DevX Home    Today's Headlines   Articles Archive   Tip Bank   Forums   

Page 1 of 9 123 ... LastLast
Results 1 to 15 of 121

Thread: Re: Summary of This Discussion :re: Patrice's latest2

  1. #1
    don Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    Hi Patrice,

    I have been avoiding looking in this group since it uses up valuable programming
    time but I couldn't resist answering you. Thanks for your rational discourse.

    "compiling behind the scenes" I suspect means 'compiling everytime it runs'.
    In other words, the ASP script is not converted permanently into binaries.
    It is instead, "interpreted code" in the same sense that VB 3.0 was interpreted
    code, java is interpreted code and ASP 2.5 is interpreted code.
    "Interlacing ASP with HTML" is a coding choice, not a requirement. Most of
    the time, there is far too much processing appearing in the ASP when it should
    be in a DLL (read binary).

    Again, messaging is a way of sending text between 2 computers. enough said.

    Performance counters? why? most if not all VB programming has been business
    programming on the clock. Most of us produce code as rapidly and as sloppily
    as necessary to get the job done. (some of us do not know better because
    we use the idiotic examples in MSDN as a guide but that is another subject.)
    Who the heck needs performance counters? To produce 'fully optimized code'
    write one exposed function, a block of exposed constants, avoid the use of
    properties and keep everything else private and IIF you destroy any objects
    you create, you will create the best possible VB code.(I am over-generalizing
    here but you get my drift.) I don't need a peformance counter to get the
    job done. I need a tool that quickly delivers what business wants. .NUTS
    does .NOT do THAT!

    Why .NET is .NUTS is simple. Here are 5 quick reasons but there are more:

    1) It harms the marketplace for VB programmers.
    2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    They never cared about inheritance, polymorphism and the like and they haven't
    wanted to learn it regardless of what MS and the focus groups say.
    3) The only people whining about these things have been C++ programmers who
    lost work to VB programmers and had to use VB by virtue of the market demand.
    They are the real winners in this change; not MS or the VB community.
    4) MS has forced a massive retooling expense upon a business community that
    depended upon VB.
    5) MS has forced a massive retraining expense upon a business community and
    the programmers that used VB.

    The fact remains that the introduction of inheritance, polymorphism and subclassing,
    etc... into VB raises the complexity, increases debug time and elevates cost
    regardless of marketing claims to the contrary.

    As for what I would want in .NET - nobody asked way back when and what I
    think really doesn't matter. --.NUTS IS GOING TO MARKET.-- The features in
    .NET are in many ways redundant of features I can find in other sources.
    Regardless of what is in .NET, I want to possess skills that are marketable.
    With the advent of .NUTS, my VB skills are irrelevant or will be in less
    than 18 months. So are yours.

    Every VB programmer must decide if he/she wants to continue programming.
    If he/she does, --FORGET VB.NET! IT IS IRRELEVANT-- he/she MUST learn C/C++
    because that is the foundation of Java, C#, Visual C++, Borland C++, Borland
    J Builder and VB.NET regardless of the noise coming from MS, Borland or Sun
    about 'ease of use'. Company specific intermediate languages and compilers
    will change (see the shift to .NET as proof) where C++ is always present
    regardless of system or context. Once you have some C++, you are not bound
    to MS or Sun or Borland - you can acquire any of the intermediate languages
    with the same relative ease. Why bust your gut learning a twisted form of
    Basic that really isn't basic at all?

    My personal technical improvement is now and will continue to be on C++.
    I am reading a very interesting book by Bjarne Stroustrup called "The C++
    Programming Language". He invented it. really he did. see http://www.research.att.com/~bs/C++.html.
    I like reading his definitions of C++ and then discovering how MS or Borland
    violated ANSI C++ standards. It really isn't that hard to understand.
    no more so than VB was in 1992.


    Thanks

    Don

    "Patrice Scribe" <scribe@chez.com> wrote:
    >I forgot to address some points :
    >- ASP.NET compiles behind the scene. I expect the same thing for Windows

    Scripting.
    >That
    >is it's look like scripting but the code is actually compiled for you.
    >- ASP.NET provides server side controls. In ASP you have to interlace HTML

    and code
    >(or to
    >output HTML tags using code). This is no more needed in ASP.NET. You have

    an HTML part
    >that includes server side control and you have a code behind file that allows

    to set
    >properties and react to events.
    >- the Framework Class Library gives access to most underlying features including

    messaging
    >(no big deal here). For "monitoring" I was specifically thinking about adding

    performance
    >counters to an application. This a matter of minutes with .NET.
    >
    >I can understand it doesn't fit your needs but is it a reason to quality

    this a .NUTS
    >strategy.
    >
    >What would be your expectations about the ideal .NET you are dreaming of

    ?
    >
    >--
    >Patrice Scribe


  2. #2
    David Rothgery Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    "don" <donaldg@varysoft.com> wrote in message
    news:3b584851$1@news.devx.com...
    > "compiling behind the scenes" I suspect means 'compiling everytime
    > it runs'.


    No, it means 'compiling the first time it runs'. Which is how JSP works,
    incidentally.

    > "Interlacing ASP with HTML" is a coding choice, not a requirement.
    > Most of the time, there is far too much processing appearing in the
    > ASP when it should be in a DLL (read binary).


    I've worked on several ASP projects where, IMO, there was far too much
    processing unloaded unto DLLs, which are a lot more work to update and not
    significantly faster for the 'run a stored procedure and return a recordset'
    functions that were in the DLL.

    People tend to assume that putting code in a DLL will make it faster, but
    forget about the overhead of COM object creation, and the maintence
    headaches it causes.





  3. #3
    Rob Teixeira Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    "don" <donaldg@varysoft.com> wrote:
    >
    >"compiling behind the scenes" I suspect means 'compiling everytime it runs'.
    >In other words, the ASP script is not converted permanently into binaries.
    >It is instead, "interpreted code" in the same sense that VB 3.0 was interpreted
    >code, java is interpreted code and ASP 2.5 is interpreted code.


    Hmmm... sorta

    I think the intent was to say that ASP.NET (the code part) is no longer a
    script when deployed, as ASP used to be. It is compiled to IL though, so
    that's somewhat "interpreted code", but then you have the sticky JIT issues,
    which to some degree, can make it permanently native binary code.

    >"Interlacing ASP with HTML" is a coding choice, not a requirement. Most

    of
    >the time, there is far too much processing appearing in the ASP when it

    should
    >be in a DLL (read binary).


    While I agree whole-heartedly with you, the fact is that some companies make
    a conscious choice to place that logic in the ASP script because they view
    that as an easier maintenance path than dealing with COM components.

    >Performance counters? why? most if not all VB programming has been business
    >programming on the clock. Most of us produce code as rapidly and as sloppily
    >as necessary to get the job done. (some of us do not know better because
    >we use the idiotic examples in MSDN as a guide but that is another subject.)


    <G>

    >Who the heck needs performance counters? To produce 'fully optimized code'
    >write one exposed function, a block of exposed constants, avoid the use

    of
    >properties and keep everything else private and IIF you destroy any objects
    >you create, you will create the best possible VB code.(I am over-generalizing
    >here but you get my drift.) I don't need a peformance counter to get the
    >job done. I need a tool that quickly delivers what business wants. .NUTS
    >does .NOT do THAT!


    I understand exactly what you mean, but also take into consideration that
    optimization is a lot tricker than that. It is also very crucial to businesses
    that deal with critical B2B or B2C applications that need to deal with thousands
    of transactions per server per second.

    >Why .NET is .NUTS is simple. Here are 5 quick reasons but there are more:
    >
    >1) It harms the marketplace for VB programmers.


    How so? I'm not trying to argue, I'm just interested in the specifics of
    why think that is the case.

    >2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    >They never cared about inheritance, polymorphism and the like and they haven't
    >wanted to learn it regardless of what MS and the focus groups say.


    Let's not generalize too quickly or too broadly. There in fact VB programmers
    who want those features. Some say nobody wants it, and others say everyone
    wants it, but neither side is right with those types of statements.

    >3) The only people whining about these things have been C++ programmers

    who
    >lost work to VB programmers and had to use VB by virtue of the market demand.


    That's also a bit too broad of a generalization to be realistic.

    >They are the real winners in this change; not MS or the VB community.


    That's yet to be determined. In the worst case, nobody wins and VB dies.
    At the same time, it's also possible that many people will benifit, and not
    necessarily one narrow generalized group.

    >4) MS has forced a massive retooling expense upon a business community that
    >depended upon VB.


    Yes

    >5) MS has forced a massive retraining expense upon a business community

    and
    >the programmers that used VB.


    >The fact remains that the introduction of inheritance, polymorphism and

    subclassing,
    >etc... into VB raises the complexity, increases debug time and elevates

    cost
    >regardless of marketing claims to the contrary.


    True, but you don't *have* to use those features.
    On the other hand, those who are used to OOP will find their work easier
    to do than trying to find a hack alternative for what they're attempting.

    [snip rant]

    >Every VB programmer must decide if he/she wants to continue programming.
    >If he/she does, --FORGET VB.NET! IT IS IRRELEVANT-- he/she MUST learn C/C++
    >because that is the foundation of Java, C#, Visual C++, Borland C++, Borland
    >J Builder and VB.NET regardless of the noise coming from MS, Borland or

    Sun
    >about 'ease of use'.


    C++ being the foundation for C++ is a bit redundant.

    I also don't think that C++ is required knowledge for VB or VB.NET.
    And I think you'll find that a good portion of Java programmers have never
    used C++, and are probably as anti-C++ as some of the VB programmers.

    >Company specific intermediate languages and compilers
    >will change (see the shift to .NET as proof) where C++ is always present
    >regardless of system or context.


    What I see as being the "real" issue, is that most of the critical bits that
    currently run the foundation software is now in C++. Therefore, as you say,
    there is a need for it. However, if that trend changes, and I actually do
    see it changing slowly now, then at some point, whatever language is used
    to implement the majority of those bits will become the stable dependant
    language.

    >Once you have some C++, you are not bound
    >to MS or Sun or Borland - you can acquire any of the intermediate languages
    >with the same relative ease.


    I think that's only because C++ forces you to learn all sorts of CS practices.
    It's not necessarily that learning the language itself enables you to work
    with other languages. After all, I personally know quite a few C++ programmers
    that have difficulty with other languages (and Karl is pretty quick to point
    them out when he sees some too!)

    >Why bust your gut learning a twisted form of
    >Basic that really isn't basic at all?


    That depends on what BASIC is. And I don't think anyone can easily define
    it in a way that everyone will agree.
    After all, VB1-6, with its procedures, functions, and event model is certainly
    a very far step from what BASIC was for a decade prior to that.

    >My personal technical improvement is now and will continue to be on C++.


    The more you know, the better.

    >I am reading a very interesting book by Bjarne Stroustrup called "The C++
    >Programming Language".


    I also recommend Bruce Eckle "Thinking in C++".

    >He invented it. really he did. see http://www.research.att.com/~bs/C++.html.


    That depends on what you mean by "it". Conceptually, he "borrowed" quite
    a bit

    >It really isn't that hard to understand.
    >no more so than VB was in 1992.


    No more so than VB.NET will be in 2002.

    -Rob

  4. #4
    Michael Q. Gautier Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    Some people write the entire site in vb (multiple classes in one project)
    with an asp entry point to instantiate the thing. Others go to the other
    extreme and write nothing but global code and resist functions and vbscript
    classes with a passion.

    As always, balance is the key to maintainbility and acceptable performance.
    I do feel that you can obtain the "organizational" benefits of compiled vb6
    classes through vbscript classes without the maintenance headaches. In some
    cases though, you loose the ease of debugging.

    I have personally seen that compiled vb code to run a db query being slower
    than already optimized ado objects running from an already optimized asp
    environment. I would think that I would make sense for compiled code to run
    faster period.

    VB.NET or C# may render this observation or principle moot.

    Professional Microsoft Web Developer,
    3 years and counting


    "David Rothgery" <drothgery@myrealbox.com> wrote:
    >
    >"don" <donaldg@varysoft.com> wrote in message
    >news:3b584851$1@news.devx.com...
    >> "compiling behind the scenes" I suspect means 'compiling everytime
    >> it runs'.

    >
    >No, it means 'compiling the first time it runs'. Which is how JSP works,
    >incidentally.
    >
    >> "Interlacing ASP with HTML" is a coding choice, not a requirement.
    >> Most of the time, there is far too much processing appearing in the
    >> ASP when it should be in a DLL (read binary).

    >
    >I've worked on several ASP projects where, IMO, there was far too much
    >processing unloaded unto DLLs, which are a lot more work to update and not
    >significantly faster for the 'run a stored procedure and return a recordset'
    >functions that were in the DLL.
    >
    >People tend to assume that putting code in a DLL will make it faster, but
    >forget about the overhead of COM object creation, and the maintence
    >headaches it causes.
    >
    >
    >
    >



  5. #5
    Mike Mitchell Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    On 20 Jul 2001 09:05:12 -0700, "Rob Teixeira" <RobTeixeira@@msn.com>
    wrote:

    >>2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    >>They never cared about inheritance, polymorphism and the like and they haven't
    >>wanted to learn it regardless of what MS and the focus groups say.

    >
    >Let's not generalize too quickly or too broadly. There in fact VB programmers
    >who want those features. Some say nobody wants it, and others say everyone
    >wants it, but neither side is right with those types of statements.


    Okay, the gospel truth is, a high percentage of classic VB programmers
    DO NOT want it. You can argue till the cows come home about the actual
    percentage, but I would place it at higher than 60%.

    MM

  6. #6
    Jonathan Allen Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    > 2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    > They never cared about inheritance, polymorphism and the like and they

    haven't
    > wanted to learn it regardless of what MS and the focus groups say.


    Reality check. #1

    Professional C programmers don't care about inheritance, polymorphism, or
    classes either. They use C (or FORTRAN) because it is faster than an object
    orientated language. Granted that the number of people who use it
    professionally are small, but they are using it on the fastest super
    computers in existence.

    Reality Check #2

    VB is based on the concepts of polymorphism. We have always believed one
    should be able to treat something as something else, to the extend that we
    created Evil Type Coercion. What other language has embraced polymorphism so
    completely that it screwed itself?


    --
    Jonathan Allen




  7. #7
    Joe \Nuke Me Xemu\ Foster Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    "Jonathan Allen" <greywolf@cts.com> wrote in message <news:3b5b21b6@news.devx.com>...

    > > 2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    > > They never cared about inheritance, polymorphism and the like and they

    > haven't
    > > wanted to learn it regardless of what MS and the focus groups say.

    >
    > Reality check. #1
    >
    > Professional C programmers don't care about inheritance, polymorphism, or
    > classes either. They use C (or FORTRAN) because it is faster than an object
    > orientated language. Granted that the number of people who use it
    > professionally are small, but they are using it on the fastest super
    > computers in existence.


    I thought the whole point of C++ was to not pay for features you don't
    actually use. If a C++ program doesn't use inheritance, or RTTI, or
    structured exception handling, then it shouldn't have their overhead.
    Has Bjarne's original vision been flushed down the toilet along with VB?

    > Reality Check #2
    >
    > VB is based on the concepts of polymorphism. We have always believed one
    > should be able to treat something as something else, to the extend that we
    > created Evil Type Coercion. What other language has embraced polymorphism so
    > completely that it screwed itself?


    And of course replacing all occurrences of "Variant" with "Object" makes
    it all better now? Remember, VB's Martians have now infected the whole
    CLR.

    --
    Joe Foster <mailto:jfoster@ricochet.net> Sacrament R2-45 <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!



  8. #8
    Jonathan Allen Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    > I thought the whole point of C++ was to not pay for features you don't
    > actually use. If a C++ program doesn't use inheritance, or RTTI, or
    > structured exception handling, then it shouldn't have their overhead.
    > Has Bjarne's original vision been flushed down the toilet along with VB?


    So? We were discussing C programmers, not C++ programmers.

    Anyways, you still have to account for them when you write the compiler. Why
    spend all the time adding those features to a custom compiler if you don't
    intend on using it? It isn't like super computers can use off the self
    compilers.

    > And of course replacing all occurrences of "Variant" with "Object" makes
    > it all better now?


    Um, what are you talking about? I don't see what that has to do with whither
    or not VB programmers used polymorphism.

    --
    Jonathan Allen


    "Joe "Nuke Me Xemu" Foster" <joe@bftsi0.UUCP> wrote in message
    news:3b5b45c4@news.devx.com...
    > "Jonathan Allen" <greywolf@cts.com> wrote in message

    <news:3b5b21b6@news.devx.com>...
    >
    > > > 2) VB programmers are VB programmers BECAUSE they are NOT C

    programmers.
    > > > They never cared about inheritance, polymorphism and the like and they

    > > haven't
    > > > wanted to learn it regardless of what MS and the focus groups say.

    > >
    > > Reality check. #1
    > >
    > > Professional C programmers don't care about inheritance, polymorphism,

    or
    > > classes either. They use C (or FORTRAN) because it is faster than an

    object
    > > orientated language. Granted that the number of people who use it
    > > professionally are small, but they are using it on the fastest super
    > > computers in existence.

    >
    > I thought the whole point of C++ was to not pay for features you don't
    > actually use. If a C++ program doesn't use inheritance, or RTTI, or
    > structured exception handling, then it shouldn't have their overhead.
    > Has Bjarne's original vision been flushed down the toilet along with VB?
    >
    > > Reality Check #2
    > >
    > > VB is based on the concepts of polymorphism. We have always believed one


    > > should be able to treat something as something else, to the extend that

    we
    > > created Evil Type Coercion. What other language has embraced

    polymorphism so
    > > completely that it screwed itself?

    >
    > And of course replacing all occurrences of "Variant" with "Object" makes
    > it all better now? Remember, VB's Martians have now infected the whole
    > CLR.
    >
    > --
    > Joe Foster <mailto:jfoster@ricochet.net> Sacrament R2-45

    <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!
    >
    >




  9. #9
    Joe \Nuke Me Xemu\ Foster Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    "Jonathan Allen" <greywolf@cts.com> wrote in message <news:3b5ba94b@news.devx.com>...

    > > I thought the whole point of C++ was to not pay for features you don't
    > > actually use. If a C++ program doesn't use inheritance, or RTTI, or
    > > structured exception handling, then it shouldn't have their overhead.
    > > Has Bjarne's original vision been flushed down the toilet along with VB?

    >
    > So? We were discussing C programmers, not C++ programmers.
    >
    > Anyways, you still have to account for them when you write the compiler. Why
    > spend all the time adding those features to a custom compiler if you don't
    > intend on using it? It isn't like super computers can use off the self
    > compilers.


    My point is that C++ isn't necessarily doomed to be slower than C.

    > > And of course replacing all occurrences of "Variant" with "Object" makes
    > > it all better now?

    >
    > Um, what are you talking about? I don't see what that has to do with whither
    > or not VB programmers used polymorphism.


    You asked, "What other language has embraced polymorphism so completely
    that it screwed itself?" The answer is, anything using the CLR.

    --
    Joe Foster <mailto:jfoster@ricochet.net> On the cans? <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!



  10. #10
    Jonathan Allen Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    > You asked, "What other language has embraced polymorphism so completely
    > that it screwed itself?" The answer is, anything using the CLR.


    Ok, I understand what your getting at now.

    --
    Jonathan Allen


    "Joe "Nuke Me Xemu" Foster" <joe@bftsi0.UUCP> wrote in message
    news:3b5bb803@news.devx.com...
    > "Jonathan Allen" <greywolf@cts.com> wrote in message

    <news:3b5ba94b@news.devx.com>...
    >
    > > > I thought the whole point of C++ was to not pay for features you don't
    > > > actually use. If a C++ program doesn't use inheritance, or RTTI, or
    > > > structured exception handling, then it shouldn't have their overhead.
    > > > Has Bjarne's original vision been flushed down the toilet along with

    VB?
    > >
    > > So? We were discussing C programmers, not C++ programmers.
    > >
    > > Anyways, you still have to account for them when you write the compiler.

    Why
    > > spend all the time adding those features to a custom compiler if you

    don't
    > > intend on using it? It isn't like super computers can use off the self
    > > compilers.

    >
    > My point is that C++ isn't necessarily doomed to be slower than C.
    >
    > > > And of course replacing all occurrences of "Variant" with "Object"

    makes
    > > > it all better now?

    > >
    > > Um, what are you talking about? I don't see what that has to do with

    whither
    > > or not VB programmers used polymorphism.

    >
    > You asked, "What other language has embraced polymorphism so completely
    > that it screwed itself?" The answer is, anything using the CLR.
    >
    > --
    > Joe Foster <mailto:jfoster@ricochet.net> On the cans?

    <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!
    >
    >




  11. #11
    Ed Courtenay Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    Hmmm - strange that. You would have thought that with numbers like that Mike
    that I would have actually *met* a VB programmer who doesn't want these
    features.

    However, among all the developers I know I can honestly tell you that the
    resistance factor is 0%. But, maybe that's because they're professionals....

    "Mike Mitchell" <kylix_is@hotmail.com> wrote in message
    news:3b5abcd3.7747485@news.devx.com...
    > On 20 Jul 2001 09:05:12 -0700, "Rob Teixeira" <RobTeixeira@@msn.com>
    > wrote:
    >
    > >>2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    > >>They never cared about inheritance, polymorphism and the like and they

    haven't
    > >>wanted to learn it regardless of what MS and the focus groups say.

    > >
    > >Let's not generalize too quickly or too broadly. There in fact VB

    programmers
    > >who want those features. Some say nobody wants it, and others say

    everyone
    > >wants it, but neither side is right with those types of statements.

    >
    > Okay, the gospel truth is, a high percentage of classic VB programmers
    > DO NOT want it. You can argue till the cows come home about the actual
    > percentage, but I would place it at higher than 60%.
    >
    > MM




  12. #12
    Ian R Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    "Mike Mitchell" <kylix_is@hotmail.com> wrote in message
    news:3b5abcd3.7747485@news.devx.com...
    >
    > Okay, the gospel truth is, a high percentage of classic VB programmers
    > DO NOT want it. You can argue till the cows come home about the actual
    > percentage, but I would place it at higher than 60%.
    >


    I think you're confusing things with what you want Mike. Most VB developers
    I've talked to are excited with most of the features in VB.NET. The few who
    don't seem to care simply won't use it.



  13. #13
    Patrice Scribe Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    > "compiling behind the scenes" I suspect means 'compiling everytime it runs'.
    > In other words, the ASP script is not converted permanently into binaries.
    > It is instead, "interpreted code" in the same sense that VB 3.0 was interpreted
    > code, java is interpreted code and ASP 2.5 is interpreted code.


    AFAIK it is compiled when the page is first loaded and all subsequent requests to this
    page will use the compiled code. You could probably also compile this yourself beforehand.
    I still have to investigate offered options.

    > "Interlacing ASP with HTML" is a coding choice, not a requirement. Most of
    > the time, there is far too much processing appearing in the ASP when it should
    > be in a DLL (read binary).


    Designers are able to work on the HTML part while developers are able to work on the code
    behind file. AFAIK you have not even a function call inside your HTML tags, you just have
    "controls". When using an ASP or a COM app. it's quite difficult for non developers to
    update the layout... I admit I still have to investigate but it seems to make sense.

    >
    > Again, messaging is a way of sending text between 2 computers. enough said.


    OK, meant that the Framework Class Library provides this among numerous other things.

    > Performance counters? why? most if not all VB programming has been business
    > programming on the clock. Most of us produce code as rapidly and as sloppily
    > as necessary to get the job done. (some of us do not know better because
    > we use the idiotic examples in MSDN as a guide but that is another subject.)
    > Who the heck needs performance counters? To produce 'fully optimized code'
    > write one exposed function, a block of exposed constants, avoid the use of
    > properties and keep everything else private and IIF you destroy any objects
    > you create, you will create the best possible VB code.(I am over-generalizing
    > here but you get my drift.) I don't need a peformance counter to get the
    > job done. I need a tool that quickly delivers what business wants. .NUTS
    > does .NOT do THAT!


    With a very miminal amount of work, it allows to have your administrator use its usual
    performance monitoring tool to check that all performs well. I don't specifically push to
    use this particular feature but I meant that you have now a whole lot of things that are
    much easier and quicker to do thanks to the Framework Class Library. Before the
    complexity/usage ratio was not good enough to give this even a try...

    I don't see what prevents to use VB.NET the same way you did with VB Classic...

    >
    > Why .NET is .NUTS is simple. Here are 5 quick reasons but there are more:
    >
    > 1) It harms the marketplace for VB programmers.
    > 2) VB programmers are VB programmers BECAUSE they are NOT C programmers.
    > They never cared about inheritance, polymorphism and the like and they haven't
    > wanted to learn it regardless of what MS and the focus groups say.
    > 3) The only people whining about these things have been C++ programmers who
    > lost work to VB programmers and had to use VB by virtue of the market demand.
    > They are the real winners in this change; not MS or the VB community.
    > 4) MS has forced a massive retooling expense upon a business community that
    > depended upon VB.
    > 5) MS has forced a massive retraining expense upon a business community and
    > the programmers that used VB.
    >
    > The fact remains that the introduction of inheritance, polymorphism and subclassing,
    > etc... into VB raises the complexity, increases debug time and elevates cost
    > regardless of marketing claims to the contrary.


    It doesn't seems that a big deal to me but I admit I made some OOP programming 10 years
    ago with Turbo Pascal 6.0 and its excellent Turbo Vision object oriented framework. You
    don't (and you shouldn't) have to use it if you don't need it. Check how to write a
    Windows service using the .NET Framework. This is really very easy thanks to OOP. As for
    any feature, you have to use these wisely and not to use this because it's just available
    but if you need to such a thing this is really easy.

    In general it seems to me that VB changes are often quite largely overweighted but of
    course this is always a matter of personal reaction. I can understand this reaction if you
    expected to run your code without any change as you could do with previous upgrades. But
    then why upgrade ?

    If you don't need these features it doesn't seem that difficult to me to just ignore them.
    It is likely that you'll have perhaps to spend a bit more time to grasp the new form
    package but it seems quite in the line of what was previously available.

    >
    > As for what I would want in .NET - nobody asked way back when and what I
    > think really doesn't matter. --.NUTS IS GOING TO MARKET.-- The features in
    > NET are in many ways redundant of features I can find in other sources.
    > Regardless of what is in .NET, I want to possess skills that are marketable.
    > With the advent of .NUTS, my VB skills are irrelevant or will be in less
    > than 18 months. So are yours.


    Language is almost irrelevant to me (personaly I plan to use C# as a default while my
    company will probably use VB.NET as a default). But the Framework Class Library seems to
    me really the big win. Also the fact that you use the same language to build either a
    Windows or a Web app. makes all this more marketable...

    > Every VB programmer must decide if he/she wants to continue programming.
    > If he/she does, --FORGET VB.NET! IT IS IRRELEVANT-- he/she MUST learn C/C++
    > because that is the foundation of Java, C#, Visual C++, Borland C++, Borland
    > J Builder and VB.NET regardless of the noise coming from MS, Borland or Sun
    > about 'ease of use'. Company specific intermediate languages and compilers
    > will change (see the shift to .NET as proof) where C++ is always present
    > regardless of system or context. Once you have some C++, you are not bound
    > to MS or Sun or Borland - you can acquire any of the intermediate languages
    > with the same relative ease. Why bust your gut learning a twisted form of
    > Basic that really isn't basic at all?


    Language is almost a non issue. I'm experienced in VB, I never had to use C++ but reading
    this code is not a real problem and JavaScript is mandatory for client side scripting so I
    hope to be quite comfortable with VB.NET, C# and JScript.NET. Anyway the Framework Class
    Library should have more attention than language (non IMO) issues.

    > My personal technical improvement is now and will continue to be on C++.
    > I am reading a very interesting book by Bjarne Stroustrup called "The C++
    > Programming Language". He invented it. really he did. see

    http://www.research.att.com/~bs/C++.html.
    > I like reading his definitions of C++ and then discovering how MS or Borland
    > violated ANSI C++ standards. It really isn't that hard to understand.
    > no more so than VB was in 1992.


    I hope you still plan to take advantage of the Framework Class Library ;-)

    IMO we have now *much more* capabilities for a *slightly* increased complexity and what
    you learn about the Framework Class remains the same whatever language you may use in the
    future...

    I believe we have pretty much exhausted the subject. No surprise, we just not rate things
    the same, especially the language issue.


    > Thanks
    >
    > Don
    >
    > "Patrice Scribe" <scribe@chez.com> wrote:
    > >I forgot to address some points :
    > >- ASP.NET compiles behind the scene. I expect the same thing for Windows

    > Scripting.
    > >That
    > >is it's look like scripting but the code is actually compiled for you.
    > >- ASP.NET provides server side controls. In ASP you have to interlace HTML

    > and code
    > >(or to
    > >output HTML tags using code). This is no more needed in ASP.NET. You have

    > an HTML part
    > >that includes server side control and you have a code behind file that allows

    > to set
    > >properties and react to events.
    > >- the Framework Class Library gives access to most underlying features including

    > messaging
    > >(no big deal here). For "monitoring" I was specifically thinking about adding

    > performance
    > >counters to an application. This a matter of minutes with .NET.
    > >
    > >I can understand it doesn't fit your needs but is it a reason to quality

    > this a .NUTS
    > >strategy.
    > >
    > >What would be your expectations about the ideal .NET you are dreaming of

    > ?
    > >
    > >--
    > >Patrice Scribe




  14. #14
    Joe \Nuke Me Xemu\ Foster Guest

    Re: Summary of This Discussion :re: Patrice's latest2

    "Ian R" <ianr@na.net> wrote in message <news:3b5c258e$1@news.devx.com>...

    > "Mike Mitchell" <kylix_is@hotmail.com> wrote in message
    > news:3b5abcd3.7747485@news.devx.com...
    > >
    > > Okay, the gospel truth is, a high percentage of classic VB programmers
    > > DO NOT want it. You can argue till the cows come home about the actual
    > > percentage, but I would place it at higher than 60%.


    > I think you're confusing things with what you want Mike. Most VB developers
    > I've talked to are excited with most of the features in VB.NET. The few who
    > don't seem to care simply won't use it.


    How did you select your sample? Did you poll only those who had been
    known to gripe about how VB wasn't Java, since those who didn't couldn't
    really be "VB developers" but mere lowly "macro coders"?

    --
    Joe Foster <mailto:jfoster@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!



  15. #15
    Larry Linson Guest

    Re: Summary of This Discussion :re: Patrice's latest2


    "Ian R" <ianr@na.net> wrote:

    > The _few_ who don't seem to care simply won't use it.


    (LL says: emphasis on _few_ is mine, not Ian's)

    Ohmigawd, approximately 100% of the developers in the VB SIG of my User Group
    are "few" -- they are going to be soooo disappointed. I've personally discussed
    this with around 100 of them, and not one of those is enthusiastic about
    .NET. But, then, few if any of them deal with the kind of applications at
    which .NET is targeted.


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
HTML5 Development Center
 
 
FAQ
Latest Articles
Java
.NET
XML
Database
Enterprise
Questions? Contact us.
C++
Web Development
Wireless
Latest Tips
Open Source


   Development Centers

   -- Android Development Center
   -- Cloud Development Project Center
   -- HTML5 Development Center
   -- Windows Mobile Development Center