To those who want to see VB become a real programming language
Jonathan Allen
Now is the time to stop lurking and demand that VB.Net not become yet
another hack.
There was a time when the MVP's were cutting edge programmers that helped
others moved forward. Now they are just a bunch little old men scared of the
dark.
And because of their fears, they have screwed over the rest of us. It
appears they are winning their battle to keep VB the same, which will make
life much harder on the rest of us.
How many of you want to write...
If A AndAlso B Then
...for the next 5 to 10 years?
If they have it their way, that is the kind of BS we have to put up with.
And "Why? you ask. Because of simple greed.
Those so called MVP's want to be able to upgrade VB6 code to VB.Net without
any work. That way they can charge big bucks for the "new version" without
actually doing anything. If they really cared about their clients, they
would leave their old code in VB6 and just write new stuff in VB.Net.
Stand up and say to MS that these losers don't represent us. Tell MS that
you were quite because you were happy with how things were going and thought
it was a done deal. Don't let them take away our last chance to see VB
become a truly good language instead of just a hack.
--
Jonathan Allen
Justin Spindler
We don't need hacks that complicate .NET programming by introducing
inconsistencies with the .NET platform. We don't need ambigious operators
that break correct precedence. We don't need VB.NET using a different True
than everyone else just because some morons treated booleans like numbers
instead of booleans, which will mean a True from C# or MC++ or COBOL.NET
might not be True in VB.NET.
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Patrick Steele
<grey...@cts.com>),
> There was a time when the MVP's were cutting edge programmers that helped
> others moved forward. Now they are just a bunch little old men scared of the
> dark.
>
> And because of their fears, they have screwed over the rest of us. It
> appears they are winning their battle to keep VB the same, which will make
> life much harder on the rest of us.
>
> How many of you want to write...
>
> If A AndAlso B Then
No need. Just write:
if( A && B ) {
:)
> Those so called MVP's want to be able to upgrade VB6 code to VB.Net without
> any work. That way they can charge big bucks for the "new version" without
> actually doing anything. If they really cared about their clients, they
> would leave their old code in VB6 and just write new stuff in VB.Net.
I don't want to presume to know any motives behind it. All I know is
that if these changes are due to MVP pressure, than these people carry
more weight than I thought and I don't want to use a language that can
be "pushed" in any direction (including backwards) by this subset of VB
users.
--
Patrick Steele
(pst...@ipdsolution.com)
Lead Software Architect
Image Process Design
Jacob Grass
These changes make my excitement about VB.NET dwindle immensely.
Jacob
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
AdrianL
VB.NET the way they planned it
from the very beginning ? If they want VB 6 compatibility, let them make
some compiler switches, but keep
the new features as default.
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Bob Butler
It is. It has been for a long time now.
> Now is the time to stop lurking and demand that VB.Net not become yet
> another hack.
it's not. VB.Net is a hacked version of VB.
> There was a time when the MVP's were cutting edge programmers that helped
> others moved forward. Now they are just a bunch little old men scared of
the
> dark.
The MVPs are certainly all capable of learning VB.Net and still being
cutting edge. None I have read seem scared, just disgusted.
> And because of their fears, they have screwed over the rest of us. It
> appears they are winning their battle to keep VB the same, which will make
> life much harder on the rest of us.
Nothing is being made harder for people writing new code in VB.Net
> How many of you want to write...
>
> If A AndAlso B Then
>
> ...for the next 5 to 10 years?
Now there I agree with you. That's an ugly construct. Even "AndIf" would
be better but the concept of using a new keyword for new functionality makes
complete sense.
> If they have it their way, that is the kind of BS we have to put up with.
>
> And "Why? you ask. Because of simple greed.
>
> Those so called MVP's want to be able to upgrade VB6 code to VB.Net
without
> any work. That way they can charge big bucks for the "new version" without
> actually doing anything. If they really cared about their clients, they
> would leave their old code in VB6 and just write new stuff in VB.Net.
because they don't want to waste man-hours rewriting code that works now?
It seems more like the greed is on the side of those who want the changes so
they can charge to update code for people. Nobody expects to upgrade
without any work - just without pointless work
Patrick Steele
<butl...@earthlink.net>),
> > Now is the time to stop lurking and demand that VB.Net not become yet
> > another hack.
>
> it's not. VB.Net is a hacked version of VB.
No, VB.NET was a brand new language. It is now being hacked to work
like the old VB.
> Nothing is being made harder for people writing new code in VB.Net
and then you say...
> Now there I agree with you. That's an ugly construct.
Won't ugly constructs make it more difficult for new code/coders?
Justin Spindler
"Bob Butler" <butl...@earthlink.net> wrote in message
news:#gz3QKfvAHA.712@tkmsftngp04...
>
> "Jonathan Allen" <grey...@cts.com> wrote in message
> news:e8dhcuevAHA.2228@tkmsftngp02...
> > To those who want to see VB become a real programming language,
>
> It is. It has been for a long time now.
No, it's always been a second-class citizen, which is exactly why Microsoft
has yet to release any functional commercial applications written in VB.
>
> > Now is the time to stop lurking and demand that VB.Net not become yet
> > another hack.
>
> it's not. VB.Net is a hacked version of VB.
>
No, VB is a hack of attempting to make a 32-bit native compiled language out
of an interpretted language while maintaining fragments of functionality
from 25 years ago. VB.NET was written from the ground up with the new
platform in mind. Many of the new syntax features have been requested since
VB4.
Code that works on a platform should stay on that platform. What happened
to those people who had freethreaded programs written in VB5SP3? Did
Microsoft provide them a way to continue freethreading in VB6? No.
Patrice Scribe
minor thing such as True being 1 instead of -1, I hardly see what you'll do
in this app to exploit VB NET new features... Do you plan to use the NET
framework or ADO.NET in this upgraded code which is really much more work
IMHO ? This is much more work than the time you'll "save" by keeping -1 for
True.
I've seen *now* VB users stuck because of True being -1 in VB. For example
to enable a "RenderState" using the DirectX API, with VB6 you must use :
Direct3DDevice8.SetRenderState RenderState, 1
True doesn't work as the value is -1 and the API match the API exposed to
C++ users that are using TRUE (that is 1). I certainly don't want this in
the whole NET framework and any other new components. Is there still a
dedicated team that will work to write wrappers for use by Visual Basic.NET
so that the values for those boolean are changed appropriatly.
It would be really STUPID to have still those issues just to avoid a minor
thing like this "boolean used as a number issue".
--
Patrice Scribe
"Bob Butler" <butl...@earthlink.net> a écrit dans le message news:
#gz3QKfvAHA.712@tkmsftngp04...
Paul Speranza
I totally agree with your points but think you were a little harsh.
How about this - for all of the people out there that can't remember simple
things like true doesn't equal -1 or that arrays start at 0 maybe Microsoft
can put stickers for their foreheads or giant posters for their cubicles
with the new rules right in the box!
What a shame we have to hack VB.Net. Here was a chance for a clean start and
they are going to blow it.
Regards,
Paul Speranza
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Jeff Key
Enough waiting around. You know more changes are coming down the line no
matter what comes of this current debacle. MS's push is obviously in the
direction of C#. Naturally I'll need to keep coding VB.NET, but I'm going
with C# whenever it's possible from this point on.
-jk
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Rockford Lhotka
rhetoric (Visual Fred, .NOT and so forth). However, I think that you are
directly doing the MVPs serious injustice here - even assuming that MS is
doing these changes merely due to the their input (which I doubt).
No one is getting screwed here - except for those tiny few people who
honestly want to be able to copy-paste VB between C# and add/remove
semi-colons or some lunacy like that. (And those people are NUTS. Are they
also going to demand that Smalltalk.NET conform such that we can copy-paste
to C#? Let's get real! Each language is SUPPOSED to be unique and
different.)
The True value remains a .NET Boolean and is totally compatible with the
rest of .NET. It is just represented as a -1 numeric _within_ VB. We get
backward compatibility without sacrificing .NET compatibility.
Arrays in .NET are more powerful than VB was allowing for. This change
continues to exploit .NET array capabilities while simplifying some
conversion issues.
Personally I believe that VB is so great because it more clearly shows
programmer intent. The addition of keywords that explicitly indicate where I
do and do not want short-circuiting in my conditionals is a GOOD thing. It
is particularly good since it also simplifies some conversion issues while
simultaneously providing us with new capabilities.
It is not just the MVPs who are worried about upgrade issues you know. There
are a LOT of people out there too busy working to spend time on these
newsgroups who are going to _expect_ Microsoft to provide them an upgrade
path from where they are to where we're going.
At the same time, VB.NET _should_ be a first-class language and _should_
provide all the power and capabilities that we've longed for over the past
decade.
In between here there is a delicate balance. When this is done and released
not everyone will be totally happy or totally satisfied, but then that's not
really the goal is it? The goal is to have a practical, pragmatic, powerful
and capable language that provides the maximum possible backwards
compatibility and the maximum possible .NET compatibility.
Are some of the MVPs a bit too conservative? Yes, I think so (sorry guys).
Are some of the pro-change folks too ready to sacrifice the past? Yes, I
think so (and I am pretty damn pro-change).
Before flying off the handle and getting all reactionary (from either side),
it is important to sit back and evaluate the _real_ impact of these changes
in terms of both backwards and .NET compatibility. Personally I am
reasonably pleased with the balance in this case.
Rocky
--
Rockford Lhotka
ro...@lhotka.net http://www.lhotka.net
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Daniel Pratt
I tend to agree, for the most part. For the same reason I didn't think
that it mattered that the integer representation of true was 1, I don't
think it matters that it goes back to -1. I'd be disappointed if
short-circuiting was sacrificed in the bargain, but again, not killer.
What _does_ bother me is that MS is taking more time and spending more
resources to make trivial changes that will hardly satiate the .NOT folks.
Many will have to pay/wait for changes they would rather not have. If it's
really true that the majority of potential VB.NET users want these changes,
then it's only fair. Unfortunately, people who are satisfied with the status
quo tend not to be as vocal as those who aren't.
Regards,
Dan
"Rockford Lhotka" <ro...@lhotka.net> wrote in message
news:OEL3AxfvAHA.2052@tkmsftngp05...
Justin Spindler
as a Boolean, but as a number. VB.NET will coerce True to and from 1 when
it knows you're passing a Boolean True between components, but if we were
only concerned about True and not it's underlying value, that wouldn't be a
problem. People, for some reason, want to pass it's underlying value, and
assume True means anything more than not-zero. So, if someone passes the
underlying value of True to and from VB.NET, they will not get the .NET
equivalent to True, unless Microsoft is going to make VB.NET change every
instance of -1 to 1 and vice versa. If you want compatibility, provide
Microsoft.VisualBasic.Compatibility.VB6.True and System.Environment.True, to
which we can add the namespace of our choice (and add MS.VB.C.VB6 namespace
by default, I don't care.)
As for bitwise operators, the VB5/6 method is a hoaky pile of garbage now,
requiring that you throw parenthesis all over the place because the
operators are ambigious and do not properly follow operator precedence, not
in VB or any other language. Of course, it might appeal to others better to
allow the logical-bitwise mix by default, and support an Option statement.
The array declaration problem will probably only rear it's ugly head in
maintenance and documentation, since I'm sure once compiled Dim a(10) As
Integer will look like 11 integers. In documentation and code, a .NET
developer in another language may misinterpret this. Another contestant for
an Option statement.
VB.NET is a tough cookie. On the one hand, we have the desire to push onto
a platform which looks and acts nothing like VB's framework of the past, but
on the other hand we have the desire to maintain a form of standard with
older versions of BASIC. 99% of the change, and 99% of the learning curve
of VB.NET lies in .NET itself, it's framework, and what it requires of a
language in order to operate well with it. These three changes wont make
any of that easier, instead it poses the opportunity for logic errors and
discrepencies in a supposedly language agnostic environment. VB is a
powerful language, but traditionally many features are just omitted to
support ease of use. Option exists, and maybe it should be given more of a
workout. Then, of course, you run the risk of having developers use
different Options, and writing code that becomes even harder to maintain.
"Rockford Lhotka" <ro...@lhotka.net> wrote in message
news:OEL3AxfvAHA.2052@tkmsftngp05...
Patrick Steele
<dprREMOVE...@hotmail.com>),
> What _does_ bother me is that MS is taking more time and spending more
> resources to make trivial changes that will hardly satiate the .NOT folks.
I would also like to know where they found the time to add new features
(like the rumored "AndAlso"), change existing ones (like no more short-
circuiting of IF's), and new functionality (easier implementation of
non-zero based arrays)...
Too bad this time couldn't have been spent moving the language forward
with operator overloading, unsigned ints or XML documentation. :(
Michael Venuto
talking up .NET myself, and VB.NET in particular. Full stop.
Michael J. Venuto
Gregor R. Peisker
What they're doing now is ridiculous. There are a number of things that
could be done to ease migration without sacrificing new features and
interoperability. Perhaps some things are being done in this direction right
now, but this thread is about the recent Boolean, array, and operator
changes.
This will not help migrating. Instead, now things are complicated further,
in order to pay respect to Worst Practices. Hey, if you're writing a VB6 app
now, just go ahead and test for -1 instead of True! No worries, until you
remember all the other items on the incompatibility list.
IMO, the discussions about the true meaning of True, logical expressions,
and GoSub got totally out of hand. What are the longest running threads in
this ng? Right, those with "GoSub" in the subject line.
So is the .NOT camp happy now? If so, then I seriously doubt that language
stability or portability was the motivation behind all the noise.
Regards,
Gregor
Rick Brandt
news:OEL3AxfvAHA.2052@tkmsftngp05...
Kudos, Rockford, for being a sensible voice here. And kudos, Microsoft,
for listening to developers then engineering some smart improvements.
Patrick Steele
no_spam!mi...@grsinc.com>),
> I second, third, and fourth that. I have to admit that I was going around,
> talking up .NET myself, and VB.NET in particular. Full stop.
.NET is still a great platform to develop in. It just seems MS wants to
keep VB as a "second class" citizen. C# seems to be the safest bet for
the future.
Cowboy
I have embraced the changes of .Net, having spent many nights hammering out
code to fit the new paradigm. Overall, I like the fact that the language has
become more OO and is compatible with other languages. I think most MVPs
would agree with the fact that change is inevitable and that VB ultimately
had to lose some compatibility to enter the arena at the same time as other
languages. And, I agree wholeheartedly to this changes and have joined the
bandwagon.
However, there are some changes that are just plain stupid. Some are
arbitrary decisions that do nothing to help further the language, while
others emphasize the fact that C# is king and VB.Net is considered a red
headed stephchild.
It is not, IMHO, a fair analysis to assume that criticism of some of the
changes in VB.Net is the same as trying to hold VB behind. Also, it is
unfair to label all MVPs as a "bunch of little old men scared of the dark."
The analysis may hold true in some instances, but hasty generalizations are
considered a form of logical falicy. Better to attack the messenger with the
message than to group all individuals under a single label.
--
Gregory A. Beamer
MVP; MCP: +I, SE, SD, DBA
************************************************
Think Outside the Box!
************************************************
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
Bob Butler
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
news:MPG.153671ecd...@msnews.microsoft.com...
> In article <#gz3QKfvAHA.712@tkmsftngp04> (from Bob Butler
> <butl...@earthlink.net>),
> > > Now is the time to stop lurking and demand that VB.Net not become yet
> > > another hack.
> >
> > it's not. VB.Net is a hacked version of VB.
>
> No, VB.NET was a brand new language. It is now being hacked to work
> like the old VB.
>
> > Nothing is being made harder for people writing new code in VB.Net
>
> and then you say...
>
> > Now there I agree with you. That's an ugly construct.
>
> Won't ugly constructs make it more difficult for new code/coders?
LOL. Not to the extent that having to review reams of existing code does.
I like that they are adding a new keyword for new functionality. I dislike
their particular choice.
Bob Butler
"Justin Spindler" <halo...@bellatlantic.net> wrote in message
news:#C$OVSfvAHA.940@tkmsftngp04...
>
> "Bob Butler" <butl...@earthlink.net> wrote in message
> news:#gz3QKfvAHA.712@tkmsftngp04...
> >
> > "Jonathan Allen" <grey...@cts.com> wrote in message
> > news:e8dhcuevAHA.2228@tkmsftngp02...
> > > To those who want to see VB become a real programming language,
> >
> > It is. It has been for a long time now.
>
> No, it's always been a second-class citizen, which is exactly why
Microsoft
> has yet to release any functional commercial applications written in VB.
It's always been perceived as such by many. It's also been an extremely
useful tool for many developers. I don't worry about what others think of
the tool if it gets the job done for me.
Rick Brandt
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
> (like the rumored "AndAlso"), change existing ones (like no more short-
> circuiting of IF's), and new functionality (easier implementation of
> non-zero based arrays)...
Maybe they took it away from the time allocated for the VB6 conversion
wizard, which needn't be so complicated now. Think about it.
Bob Butler
"Patrice Scribe" <http://www.chez.com/scribe/en/mail.htm> wrote in message
news:e2CkQVfvAHA.1452@tkmsftngp02...
> Really Bob, if you want to be able to upgrade without having to handle a
> minor thing such as True being 1 instead of -1,
The particular issue is minor and I do not expect it to affect me or any
migrations I do personally since I do not rely on the value of True. I see
this as representative of the disdain that MS has for the language and their
willingness to make unnecessary changes at any time leaving any/all VB code
vulnerable.
Justin Spindler
issues posed no problems for upgrading. Logical-Bitwise operations were
usually migrated successfully, and those that were ambiguous marked with a
'UPDAGE statement in the task list. VB arrays used a class in the
Compatibility.VB6 namespace to emulate SAFEARRAY behavior, whether that be
lowerbounds other than 0, etc. True is True is True, but if you tried to
stick it into a number, an 'UPGRADE statement was added to the task list.
So, in actuality, they're spending more time removing things that were
already completed.
"Rick Brandt" <ri...@dbmv.com> wrote in message
news:u7BjRsgvAHA.1108@tkmsftngp03...
Patrick Steele
<butl...@earthlink.net>),
> LOL. Not to the extent that having to review reams of existing code does.
> I like that they are adding a new keyword for new functionality. I dislike
> their particular choice.
If they're bent on getting backwards compatibility, why not introduce an
"IF" that is short-circuited:
IFsc [expression]|[expression]|...
while leaving the existing IF:
If [expression]|[expression]|...
to fully evaluate each expression?
Patrick Steele
<ri...@dbmv.com>),
>
> Maybe they took it away from the time allocated for the VB6 conversion
> wizard, which needn't be so complicated now. Think about it.
LOL!! :)
Larry Serflaten
"Bob Butler" <butl...@earthlink.net> wrote
> Nothing is being made harder for people writing new code in VB.Net
You could say the same thing about Empty and Null, but I don't think it
helps the argument any :-P
I thnk it would be better to address the AND OR and NOT properties,
than say the -1 issue is the key. They are linked, and should be addressed
accordingly.
The way I see it, we have the VB group saying, look, this -1 works well
in all these logical and bitwise combinations. Dividing up that functionality
into separate operations adds complexity.
Then the C types say that they have always used +1, and now the entire
framework uses +1, so you have to use +1 too.
Perhaps the answer will be a few compiler switches, I don't know, but
I do want to say that I would rather see VB perform well with the framework
it runs on, than to surrender major functionality due to backward compatablitly.
Taking 'time to market' out of the equation, do we know where the problems
actually are, prohibiting backward compatability? That type of dicussion would
easitly get out of my league, (I still know very little of the OO framework) but a
summary would be interesting....
LFS
Bob Butler
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
> In article <eVVaUrgvAHA.436@tkmsftngp04> (from Bob Butler
> <butl...@earthlink.net>),
> > LOL. Not to the extent that having to review reams of existing code
does.
> > I like that they are adding a new keyword for new functionality. I
dislike
> > their particular choice.
>
> If they're bent on getting backwards compatibility, why not introduce an
> "IF" that is short-circuited:
>
> IFsc [expression]|[expression]|...
>
> while leaving the existing IF:
>
> If [expression]|[expression]|...
>
> to fully evaluate each expression?
I thought about that but it would not help in other expressions:
dim result As Boolean
result = (a And b)
Maybe sc on the keywords?
If a scAnd b Then <g> almost as bad as AndAlso...
If a AndIf b Then
If a OrIf b Then
To be honest I wish they had made And/Or/Not/Xor logical and added the
BitXXX versions from the start since the logical operators are probably more
used. Unfortunately they didn't and I value the backward consistency more
than the language "cleansing".
Patrick Steele
<butl...@earthlink.net>),
> I thought about that but it would not help in other expressions:
>
> dim result As Boolean
> result = (a And b)
Doh! :)
> Maybe sc on the keywords?
> If a scAnd b Then <g> almost as bad as AndAlso...
> If a AndIf b Then
> If a OrIf b Then
Hmmmm... This is not going to be easy... :(
Bob Butler
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
> In article <Od0Mp8gvAHA.2304@tkmsftngp02> (from Bob Butler
> Hmmmm... This is not going to be easy... :(
I'll bet that's what MS has been saying <g>
Justin Spindler
wrappers for DirectX consider this "adding back VB6 bugs," especially where
the bitwise operations are concerned
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
> To those who want to see VB become a real programming language,
>
> Now is the time to stop lurking and demand that VB.Net not become yet
> another hack.
>
Corrado Cavalli
years of feeling like a 2nd choice programmer i'd like to have a #1 language
knowledge.
Just wonder why M$ always wants VB community to be a step back :(
Corrado
John Timney (MVP)
Thats a very generalised statement about the MVP group, of which over 500+
of the 600 are not VB MVP's. I for one have never complained about any
changes to VB to make it become VB.NET and I actually encourage the new
language syntax and structures in .NET.
The requests for changes did not only come from the MVP's alone - MS
recieved an awful lot of feedback to implement these changes for the beta
groups, conferences and many other sources.
Regards
John Timney
Microsoft MVP
(http://support.microsoft.com/support/mvp/program.asp)
Co-Author Professional JSP
ISBN: 1-861003-62-5
Randy Birch
without
: any work. That way they can charge big bucks for the "new version" without
: actually doing anything. If they really cared about their clients, they
: would leave their old code in VB6 and just write new stuff in VB.Net.
Well Jonathan I'm convinced. The rumors were true .. you are an idiot.
--
Randy
Kathleen Dollard-Joeris
You are a fool speaking from absolutely no knowledge of the real situation.
None.
Nada.
I am not a little old man, nor do I ever intend to be.
--
Kathleen
(MS-MVP)
Reply in the newsgroup so everyone can benefit
--
Kathleen Dollard-Joeris
I can't comment beyond repeating what was said in the releases and asking
you to read more carefully.
> We don't need hacks that complicate .NET programming by introducing
> inconsistencies with the .NET platform. We don't need ambigious operators
> that break correct precedence. We don't need VB.NET using a different
True
> than everyone else just because some morons treated booleans like numbers
> instead of booleans, which will mean a True from C# or MC++ or COBOL.NET
> might not be True in VB.NET.
The value of True is True. In the CLR and all interop it is as defined in
the spec (1). True in C#, MC++, Cobol.Net and VB.Net will ALL be the same.
True (which happens to currently be defined as 1). There has been no
proposal to change this!
The only change is the value of True when converted to an signed integer
within VB. Do you know of any scenario outside compatibility with past VB
behavior that would be effected by the value of True when such a conversion
is performed. It seems a bizarre thing to do outside this context, but if
there is a reason for someone to do it in another context, I'd like to hear
about it.
Let's PLEASE keep this change in the correct context. It does _not_ affect
interop with other .Net languaes. (If you think it does, please explain). No
one has suggested changing the underlying, CLR value of True.
True is True.
Justin Spindler
coerced into a number are different. If people were using True and not
worrying about the numbers, which wouldn't be an issue. Apparently people
have relied on the underlying values of True for processing, so they will
likely continue. Therefore, when passing the underlying value of True to
another component, it's no longer True. This happens now with COM and VB.
I have several third-party COM components where Boolean values do not
evaluate correctly to True, and require a spin through CBool(Cint(x)) in
order to work.
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message
news:OLQ2tkivAHA.1824@tkmsftngp02...
Kathleen Dollard-Joeris
This is true _only_ if you pass the CInt value of True. You planning to do
that? (guessing at final syntax in the following)
ToMyCSharpFunction Convert.ToInt32(bMyBooleanValue)
Seriously? You anticipate doing that?? Can you tell me _why_ you would use
the code above rather than the code below? Just one scenario?
ToMyCSharpFunction bMyBooleanValue
Passes 1, or at least sure as hell should (I haven't see final details on
this any more than you have).
Justin Spindler
with? If they're not, why is this an issue, and why can't the underlying
value remain 1?
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message
news:upBYruivAHA.2252@tkmsftngp02...
Kathleen Dollard-Joeris
You know if you weren't there, I might be able to dredge up some benefit of
doubt and simply think you were ignorant. Since you _were there, the
complete fabrication in this message is nothing short of disgusting.
Trivially: MVPs are not afraid of vb.net. None of us would gain by our
clients/ companies being able to port VB6 code more smoothly. I don't think
any wish VB.Net to be like VB6, because it can't be.
Significantly: you know that what was presented to them verbally by myself
and others and what was described in the press release differ. I am going to
find out what they are really saying on a technical level so I understand
what they really intend to do and what it means before I comment.
There are technical issues here. Things that _should_ be discussed. Details
are not available yet. Crawl back in your hole until you have real
information and can discuss these issues professionally.
Kathleen Dollard-Joeris
Read my lips!!!
The underlying value of True is True, which happens to be defined as 1 in
the CLR spec.
It is an issue in some conversions from VB6. Otherwise it just doesn't
matter. If there is a reason someone in their right mind would pass a
boolean converted to an unsigned issue to C# I would like to hear it.
This allows VB6 converted code with Option Strict Off to still behave as
expected in bitwise operations. That is it. Who else cares? Who does it
hurt?
Joe "Nuke Me Xemu" Foster
> That's the problem, the value of the CLR True and the VB.NET True when
> coerced into a number are different. If people were using True and not
> worrying about the numbers, which wouldn't be an issue. Apparently people
> have relied on the underlying values of True for processing, so they will
> likely continue.
We wouldn't have had to if IIf and similar functions weren't so completely
buggered. It really seemed to say something when my Iff function usually
outperformed IIf:
Function Iff (ByVal X As Boolean, ByVal Y As Variant, ByVal Z As Variant) _
As Variant
If X Then Iff = Y Else Iff = Z
End Function
What the heck is IIf doing under the covers? Whatever it is, its hands
definitely aren't visible!
--
Joe Foster <mailto:jfo...@ricochet.net> Greed = God? <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!
Kathleen Dollard-Joeris
This is Dan Barclay's explanation of the issues, if you have not seen it.
http://www.mvps.org/vb/tips/truth.htm
Justin Spindler
it evaluates the True and False portions in their entirity, in such a way
that the following code would not prevent a DivideByZero exception:
x = IIf(Not y = 0, x / y, 0)
"Joe "Nuke Me Xemu" Foster" <j...@bftsi0.UUCP> wrote in message
news:eILOPBjvAHA.2292@tkmsftngp02...
Joe "Nuke Me Xemu" Foster
> IIf is a bad construct in general considering it uses all Variants and that
> it evaluates the True and False portions in their entirity, in such a way
> that the following code would not prevent a DivideByZero exception:
>
> x = IIf(Not y = 0, x / y, 0)
That's part of what I mean by its being buggered. The compiler should be
smart enough to detect when it's safe to transparently replace IIf with
something that certainly doesn't involve pissing about with Variants or
maybe even replace it with something like this:
if not y = 0 then x = x / y else x = 0
Perhaps all three arguments should always be evaluated, perhaps not.
Perhaps there should be an "Option LazyEvaluation"? I like Option. It
lets me see what's going on without clicking around in tabbed dialogs.
--
Joe Foster <mailto:jfo...@ricochet.net> Sacrament R2.45 <http://www.xenu.net/>
Bill McCarthy
What is this rubbish. not only are your facts wrong, and I mean seriously
wrong, you also decide to start posting this kind of crap throwing insults
at people, rather than try to focus on the technical issues.
Grow up Jonathan, and please try to keep your posts of a technical nature in
the future. It might also pay to try to get your facts straight too. Because
you are wrong, so very very wrong !!
JFTR: What I have asked for is to make the language clearer , less
ambiguous, less bug prone. Your claims are just crap, and it seems you
resort to this type of childish insult throwing because you can't handle the
technical issues.
Grow up Jonathan.
"Jonathan Allen" <grey...@cts.com> wrote in message
news:e8dhcuevAHA.2228@tkmsftngp02...
> To those who want to see VB become a real programming language,
>
> Now is the time to stop lurking and demand that VB.Net not become yet
> another hack.
>
> There was a time when the MVP's were cutting edge programmers that helped
> others moved forward. Now they are just a bunch little old men scared of
the
> dark.
>
> And because of their fears, they have screwed over the rest of us. It
> appears they are winning their battle to keep VB the same, which will make
> life much harder on the rest of us.
>
> How many of you want to write...
>
> If A AndAlso B Then
>
> ...for the next 5 to 10 years?
>
> If they have it their way, that is the kind of BS we have to put up with.
>
> And "Why? you ask. Because of simple greed.
>
> Those so called MVP's want to be able to upgrade VB6 code to VB.Net
without
> any work. That way they can charge big bucks for the "new version" without
> actually doing anything. If they really cared about their clients, they
> would leave their old code in VB6 and just write new stuff in VB.Net.
>
Patrick Steele
<kjo...@noemailplease.com>),
> Justin,
>
> This is true _only_ if you pass the CInt value of True. You planning to do
> that? (guessing at final syntax in the following)
>
> ToMyCSharpFunction Convert.ToInt32(bMyBooleanValue)
>
> Seriously? You anticipate doing that?? Can you tell me _why_ you would use
> the code above rather than the code below? Just one scenario?
I'm sure Justin doesn't have a scenario where *he* would use it.
However, if the integer value of True is such a non-issue, why change it
at this point in VB?
In fact, this true/false/-1/0/+1 issue has come about from the fact that
people (for whatever, unknown reason) are *using* the integer value of
true. If they start passing that around (again, I don't know why), it
could cause interop problems.
And VB.NET programmers will have this to deal with:
CInt(boolean_of_true) <> boolean_of_true.ToInteger()
Again -- for those of us who never convert True to an integer, this will
never be a problem. But, based on the changes that *appear* to be in
the works, the integer value is something people care about -- probably
for a reason... :O
Bob O`Bob
> "Jonathan Allen" <grey...@cts.com> wrote
> > To those who want to see VB become a real programming language,
>
> It is. It has been for a long time now.
>
> > Now is the time to stop lurking and demand that VB.Net not become yet
> > another hack.
>
> it's not. VB.Net is a hacked version of VB.
>
Not even that. It's a hacked-up version of C# that somewhat resembles VB.
I was with a group of 'softies at VBITS-SF this year, and attempted to
make a **joke** about VB.Net being reduced to the point where all it
would take is a preprocessor and the C# compiler.
Suddenly I thought I was in a commercial for Staples...
"yeah - we've got that" they said, and they were perfectly serious.
That's not what I consider a "first-class" language. It's a totally
neutered port of some other language. MS have been going on and on
about how one's choice of language should be a personal preference,
and all languages should be able to do the same things.
DOES NO ONE REMEMBER Ada?
Languages are *supposed* to have strengths, differences, and yes,
weaknesses, at least relative to one another.
Otherwise, there's no point in having more than one.
Bob O`Bob
--
Microsoft MVP - VB
Bob O`Bob
>
> In article <us6$p9fvAHA.2228@tkmsftngp02> (from Daniel Pratt
> <dprREMOVE...@hotmail.com>),
> > What _does_ bother me is that MS is taking more time and spending more
> > resources to make trivial changes that will hardly satiate the .NOT folks.
>
> I would also like to know where they found the time to add new features
> (like the rumored "AndAlso"), change existing ones (like no more short-
> circuiting of IF's), and new functionality (easier implementation of
> non-zero based arrays)...
>
> with operator overloading, unsigned ints or XML documentation. :(
No it would take almost ZERO time.
The "And" of dotnet beta 1 is now spelled "AndAlso", has the short-circuting, etc.
The "BitAnd" of dotnet beta 1 is now spelled "And", works like "And" always has, etc.
...and so on.
Very simple. A few spelling changes in the symbol table and it's done.
This isn't going to hurt you beta 1 zealots; you lose nothing, and the
product hadn't shipped yet anyway.
Bob O`Bob
> As for bitwise operators, the VB5/6 method is a hoaky pile of garbage now,
> requiring that you throw parenthesis all over the place because the
> operators are ambigious and do not properly follow operator precedence, not
> in VB or any other language.
Can you elaborate on this, please?
What is it of the current behavior that you object to?
Bill McCarthy
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
news:MPG.1536f2f52...@msnews.microsoft.com...
>
> In fact, this true/false/-1/0/+1 issue has come about from the fact that
> people (for whatever, unknown reason) are *using* the integer value of
> true. If they start passing that around (again, I don't know why), it
> could cause interop problems.
>
Well see this is the problem with disinformation. The representation of
Boolean is False or Not(False), as in only two states. This does not
correspond to any particular values until it is expressed in another type.
For exampel, a 32 bit signed integer, the two states are 0 and not(0) which
equates to 0 and -1. In a unsigned byte however the two states are 0 and
not(0) which equates to 0 and 255. The bitwise not for the type that
represents the bistates is the most robust representation you can get. It
ensures that a bitmaks test with the default not(0) value will always return
True is the value being tested is non zero. With 0,1, you cannot do that
instead yo unarrow the possibilities to an equality test, and hence are
testing for tri-state or a bistate that breaks the Boolean bistate rules.
> And VB.NET programmers will have this to deal with:
>
> CInt(boolean_of_true) <> boolean_of_true.ToInteger()
>
Well you already have that problem with Cstr compared to .ToString .
Oh and not only was there never any .ToInteger, there won't be either.
Jens Samson
I just sent an e-mail to ms...@microsoft.com and vbw...@microsoft.com to
express my discontent.
Jens
"Justin Spindler" <halo...@bellatlantic.net> wrote in message
news:OHFzW3evAHA.2228@tkmsftngp02...
> VB POWER!
>
> We don't need hacks that complicate .NET programming by introducing
> inconsistencies with the .NET platform. We don't need ambigious operators
> that break correct precedence. We don't need VB.NET using a different
True
> than everyone else just because some morons treated booleans like numbers
> instead of booleans, which will mean a True from C# or MC++ or COBOL.NET
> might not be True in VB.NET.
>
> "Jonathan Allen" <grey...@cts.com> wrote in message
> news:e8dhcuevAHA.2228@tkmsftngp02...
> > To those who want to see VB become a real programming language,
> >
> > Now is the time to stop lurking and demand that VB.Net not become yet
> > another hack.
> >
Patrice Scribe
languages (easing passing parameters between cross languages modules). I
have still to look into the other options to make up my mind...
--
Patrice Scribe
"Bob Butler" <butl...@earthlink.net> a écrit dans le message news:
#AreGtgvAHA.1400@tkmsftngp05...
>
> "Patrice Scribe" <http://www.chez.com/scribe/en/mail.htm> wrote in message
> news:e2CkQVfvAHA.1452@tkmsftngp02...
> > Really Bob, if you want to be able to upgrade without having to handle a
> > minor thing such as True being 1 instead of -1,
>
> The particular issue is minor and I do not expect it to affect me or any
> migrations I do personally since I do not rely on the value of True. I
see
> this as representative of the disdain that MS has for the language and
their
> willingness to make unnecessary changes at any time leaving any/all VB
code
> vulnerable.
>
>
>
Jens Samson
zillion times.
I don't care about the value of True, you can have your way if you want to,
but I DO NOT WANT TO WRITE
' If A AndAlso B Then
...for the next 5 to 10 years'
That is the real point.
This syntax is UGLY. We'll be laughed at by everyone.
And every mature language has short-circuiting with the simple boolean
operators. When I first did some coding in VB I did not even understand why
VB did no short-circuiting. I thought that every language was smart enough
to know that if the first operand is false the complete expression can never
be true with an AND.
As everybody thinks VB.NET is a rewrite, it was the time to clean this up.
If we had to take all this bagage with us for ever you would in 200 years
still see MS OS's that support DOS. Would that be logical ?
Anyway, they had the chance but chose not to take it. I fear it'll be
screwed up forever now.
Jens
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message
news:eccpaCjvAHA.2012@tkmsftngp05...
Jens Samson
can all teel then just what is the truth ?
A. Is the -1 stroy True ?
B. Is the Dim x(10) as Integer being an array of 11 elements true ?
C. Is the AndAlso story true ?
D. If all those changes are being made and you MVPs did not request then
did you oppose to them ?
I for one do not know anything about the facts except for what I have read
here. And I have read that this will happen. Nobody yet has said that it
will not happen.
Instead of attacking Jonathan you would all get us much further if you were
to tell your side of the story.
As for me, I just would like VB to be a serious alternative to C(++/#) and
things like AND being bitwise instead of logical and things like AndAlso
will certainly not help this matter.
Sincerely,
Jens
Bill McCarthy
"Jens Samson" <je...@esalar.be> wrote in message
news:#6M$j$mvAHA.1732@tkmsftngp03...
> Instead of attacking Jonathan you would all get us much further if you
were
> to tell your side of the story.
Let's get some facts straight first please Jens. It was Jonathan Allen who
said :
"There was a time when the MVP's were cutting edge programmers that helped
others moved forward. Now they are just a bunch little old men scared of the
dark. And because of their fears, they have screwed over the rest of us."
so why aren't you asking Jonathan for proof, for facts. You seem so willing
to accept his point of view, willing to happily ignore his slanderous
insults, yet when I denounce what his has said, you then tell me I am
attacking Jonathan. Please have a relook at this. It was jonathan who made
the slanderous comments based on pure fiction and speculation. I have
clearly rebutted what he has said.
I cannot at present comment on what MS may or may not be intending to
implement. If you do have a specific question as to my opinion on any
certain language issue I would be most happy to answer that here in public,
but cannot comment on MS's behalf, only on what my stance is.
I hope that makes the situation somewhat clearer for you. What Jonathan has
said is false, lies, insults and devoid of any technical merit.
Jens Samson
> "There was a time when the MVP's were cutting edge programmers that helped
> others moved forward. Now they are just a bunch little old men scared of
the
> dark. And because of their fears, they have screwed over the rest of us."
I think that impression comes from the fact that Dan Barclay is the one in
this NG that is opposed to most of the changes in VB.NET combined with the
fact that the other MVPs don't say that they do not agree with him.
It is the silence of the other MVPs, combined with the fact (or is it not a
fact yet, it still is not clear for me) that due to the attitude toward
change of people like Dan Barclay (this is not a personal attack, he may
have his reasons, but he does not want that kind of a change) some of the
language changes will be rolled back that makes the MVP group as a whole
look like they were the ones asking for it.
The MVP group as a whole has done nothing to get rid of that impression,
neither have any MVP members (I have to be careful with this because I don't
really know who is MVP or not, I, and most of the people in this NG, can
only know it by looking if the word MVP appears in the signing of the mail)
clearly stated that they did not want the rollback of the changes as they
are now decided or proposed.
So if we are reading this wrong, or if some of you don't feel like we think
you feel, then I for one appologize for that, but you maybe might let it
know to us so that we (the supporters if VB.NET and not going back to VB6)
know that we are not alone.
Jens
Bill McCarthy
You raise some issues there, which I will try to address.
First let me say I am VB MVP. Sorry if that was not clear. I do not include
MVP in my sig, because I don't post a sig, and becasue I would always rather
thigns be looked at and evaluated based on technical merit regardless of who
posted it.
Your grouping of all MVPs is the kind of, dare I say, McCarthyism, that
jonathan Allen is trying to project, not lookingat the technical issues, but
rather trying to blacklist a group of individuals based on their membership
of a club or organisation. It is atypical of scare mongering tacticks used
when people are not willing to look at the facts.
It is important to realise that MVPs are all individuals. We all have our
own opinions. Sometimes these opinions may be the same. That is not to say
that is because of some conspiracy, but rather is more likely because there
is technical merit in that opinion.
MVPs are individuals who Microsoft has recognised as providing peer to peer
support with Microsoft products, and are often leaders in their fields. Each
one is given the MVP award based on their contribution to the online
communities. Each MVP is an individual, and individual Microsoft has
recognised to have given a high level of technically accurate and helpful
information/support for Microsoft products.
So when you say the MVP group as a whole, who are you referring to ?
Microsoft, as it is their program, or do you expect all the MVPs to waste
their time continually correcting the dis-information that Jonatahn is
trying to spread. I have clearly stated that what Jonathan is said is false.
I know you would like more information as to what Microsoft is intending, I
think all of us would. If that is the case you should be targetting your
questions to them, it is they who are implementing the changes (if any).
As to the issue of Dan's stance, I would have to know exactly what you are
talkign about to comment further accurately. If you are talking about
boolean values or bitwise operators, then on those two issues I would tend
to agree with Dan that MS has it wrong, and if the rumours are correct then
MS also agrees that they have it wrong. As to the best solution, I have
different opinions from others on this, and off the top of my head I can't
recall what Dan's opinion on the best way to address them is. I am in favour
of introducing the boolAnd, boolOr operators.
But more importantly, regardless of whatever solution you think may be the
best, it is important to realise why Dan, and why myself and others have
been out spoken on soem of these issues. There are problems when moving from
VB6 to VB.NET, changes in the meaning of keywords brings about a greater
risk of bugs in code, of code failing to work. It increases the amount of
work to do when maintaining code, as the ambiguities mean it is necessary to
look further to try to determine the full intent of the code's author. This
amounts to time, money, lost production, slower uptake of the product etc.
One thing you can be sure of, is most of, if not all of the VB MVPs love VB.
I persoanlly want VB.NET not only to be a huge success, but a great success.
The best damn programming language there is. We have the base, the base of
over 3.5 million VB developers, the base of the best existing RAD software
there is today. I want us to extend that base, not abondon it. I want VB.NET
to be the best damn programming language there is.
"Jens Samson" <je...@esalar.be> wrote in message
news:#fuDW#nvAHA.2120@tkmsftngp04...
Jens Samson
VB.
> I persoanlly want VB.NET not only to be a huge success, but a great
success.
> The best damn programming language there is. We have the base, the base of
> over 3.5 million VB developers, the base of the best existing RAD software
> there is today. I want us to extend that base, not abondon it. I want
VB.NET
> to be the best damn programming language there is.
Bill,
Regardless of all the rest I am sure that Jonathan also wants VB.NET to be a
huge success, so do I.
Maybe we are all shouting at the wrong people but that is because people who
state their opinion usually are targetted (so have I on this NG).
And as MS does not state anything for the time being (the
msdn.microsoft.com/net page looks as dead as it can be) and the rumour has
been spread in this NG that the MVPs went to MS, this means that we feel
that the MVPs have a lot of influence at MS.
Given all the rumours I (I will not speak for anyone else) feel that the
MVPs (or some of them) are behind the announced changes.
Anyway, all this bitching could be stopped if all the parties were to state
an official anouncement.
My questions remain (and I do understand that maybe you can not talk about
it).
1. Is it true that those three changes will be rolled back ?
2. Who was the asking party for that ? I have a hard time believing it was
the majority of the VB6 crowd as lots of them do not code professionaly, and
even more of them have not even played around with VB.NET.
So it must have been the people who have a lot of legacy code and fear that
upgrading will be a hell for them. And that's what I am affraid of, because
if they managed to get MS to do this, they maybe can manage a whole bunch of
other rollbacks, because those three changes are hadly going to make a
difference in the conversion.
3. If the first point is true and the articles on ZDNet and CNet are true,
why did they choose to do it this way when in this NG there are a lot of
suggestions to do it another way that would not cause so much discontent on
both sides.
Jens
Bob Butler
> Good question. Why is anybody passing the underlying value of True to
begin
> with? If they're not, why is this an issue, and why can't the underlying
> value remain 1?
I do not believe that the main issue is passing the value, it's that there's
a fair amount of code that does things like this:
x = 10 - (a>b)*5
It relies on VB coercing "True" to -1 or 0 and changing the value to +1
makes it fail. Is it bad coding? Probably. Is there a lot of it and will
it cause a lot of problems for people trying to move existing code into
VB.Net? Definitely. The change MS is making keeps VB.Net consistent with
VB6 internally and consistent with the CLR externally and that seems to me
to be a very reasonable compromise.
There will still likely be issues such as people passing CInt(boolean)
values or reading/writing files and getting different values now but my
guess is that those are fewer and further between.
Justin Spindler
operators do not follow operator precedence, which becomes an issue and
propogates logic errors in bitmasking. This becomes evident when trying to
use such libraries as DX, and you have to throw around parenthesis to clean
up. Bitwise operators should happen before comparison operators and
comparision operators should happen before logical operators. Instead, we
have bitwise operations occuring after comparison operators, left to right,
in the same precedence as logical operators.
Personally, I think all of these changes should exist in something you can
turn on or off. For example, make it so that VB5/6 behavior occurs if
Microsoft.VisualBasic.Compatibility.VB6 is imported, and permit that
namespace to be imported as a default (like you can with Option Explicit in
VB5/6.)
Probably the one tradition in BASIC is that it's culled everytime it evolves
forward. QBASIC was hardly the BASIC of Apple II/e or Commodore 64, Visual
Basic is hardly like QBASIC, and the next logical step is that VB.NET is
hardly like Visual Basic, which is pretty well accomplished given the
extreme changes to the platform which require extreme changes to the code
structure and syntax. Probably the only set of these BASIC derivatives that
have been standardized (StandardBASIC) are the Apple II/e and Commodore 64
versions. Legacy is not always a good thing. It's the reason Win9x was so
easy to crash (CLI anybody?), and the reason we're all still using a
processor that has to operate at 1.5Ghz in order to meet the operational
speed of other CPUs that run a third as fast.
"Bob O`Bob" <b...@cluestick.org> wrote in message
news:3ACD48...@cluestick.org...
Larry Serflaten
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message news:eccpaCjvAHA.2012@tkmsftngp05...
> Justin,
>
> This is Dan Barclay's explanation of the issues, if you have not seen it.
>
> http://www.mvps.org/vb/tips/truth.htm
>
Don't forget to take a bag of salt with you....
To me, it looks a lot like Dan learned how to disassemble VB statements,
and then put his own spin on what he saw the compiler doing. I am not
trying to slight his efforts any, but where do you find any more information,
than what can be learned by looking at the disassembled code?
An important fact I can agree with is:
"Still, it's bits. Only zero and 1 are allowed (exist). The key here is that Boolean
operators are bitwise and it is important what all the bits do. With the Boolean
data type you are simply assured they are all doing the same thing."
But this comment doesn't sit well with me.;
"Knowledge of the behavior includes both the defined and undefined characteristics."
Simply because there are no facts on the page that indicate there are any undefined
characteristics. When the subject is 'simple' logical operations, there are _no_ undefined
conditions, a truth table is proof of that....
Making his next statement misleading:
"Misunderstanding the defined behavior, or dependence on undefined behavior,
lead to unwanted (even if predictable) behavior."
Is that news? But, specifically the discussion is about Boolean logic, where
'understanding' is communicated via the truth table, where there is no 'undefined'
behavior. He can give me a black box, and its truth table, and, based on the
truth table, I can put that black box to use. The reason it works is because I
can predict the outcome, if I know the input.
So, for a 'general' conclusion on a page about boolean logic, I see it as somewhat
of a disservice. Because turn-about is fairplay, I'll let you all have an opportunity
to crtique my own comments on the topic:
http://www.usinternet.com/users/serflaten/logic.htm
BTW, VB5 documentation defines True, this way:
"The True keyword has a value equal to -1."
LFS
Justin Spindler
we remove that to? (Say yes and I will hunt you down) I think coding of
that type is extremely sloppy, and I hope much rarer than you make it sound.
Code like that deserves to break.
"Bob Butler" <butl...@earthlink.net> wrote in message
news:#24l$rpvAHA.720@tkmsftngp03...
Bill McCarthy
some replies inline:
"Jens Samson" <je...@esalar.be> wrote in message
news:e0KPyeovAHA.1992@tkmsftngp04...
>
<snip>
> And as MS does not state anything for the time being (the
> msdn.microsoft.com/net page looks as dead as it can be) and the rumour has
> been spread in this NG that the MVPs went to MS, this means that we feel
> that the MVPs have a lot of influence at MS.
> Given all the rumours I (I will not speak for anyone else) feel that the
> MVPs (or some of them) are behind the announced changes.
>
Well you are basing your assumptions on rumours, not fact. Fact is MS also
Jonathan Allen was also invited by MS to go to Redmond and was there whn the
MVPs were there. So on those facts, the changes could be because of
Jonathan, no ?
> Anyway, all this bitching could be stopped if all the parties were to
state
> an official anouncement.
> My questions remain (and I do understand that maybe you can not talk about
> it).
>
Well on this one you are probably right <bg>. Best to ask MS.
> 1. Is it true that those three changes will be rolled back ?
Only MS can answer that at present.
> 2. Who was the asking party for that ? I have a hard time believing it
was
> the majority of the VB6 crowd as lots of them do not code professionaly,
and
> even more of them have not even played around with VB.NET.
> So it must have been the people who have a lot of legacy code and fear
that
> upgrading will be a hell for them. And that's what I am affraid of,
because
> if they managed to get MS to do this, they maybe can manage a whole bunch
of
> other rollbacks, because those three changes are hadly going to make a
> difference in the conversion.
No offense but I find these assumptions very annoying to say the least.
Let's look at some facts, shall we. If MS made it's decisions based entirely
on feedback from MVPs then why would such changes occur so late in the
program if MS did listen to MVPs. That is not to say any changes they are
implementing are actually what any MVP asked for. I think you may have read
my opinion on some issues which may be related to the rumoured changes.
Clearly that is not what I as an MVP asked for is it ?
As to the issue of legacy code, well that, IMO is an important one. Issues
to do with making the language clear and concise, removing potentail sources
of human error. Issues to do with how quickly and cost efficiently companies
can uptake Vb.NET, issues to do with how cost efficient it will be for them
to maintain code written in VB.NET.
Also it is important to note that there are already 3.5 million developers
using VB today, those using VB.NET will have to come from somewhere. The
issue of skills migration is an important one. To say or take an attitude
that nothign existed prior to VB.NET is not only limiting the market for
VB.NET, but also comes at a huge cost for corporations. It's a head in the
sand approach, the kids with the new toy type of attitude I refuse to take.
I like VB.NET, but that does not mean it is perfect.
> 3. If the first point is true and the articles on ZDNet and CNet are
true,
> why did they choose to do it this way when in this NG there are a lot of
> suggestions to do it another way that would not cause so much discontent
on
> both sides.
>
LOl. Yes that is the million dollar question. And I too would like to know
the answer to that. I would like to know why things liek the To keyword is
not being supported in arrays. Must be Jonathan's fault hey ?? <bg>
BTW: (Actually, if you consider that if each VB developer spends about1
minute extra due to ambiguities in the language per day , it's actually a
3.5 million doallar question per day <g>)
Kathleen Dollard-Joeris
MS did meet with the MVPs.
The MVPs did give their opinions.
MS did meet with other people
Other people did give their opinions.
MS made their decision as to what changes to make in Beta2 based on all
input, a great deal of internal input I expect, and technical issues.
MS announced three of these changes on Wednesday. Only 3. The three changes
announced were to help COM interop/backward compatibility according to Ari
in the releases.
The changes announced do not match some unified voice from the MVPs. To the
extent there was agreement among MVPs, these changes at best are a partial
match.
The history of this seems trivial to me in a technical group. If it matters,
I certainly don't think I have an NDA on my thoughts or actions. I don't
have a problem talking about it, I just don't see it as helpful.
MVPs are not currently free to talk about changes. I and others are begging
to be free to talk about these changes.
Bob Butler
> Option Strict would catch that and not permit the code to compile. Should
> we remove that to? (Say yes and I will hunt you down) I think coding of
> that type is extremely sloppy, and I hope much rarer than you make it
sound.
> Code like that deserves to break.
No, not at all. Option Strict can be turned off if individuals want to but
I would always recommend leaving it on and adding the necessary
cast/conversion functions. This example would need a CInt or equivalent
added to make it compile. It's just that CInt should return -1 for
compatibility with CInt in VB6.
BTW, after spending a lot of time with the newsgroups and the local VB user
group I'm very convinced that there's a lot of that kind of code out there.
I always advocate explicit If/Then instead but that code is perfectly legal,
documented behaviour in VB6 and earlier.
Larry Serflaten
"Bob Butler" <butl...@earthlink.net> wrote
>
> "Patrick Steele" <pst...@ipdsolution.com_> wrote in message
>
> I'll bet that's what MS has been saying <g>
>
But.. ...OrElse ???
It sure sounds like someone got pissed off....
LFS
Kathleen Dollard-Joeris
I don't have context here, sure he didn't mean documented and undocumented?
Does he have it in latest version?
> "Knowledge of the behavior includes both the defined and undefined
characteristics."
Your page is definitely shorter.
Larry Serflaten
"Bill McCarthy" <bill...@iprimus.com.au> wrote
> so why aren't you asking Jonathan for proof, for facts.
His broad brush was certainly not called for, and plenty have called him on it....
>
> I cannot at present comment on what MS may or may not be intending to
> implement. If you do have a specific question as to my opinion on any
> certain language issue I would be most happy to answer that here in public,
> but cannot comment on MS's behalf, only on what my stance is.
But, the MVP's enjoy a 'closer link' to MS counterparts than 'we group members'
can attain. Can you ask them....
Will there be a published announcement from MS, of their intent, anytime soon?
(Anyone from MS care to comment?) So as to provide actual facts, instead of
the hearsay, we all are repeating...
;-)
LFS
Kathleen Dollard-Joeris
Don't know details, but it would be in the spirit of Option Strict for this
code to still throw a compiler error, and the announcements imply that it
will retain the VB6 behavior with Option Strict Off.
Work for you?
Larry Serflaten
"Jens Samson" <je...@esalar.be> wrote
I agree, I want a first class, all purpose language, made for a man, yet
gentle enough for a woman <g>
Some added info:
> The MVP group as a whole has done nothing to get rid of that impression,
> neither have any MVP members (I have to be careful with this because I don't
> really know who is MVP or not, I, and most of the people in this NG, can
> only know it by looking if the word MVP appears in the signing of the mail)
Check here for a list:
http://www.mvps.org/vb/index2.html?links.htm
LFS
Karl Hering
I, for one, certainly would not mind eliminating some of the garbage that seems to have collected from prior versions of visual basic. Memorizing 3 different techniques for database access (namely DAO, RDO & ADO), or the properties, methods & events for three sets of data access controls is not exactly the most efficient use of time. VB must be in-line with the trends in other languages.
As to making it easily up-gradable, who would want to simply upgrade code developed in past versions? There are major changes in .NET, and re-writing code can be great practice.
*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!
Patrick Steele
<bill...@iprimus.com.au>),
> > And VB.NET programmers will have this to deal with:
> >
> > CInt(boolean_of_true) <> boolean_of_true.ToInteger()
> >
>
> Well you already have that problem with Cstr compared to .ToString .
> Oh and not only was there never any .ToInteger, there won't be either.
Yes, I meant one of the .ToIntXX().
And what's this about those going away?
--
Patrick Steele
(pst...@ipdsolution.com)
Lead Software Architect
Image Process Design
Patrick Steele
<b...@cluestick.org>),
> The "And" of dotnet beta 1 is now spelled "AndAlso", has the short-circuting, etc.
> The "BitAnd" of dotnet beta 1 is now spelled "And", works like "And" always has, etc.
> ...and so on.
If it's that easy, why didn't they change "End While" to "Wend"??
Wouldn't that have helped out migration? That was the goal, wasn't it?
Kathleen Dollard-Joeris
The conversion methods are being shifted to the Convert class, where
arguably they should have been all along.
i = SomeVariable.ToInt32
becomes
i = Convert.ToInt32(SomeVariable)
Patrick Steele
<kjo...@noemailplease.com>),
> The conversion methods are being shifted to the Convert class, where
> arguably they should have been all along.
I agree. Much cleaner. Thanks for the clarification.
Rockford Lhotka
i = CInt(SomeVariable)
I still maintain that abandoning these core and common 'keywords' of VB are
what really could turn VB into a Visual Fred.
--
Rockford Lhotka
ro...@lhotka.net http://www.lhotka.net
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message
news:u4SL35rvAHA.1424@tkmsftngp03...
Justin Spindler
Microsoft.VisualBasic.DLL
"Rockford Lhotka" <ro...@lhotka.net> wrote in message
news:#Y9l5osvAHA.1880@tkmsftngp03...
Rockford Lhotka
Yet were you to totally avoid the use of all the VB keywords located in
Microsoft.VisualBasic I really am left wondering why you'd bother to do VB
at all? By the time you get rid of that much of the language (or at least
what most people consider to be part of the language) you aren't really left
with much. Just a hollow shell that is basically C# without the semicolons.
Saying you can do 'VB' without Microsoft.VisualBasic is much like asserting
that you can do 'C' without the standard libraries. Technically possible,
but hardly the experience anyone would expect.
Following this train of thought - I imagine that Smalltalk.NET programmers
will be expected to avoid the smalltalk runtime library as well? And the
same for all the other languages as well?
To most people a language is composed of a lot more than the core keywords
from which it is composed. Instead a language is composed of those keywords
and whatever standard function set is included in that languages equivalent
to the C standard libraries - and in VB's case this includes a great deal of
functionality found in Microsoft.VisualBasic.
I imagine there will be those who attempt to use the quasi-VB language
without relying on anything in Microsoft.VisualBasic. I pity them already.
By doing so they choose to isolate themselves from the mainstream VB
community by creating non-VB apps. At the same time, since they are sort of
doing VB they also choose to be looked down upon by the C-style bigots of
the world. What a nasty lose-lose situation...
--
Rockford Lhotka
ro...@lhotka.net http://www.lhotka.net
"Justin Spindler" <halo...@bellatlantic.net> wrote in message
news:O9SMnssvAHA.956@tkmsftngp03...
Bob O'Brien
> I object to the mix of bitwise and logical operators where the bitwise
> operators do not follow operator precedence, which becomes an issue and
> propogates logic errors in bitmasking. This becomes evident when trying
to
> use such libraries as DX, and you have to throw around parenthesis to
clean
> up. Bitwise operators should happen before comparison operators and
> comparision operators should happen before logical operators. Instead, we
> have bitwise operations occuring after comparison operators, left to
right,
> in the same precedence as logical operators.
Specifics, please, with at least one example.
You have provided no information here.
Bob
--
Bob O'Brien
"Patrick Steele" <pst...@ipdsolution.com_> wrote in message
> In article <eephkDgvAHA.1456@tkmsftngp02> (from Michael Venuto <!
> no_spam!mi...@grsinc.com>),
> > I second, third, and fourth that. I have to admit that I was going
around,
> > talking up .NET myself, and VB.NET in particular. Full stop.
>
> .NET is still a great platform to develop in. It just seems MS wants to
> keep VB as a "second class" citizen. C# seems to be the safest bet for
> the future.
The essential truth you note has not changed at all, only your awareness
has.
Congratulations on your enlightenment.
Bob O`Bob
--
Justin Spindler
library that redefines functions already found in .NET, wasting both
harddrive space and RAM?
"Rockford Lhotka" <ro...@lhotka.net> wrote in message
news:#lCjh9svAHA.1420@tkmsftngp03...
Bob Butler
"Kathleen Dollard-Joeris" <kjo...@noemailplease.com> wrote in message
news:u4SL35rvAHA.1424@tkmsftngp03...
>
> The conversion methods are being shifted to the Convert class, where
> arguably they should have been all along.
>
> i = SomeVariable.ToInt32
>
> becomes
>
> i = Convert.ToInt32(SomeVariable)
With the exception of .ToString, correct?
Bob O'Brien
"Larry Serflaten" <serf...@usinternet.com> wrote in message
news:#gQUHlqvAHA.2120@tkmsftngp04...
> But, the MVP's enjoy a 'closer link' to MS counterparts than 'we group
members'
> can attain. Can you ask them....
>
> Will there be a published announcement from MS, of their intent,
anytime soon?
> (Anyone from MS care to comment?) So as to provide actual facts,
instead of
> the hearsay, we all are repeating...
I am reminded of John Cleese and the Dead Parrot sketch.
Yes, we have asked. We have asked. Boy-howdy have we ever asked.
Bob O`Bob
--
Gregor R. Peisker
> To most people a language is composed of a lot more than the core keywords
> from which it is composed. Instead a language is composed of those
keywords
> and whatever standard function set is included in that languages
equivalent
> to the C standard libraries - and in VB's case this includes a great deal
of
> functionality found in Microsoft.VisualBasic.
Yes, ...
> I imagine there will be those who attempt to use the quasi-VB language
> without relying on anything in Microsoft.VisualBasic. I pity them already.
> By doing so they choose to isolate themselves from the mainstream VB
> community by creating non-VB apps. At the same time, since they are sort
of
> doing VB they also choose to be looked down upon by the C-style bigots of
> the world. What a nasty lose-lose situation...
... but this conclusion is a little radical for my taste. Core syntax still
has some value.
Regards,
Gregor
Rockford Lhotka
news:O8jaH0tvAHA.1348@tkmsftngp02...
> [...]
> > I imagine there will be those who attempt to use the quasi-VB language
> > without relying on anything in Microsoft.VisualBasic. I pity them
already.
> > By doing so they choose to isolate themselves from the mainstream VB
> > community by creating non-VB apps. At the same time, since they are sort
> of
> > doing VB they also choose to be looked down upon by the C-style bigots
of
> > the world. What a nasty lose-lose situation...
>
> ... but this conclusion is a little radical for my taste. Core syntax
still
> has some value.
Yeah, it may be a bit radical.
Yet I am of the belief that VB without the Microsoft.VisualBasic contents
really IS Visual Fred. A perfectly wonderful and useful .NET language - but
most definitively not VB... Unfortunately since it has no semicolons or
squiggly brackets it won't be considered valid by some measurable percentage
of the population whether we forgo the familiar convenience of the VB
functionality or not...
I find that I am substantially more productive working in VB.NET than in C#.
This is specifically true because I don't have to go hunting through help
finding out how to do mundane simple things that are 'built in' to VB via
the Microsoft.VisualBasic functionality. Were I to abandon all the 'normal'
VB functionality I imagine I'd be about as productive in VB as C# - and
that, to me, makes no sense at all...
Kathleen Dollard-Joeris
Right, ToString will be accessible from both Convert and the base type
class. Both
SomeVariable.ToString
and
Convert.ToString(SomeVariable)
are valid
Kathleen Dollard-Joeris
Right, ToString will be accessible from both Convert and the base type
class. Both
SomeVariable.ToString
and
Convert.ToString(SomeVariable)
are valid
--
Kathleen Dollard-Joeris
Right. ToString is in both classes
Both
SomeVariable.ToString
and
Convert.ToString(SomeVariable)
are valid.
Bob Butler
> Right, ToString will be accessible from both Convert and the base type
> class. Both
> SomeVariable.ToString
> and
> Convert.ToString(SomeVariable)
> are valid
Thanks for verifying that.
Peter McMahon
because pretty much all their programmers are C++ people, and their C++
knowledge really comes in handy when dealing with Win32 API functions. I'm a
Delphi/VB guy, and if I didn't know the little bit of C++ I did I'd never be
able to use the Win32 API, because all the docs give samples in C++ etc.
My 2 cents...
Peter
"Bob Butler" <butl...@earthlink.net> wrote in message
news:e95BFsgvAHA.2304@tkmsftngp02...
>
> "Justin Spindler" <halo...@bellatlantic.net> wrote in message
> news:#C$OVSfvAHA.940@tkmsftngp04...
> >
> > "Bob Butler" <butl...@earthlink.net> wrote in message
> > news:#gz3QKfvAHA.712@tkmsftngp04...
> > >
> > > "Jonathan Allen" <grey...@cts.com> wrote in message
> > > news:e8dhcuevAHA.2228@tkmsftngp02...
> > > > To those who want to see VB become a real programming language,
> > >
> > > It is. It has been for a long time now.
> >
> > No, it's always been a second-class citizen, which is exactly why
> Microsoft
> > has yet to release any functional commercial applications written in VB.
>
> It's always been perceived as such by many. It's also been an extremely
> useful tool for many developers. I don't worry about what others think of
> the tool if it gets the job done for me.
>
>
>
>
Jonathan Allen
> Well Jonathan I'm convinced. The rumors were true .. you are an idiot.
>
Maybe, but look at all the fools that followed me. If I'm an idiot, what do
you call them?
--
Jonathan Allen
Bob O`Bob
>
> Maybe, but look at all the fools that followed me. If I'm an idiot, what do
> you call them?
>
You've already handled that detail nicely. I hope they appreciate it.
Bob O`Bob
--
Bill McCarthy
Well JA, as you pointed out they followed you. Hopefully they have learnt
and won't make the same mistake again. BTW: you do know the difference
between rioting and inciting to riot don't you ?
Bob O`Bob
>
> I object to the mix of bitwise and logical operators where the bitwise
> operators do not follow operator precedence, which becomes an issue and
> propogates logic errors in bitmasking. This becomes evident when trying to
> use such libraries as DX, and you have to throw around parenthesis to clean
> up. Bitwise operators should happen before comparison operators and
> comparision operators should happen before logical operators. Instead, we
> have bitwise operations occuring after comparison operators, left to right,
> in the same precedence as logical operators.
VB has no "logical operators"
So, essentially, you have some vague objection somewhere to something
involving logic and parentheses, perhaps due to some article or posting
that you once read, but you can't be specific because you really don't
know what it is that you are writing about?
Is that really it, or might you actually be able to identify something specific?
Bob O`Bob
--
Jonathan Allen
> and won't make the same mistake again.
We'll see. In the mean time I have a new game; how long will it take for me
to get my credibility back with everybody (expect of course the ones who
never believed me in the first place).
> BTW: you do know the difference
> between rioting and inciting to riot don't you ?
Yea, the later is done by the idiot in the front of the crowd. The former is
done by the remainder.
--
Jonathan Allen
"Bill McCarthy" <bill...@iprimus.com.au> wrote in message
news:etZ7Qf#vAHA.1908@tkmsftngp05...
Peter McMahon
> > Hopefully they have learnt
> > and won't make the same mistake again.
>
> We'll see. In the mean time I have a new game; how long will it take for
me
> to get my credibility back with everybody (expect of course the ones who
> never believed me in the first place).
A 'new game'? So this is all a game? :-)
P
Bill McCarthy
>
> We'll see. In the mean time I have a new game; how long will it take for
me
> to get my credibility back with everybody (expect of course the ones who
> never believed me in the first place).
>
you know JA, I think this is one of the saddest and sickest posts I have
ever seen you made. Sad, because it appears now your true colors are really
showing up.
You posted here, bad mouthing all MVPs. You attempted to falsly tarnish
their reputations, the reputations of people who have contributed so much to
these forums over the years. And rather than you actually apologise, all
you are worried about is your own reputation.
That kind os selfishness is what is the difference between you and a MVP.
MVP's as has been recognised by Microsoft, are those who give selflesly to
these communities.
And it goes further than just that, you also went on to insult all those who
even listened to you, as per your post in the devx newsgroup yeaterday.
> What can I say, I was bored. I felt like seeing what kind of effect a few
> choice rants could have, and boy did it work. It always amazes me how
> gullible people are.
And as i look back now on all those arguements we have had over VB6 to
VB.NEt compatibility, one thing rings clear, they way you kept saying it
didn't effect you so you didn't want if fixed. I think that truely shows
your selfish nature.
I do feel soory for you. You critise MVPs saying they are scared of the
dark. Well to be scared of the dark in fact is a natural thing, better than
tripping over something you can't see, something as in your case you refuse
to see. MVPs are renowned for shedding light onto things that are in the
dark. They confront their ears are help remove thefears others may have.
You, on the other hand seem to be scared of the light, instead you hide away
in your bedroom, trying to build a reputation by falsly critising others,
and displaying your selfish attitudes.
As to the comments you made about the age of MVPs, be aware JA that many of
them are actually younger than you. They contribute to this forum and
others, but never seem to stoop to the dispicable levels you do. It isn't a
question of age JA, but a question of maturity.
So grow up JA, and learn to at least apologise *without* having been told
to.
Karl E. Peterson
>
> "Bob Butler" <butl...@earthlink.net> wrote
> >
> > "Patrick Steele" <pst...@ipdsolution.com_> wrote in message
> >
> > I'll bet that's what MS has been saying <g>
>
> But.. ...OrElse ???
>
> It sure sounds like someone got pissed off....
"Someone" did, yes.
--
[This space intentionally left blank.]