Design Ideas

[John Lee]

Hi,

We are planning to use SQL 2005 as database backend, VS.NET 2005 to develop
service layer component (accessing SQL 2005 and taking advantage of
Transactions namespace, etc.) and create web services then the User
interface will be developed using VS.NET 2003 (I wanted to use VS.NET 2005
to develop the UI part too but my concern is there might be break changes in
next beta or release version makes the current developed UI code
non-compileable) - Our new system must go live end of January 2006 and I do
not know if .NET 2.0 along with SQL 2005/VS.NET 2005 could be RTM that time.

Is this a way to go or do you have any suggestion? We are starting to
develop in Mid-June.

Thanks very much!
John

[orekinbck]

Hi John

Call me conservative, but before taking on ANY new technology I try to
ensure that the first service pack is out (or at a minimum, that there
is a good body of knowledge built up on the web).

Even if everything is RTM in the time frame you require, you may find
yourself spending many sleepless nights trying to work out errors that
turn out to be issues with the new technologies themselves, not issues
with your code.

Cheers
Bill

John Lee wrote:

Hi,

We are planning to use SQL 2005 as database backend, VS.NET 2005 to develop service layer component (accessing SQL 2005 and taking advantage of
Transactions namespace, etc.) and create web services then the User
interface will be developed using VS.NET 2003 (I wanted to use VS.NET 2005 to develop the UI part too but my concern is there might be break changes in next beta or release version makes the current developed UI code
non-compileable) - Our new system must go live end of January 2006 and I do not know if .NET 2.0 along with SQL 2005/VS.NET 2005 could be RTM that time.
Is this a way to go or do you have any suggestion? We are starting to develop in Mid-June.

Thanks very much!
John

[Gabriel Lozano-Morán]

If you need to deliver the product in january 2006 I would go for Beta 2 of
Visual Studio .NET 2005. If everybody would wait until the first service
pack there wouldn't be a first service pack or it might take 2 years for the
first service pack :)

Gabriel Lozano-Morán

<or*******@yahoo.com.au> wrote in message
news:11**********************@o13g2000cwo.googlegr oups.com...

Hi John

Call me conservative, but before taking on ANY new technology I try to
ensure that the first service pack is out (or at a minimum, that there
is a good body of knowledge built up on the web).

Even if everything is RTM in the time frame you require, you may find
yourself spending many sleepless nights trying to work out errors that
turn out to be issues with the new technologies themselves, not issues
with your code.

Cheers
Bill

John Lee wrote:

Hi,

We are planning to use SQL 2005 as database backend, VS.NET 2005 to

develop

service layer component (accessing SQL 2005 and taking advantage of
Transactions namespace, etc.) and create web services then the User
interface will be developed using VS.NET 2003 (I wanted to use VS.NET

2005

to develop the UI part too but my concern is there might be break

changes in

next beta or release version makes the current developed UI code
non-compileable) - Our new system must go live end of January 2006

and I do

not know if .NET 2.0 along with SQL 2005/VS.NET 2005 could be RTM

that time.

Is this a way to go or do you have any suggestion? We are starting to

develop in Mid-June.

Thanks very much!
John

[Willy Denoyette [MVP]]

<or*******@yahoo.com.au> wrote in message
news:11**********************@o13g2000cwo.googlegr oups.com...

Hi John

Call me conservative, but before taking on ANY new technology I try to
ensure that the first service pack is out (or at a minimum, that there
is a good body of knowledge built up on the web).

Even if everything is RTM in the time frame you require, you may find
yourself spending many sleepless nights trying to work out errors that
turn out to be issues with the new technologies themselves, not issues
with your code.

Cheers
Bill

John Lee wrote:

With due respect but , if everyone was as conservative as you are, would we
ever see the first SP? I guess not.
Given the time the next version is in beta and the number of beta tester out
there, there is already a good body of knowledge built up on the web. Also
note that we are talking about a new release, not a new technology (there is
no such thing in this busyness - new technologies are mostly new names for
existing stuff).
IMO it's just a matter of trust and planning. Planning means take into
account when developing against a new release that you will have to deal
with bugs outside your direct control, if you stay with the current release,
consider the impact of upgrading later. Trust means take the calculated risk
but consider a fallback scenario, after all your initial customers must do
the same don't they?

Just my 2¤cent.
Willy.

[orekinbck]

Your 100% right, if everyone took my approach there would be no SP !!
.... the industry needs 'trail blazers' that take on the new releases
and work out the bugs.

Here's my rant .... strictly my personal opinion of course ....

Realistically, the way most IT companies operate these days (Microsoft
included), there are plenty of bugs for the trailblazers to find ...
they rush rush rush to get the product out, put pressure on the
programming team (which increases the error rate) and cut back on
planning and testing.

The number of bugs in released software is one of my major annoyances.
The other one is the number of errors in technical books these days.
It is SO rare to actually find a technical book that is not loaded down
with errors. Again it is the same thing ... rush rush rush, get the
product out, and cut back on proof reading.

Personally, I have spent too many long nights (and weeekends) trying to
work out problems with new software releases, that I have decided to
stay away from the bleeding edge. I would prefer to have a reasonable
stable working day which gives me time for more important things.

Anyways, that is my rant over with. I don't really want to get into an
argument about the whole thing, as I can see your points of view and
respect them. As an aside, I wonder how the actual percentage of
'early adopters' such as yourselves compares to the 'wait and see' camp
? I know I am certainly not alone in my approach!

Cheers
Bill

[rossum]

On Sun, 1 May 2005 12:13:48 +0200, "Willy Denoyette [MVP]"
<wi*************@telenet.be> wrote:

<or*******@yahoo.com.au> wrote in message
news:11**********************@o13g2000cwo.googleg roups.com...

Hi John

Call me conservative, but before taking on ANY new technology I try to
ensure that the first service pack is out (or at a minimum, that there
is a good body of knowledge built up on the web).

Even if everything is RTM in the time frame you require, you may find
yourself spending many sleepless nights trying to work out errors that
turn out to be issues with the new technologies themselves, not issues
with your code.

Cheers
Bill

John Lee wrote:

With due respect but , if everyone was as conservative as you are, would we
ever see the first SP? I guess not.

It depends on how business critical the application is. If the
success of the business is dependant on the success of the
application, then most businesses are going to be conservative. If
the application is a "nice to have" then it is a good candidate for
early adoption of a new technology - if it fails then it is not going
to take the business down with it.

This also allows the business to develop some internal expertise in
the new technology so it is ready when it does want to use the SP2
version for something important. Even if the internal expertise is
"Don't touch it with a barge pole" then that is useful.
Given the time the next version is in beta and the number of beta tester out
there, there is already a good body of knowledge built up on the web. Also
note that we are talking about a new release, not a new technology (there is
no such thing in this busyness - new technologies are mostly new names for
existing stuff). I am just waiting for that to appear in the nexr Microsoft press
release: "This product has been completely re-engineered on the name
level."

rossum
IMO it's just a matter of trust and planning. Planning means take into
account when developing against a new release that you will have to deal
with bugs outside your direct control, if you stay with the current release,
consider the impact of upgrading later. Trust means take the calculated risk
but consider a fallback scenario, after all your initial customers must do
the same don't they?

Just my 2¤cent.
Willy.

The ultimate truth is that there is no ultimate truth