- Karl Peterson, 2001-07-11, news://news.devx.com/3b4cda9b$1...@news.devx.com
Now kindly just FOAD, mmmkay?
--
Joe Foster <mailto:jlfoster%40znet.com> L. Ron Dullard <http://www.xenu.net/>
Joe "Nuke Me Xemu" Foster
> > VB is dead. Long live VB!
>
> I see you finally decided to accept the new VB. Welcome aboard.
Still rewriting that sarcasm detector in VB.NET, are you? Perhaps in
several years it'll work as well as the VB Classic version...
http://joelonsoftware.com/stories/storyReader$47
Zane Thomas
> That's a *lie*, and you know it.
One of your standard responses, name-calling. Weenies, liars, etc.
> Now kindly just FOAD, mmmkay?
I notice you didn't address the off-by-one issue, no surprise there. Btw,
so you think it's ok for you to follow me around the newsgroups and reply
with StandardRants.ChooseOne() but it's not ok for me to reply to your
posts? No surprise there either, zealots can't tolerate differences of
opinion.
Zane
Jim Baker
I'm sick of searching for legitimate information and running into 3rd grade
arguments.
Zane Thomas
> http://joelonsoftware.com/stories/storyReader$47
Still relying on whole-sale quotes instead of something resembling reason
eh?
Btw, Joel is flatout *wrong* about that. There are, as I'm sure most here
recognize, times when the very best thing you can do is through a mess on
the trash-heap and rewrite from scratch.
Zane
Jonathan Allen
> recognize, times when the very best thing you can do is through a mess on
> the trash-heap and rewrite from scratch.
I think he made some good points, and I even tried to live by it in a
project. But when it finally came down it, a total rewrite ended up being
much better than trying to fix an inherently flawed architecture. With a
rewrite (keeping bits and pieces of course), I was able to do in a month
what 10 months worth of patches couldn't.
As for his example, things are still up in the air. Nav6 may have had some
problems, but I heard that they are being addressed in a new release.
Hopefully we can soon see if the rewrite is truly an improvement.
--
Jonathan Allen
"Zane Thomas" <za...@mabry.com> wrote in message
news:#b$PiDQUBHA.1712@tkmsftngp03...
Zane Thomas
> I think he made some good points ...
Sure he did, but he takes such an absolutist position that it detracts from
his otherwise worthwhile comments. Maybe the PHBs like to see that before
they hire him as a consultant. <shurg><g>
> But when it finally came down it, a total rewrite ended up being
> much better than trying to fix an inherently flawed architecture. With a
> rewrite (keeping bits and pieces of course), I was able to do in a month
> what 10 months worth of patches couldn't.
I think there's a pretty fundamental, and obvious, flaw in his argument (at
least in the overly absolute form he presented it):
A design should flow out of *requirements* - an application which meets
those requirements should follow. I think Joel's argument can be seen to be
false if you reduce it to the case where there is some simple application
which initially has a single requirement. Some time goes by, new
requirements come along, and the application is "adjusted" to accomodate
them. We have all experienced what happens as that process is iterated: at
some point the new requirements strain the implementation of the preceeding
requirements to the breaking point. I think there must be a physical law at
work there. Maybe I'll try to identify it and lay my claim to usenet fame.
:-)
Anyway, whether you start with 100 requirments or 1, the unending arrival of
subsequent requirements will eventually break the design.
Zane
Michael Giagnocavo
(less memory, far faster, etc.).
What *is* your problem?
-mike
"Joe "Nuke Me Xemu" Foster" <j...@bftsi0.UUCP> wrote in message
news:uBuJo#PUBHA.2124@tkmsftngp03...
Jonathan Allen
> some point the new requirements strain the implementation of the
preceeding
> requirements to the breaking point. I think there must be a physical law
at
> work there. Maybe I'll try to identify it and lay my claim to usenet
fame.
It sounds like you are already half-way there. You may just need to find a
way to express it in a mathematical formula.
--
Jonathan Allen
"Zane Thomas" <za...@mabry.com> wrote in message
news:uKEm5oQUBHA.1916@tkmsftngp07...
Joe "Nuke Me Xemu" Foster
> Joe,
>
> > That's a *lie*, and you know it.
>
> One of your standard responses, name-calling. Weenies, liars, etc.
Yes, when it's called for, such as when you made your false and
unsupportable assertion about my attitudes toward VB6, liar-boy.
After all, you even *knew* I considered adding inheritance to VB
a good idea! (It's just too bad they didn't add it to *VB*...)
news://news.devx.com/3bba...@news.devx.com
news://news.devx.com/3bba...@news.devx.com
news://news.devx.com/3b97...@news.devx.com
news://news.devx.com/3b3f...@news.devx.com
news://news.devx.com/3b22...@news.devx.com
news://news.devx.com/3ab6...@news.devx.com
news://news.devx.com/3aa0...@news.devx.com
news://news.devx.com/3a7c...@news.devx.com
news://news.devx.com/137_3a...@news.devx.com
> > Now kindly just FOAD, mmmkay?
>
> I notice you didn't address the off-by-one issue, no surprise there. Btw,
What's the matter, never got the hang of SAFEARRAY? Were you simply
incapable even of wrapping it in a class, perhaps even in a template,
that would allow you to continue using your precious a[x] syntax, or
perhaps even STL's for_each? Were lbound() and ubound() methods just
too much for you to cope with?
http://support.microsoft.com/support/kb/articles/Q131/0/86.ASP
Your claims that moderately
skilled programmers can't cope with vb arrays are unfounded, and
likely apply more accurately to you.
"...Blitz++ allows you to choose arbitrary bases, possibly different
for each dimension. For example, this declaration creates an array
whose indices have ranges i=5..8 and j=2..5..."
-- http://oonumerics.org/blitz/manual/blitz02.html#arrays-storage-detailed
"I do think Bruce has a point when it comes to arrays. There are simple
uses for arrays, and a high-level language should - imo - support such
uses. After the compiler gets done there is little cost. For instance,
dimming an array (FirstPartNum, FirstPartNum + TotalParts - 1) (or
something like that) is a useful approach for people writing simple
applications."
-- Zane Thomas, in news://news.devx.com/3b17220b....@news.devx.com
"There are some things to not like in .net, I do think MS changed VB too
much. BitAnd? Zero-based arrays? Why?" -- Zane Thomas, in
news://news.devx.com/39e358ba...@news.devx.com
But that's OK, since everything will be just as easy in vb.net, right?
"It's not the OOPness of it, but the difference in the way common tasks are
done now. Converting a string to an array, for instance, is not nearly as
simple as it used to be and I had to dig in the docs a bit to figure out
what to do." -- Zane Thomas, in news://news.devx.com/3a9b2d6b....@news.devx.com
> so you think it's ok for you to follow me around the newsgroups and reply
> with StandardRants.ChooseOne() but it's not ok for me to reply to your
> posts? No surprise there either, zealots can't tolerate differences of
> opinion.
I'm following *you* around? Just what sort of weasel are you, anyway?
The first post of yours I saw in here was your "Another Good .Net Day".
You really can't stop lying, can you?
Jonas Dygd
> which held on to it and then later passed some integer to a method which
was
> then used as an index into the previously-passed array. Off by one error.
> There is no general way to address that issue other than by requiring
> consistency between languages (well, unless you would resort to the absurd
> suggestion that all languages be aware of and track array-indexing issues
> presented by all other languages).
>
> IMO if you have to choose between zero-based and one-based arrays, then
> zero-based is the best choice because a) it's more efficient in the first
> place and b) calculations of indices into the array are generally more
> efficient since they don't require +1 all over the place.
The simple solution to this problem of passing a non-zero based array to a
C# function is that you don't. The non-zero-based arrays would be internal
to VB.
And, to pass them to other language, you'd use a ToBaseZero keyword to
indicate that you're aware of the problem.
Joe "Nuke Me Xemu" Foster
> Joe,
>
> > http://joelonsoftware.com/stories/storyReader$47
>
> Still relying on whole-sale quotes instead of something resembling reason
> eh?
Just trying to head off yet another of your games of Where's The Cite?
news://news.devx.com/3ae95388....@news.devx.com
news://news.devx.com/3bb3...@news.devx.com
news://news.devx.com/3bacdecc....@news.devx.com
news://news.devx.com/3bb3fb25....@news.devx.com
> Btw, Joel is flatout *wrong* about that. There are, as I'm sure most here
> recognize, times when the very best thing you can do is through a mess on
> the trash-heap and rewrite from scratch.
And salvage *nothing*? Can't even refactor it, huh? (scroll to the end:)
http://greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=006ZLJ
BTW, just what got thrown on the trash-heap when the plug was finally
pulled on "Pyramid"?
Zane Thomas
Zane
Michael (michka) Kaplan
> What language the VB.NET compiler
> was actually written in is irrelevant.
Perhaps to you.... others, who are looking at what language to use, may care
quite a bit about what Microsoft's own choices were, in a world where VB and
C# are both claimed to have so much power? :-)
--
MichKa
Michael Kaplan
(principal developer of the MSLU)
Trigeminal Software, Inc. -- http://www.trigeminal.com/
the book -- http://www.i18nWithVB.com/
Michael (michka) Kaplan
> This is just the tip of the ice berg. There are many things that VB.Net
can
> do that C# can't. Some are just syntactic sugar and automagical
constructs,
> others are actual CLR features that C# didn't implement.
One of the projects I have been working on for MS needed WinForms and
especially GDI+ (for its Uniscribe-ish fuctionality), and I could have gone
with either VB.Net or C#. This was early enough in the product cycle that
regulat builds were not always terrily stable, and it just so happened that
major parts of VB.Net was broken in a particular build I had installed
(fixed in the next build), so I arbitrarily ended up using C#.
It was an easy experience, the most notable feature was UNSIGNED TYPES -- I
had to do a ton of casting between Unicode code points and characters, and
being able to do that without having to lie and sign-extend was
*wonderful* -- so much so that the next thing I was writing was using C#
too.
I had a similar (perhaps more common, perhaps less) need for unsafe code in
a related project.
Now I am not anti-anything here, but its worth pointing out that there are
differences between the languages, and although everyone could have real
reasons to prefer one to the other, all of the actual reasons will fall into
one or more of the following camps:
1) Personal preference for syntax
2) Requirements of the client/boss dictating language choice
3) Actual features of the language
Of the three, I think #1 will be the most common, and #3 will be the least
common. For most apps, it really does not matter. But when it does, its a
big deal....
Jonas Dygd
"Michael (michka) Kaplan" <forme...@spamfree.trigeminal.nospam.com> wrote
in message news:OTbQn0WUBHA.1696@tkmsftngp03...
> "Jonas Dygd" <jo...@novasoftware.se> wrote...
>
> > What language the VB.NET compiler
> > was actually written in is irrelevant.
>
> Perhaps to you.... others, who are looking at what language to use, may
care
> quite a bit about what Microsoft's own choices were, in a world where VB
and
> C# are both claimed to have so much power? :-)
>
I meant irrelevant in this discussion. Not irrelevant in general. :)
Michael Giagnocavo
-Mike
"Michael (michka) Kaplan" <forme...@spamfree.trigeminal.nospam.com> wrote
in message news:OTbQn0WUBHA.1696@tkmsftngp03...
Michael (michka) Kaplan
real relevance to the conversation.
People use C#, or VB.Net, or whatever. Claiming that they are the same is
like saying that men and women both breathe so it makes no difference who
you marry. Many people would beg to differ!
--
MichKa
Michael Kaplan
(principal developer of the MSLU)
Trigeminal Software, Inc. -- http://www.trigeminal.com/
the book -- http://www.i18nWithVB.com/
"Michael Giagnocavo" <nos...@spam.com> wrote in message
news:Oo8IVraUBHA.2120@tkmsftngp05...
Jonas Dygd
> Since almost no one will WRITE such disassmbled code, your comment has no
> real relevance to the conversation.
>
> People use C#, or VB.Net, or whatever. Claiming that they are the same is
> like saying that men and women both breathe so it makes no difference who
> you marry. Many people would beg to differ!
>
Interesting metaphor. But remember that .NET allows you to be
bisexual/-lingual and you don't even have to be monogam!
Michael (michka) Kaplan
human nature and the fact that most people WILL have one companion and stick
with what they know.
--
MichKa
Michael Kaplan
(principal developer of the MSLU)
Trigeminal Software, Inc. -- http://www.trigeminal.com/
the book -- http://www.i18nWithVB.com/
"Jonas Dygd" <jo...@novasoftware.se> wrote in message
news:#Oq$zikUBHA.1412@tkmsftngp03...
Jonas Dygd
> NO, it allows you the freedom to do what you want. But it does not ignore
> human nature and the fact that most people WILL have one companion and
stick
> with what they know.
>
Hmm.. I don't know how you figured that big capital 'NO' into there when you
actually agreed with what I said.
Unless you're trying to tell me that you're not able to write one component
in VB and another one in C# and have them talk to eachother without
problems.
Michael (michka) Kaplan
fact commonly DO that.
Generally speaking, they will not. They will use the language they can or
the one they are directed to.
"Jonas Dygd" <jo...@novasoftware.se> wrote in message
news:#syIuIlUBHA.716@tkmsftngp05...
Michael Giagnocavo
compiler was written in because they generate pretty close IL. So, I don't
see how anyone would care what they were written in, because in the end,
it's the output that counts.
Where did you get the idea about people writing IL? Also, IL isn't
"dissasembled" code.
About your metaphor, it's like saying both men and women breath, and I
need someone to breath into a bag, so their sex would be irrelevant. Better
put, its like saying both men and women can program, so it truly wouldn't
matter who I contract to write a program.
-Mike
"Michael (michka) Kaplan" <forme...@spamfree.trigeminal.nospam.com> wrote
in message news:erd3ZMfUBHA.1512@tkmsftngp05...
Joe "Nuke Me Xemu" Foster
> The issue was not interoperability, it was about whether people would in
> fact commonly DO that.
They may try a phased migration, in which case B# and C# programs and
components will have to interoperate with COM components, giving rise
to creeping NDF catastrophes. Developers' heads are going to roll.
> Generally speaking, they will not. They will use the language they can or
> the one they are directed to.
Then why are certain induhviduals ranting and raving about how array
lower bounds MUST be ZERO in the name of INTEROPERABILITY!!!1!!
--
Joe Foster <mailto:jlfoster%40znet.com> On the cans? <http://www.xenu.net/>
Michael (michka) Kaplan
why. Interoperability is still important to them, though... they want it to
work.
No mysteries here, cards are all down on the table.
--
MichKa
Michael Kaplan
(principal developer of the MSLU)
Trigeminal Software, Inc. -- http://www.trigeminal.com/
the book -- http://www.i18nWithVB.com/
"Joe "Nuke Me Xemu" Foster" <j...@bftsi0.UUCP> wrote in message
news:#waEXNnUBHA.1196@tkmsftngp03...
Jonas Dygd
> The issue was not interoperability, it was about whether people would in
> fact commonly DO that.
>
> Generally speaking, they will not. They will use the language they can or
> the one they are directed to.
>
Huh? I was talking about the restrictions in .NET (or lack there of), not
anything else.
Michael (michka) Kaplan
--
MichKa
Michael Kaplan
(principal developer of the MSLU)
Trigeminal Software, Inc. -- http://www.trigeminal.com/
the book -- http://www.i18nWithVB.com/
"Jonas Dygd" <jo...@novasoftware.se> wrote in message
news:u7N2DHwUBHA.2296@tkmsftngp05...
Joe "Nuke Me Xemu" Foster
> Joe,
>
> Zane
Translation: Irrefutable.
And four seconds of research back it up.
Zane Thomas
> Translation: Irrefutable.
No, that's not it. The truth of the matter is that I don't feel wasting my
time with your silly games. There are people here who are interested in
learning and using vb.net.
Zane
Dan Barclay
<jo...@novasoftware.se> wrote:
>I've been using .NET for a while but I really haven't bothered to look at VB
>until now, since I'm a C++ programmer (C# nowadays).
You may want to keep it that way.
>The one thing that struck me the most was the forcing of paranthesis to not
>only functions with no parameters, but with subs as well. How hard could it
>have been to have kept the old syntax?
No, it would have been trivial for them to do it right.
>And non-zero based arrays are gone, with the CLR standard as an excuse. That
>is stupid, and could have be worked around by just substracting the lower
>index in all references. That is, if the developers had actually wanted to
>create a new version of the VB language.
IL supports non zero based arrays natively. All they really had to do
was use them, straight out. Instead they "redesigned" VB arrays and
implemented them as vectors.
>
>After taking a look at this new language, it looks to me as it was written
>by C++ programmers, who's trying to show VB programmers what's best for
>their own good.
You get the prize. Another lightbulb clicks on. Unfortunately, this
has all been going on since *last* year.
>It wouldn't surprise me if the all the VB compiler does is
>parse through the VB code, converts it to C# and compiles it using the C#
>compiler.
No, they actually went to the trouble of creating a specific, screwed
up, VB compiler. Go figure.
Dan
MS Dev MVP - Visual Basic
Joe "Nuke Me Xemu" Foster
> I've already re-written many things from VB6 to VB.NET, and they work better
> (less memory, far faster, etc.).
>
> What *is* your problem?
That you most likely simply can't write decent code?
--
Joe Foster <mailto:jlfoster%40znet.com> Sign the Check! <http://www.xenu.net/>
Zane Thomas
> > What *is* your problem?
Can't stand the fact that some people are really enjoying .net?
--
Zane Thomas
www.mabry.com/dotnet
Jonas Dygd
> Joe,
>
> > > What *is* your problem?
>
> Can't stand the fact that some people are really enjoying .net?
>
Not only am I enjoying it, I'm living it baby! Life as an ASP programmer has
never been this sweet!