From: Bruno van Dooren [MVP VC++] on
> Nishant is my hero in thinking C# could be vastly improved.

It could, but then it would be C++/CLI :-)

> Your blog says there's more good C# programmers than C++ programmers,
> which surprised me. Perhaps due to my focus to Windows internals and
> hardware, I've not met many of them! :-) I think it's true in enterprise
> that .NET is taking over, less sure outside of big corps.

Most of my clients are Megacorps. Or at least, companies willing to shill
out $$$ per day for my services.
They employ hordes of C# programmers, but have only a couple of good C++
programmers.
And those C++ programmers -without exception- have no clue about C++/CLI.
They MIGHT know it exists, but that's where it stops.

I meddle in device drivers from time to time, mostly as a hobby, and in that
world, There is only C++ and C. That will be the case for a long time to
come. But that world is tiny, compared to the world of custom enterprise
software.
Megacorps love .NET and C# like they haven't loved anything before.
good C++ programmers are hard to come by.
good C# programmers are easier to find.
If I were an ICT manager, responsible for internal software development, I'd
love C# too. Less cost on developers, and better guarantees for maintenance.

--

Kind regards,
Bruno van Dooren
bruno_nos_pam_van_dooren(a)hotmail.com
Remove only "_nos_pam"


From: Joseph M. Newcomer on
That's one of the reasons I haven't moved to C#; my code will be taken over by in-house
MFC programmers. They don't want to see the Standard C++ library (I have no idea why; it
has lots of good features), they don't want third-party libraries (unless they come with
source--for very good reasons, in fact), and as you point out, they probably don't want to
see C++/CLI. They certainly don't want to see C# (I've even asked if they want C#)

In the device driver world, unless you are writing a streaming driver, you should avoid
C++. I spent a half-hour at Driver DevCon talking to the C++ compiler lead, who explained
in exquisite detail why C++ cannot work in the kernel (it is nothing short of a miracle
that streaming drivers work at all; move outside a limited subset of C++ and there will be
problems). Any general driver that uses C++ is essentially unsupported as a concept.

I don't recall all the details, but I went away from that conversation with one important
bit set: don't use C++ in drivers!
joe

On Mon, 12 Feb 2007 09:27:05 +0100, "Bruno van Dooren [MVP VC++]"
<bruno_nos_pam_van_dooren(a)hotmail.com> wrote:

>> Nishant is my hero in thinking C# could be vastly improved.
>
>It could, but then it would be C++/CLI :-)
>
>> Your blog says there's more good C# programmers than C++ programmers,
>> which surprised me. Perhaps due to my focus to Windows internals and
>> hardware, I've not met many of them! :-) I think it's true in enterprise
>> that .NET is taking over, less sure outside of big corps.
>
>Most of my clients are Megacorps. Or at least, companies willing to shill
>out $$$ per day for my services.
>They employ hordes of C# programmers, but have only a couple of good C++
>programmers.
>And those C++ programmers -without exception- have no clue about C++/CLI.
>They MIGHT know it exists, but that's where it stops.
>
>I meddle in device drivers from time to time, mostly as a hobby, and in that
>world, There is only C++ and C. That will be the case for a long time to
>come. But that world is tiny, compared to the world of custom enterprise
>software.
>Megacorps love .NET and C# like they haven't loved anything before.
>good C++ programmers are hard to come by.
>good C# programmers are easier to find.
>If I were an ICT manager, responsible for internal software development, I'd
>love C# too. Less cost on developers, and better guarantees for maintenance.
Joseph M. Newcomer [MVP]
email: newcomer(a)flounder.com
Web: http://www.flounder.com
MVP Tips: http://www.flounder.com/mvp_tips.htm
From: Bob Moore on
On Mon, 12 Feb 2007 05:40:21 GMT, David Ching wrote:

>Yeah, I think C# definitely is too much like Delphi

Some of us think that's a GOOD thing. I've spent most of the last two
years working part-time in Delphi and it was a breath of fresh air.
Decent RAD, but with some real performance oomph behind it.

One of our guys even wrote a complex multi-threaded service in Delphi,
but that's just *perverse* :-)

--
Bob Moore
http://bobmoore.mvps.org/
(this is a non-commercial site and does not accept advertising)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Do not reply via email unless specifically requested to do so.
Unsolicited email is NOT welcome and will go unanswered.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
From: David Ching on
"MrAsm" <mrasm(a)usa.com> wrote in message
news:teius2548cfe6qde07a84i4nkb5731da93(a)4ax.com...
> In your opinion, do you think that would be possible to buil some
> "important" software like Office or Visual Studio or Mathematica using
> C#? (...and don't require to user wait long time for app start-up or
> computations.)
>

The 5 second startup time seemed insane for me too, but it's funny that when
testing my own .NET app under VS2005 in Vista, it starts right up. Perhaps
because the .NET libraries are already cached.

In the tech note accompaning the .NET framework item in Windows Update, it
says, "Some applications that depend on the .NET framework for support
include the following:

* Microsoft Windows Live OneCare
* Intuit QuickBooks
* HP Image Zone
* Streamload
* Paypal
* Grouper


Not exactly a stellar list, but the QuickBooks item shows one business class
app.

-- David (MVP)




From: MrAsm on
On Mon, 12 Feb 2007 23:08:40 GMT, "David Ching"
<dc(a)remove-this.dcsoft.com> wrote:

>"MrAsm" <mrasm(a)usa.com> wrote in message
>news:teius2548cfe6qde07a84i4nkb5731da93(a)4ax.com...
>> In your opinion, do you think that would be possible to buil some
>> "important" software like Office or Visual Studio or Mathematica using
>> C#? (...and don't require to user wait long time for app start-up or
>> computations.)
>>
>
>The 5 second startup time seemed insane for me too, but it's funny that when
>testing my own .NET app under VS2005 in Vista, it starts right up. Perhaps
>because the .NET libraries are already cached.

I thought that C# and .NET code was just-in-time compiled, so I tought
that startup times would be longer for bigger apps (when more code
must be compiled), like Office or Visual Studio...

MrAsm