J
G
Gerry Hickman
Hi Jim,
I see two sides to this. On the one hand I have no sympathy for anyone
that chose VB6. I warned against it six years ago, and people who
ignored the advice did so at their own risk. The thought that it would
be resurrected fills me with dread!
In the wider picture, however, I agree with the MVPs over the current
obsession of Microsoft to discontinue everything that they used to do
well, and replace it with Disneyland nonsense targetted at home users
and kids.
I see two sides to this. On the one hand I have no sympathy for anyone
that chose VB6. I warned against it six years ago, and people who
ignored the advice did so at their own risk. The thought that it would
be resurrected fills me with dread!
In the wider picture, however, I agree with the MVPs over the current
obsession of Microsoft to discontinue everything that they used to do
well, and replace it with Disneyland nonsense targetted at home users
and kids.
R
Rob R. Ainscough
As far as I can tell, VB6 vs. VB.NET has NOT actually been able to produce a
Windows application that is more stable, faster to produce, and easier to
maintain -- at least not a significant difference that affects the funding
of such development projects.
VB6 & web development vs. VB.NET with ASP for web development, well that is
a different story -- VB.NET has actually reduced time to produce the final
product for web apps, but certainly no less buggy.
But I think we still have a LONG way to go before software development
becomes truely "reuseable" and we see a reduced development cycle from
concept to final product. We are still very much bonded to the syntax -- I
hope this will change and once and for all remove the language camps that so
many developers get hung up on.
The implications are obvious, good designers will prosper, code monkeys (aka
work with the spec and only the spec) will fall the way of accountants. The
days of big salaries are long gone just because someone can say "web".
Windows application that is more stable, faster to produce, and easier to
maintain -- at least not a significant difference that affects the funding
of such development projects.
VB6 & web development vs. VB.NET with ASP for web development, well that is
a different story -- VB.NET has actually reduced time to produce the final
product for web apps, but certainly no less buggy.
But I think we still have a LONG way to go before software development
becomes truely "reuseable" and we see a reduced development cycle from
concept to final product. We are still very much bonded to the syntax -- I
hope this will change and once and for all remove the language camps that so
many developers get hung up on.
The implications are obvious, good designers will prosper, code monkeys (aka
work with the spec and only the spec) will fall the way of accountants. The
days of big salaries are long gone just because someone can say "web".
G
Guest
Jim,
Computerworld ran an article which stated at the end that perhaps MS should
focus on improving the migration tool. I'd like that more than having MS try
to support a legacy application. Seems like legacy support is a time waster.
Jamie
Computerworld ran an article which stated at the end that perhaps MS should
focus on improving the migration tool. I'd like that more than having MS try
to support a legacy application. Seems like legacy support is a time waster.
Jamie
J
Jim Hubbard
thejamie said:
I agree. This would be a better solution. The problem is that we don't
really have a tool to do migration from VB6 to VB.Net. If Microsoft could
give us a tool that worked (or even told us how to fix the code instead of
only why it won't work) that would be great.
It would reduce the costs of porting applications below the current 60% of
the original development costs, and would go a long way towards quelling the
furor over the end of VB6 support.
We can dream, can't we?
Jim Hubbard
G
Guest
I haven't found the migration tool to be totally useless though. For
example, I have trouble, I assume I am not alone, finding simple things such
as the difference between say -
ListIndex and SelectedIndex
The way you find it is the problem... To find this information in the help,
took longer than it did to go into VB6, create a simple listbox routine,
reopen and migrate the project to .NET 2003. Once I see the conversion, it
become obvious the way it works and I can move on. This process works well.
But it is slow and tedious.
So, IMHO, I think the migration tool "works" fine. But the documentation is
so completely different in Net 2003... you can't just hit F1, find an
example, and move on. Instead you have to create your own examples... I
think I can safely say that there is a serious problem with the migration
tools. They work well, but they don't work easily. My head hurts just
thinking about it. Try doing this under pressure.
Jamie
example, I have trouble, I assume I am not alone, finding simple things such
as the difference between say -
ListIndex and SelectedIndex
The way you find it is the problem... To find this information in the help,
took longer than it did to go into VB6, create a simple listbox routine,
reopen and migrate the project to .NET 2003. Once I see the conversion, it
become obvious the way it works and I can move on. This process works well.
But it is slow and tedious.
So, IMHO, I think the migration tool "works" fine. But the documentation is
so completely different in Net 2003... you can't just hit F1, find an
example, and move on. Instead you have to create your own examples... I
think I can safely say that there is a serious problem with the migration
tools. They work well, but they don't work easily. My head hurts just
thinking about it. Try doing this under pressure.
Jamie
J
Jim Hubbard
Rob R. Ainscough said:
You are right.
Again, you are right. But, who is really in the market to produce web
applications? Not many companies - compared to the based of companies that
produce applications.
The web is unstable. If your internet connection is down, what happens to
the web-based applications you need to do your job? If they are written
correctly they can run offline as well as on.....but current applications
can do that wihout .Net
So, what's the big deal with .Net? It's a lttle long-winded, but check out
my blog at http://poderthis.blogspot.com/ and I'll show you.
That'd be nice. But, didn't we get awfully close with activeX components?
You could use them with C++, Visual Basic or Borland's C++ or Delphi. In
order to have true interoperability across languages, there is a significant
trade-off in security.
Not the security of the environment, but the security of intellectual
property rights. In .Net, it is quite easy to dissassemble another person's
work and steal their ideas and code - even if an obfuscator is used. Binary
code was much more secure in this aspect.
This one flaw in .Net virtually forces you to make your applications
web-applications to hide your proprietary code on the servers. But, with
the security breaches in Microsoft server products, even this is not very
secure.
Ahhhhh, the "good ol' days". I remember them well.
Jim Hubbard
G
Gerry Hickman
Hi Jim,
In my view, these are the only apps to bother with. The problem for me,
is that people all over the world need to use my apps at anytime of day
or night - they all have different operating systems
Windows/Mac/Linux/Sun and different browsers IE/Safari/WebTV/Mozilla.
I did all the internal company stuff in 3 tier web apps, and it's great
because it works just the same whether they're at their desk or on the
other side of the world. I also never need to worry about re-installing
or patching every time someone buys a new computer/laptop - they just
type in the URL and off they go.
Uh?? I wonder if Google and Amazon are aware of this fact?
Well you could argue "email is unreliable" because "what if the internet
is down".
I don't really have time to read blogs, but can you give us a short summary?
Let's not forget JScript.
Security is NOT the right word to use here!
I see it the other way round. I have all my stuff coded up in JScript; 3
tier apps, Enterprise management, Change and Configuration management,
Software deployment (GPO Alternate), and W3C compliant front-ends. If
anyone wants the source code, just ask, or find me on groups like WMI. I
post it all the time. Why should I care who can see the code? The way I
see it is we should all post our code, then we learn from others.
In my view, these are the only apps to bother with. The problem for me,
is that people all over the world need to use my apps at anytime of day
or night - they all have different operating systems
Windows/Mac/Linux/Sun and different browsers IE/Safari/WebTV/Mozilla.
I did all the internal company stuff in 3 tier web apps, and it's great
because it works just the same whether they're at their desk or on the
other side of the world. I also never need to worry about re-installing
or patching every time someone buys a new computer/laptop - they just
type in the URL and off they go.
Uh?? I wonder if Google and Amazon are aware of this fact?
Well you could argue "email is unreliable" because "what if the internet
is down".
I don't really have time to read blogs, but can you give us a short summary?
Let's not forget JScript.
Security is NOT the right word to use here!
I see it the other way round. I have all my stuff coded up in JScript; 3
tier apps, Enterprise management, Change and Configuration management,
Software deployment (GPO Alternate), and W3C compliant front-ends. If
anyone wants the source code, just ask, or find me on groups like WMI. I
post it all the time. Why should I care who can see the code? The way I
see it is we should all post our code, then we learn from others.
J
Jim Hubbard
Hi Gerry,
I agree....to a point. Web applications are definitely preferrable to the
installation headaches of multiple user configurations - especially when
looked at globally. However, web applications lack the power of desktop
apps if you are talking about a graphics-oriented application or very rich
UI and they lack the ease-of-use that you can achieve by settling on one
desktop and coding to take advantage of said desktop OS (whether it be XP,
OSX or Linux).
That works fine for databases, simple text and financial applications, but a
rich user interface that could be built in a RAD environment (until .Net)
was not really possible on the web. Of course you can do Flash or
Shockwave to create rich UIs, but that takes more art talent than most
programmers have on tap. And, they are not exactly RAD tools.
Sure they are. That's why their UIs and applications are so simple (when
compared to an N-tier enterprise application).
But, email is hardly a component that will shut down a company if it isn't
working at the moment. On the other hand, of you use a web-based
point-of-sale terminal to ring up your customerss and the internet
connection is lost - now you have a problem.
While Amazon can handle this down time (if it doesn;t last too long), it is
an entirely different matter for a sales clerk with a customer standing in
line at a small business to deal with. It is also far more financially
devistating to the small business owner to be down for a day than it is for
Amazon or Google.
No problem.....
1) .Net was not written for you or I, it was written by and for
Microsoft to further their "software as a service" goals.
2) .Net is using the Microsoft programming community as the world's
largest beta test group for it's new programming tools that it will then use
to wring more money out of those same users by changing to "rentalware"
instead of software as we understand and purchase it today.
3) .Net does have value for those individuals that want to do "software
as a service" - but they are a minority at this point.
4) Microsoft doesn't care what's good for programmers outside Microsoft
or the world IT community in general. It ignores the pleadings of it's
customer base and can do so simply beacause it has become too powerful.
(Breaking up Microsoft is still a good idea.)
Ahhhh.....sorry about that.
What is? I consider protecting my intelectual property security.
Do you post the source for your entire application (all n tiers)?
Most companies will not. It is not in their best financial interests to
give away their corporate model or source code.
Open source - as a model - will fail. You can already see it with Linux.
Linux has fragmented into so many distros that software written on one
distro (especially if it takes advantage of hooks or APIs built into that
distro) will not run on many others.
This is where Linux went wrong (and where Java is soon to screw up). Even
Microsoft talking about opening up the source code is a big mistake. But,
there is a right way to accomplish the goal of interoperability without
sharing the source code.
Simply share ALL of the OS's APIs. If Linux had opened ALL APIs to
developers and kept the source code patented for its use and the use of its
customers only, Linux would not be fragmented into so many distros and would
have already put Microsoft on the ropes through the use of more
simplification through APIs and hooks into the dekstop and underlying
kernel.
This would've also allowed Linux to grow much faster because all Linux
developers would be working on the same desktop - making the market of
supported applications available much larger than it currently is. (Yes, I
know there are thousands of Linux titles out there, but there are'nt nearly
that many that have professional teams backing them up - and that is
required by any serious company. Freeware just doesn't cut it in most
business scenarios in the US.)
If Microsoft would open up ALL of their code's APIs, they wouldn't need to
show their source code and people could write more compatible code to work
with the OS and Office suites. And, that's all developers really want....to
have a level playing field with Microsoft APIs. We don't even need the
source code.
Open source leads to fragmentation of the application and user base -
resulting in higher application costs in the long run.
This is just my opinion, of course.
Jim Hubbard
I agree....to a point. Web applications are definitely preferrable to the
installation headaches of multiple user configurations - especially when
looked at globally. However, web applications lack the power of desktop
apps if you are talking about a graphics-oriented application or very rich
UI and they lack the ease-of-use that you can achieve by settling on one
desktop and coding to take advantage of said desktop OS (whether it be XP,
OSX or Linux).
That works fine for databases, simple text and financial applications, but a
rich user interface that could be built in a RAD environment (until .Net)
was not really possible on the web. Of course you can do Flash or
Shockwave to create rich UIs, but that takes more art talent than most
programmers have on tap. And, they are not exactly RAD tools.
Sure they are. That's why their UIs and applications are so simple (when
compared to an N-tier enterprise application).
But, email is hardly a component that will shut down a company if it isn't
working at the moment. On the other hand, of you use a web-based
point-of-sale terminal to ring up your customerss and the internet
connection is lost - now you have a problem.
While Amazon can handle this down time (if it doesn;t last too long), it is
an entirely different matter for a sales clerk with a customer standing in
line at a small business to deal with. It is also far more financially
devistating to the small business owner to be down for a day than it is for
Amazon or Google.
No problem.....
1) .Net was not written for you or I, it was written by and for
Microsoft to further their "software as a service" goals.
2) .Net is using the Microsoft programming community as the world's
largest beta test group for it's new programming tools that it will then use
to wring more money out of those same users by changing to "rentalware"
instead of software as we understand and purchase it today.
3) .Net does have value for those individuals that want to do "software
as a service" - but they are a minority at this point.
4) Microsoft doesn't care what's good for programmers outside Microsoft
or the world IT community in general. It ignores the pleadings of it's
customer base and can do so simply beacause it has become too powerful.
(Breaking up Microsoft is still a good idea.)
Ahhhh.....sorry about that.
What is? I consider protecting my intelectual property security.
Do you post the source for your entire application (all n tiers)?
Most companies will not. It is not in their best financial interests to
give away their corporate model or source code.
Open source - as a model - will fail. You can already see it with Linux.
Linux has fragmented into so many distros that software written on one
distro (especially if it takes advantage of hooks or APIs built into that
distro) will not run on many others.
This is where Linux went wrong (and where Java is soon to screw up). Even
Microsoft talking about opening up the source code is a big mistake. But,
there is a right way to accomplish the goal of interoperability without
sharing the source code.
Simply share ALL of the OS's APIs. If Linux had opened ALL APIs to
developers and kept the source code patented for its use and the use of its
customers only, Linux would not be fragmented into so many distros and would
have already put Microsoft on the ropes through the use of more
simplification through APIs and hooks into the dekstop and underlying
kernel.
This would've also allowed Linux to grow much faster because all Linux
developers would be working on the same desktop - making the market of
supported applications available much larger than it currently is. (Yes, I
know there are thousands of Linux titles out there, but there are'nt nearly
that many that have professional teams backing them up - and that is
required by any serious company. Freeware just doesn't cut it in most
business scenarios in the US.)
If Microsoft would open up ALL of their code's APIs, they wouldn't need to
show their source code and people could write more compatible code to work
with the OS and Office suites. And, that's all developers really want....to
have a level playing field with Microsoft APIs. We don't even need the
source code.
Open source leads to fragmentation of the application and user base -
resulting in higher application costs in the long run.
This is just my opinion, of course.
Jim Hubbard
M
Marc Scheuner [MVP ADSI]
http://www.eweek.com/article2/0,1759,1774642,00.asp
Mind you: **SOME** Microsoft MVP - but far not ALL !
Marc
================================================================
Marc Scheuner May The Source Be With You!
Berne, Switzerland m.scheuner -at- inova.ch
Mind you: **SOME** Microsoft MVP - but far not ALL !
Marc
================================================================
Marc Scheuner May The Source Be With You!
Berne, Switzerland m.scheuner -at- inova.ch
G
Gerry Hickman
Hi Jim,
OK, this is very interesting. I hear this argument day and night, but I
have to say I'm having trouble agreeing. We live in a world where office
workers are obsessed with "the internet", even the non-technical ones!
One minute they're on Yahoo checking email, the next they're on Google
searching for things, suddenly they're using their "on-line banking",
and the rest of the time they're booking cheap-ass holidays. For these
users, the "web interface" is the most natural thing in the world. They
don't like Flash and they don't like pop-ups. They want simple W3C
compliant forms where they can just click a few drop downs and hit the
"submit" button. This is exactly how I design all my web applications
for my clients, and the users LOVE it. For them, it's just like being on
a holiday booking site. I use super-fast cross-browser scripting to
provide things like expand/collapse tree style navigation systems
(without a server round trip as per .NET), use frames so that they can
input to one pane while seeing real-time results in an other, and use
things like client-side filtering and sorting. I use cross-browser DHTML
so that elements can collapse and expand as the content changes. I can't
think of any advantage to suddenly having all this converted to "Avalon"
or any other bungling UI technology, except that it would slow the user
down, confuse them, and cease to work on many devices.
Don't believe me?
Well consider this simple example. Two years ago everyone would be
"laughing out loud" if you'd proposed a 1980s style HTML page with just
one graphic and one text box as the UI to your world facing web-site,
but while Microsoft were busy trying to make their enormous web graphics
look like something from Macromedia hell, Google went ahead with the
1980s concept, and the rest as they say, is history! What Google proved,
is that people can't stand Flash and graphics that take ages to
download. They also don't like sites over-loaded with eye-candy. They
also don't want to "take" one of Microsoft's mind-numbing "surveys".
They just want to get straight to the information and get on with their
work.
Don't believe me?
Well, it's strange so many people switched to Mozilla, fed up with all
the Flash, pop-ups and ActiveX nonsense. Now they can get a clean and
fast W3C web experience from Mozilla, and avoid the lag they get when
using Microsoft's browser. One problem, is that IE downloads plug-ins
BEFORE it asks you if you want the plug-in, so that kills the user
experience too, if you don't have the plug-in it fires an error for
every page thereafter that tries to use it. Mozilla simply ignores such
things until you say you want it. Many sites have hidden Flash and other
types of rogue content designed to spy on the user.
With all these things added together, I just don't buy it that the "rich
UI" is necessary for people to get their work done. I also can't agree
that small companies somehow "need rich UIs" when big companies clearly
don't?
I can't agree here. Are you saying Google's indexing system and Amazon's
order processing system are less complex or less capable than the
average internal corporate application. Amazon's front-end is not
exactly what I'd call simple anyway. I just don't think the fron-end and
back-end of these systems are closely related. Google's front-end was
not put there through lack of technical ability, but because they knew
simplicity was the key to success.
OK, but it's strange I'm not out of a job in that case? No one has ever
complained to me about this, in fact it's the opposite - they can now
use their apps while visiting places like Africa and China from hotel
rooms, internet cafe's and so on - not to mention lazy-ass directors who
can now "work from home". My work offers are flooding in, and no one has
ever said the UI is "too basic".
This is what I call "vendor tie in", and I think it will backfire. Up
until around 2000, I had no interest in anything that wasn't Microsoft!
The marketing was honest, the product pitch was realistic, the interop
was good, the non-stop give-aways of tools and SDKs, excellent docs,
reskits and KBs. I.E. had excellent W3C and ECMA implementations.
Now they've thrown all that away. The recent marketing papers on
smart-clients are misinformation at best and lies at worse. The reskits
have gone, the good white papers deleted, the MSDN Jan2005 is broken,
the docs have degenerated into kiddie-speak, and are now being authored
in Word! The utter nonsense about Avalon is hard to read without
vomiting. It's no wonder more and more people are looking elsewhere.
I just think when you say "security" it implies "buffer over-runs" etc.
It would be a bit big to fit on here. There's no secrets; the back end
has views and stored procedures, there's a transactional layer in the
middle, then server-side web, and then client side web. Separate to
this, there's all the change and configuration management stuff and
deployment stuff - I've already posted most of the code for that over
the last year on so, on groups like WMI and WSH.
Thing is, the back-end/middle tier code is of no use to anyone unless
they have the same business models, same network topology etc. Have a
look at your own code-base, do you think it would plug straight into a
different company without hours of work?
When I go into different companies, you may think I could "steal" the
source code and "get rich quick"?
Errr, not quite. If you'd seen the bungling mess these companies systems
are in, you'd be running screaming from the building! The _last_ thing
you'd want to do is steal this source code, the only thing you'd want to
do is demand an instant re-write from the ground up.
I don't agree with this, but see next part.
Yup, I certainly agree with this! Fragmentation!
There are some major problems with open source and free apps, but I
don't think it will "fail" outright. Look at PERL and CPAN for example.
That's been around for years, the categorization is clear, and it
certainly hasn't "failed". What about BSD Unix, surely you are not
saying it's "failed". Isn't that what Microsoft had to use when Hotmail
fell over on Windows?
I'm actually thinking of moving all my web apps and database back ends
to UNIX. I could get something like Sun 4 way UltraSPARC III servers
with fibre SANs, the new superfast TCP/IP stack, and get true 64 bit 4
way processing of things like XML parsing and full-text indexing. Couple
that to CORBA middle-ware and some Apache modules and WHOA! Compare that
to what you can do with bloated Windows 2003 server where they're more
worried about the "XP look and feel" than the interface for the fibre
and SAN.
Hmmmm, food for thought.
OK, this is very interesting. I hear this argument day and night, but I
have to say I'm having trouble agreeing. We live in a world where office
workers are obsessed with "the internet", even the non-technical ones!
One minute they're on Yahoo checking email, the next they're on Google
searching for things, suddenly they're using their "on-line banking",
and the rest of the time they're booking cheap-ass holidays. For these
users, the "web interface" is the most natural thing in the world. They
don't like Flash and they don't like pop-ups. They want simple W3C
compliant forms where they can just click a few drop downs and hit the
"submit" button. This is exactly how I design all my web applications
for my clients, and the users LOVE it. For them, it's just like being on
a holiday booking site. I use super-fast cross-browser scripting to
provide things like expand/collapse tree style navigation systems
(without a server round trip as per .NET), use frames so that they can
input to one pane while seeing real-time results in an other, and use
things like client-side filtering and sorting. I use cross-browser DHTML
so that elements can collapse and expand as the content changes. I can't
think of any advantage to suddenly having all this converted to "Avalon"
or any other bungling UI technology, except that it would slow the user
down, confuse them, and cease to work on many devices.
Don't believe me?
Well consider this simple example. Two years ago everyone would be
"laughing out loud" if you'd proposed a 1980s style HTML page with just
one graphic and one text box as the UI to your world facing web-site,
but while Microsoft were busy trying to make their enormous web graphics
look like something from Macromedia hell, Google went ahead with the
1980s concept, and the rest as they say, is history! What Google proved,
is that people can't stand Flash and graphics that take ages to
download. They also don't like sites over-loaded with eye-candy. They
also don't want to "take" one of Microsoft's mind-numbing "surveys".
They just want to get straight to the information and get on with their
work.
Don't believe me?
Well, it's strange so many people switched to Mozilla, fed up with all
the Flash, pop-ups and ActiveX nonsense. Now they can get a clean and
fast W3C web experience from Mozilla, and avoid the lag they get when
using Microsoft's browser. One problem, is that IE downloads plug-ins
BEFORE it asks you if you want the plug-in, so that kills the user
experience too, if you don't have the plug-in it fires an error for
every page thereafter that tries to use it. Mozilla simply ignores such
things until you say you want it. Many sites have hidden Flash and other
types of rogue content designed to spy on the user.
With all these things added together, I just don't buy it that the "rich
UI" is necessary for people to get their work done. I also can't agree
that small companies somehow "need rich UIs" when big companies clearly
don't?
I can't agree here. Are you saying Google's indexing system and Amazon's
order processing system are less complex or less capable than the
average internal corporate application. Amazon's front-end is not
exactly what I'd call simple anyway. I just don't think the fron-end and
back-end of these systems are closely related. Google's front-end was
not put there through lack of technical ability, but because they knew
simplicity was the key to success.
OK, but it's strange I'm not out of a job in that case? No one has ever
complained to me about this, in fact it's the opposite - they can now
use their apps while visiting places like Africa and China from hotel
rooms, internet cafe's and so on - not to mention lazy-ass directors who
can now "work from home". My work offers are flooding in, and no one has
ever said the UI is "too basic".
This is what I call "vendor tie in", and I think it will backfire. Up
until around 2000, I had no interest in anything that wasn't Microsoft!
The marketing was honest, the product pitch was realistic, the interop
was good, the non-stop give-aways of tools and SDKs, excellent docs,
reskits and KBs. I.E. had excellent W3C and ECMA implementations.
Now they've thrown all that away. The recent marketing papers on
smart-clients are misinformation at best and lies at worse. The reskits
have gone, the good white papers deleted, the MSDN Jan2005 is broken,
the docs have degenerated into kiddie-speak, and are now being authored
in Word! The utter nonsense about Avalon is hard to read without
vomiting. It's no wonder more and more people are looking elsewhere.
I just think when you say "security" it implies "buffer over-runs" etc.
It would be a bit big to fit on here. There's no secrets; the back end
has views and stored procedures, there's a transactional layer in the
middle, then server-side web, and then client side web. Separate to
this, there's all the change and configuration management stuff and
deployment stuff - I've already posted most of the code for that over
the last year on so, on groups like WMI and WSH.
Thing is, the back-end/middle tier code is of no use to anyone unless
they have the same business models, same network topology etc. Have a
look at your own code-base, do you think it would plug straight into a
different company without hours of work?
When I go into different companies, you may think I could "steal" the
source code and "get rich quick"?
Errr, not quite. If you'd seen the bungling mess these companies systems
are in, you'd be running screaming from the building! The _last_ thing
you'd want to do is steal this source code, the only thing you'd want to
do is demand an instant re-write from the ground up.
I don't agree with this, but see next part.
Yup, I certainly agree with this! Fragmentation!
There are some major problems with open source and free apps, but I
don't think it will "fail" outright. Look at PERL and CPAN for example.
That's been around for years, the categorization is clear, and it
certainly hasn't "failed". What about BSD Unix, surely you are not
saying it's "failed". Isn't that what Microsoft had to use when Hotmail
fell over on Windows?
I'm actually thinking of moving all my web apps and database back ends
to UNIX. I could get something like Sun 4 way UltraSPARC III servers
with fibre SANs, the new superfast TCP/IP stack, and get true 64 bit 4
way processing of things like XML parsing and full-text indexing. Couple
that to CORBA middle-ware and some Apache modules and WHOA! Compare that
to what you can do with bloated Windows 2003 server where they're more
worried about the "XP look and feel" than the interface for the fibre
and SAN.
Hmmmm, food for thought.
A
Alvin Bruney [ASP.NET MVP]
However, web applications lack the power of desktop apps if you are talking
desktop machines fall under that category.
issue
--
Regards,
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
_________________________
nope. standard servers run dual or quad cpu's at the very least. very few
desktop machines fall under that category.
nope. that's a design/architecture issue. very distance from a platform
issue
--
Regards,
Alvin Bruney
[Shameless Author Plug]
The Microsoft Office Web Components Black Book with .NET
available at www.lulu.com/owc
_________________________
C
Cor Ligthert
Gerry,
We don't disagree, however certainly not agree.
When we have to deal with a simple "Give me pages" application, than
straight HTML with JavaScript has my preference. I don't than like all the
things as you mention as well.
However this preference change, when we start in large dialogs where
databases are involved. For that we see around all kind of clumsy HTML
pages, which force me to enter fields and than at the end tells to re-enter
the page because the year is written as yy instead of yyyy.
This is the place where Net has to come in and with the Richest Dialog it
can supply. In that should be the evaluated windowsform dialog be at least
the goal (what is at the moment not to reach for worldwide used pages,
where I am talking about, otherwise my choose are webservice applications).
Beside that, because this goes about VBNet, the time needed to make this
kind of pages shrinks enormous by using VNNet.
(Did you try it ever, you will be suprissed)
Just my opinion.
Cor
We don't disagree, however certainly not agree.
When we have to deal with a simple "Give me pages" application, than
straight HTML with JavaScript has my preference. I don't than like all the
things as you mention as well.
However this preference change, when we start in large dialogs where
databases are involved. For that we see around all kind of clumsy HTML
pages, which force me to enter fields and than at the end tells to re-enter
the page because the year is written as yy instead of yyyy.
This is the place where Net has to come in and with the Richest Dialog it
can supply. In that should be the evaluated windowsform dialog be at least
the goal (what is at the moment not to reach for worldwide used pages,
where I am talking about, otherwise my choose are webservice applications).
Beside that, because this goes about VBNet, the time needed to make this
kind of pages shrinks enormous by using VNNet.
(Did you try it ever, you will be suprissed)
Just my opinion.
Cor
G
Gerry Hickman
Hi Cor,
Yes, but to me this is not a limitation of web applications, but simply
a case of BAD design. I could also cite a hundred cases of Microsoft
Access forms (or VB forms) that are inferior and more clumsy than a well
designed web interface. In fact most companies I'm dealing with are
trying to get rid of these "rich" forms because they are less intuitive
than the web equivalents. The other thing with the web, is that you can
make the navigation really nice and have much better text handling to
help the user as they move through the app. You can use text in VB forms
of course, but there's something about web text and CSS that just seems
nicer. The web also guarantees Accessibility, which is becoming a hot
policy issue.
Did I try what? I've been using .NET since the first betas and I'm
evaluating .NET 2.0 on a new server. I've never used VB.NET (or VB
Anything) because I don't consider VB to be a professional grade
language. I use JScript for COM, C++ for Win32, and C# for the .NET
Framework. For non-Microsoft servers I use PERL.
Yes, but to me this is not a limitation of web applications, but simply
a case of BAD design. I could also cite a hundred cases of Microsoft
Access forms (or VB forms) that are inferior and more clumsy than a well
designed web interface. In fact most companies I'm dealing with are
trying to get rid of these "rich" forms because they are less intuitive
than the web equivalents. The other thing with the web, is that you can
make the navigation really nice and have much better text handling to
help the user as they move through the app. You can use text in VB forms
of course, but there's something about web text and CSS that just seems
nicer. The web also guarantees Accessibility, which is becoming a hot
policy issue.
Did I try what? I've been using .NET since the first betas and I'm
evaluating .NET 2.0 on a new server. I've never used VB.NET (or VB
Anything) because I don't consider VB to be a professional grade
language. I use JScript for COM, C++ for Win32, and C# for the .NET
Framework. For non-Microsoft servers I use PERL.
C
Cor Ligthert
Gerry,
Our thoughts are not so wide from each other.
I don't agree this statement with you about VBNet, however I used the text
VBNet because it was in the trend of this threath, it could have been C# as
well.
Cor
Our thoughts are not so wide from each other.
I don't agree this statement with you about VBNet, however I used the text
VBNet because it was in the trend of this threath, it could have been C# as
well.
Cor