Right time to move to .NET??

  • Thread starter Thread starter Ray Lavelle
  • Start date Start date
R

Ray Lavelle

I have an application I have developed and been working on for about 2
years. It was written in VB6 with a SQL Server database. Things have
been going well but I am reaching the point where I need to add a
bunch of new features as well as rework some of the areas of the
backend. I have been thinking that in the process of doing this work
I would try to migrate the application to .NET. After looking into
..NET and trying a conversion of my application it looks like it's
going to take quite a bit of rewriting to get this running in .NET and
may even turn out to be a total rewrite judging from all the errors I
got during the conversion. My question is if I should take the time
now and rewrite/port my application to .NET or do this at a future
date or when really necessary. I feel it is going to be a
considerable amount of work on top of the new features I plan to add
and will probably add several months of extra development time so I'm
wondering if it's really worth it? What do I really gain by going to
..NET now? Would I be better off implementing my changes now then
moving to .NET at a future date? For those of you who have moved your
VB6 applications to .NET, how much work was it and are you glad you
made the switch?
Any other suggestions?
 
Ray,

Don't get me wrong here - I think .NET is a great development environment
with real OOP attributes that could have saved me a ton of "clone and edit"
in VB6. As soon as this version of the app I'm building hits the market and
I'm making some money from it, I'm rewriting the whole thing to VB.NET.
However ...

I'd also spent a couple of years with a VB6/ SQL Server/Access app that's
very structured (custom support dll's, custom ocx-based with custom child
ocx's, the ability to dynamically select the type of database and the actual
database at run-time, etc.). The Data Report in VB 6 didn't give me the
functionality I needed for a couple of the remaining reports. So, seeing
that CR.NET could do what I needed and having an MSDN subscription (free
CR.NET!), I decided to move all of the reports to CR.NET and build wrapper
classes and collections in VB.NET for them. I also thought it would be a
good idea to migrate a complex schema interrogator dll from VB6 to VB.NET
and use ADO.NET instead of ADO and ADOX as a kind of test of the
migration/re-write issues. I'm almost finished the reports and the schema
interrogator dll now works. The rest of the app is still in VB6. I now know
enough to be able to re-write the rest of the app into VB.NET and deal with
the deployment problems of a hybrid application. It took me 8 months on top
of the 2 years I'd already spent and I could probably finish in another 6-8
months.

I'm sure it depends on how your VB6 app is written and others will, I'm also
sure, have different experiences and opinions but, in retrospect, it would
have been cheaper for me to buy CR 9 eight months ago and defer the move to
..NET. I'd advise against it unless, of course, someone else is funding the
development effort.

David Straker
 
Back
Top