How would you do a time estimate of a conversion from VB6 to dot net?

[tom c]

I hate doing estimates of how long things will take, but of course
people want them. I am doing some modifications on a VB6 system and
the owner wants an estimate on how long it will take to convert it to
VB dot net. How would you approach doing a time estimate on this?

I have worked in VB dot net, but I have never converted a system from
VB6 to dot net.

 

[Carl Daniel [VC++ MVP]]

I hate doing estimates of how long things will take, but of course
people want them. I am doing some modifications on a VB6 system and
the owner wants an estimate on how long it will take to convert it to
VB dot net. How would you approach doing a time estimate on this?

I have worked in VB dot net, but I have never converted a system from
VB6 to dot net.

Estimate how much time it would take to build it from scratch in VB.Net.
That's at least a reasonable upper bound on the time to convert it.
Depending on the app, it may not be far off (although I gather that there's
an elaborate conversion tool that may or may save you large amounts of
time).

-cd

 

[Scott M.]

That's a very open-ended question you ask. Answering it requires a
knowledge of how many lines of code we are talking about, how many
developers there are to work on the conversion, what the architecture of the
application is, what does the application do, are there any changes that are
desired in the new replacement and more.

Without all this info., it would be impossible to give you any *reasonable*
idea of how long it would take.

 

[tom c]

Thanks for your thoughts Scott. I understand that no one could
possibly give an estimate from the info I posted. I just wanted some
ideas about how to do an estimate myself. For example Carl said to
extimate how long it would take to build from scratch and use that as
the top end. That was the kind of tip I was looking for, not for
someone to tell me how many hours it would take.

 

[PS]

I hate doing estimates of how long things will take, but of course
people want them. I am doing some modifications on a VB6 system and
the owner wants an estimate on how long it will take to convert it to
VB dot net. How would you approach doing a time estimate on this?

I have worked in VB dot net, but I have never converted a system from
VB6 to dot net.

The best thing to do is spend some time converting some of the existing code
then based on the amount you converted you can make a reasonable guess as to
how long it would take to convert the complete project. Some people use
lines of code (LOC) as a metric but some things like forms will skew this.
You may want to break up the current software into different value units
like a form = 1, 100 LOC = 1, while some code that cannot be converted might
be 20 LOC = 1.

You can use this when you present your estimate to the owner to show that
you have not just come up with some random figure.

SP

 

[Damien]

I hate doing estimates of how long things will take, but of course
people want them. I am doing some modifications on a VB6 system and
the owner wants an estimate on how long it will take to convert it to
VB dot net. How would you approach doing a time estimate on this?

I have worked in VB dot net, but I have never converted a system from
VB6 to dot net.

I'd suggest getting the VB6 Code Advisor from Microsoft:

http://www.microsoft.com/downloads/...1a-b5c0-4d40-b015-0caa02634fae&DisplayLang=en

And letting it have a chug through your project. If it comes out saying
there's nothing to change, you're probably in fairly good shape. If it
comes out with thousands (or millions) of required changes, you may be
in for a rough ride.

Once it's gone through your project, you need to look at if it's one
single type of issue (possibly even one you can fix with a global find
& replace), or whether the issues are widespread and varied.

Hopefully, this will give you some ideas about the level of changes
that will be required in your application.

Bear in mind, even once you've fixed all of these issues, and
loaded/upgraded the project into .Net, it will still not be as good as
if you re-write some of the code into .Net, but it should at least
compile and be close.

Damien

 

[Scott M.]

I know, but I have to say that you can't figure out how long it would take
to build it from scratch (Carl's suggestion) unless you know the answers to
the questions I posed.