suggestions please

  • Thread starter Thread starter éric
  • Start date Start date
É

éric

I completed, about 10 months ago, a point of sale application for a dairy
genetics company in Canada. They now want me to create another application
which helps Technicians in the field determine the best bull to breed their
cows and determine what potential genetic attributes the offspring will have
(mostly regarding potential milk production)

This application has to be disconnected since the users usually operate even
out of cell phone range.

The database will contain all the cows in Western Canada (2 million records)

The genetic calculations can take quite a while to generate and can be
improved on, in some cases, by rapid data access as well as cpu speed

The project deadline is Jan 1st or a bit beyond

I read some of the benefits of SQLCE 3.0 and CF 2.0 online. I know this will
not be out for a while but I think that this technology would make this
future application much more functional than building on CF 1 and SQLCE 2.0.

What do you think? Should I work through the alpha and beta builds of sqlce3
and cf2 to get the best possible performance for my project release date?

Should I just stick with CF1 and sqlce2?

Thanks for taking the time to answer my questions



éric
 
Eric:

Personally, I think the new stuff is going to rock based on the demos I've
seen, and in the newer builds of VS 2005, some of the new CF tools are
unreal. However, a lot is going to change. VS2005 isn't even in Beta yet,
so stuff you build could conceivably not exist or be very different from
when the final version is released. Moreover, VS 2005 isn't going to be
ready in January and the most optimistic stuff I've heard for the RTM
version is in the mid part of next year. That's pretty far off. The
features will be great, but I don't know that they'll be that much better.
I'd really focus on developing a robust product that meets the customers
needs very well. once I had this, I'd start planning a build based on the
new stuff, by then the official beta should be out and the chances of
something changing notably will be much less. It's possible that nothing
will change significantly and all this is moot. However, it may change a
bunch and if it does, any work you did may be wasted depending on the
changes. As such, I'd build my product and get it bullet proof...building a
disconnected app like you mention will take some work and fine tuning so you
make sure it has no rough edges. Then, later on this year, start building
with the newer version but don't couple anything too tightly b/c some change
may still be made (but a beta is a lot less likely to change dramatically
than an alpha is) As a matter of fact, I was looking at a build of WHidbey
last week that didn't resemble the whidbey I've been using for the last
month...it had tons of changes and improvements. Anyway, you'll be able to
upgrade your app and take advantage of the new features pretty easy (the
difference between 1.1 and 2.0 is going to be much bigger than 1.0 and 1.1
but you can still do it and there will be backward compatibility). I'd hold
off a while and then upgrade when it was feasible. Building a product that's
robust and stable is the most important thing b/c even if it's really fast
if it's not stable users are going to be mad.

I'm currently playing with the 2.0 framework extensively b/c I want to learn
it early on but as much as I use it, I'm defintiely not planning on building
anything with it that'll be used in production until the later part of this
year or early next year. (During a MS demo, there was a pretty bad error in
the IDE and the quote was "If it didn't still have a lot of bugs, we'd have
released it already"). The comment was said jokingly but the point was that
the IDE still needs a lot of work. I think that says it all.

BTW, what specific features are you thinking about right now?

HTH,

Bill
 
Ryan,

Thanks for your excellent reply.
I think you are right... stability is important.
BTW, what specific features are you thinking about right now?
Click to expand...

I am mostly interested in what SQLCE 3.0 had to offer... direct table
access, linking controls to queries etc. Also any speed improvement would be
great!
Replication with the ability to display a status bar would be a big plus!

On the designer side it seems like there are some nice options for
developing sqlce appsbut I can live without those as long as the app
performs.

Displaying in landscape and anchoring are not necessary but any app coming
out a year from now whithout those features will likely have to be reworked
with that in mind.

That is all I can think of for now... but I remember watching a sqlce 3 and
a cf2 demo online a little while back and more than once I was thinking that
feature would be perfect for my application.

Thanks again

éric
 
appologies for the dislexia Bill!

éric said:
Ryan,

Thanks for your excellent reply.
I think you are right... stability is important.


I am mostly interested in what SQLCE 3.0 had to offer... direct table
access, linking controls to queries etc. Also any speed improvement would be
great!
Replication with the ability to display a status bar would be a big plus!

On the designer side it seems like there are some nice options for
developing sqlce appsbut I can live without those as long as the app
performs.

Displaying in landscape and anchoring are not necessary but any app coming
out a year from now whithout those features will likely have to be reworked
with that in mind.

That is all I can think of for now... but I remember watching a sqlce 3 and
a cf2 demo online a little while back and more than once I was thinking that
feature would be perfect for my application.

Thanks again

éric

proof...building error
Click to expand...
 
Back
Top