To Many Errors!

  • Thread starter Thread starter chris
  • Start date Start date
C

chris

Hallo,

I am now working now on a project since 1/2 a year. There are
too many bad things happening. Sure: You can get an answer for most
of the problems searching the web, and there are MVP'S like Chris
who try to help out. But, the root of the problem: Writing a fast
application
is the same like it was in eVB: IT DOES NOT WORK. You can not write a
Commercial Application with Windows CE. It does not make sense. Yes:
It does not make sense to be on the Internet 24 hours to solve trivial
problems.
It does not make sense to be a Master like Chris Tacke, just to write
a simple application. As long as Pocket PC application development is
so difficult, it does not make sense to do it! Not everybody is smart
like Chris Tacke!!!
Think about normal 8 - 18 hours programmers!!!! But this guys should
write this applications.


Please stop to promote CF and VB.NET as long as this fundamental
problem are not solved: Writing an Application in No Time!!! Writing
an application in resenable time! Every else does not matter!!!

Microsoft Please wake up. A view days ago a guy from HP came into my
Office and
said: You are the Customer we sell the most of our Pocket PC in the
whole country. (We just received 300 devices in one day) What are you
doing with all this mashines? I said, Software, but we have such a bad
time doing this!!!
Yeas, we use this Pocket PC bullshit (Not me,but my employee) Every
day they
come in by the hundreds, using my software. But, it's all bullshit. I
get
drunk in the evening to forget my programming work ...

Sorry my bad English!!!

Chris
 
I'm going to have to disagree. We have several customers (that BTW are not
Chris Tacke) that have implemented commercial applications using the CF.
I've talked with other companies that have implemented very good PPC
applications based on the CF prior to knowing me or OpenNETCF, so we're not
some "recipe for success".

OpenNETCF.org has a case study on one, and Microsoft has many others. I've
seen many CF apps outperform eVB for like functionality, and debugging is
far, far, far better.

I'm sorry to hear you're having such a bad experience. Yes there are some
limitations and areas that the CF team could have done better on, but I
still believe this is the best v 1.0 product Microsoft has ever delivered.

Writing code is simple. Writing good code is not. Add to that the
challenges of limited resources, processor power and memory, and mobile
development is some of the most challenging to do well.
 
Yeas, we use this Pocket PC bullshit (Not me,but my employee) Every
day they
come in by the hundreds, using my software. But, it's all bullshit. I
get
drunk in the evening to forget my programming work ...
Click to expand...

That's a great line. I'll have to remember it. <g>

Jamie
 
I don't agree.

We ported our object database engine to CF and everything
appears to work very good.

.....with no special hacks around CF bugs. We only had to
find workarounds for functionality that's not available.
It tooks us little more than two weeks of development time.

Performance is O.K. on CF, it's not really good but it's
acceptable, if you compare it to Java VMs.


GUI development and debugging work extraordinarily good!
Good job Microsoft!


I am sure that we will see faster versions of CF as the
platform evolves. I think CF is a good decision, if you
need to develop for PocketPC.


Kind regards,
Carl
 
Our experience is that the .NET Compact Framework does offer some
advantages. Debugging is the highest. From a coding perspective, I do have
to say that the number of lines of total lines of code as compared to
functional code is somewhere in the neighbourhood of 300:1. Typical desktop
applications we've developed tend to run somewhere around 15:1. These apps
are fully retail ready, not your inhouse 'good enoughs'. The reason for the
higher count on the PPCs is that there is significant coding to get around
the limitations in the .NET CF. Managing forms is one of the largest
consumers. This higher ratio translates directly into longer development
times but is being somewhat offset by the reduced debugging time. It still
takes longer than developing a desktop app (somewhere about 3x or higher).
With the higher line count, the potential for bugs is exponentially larger
though and we end up needing to do a much greater level of QA (which uses
more resources).

Overall, there has been a lot of frustration using .NET CF, but believe
(perhaps erroneously) that it's better than some of the alternatives that
we've tried. The biggest single improvement that we would like to see is
more accurate documentation, followed by greater capability in the CF and
lastly, more complete and tested code samples in both C# and VB (not
everyone uses C#). We can all contribute by providing responses to others in
newsgroups that are complete, accurate and correct.

Mike
 
Hi

I also disagree. I've only been programming with .NETCF for 3 months and I was able to create a working application that we are putting out in a few months. Sure it was difficult for the first month or so, but I bought a book, and I read this newsgroup, and other forums and now I've got a pretty good grasp on the basics of vb.net. I would also have to say that VS.NET 2003 is the best developing environment that I've ever used! I've been programming for over 10 years and the intelisense that you get when programming in vb is amazing! So much so that when I went back and did a little tweaking of my code in PL/SQL the other day I was lost without it! Like Chris said, .NETCF a VERY good product for v1.0, and 2.0 will be even better.

Programming isn't easy, it looks easy from the outside, My father-in-law says that all I do at work is "Sit at the computer and tap the keys, that's not work!" :) Anyways, just stick with it, it'll be worth it

skicow
 
Interesting observations! I'd be interested in seeing similar analysis of
other projects to see of the 300:1 ratio is commonplace. If you take the
SDF as "workaround code" for a project, I can see it adding a lot. I'm a
bit surprised at 300:1. How do you define a "line" and a "functional line"?

Keep in mind I don't doubt you, I think that this kind of analysis is very
good and I'd like to propose a "standardized" way to make the measurements
and then collect data from other projects.
 
Hi Chris,

For some reason the original post has gone missing.

However, it is my personal observation that the Compact Framework provides
an order of magnitude improvement (over eVB) both in performance and in my
productivity. To emphasize this, I removed the eVB chapter (well, placed it
on the accompanying CD ROM) from the next edition, the 4th, of my book and
replaced it with a chapter on the Compact Framework -- VB .NET, naturally.

Dick

--
Richard Grier (Microsoft Visual Basic MVP)

See www.hardandsoftware.net for contact information.

Author of Visual Basic Programmer's Guide to Serial Communications, 3rd
Edition ISBN 1-890422-27-4 (391 pages) published February 2002.
 
DG [Wed, 24 Mar 2004 08:51:21 -0700]:
For some reason the original post has gone missing.
Click to expand...

I removed that i to make it perfectly un-profane, but I
really doubt that was the reason for the censorship.
We'll soon find out... Now, where's that 300:1 guy,
and what is he smokin'? In my business, it's not the line
cout, but the though count; I port what takes me a day in
x86 to ppc in only a few MINUTES, using plain old, real-
programmers-don't-bake-quiche, computer languages and
real compilers and assemblers.
 
Back
Top