Oracle Forms vs Visual Studio .Net
John Smith
Visual Studio .Net. I am new to Oracle and hope those who have gone the
decision path could advise the pros and cons. We have Oracle 8 database in a
AIX unix box (plan to upgrade to 9i in future), MS SQL Server and MS IIS. We
do not run any Java web server or have any java programs. Mostly VB and ASP
stuff.
If I use Oracle Forms for form-based app and web-based app, then I would
need the resource and skill to run Oracle AS right?
Will there be any issues or problems I might faced if we standadised on .NET
and Visual Studio as platform and development tool and Oracle as the
database? Are there any company doing this?
Thanks
regards
Andrew Carruthers
Lots of companies use .NET and Oracle back ends with great success, however,
you should be aware of the tighter integration that Java gives you with
Oracle, there's an inbuilt JVM and the ability to store Java components
within the database and call them from PL/SQL (and vice versa).
If you're looking at a new architecture you should consider the overall
infrastructure and current mix of technologies rather than introduce a new
one unless you can justify the decision as being a long term investment with
a clear path to migrate existing tools.
Personally, I would steer clear of Oracle Forms, most of my clients
predominantly use Java or .NET technologies.
It comes down to what you already have, ongoing support, maintenance, skills
and migration path
"John Smith" <som...@microsoft.com> wrote in message
news:bu8h2a$76i$1...@mawar.singnet.com.sg...
Jeff
requires 9iAS (the application server), which is expensive and a pain to
install and maintain, and Forms 6i is being desupported soon. Forms,
therefore, does not offer a viable long-term client-server solution for
non-web apps, and that lack of support is the primary reason I'd choose .NET
over forms. Forms offers a much tighter level of interaction with an Oracle
database than .NET, but .NET is a stronger Windows development tool. .NET
apps are typically smaller compared to Forms apps, especially when you compare
the sizes of the .NET Framework vs. the Forms runtime (6i) or the Application
Server (9i). .NET will be easier in connecting to multiple databases,
particularly if some of them are non-Oracle.
In article <bu8h2a$76i$1...@mawar.singnet.com.sg>, "John Smith"
Daniel Morgan
I've read Andrew and Jeff's responses and want to give you some
additional arguments in favor of Oracle Forms.
1. Everybody is leaning .NET. When pretty much everyone knows something
competition for jobs becomes more difficult and the salary goes down.
There is just more money is being a Fprms expert.
2. Jeff states that Forms 9i requires the 9iAS app server which is a
pain to install and configure: He is incorrect. OC4J is required not the
app server and installing and configuring 0C4J is not that difficult.
Also, he is looking to the past and assuming that because 9iAS is
difficult therefore it is the way it will always be. This too is untrue.
The new Oracle Application Server 10g is as simple as 9iAS is
challenging. He also talks about Forms 6i which is meaningless. The
curent version is 9i and the 10g version will be here long before you
could have anything ready to deploy.
3. If you learn Forms 6i, or 9i, the learning curve to 10g will be
extremely small.
4. Far superior security to .NET
5. Far tighter integration than .NET
6. Far more sensible approach if using the database for anything other
than columns and rows.
7. Platform independence
8. More stable
9. More scalable
10. There are several few companies that sell products that convert
Oracle Forms to Java apps and Forms independence.
From my experience with .NET ... I am far less than impressed. Some of
the worst code I've ever seen has been in .NET. If you are going to work
with SQL Server ... .NET may make sense. Against Oracle I'd definitely
go with Forms. And if not forms I'd be looking at Delphi or other tools
as my second choice.
--
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damo...@x.washington.edu
(replace 'x' with a 'u' to reply)
shay
Like VS .NET it gives you a visual way to build Web applications,
Client Servr applications, Web services or anything else that you
want.
Take a look at the Oracle JDeveloper 10g demos at:
http://otn.oracle.com/products/jdev to see how it looks and works and
what the development is like.
Why you should choose a Java approach over a .NET approach?
Well this is a long discussion covered in many places but it comes
down to the word "choice"
with Java you can choose your Operating System and your software
supplier.
With .NET you are locked to MS both in the OS and Tools. (this means
that you can forget about running your .NET application on Linux).
Daniel Morgan
Good call. I should have mentioned it.
Scott Mattes
news:1074267648.488805@yasure:
> definitely go with Forms. And if not forms I'd be looking at Delphi or
> other tools as my second choice.
Not that I am an expert, but I have done both Delphi and Forms and I'd
take Delphi and a pay cut over Forms (though I haven't done 9i Forms yet)!
Niall Litchfield
news:1074267648.488805@yasure...
> I've read Andrew and Jeff's responses and want to give you some
> additional arguments in favor of Oracle Forms.
> 4. Far superior security to .NET
> 5. Far tighter integration than .NET
How well is forms integrated with the application server (IIS/COM+) that
will be in use here?
> 6. Far more sensible approach if using the database for anything other
> than columns and rows.
I don't think that I understand?
> 7. Platform independence
but this already is an MS shop. Isn't this argument a bit of a red herring?
> 8. More stable
> 9. More scalable
I wonder if you can substantiate the above claims.
> From my experience with .NET ... I am far less than impressed. Some of
> the worst code I've ever seen has been in .NET. If you are going to work
> with SQL Server ... .NET may make sense. Against Oracle I'd definitely
> go with Forms. And if not forms I'd be looking at Delphi or other tools
> as my second choice.
On the whole I disagree, ISTM that with the .Net runtime and the Visual
Studio IDE, Microsoft have just about the best development environment that
there is out there. It sure as hell beats JDeveloper. I also seem to recall
someone recently saying that you couldn't blame Oracle for the crap code out
there that runs on Oracle - to the extent that this is true it would seem
that blaming MS for poor development on .net is an equally unconvincing
argument.
--
Niall Litchfield
Oracle DBA
Audit Commission UK
*****************************************
Please include version and platform
and SQL where applicable
It makes life easier and increases the
likelihood of a good answer
******************************************
Daniel Morgan
Niall Litchfield wrote:
> "Daniel Morgan" <damo...@x.washington.edu> wrote in message
> news:1074267648.488805@yasure...
>
>>I've read Andrew and Jeff's responses and want to give you some
>>additional arguments in favor of Oracle Forms.
>
>
>>4. Far superior security to .NET
>
>
>>5. Far tighter integration than .NET
>
>
> How well is forms integrated with the application server (IIS/COM+) that
> will be in use here?
Don't recall anything about IIS/COM+: Perhaps my mistake. My
recollection was whether iAS is required and it is not.
>>6. Far more sensible approach if using the database for anything other
>>than columns and rows.
>
>
> I don't think that I understand?
Microsoft often does not support newer abilities in Oracle. So your
ability to work with some data types (especially UDT), functions, etc.
may not be supported as completely.
>>7. Platform independence
>
>
> but this already is an MS shop. Isn't this argument a bit of a red herring?
An MS shop meaning they don't have and wish to limit themselves from
ever having any other technology? Basically ... if I understand your
argument, preclude yourself from ever moving to Linux because you will
have too big an investment in a single vendor's technology.
>>8. More stable
>>9. More scalable
>
>
> I wonder if you can substantiate the above claims.
Every study I've ever seen shows _NIX platforms more stable and more
scalable than Windows. And then there is the issue of virus and hacking
attacks.
When the last big attack occured banks relying on Windows technology,
especially their ATMs fell like mosquitos in a bowl of DDT. Those
relying on _NIX continued to function.
>> From my experience with .NET ... I am far less than impressed. Some of
>>the worst code I've ever seen has been in .NET. If you are going to work
>>with SQL Server ... .NET may make sense. Against Oracle I'd definitely
>>go with Forms. And if not forms I'd be looking at Delphi or other tools
>>as my second choice.
>
>
> On the whole I disagree, ISTM that with the .Net runtime and the Visual
> Studio IDE, Microsoft have just about the best development environment that
> there is out there. It sure as hell beats JDeveloper. I also seem to recall
> someone recently saying that you couldn't blame Oracle for the crap code out
> there that runs on Oracle - to the extent that this is true it would seem
> that blaming MS for poor development on .net is an equally unconvincing
> argument.
>
> --
> Niall Litchfield
> Oracle DBA
> Audit Commission UK
> *****************************************
> Please include version and platform
> and SQL where applicable
> It makes life easier and increases the
> likelihood of a good answer
> ******************************************
Beats H out of JDeveloper? Can you get anything you build to fail-over
on RAC? With JDeveloper fail-over is just a Ctrl-C, Ctrl-V away if you
grab the TAF demo from otn.
Of course people write garbage code in almost everything. But almost all
of the .NET I've seen going against Oracle demonstrates a profound lack
of understanding of Oracle. It might be fine against SQL Server or
Sybase but against Oracle it just creates consulting hours telling the
developers how to fix what they already wrote.
Which is not to say that you couldn't build something adequate, or even
good with .NET against Oracle. I'm just saying that if I had to choose
between a proposal in .NET and one in JDeveloper ... I'd go with the
JDeveloper every time based on the technology involved.
Reasonable people may disagree. It is just my opinion.
Galen Boyer
> I'm just saying that if I had to choose between a proposal in
> .NET and one in JDeveloper ...
I don't get it. One's a platform to rival J2EE, the other is an
IDE.
--
Galen Boyer
Hans Forbrich
(I think he was referring to "Visual Studio .Net")
Daniel Morgan
Correct.
Galen Boyer
> Hans Forbrich wrote:
>
>> Galen Boyer wrote:
>>
>>>On Sat, 17 Jan 2004, damo...@x.washington.edu wrote:
>>>
>>>
>>>>I'm just saying that if I had to choose between a proposal in
>>>>.NET and one in JDeveloper ...
>>>
>>>I don't get it. One's a platform to rival J2EE, the other is
>>>an IDE.
>>>
>> (I think he was referring to "Visual Studio .Net")
>
> Correct.
Whew, thats better! I disagree with Daniel alot but I never
thought he was just plain lost.
--
Galen Boyer
Daniel Morgan
> On Sat, 17 Jan 2004, damo...@x.washington.edu wrote:
>
>>Hans Forbrich wrote:
>>
>>
>>>Galen Boyer wrote:
>>>
>>>
>>>>On Sat, 17 Jan 2004, damo...@x.washington.edu wrote:
>>>>
>>>>
>>>>
>>>>>I'm just saying that if I had to choose between a proposal in
>>>>>.NET and one in JDeveloper ...
>>>>
>>>>I don't get it. One's a platform to rival J2EE, the other is
>>>>an IDE.
>>>>
>>>
>>>(I think he was referring to "Visual Studio .Net")
>>
>>Correct.
>
>
> Whew, thats better! I disagree with Daniel alot but I never
> thought he was just plain lost.
Given the doses of Percodan I'm taking this would be the one time it
would be excusable on my part. Maybe I should have broken my leg long
ago. ;-)
Niall Litchfield
--
Niall Litchfield
Oracle DBA
Audit Commission UK
*****************************************
Please include version and platform
and SQL where applicable
It makes life easier and increases the
likelihood of a good answer
******************************************
"Daniel Morgan" <damo...@x.washington.edu> wrote in message
news:1074366905.952181@yasure...
> Comments in-line.
>
> Niall Litchfield wrote:
>
> > "Daniel Morgan" <damo...@x.washington.edu> wrote in message
> > news:1074267648.488805@yasure...
> >
> >>I've read Andrew and Jeff's responses and want to give you some
> >>additional arguments in favor of Oracle Forms.
> >
> >
> >>4. Far superior security to .NET
> >
> >
> >>5. Far tighter integration than .NET
> >
> >
> > How well is forms integrated with the application server (IIS/COM+) that
> > will be in use here?
>
> Don't recall anything about IIS/COM+: Perhaps my mistake. My
> recollection was whether iAS is required and it is not.
<quote>
We have Oracle 8 database in a
AIX unix box (plan to upgrade to 9i in future), MS SQL Server and MS IIS. We
do not run any Java web server or have any java programs. Mostly VB and ASP
stuff.
</quote>
>
> >>6. Far more sensible approach if using the database for anything other
> >>than columns and rows.
> >
> >
> > I don't think that I understand?
>
> Microsoft often does not support newer abilities in Oracle. So your
> ability to work with some data types (especially UDT), functions, etc.
> may not be supported as completely.
Fair enough, though if the Oracle Data Provider for .Net doesn't support
them I'd be bugging Oracle not MS. I'm not a fan of using ms providers to
access 3rd party db platforms.
> >>7. Platform independence
> >
> >
> > but this already is an MS shop. Isn't this argument a bit of a red
herring?
>
> An MS shop meaning they don't have and wish to limit themselves from
> ever having any other technology? Basically ... if I understand your
> argument, preclude yourself from ever moving to Linux because you will
> have too big an investment in a single vendor's technology.
Choosing heterogeneous operating systems and skillsets is a strategic
decision. It might be a correct decision, but my take on the request was
that we were commenting on which of Forms or .Net was more appropriate for
this application in this organisation. to be honest given what was said
about where they were coming from I'd be more concerned about wether Oracle
was the right choice than Forms v .Net
> Beats H out of JDeveloper? Can you get anything you build to fail-over
> on RAC? With JDeveloper fail-over is just a Ctrl-C, Ctrl-V away if you
> grab the TAF demo from otn.
I don't know, but I suspect that you can. On the other hand given that the
number of customers Oracle has on RAC is in the 'hundreds' according to
Larry in his 10g presentation, for the most part RAC is just an irrelevance.
I suspect TAF is even more so since there is no requirement to rewrite apps
for TAF so I'd be sceptical if there were more than say 150 TAF enabled
production apps in the world. I'm sure if there are someone from Oracle will
jump in.
> Reasonable people may disagree. It is just my opinion.
Don't know if I'm always reasonable but sure we have different opinions on
this.
Jeff
>1. Everybody is leaning .NET. When pretty much everyone knows something
>competition for jobs becomes more difficult and the salary goes down.
>There is just more money is being a Fprms expert.
Basically, number #1 on your list, says go with something that's less
prevalent and perhaps harder to use just to keep your salary high, nevermind
what's good for the shop. That's a piss poor reason to choose one over the
other, IMHO.
>2. Jeff states that Forms 9i requires the 9iAS app server which is a
>pain to install and configure: He is incorrect. OC4J is required not the
>app server and installing and configuring 0C4J is not that difficult.
That's funny because Oracle hasn't offered any options for 9i Forms deployment
(that I've seen anyway) other than through 9iAS. Kindly point us towards some
Oracle-supported solutions for 9i Forms deployment that don't require the iAS.
>Also, he is looking to the past and assuming that because 9iAS is
>difficult therefore it is the way it will always be. This too is untrue.
Always? No. I can only speak to what I've yet seen or heard. I can't make a
recommendation based on future products or products I've heard nothing about.
Who can? The OP asked about Forms vs. .NET, which is what my post was about.
If 10g is available now and the next great thing in database apps, then that's
news to me.
> The new Oracle Application Server 10g is as simple as 9iAS is
If this is true, that's refreshing to hear. What have they changed in 10gAS
that has so greatly simplified the bloated, expensive, nightmare that was
9iAS?
>challenging. He also talks about Forms 6i which is meaningless. The
>curent version is 9i and the 10g version will be here long before you
>could have anything ready to deploy.
Forms 6i is meaningless now that it's about to be desupported, but it wasn't
impertanent to my response. The OP didn't seem interested in Web apps (I
could be wrong), and for many shops this is true as well. 6i was the last
Forms that supported the simple client-server model... unless Oracle's
reintroduced it in a new version that I'm not aware of.
For some, Oracle's move to web-only deployment is a slap in the face. I'm not
keen on non-web apps that only run in a browser... nor am I particularly fond
of Java apps that just seem to me to be unstable, fat and slow (I've yet to
meet the Java app that wasn't). If I needed to deploy apps to customers that
aren't in-house, web-deployement would make sense (actually, .NET does this
too, but it's optional), but for in-house apps, web apps pale in comparison in
stability and robustness. Of course, that's just my opinion.
>3. If you learn Forms 6i, or 9i, the learning curve to 10g will be
>extremely small.
If you intend to go with 10g, this could be relevant, but one could say the
same thing about almost any other IDE evolution.
>4. Far superior security to .NET
I agree.
>5. Far tighter integration than .NET
I said this as well... or tried to.
>6. Far more sensible approach if using the database for anything other
>than columns and rows.
Define this, please.
>7. Platform independence
Possibly, but realities are that for some shops (particularly smaller shops),
this isn't relevant, because the vast majority (99.9%) of their clients use M$
(Windows)... and not just to run in-house apps. However, if client
platform-independance is an issue, then by all means use Forms 9i/10g, Java or
C++ or whatever else. Of course, if ASP's are supported by non-IE browsers,
then .NET is as viable as Forms is.
>8. More stable
I doubt this.
>9. More scalable
Debatable.
>10. There are several few companies that sell products that convert
>Oracle Forms to Java apps and Forms independence.
Links? I am admittedly skeptical of such claims, because as far as I've ever
seen, the conversions are buggy, maintenance nightmares. So much so, that
they get thrown out and you start over from scratch.
> From my experience with .NET ... I am far less than impressed. Some of
>the worst code I've ever seen has been in .NET. If you are going to work
>with SQL Server ... .NET may make sense. Against Oracle I'd definitely
>go with Forms. And if not forms I'd be looking at Delphi or other tools
>as my second choice.
I cannot speak to what you've seen--just as the reverse is true--but I've seen
nothing in .NET so far that makes it seem predisposed to writing bad code
either. As far as Windows apps go, .NET is very powerful, IMHO. By far,
moreso than Forms, and probably on par, at the very least, to any other
product you could name, including Delphi.
Against Oracle-only databases, Forms 6i is great... if it weren't being
desupported, I'd be using it now. I've already made my feelings known about
9i and it's web-only deployment, and I cannot speak to 10g until I know
something about it. But note that the OP indicated (I thought) that he might
be connecting to non-Oracle databases as well... and, in that case especially,
Forms isn't a very good choice, IMO.
Daniel Morgan
Jeff wrote:
> In article <1074267648.488805@yasure>, Daniel Morgan <damo...@x.washington.edu> wrote:
>
>
>>1. Everybody is leaning .NET. When pretty much everyone knows something
>>competition for jobs becomes more difficult and the salary goes down.
>>There is just more money is being a Fprms expert.
>
> Basically, number #1 on your list, says go with something that's less
> prevalent and perhaps harder to use just to keep your salary high, nevermind
> what's good for the shop. That's a piss poor reason to choose one over the
> other, IMHO.
I agree. But anytime you want to have a head-to-head competition for ROI
with me using Forms and you using .NET to build an application I will
gladly put my reputation on the line.
I'll likely be Beta testing while your still trying to create a stable
connection.
>>2. Jeff states that Forms 9i requires the 9iAS app server which is a
>>pain to install and configure: He is incorrect. OC4J is required not the
>>app server and installing and configuring 0C4J is not that difficult.
>
> That's funny because Oracle hasn't offered any options for 9i Forms deployment
> (that I've seen anyway) other than through 9iAS. Kindly point us towards some
> Oracle-supported solutions for 9i Forms deployment that don't require the iAS.
Here is one of many links:
http://dbforums.com/showthread.php?threadid=534153
I wouldn't expect Oracle sales people to make a big deal out if simply
because it isn't in their best interest.
>>Also, he is looking to the past and assuming that because 9iAS is
>>difficult therefore it is the way it will always be. This too is untrue.
>
> Always? No. I can only speak to what I've yet seen or heard. I can't make a
> recommendation based on future products or products I've heard nothing about.
> Who can? The OP asked about Forms vs. .NET, which is what my post was about.
> If 10g is available now and the next great thing in database apps, then that's
> news to me.
The 10g database is not available. But many other 10g apps are
available. If not in final release form ... as "previews".
>> The new Oracle Application Server 10g is as simple as 9iAS is
>
> If this is true, that's refreshing to hear. What have they changed in 10gAS
> that has so greatly simplified the bloated, expensive, nightmare that was
> 9iAS?
Greatly improved installation that prompts for the configuration
information. The nightmare seems to be over.
Now I can't guarantee what I've seen is what will be released ... but
I'd take what I've seen over 9iAS anytime of any day.
>>challenging. He also talks about Forms 6i which is meaningless. The
>>curent version is 9i and the 10g version will be here long before you
>>could have anything ready to deploy.
>
> Forms 6i is meaningless now that it's about to be desupported, but it wasn't
> impertanent to my response. The OP didn't seem interested in Web apps (I
> could be wrong), and for many shops this is true as well. 6i was the last
> Forms that supported the simple client-server model... unless Oracle's
> reintroduced it in a new version that I'm not aware of.
Either I misunderstood you or you misunderstood me or both. I have long
said Oracle made a huge mistake dropping client-server capabilities from
Forms. Basically they ceded the market for client server to Microsoft.
And there are a lot of Oracle shops that can save money and get what
they need without deploying the latest alphabet soup.
I am so sick of JAVA, JSP, XML, DIME, MIME, blah blah blah blah blah
when I am working with a small shop that just needs to insert records
and query a local database. Not everyone is Boeing or AT&T. Not everyone
has an IS department. And Oracle is just walking away and leaving their
business on the table.
> For some, Oracle's move to web-only deployment is a slap in the face. I'm not
> keen on non-web apps that only run in a browser... nor am I particularly fond
> of Java apps that just seem to me to be unstable, fat and slow (I've yet to
> meet the Java app that wasn't). If I needed to deploy apps to customers that
> aren't in-house, web-deployement would make sense (actually, .NET does this
> too, but it's optional), but for in-house apps, web apps pale in comparison in
> stability and robustness. Of course, that's just my opinion.
One that I agree with. I like Java ... for some things. But I don't like
Java for ... everything. Fanatics rarely make good decisions. And it
seems that somewhere in Oracle there are a bunch of Java fanatics.
>>3. If you learn Forms 6i, or 9i, the learning curve to 10g will be
>>extremely small.
>
> If you intend to go with 10g, this could be relevant, but one could say the
> same thing about almost any other IDE evolution.
True.
>>4. Far superior security to .NET
>
> I agree.
>
>>5. Far tighter integration than .NET
>
> I said this as well... or tried to.
>
>>6. Far more sensible approach if using the database for anything other
>>than columns and rows.
>
> Define this, please.
Getting past VARCHAR, NUMBER, and DATE and into TIMESTAMP, INTERVAL,
user defined data types, etc. Maybe I missed something but I don't
recall see a lot of Oracle's newer functionality supported by .NET.
>>7. Platform independence
>
> Possibly, but realities are that for some shops (particularly smaller shops),
> this isn't relevant, because the vast majority (99.9%) of their clients use M$
> (Windows)... and not just to run in-house apps. However, if client
> platform-independance is an issue, then by all means use Forms 9i/10g, Java or
> C++ or whatever else. Of course, if ASP's are supported by non-IE browsers,
> then .NET is as viable as Forms is.
True today. But many of these same companies are beginning to realize
that Linux costs a lot less. If they build a huge pile of MS reliant
infrastructure it makes the cost of moving from Windows to Linux much
more expensive.
>>8. More stable
>
> I doubt this.
I'll put my Linux, Solaris, or HP/UX machines up against any Windows
machine and let a bunch of university students try to crack them or
bring them down by any means possible. The Windows machine will be the
first one to fry.
>>9. More scalable
>
> Debatable.
Even Microsoft doesn't use Windows for their own accounting system.
Don't you think they would if they could?
>>10. There are several few companies that sell products that convert
>>Oracle Forms to Java apps and Forms independence.
> Links? I am admittedly skeptical of such claims, because as far as I've ever
> seen, the conversions are buggy, maintenance nightmares. So much so, that
> they get thrown out and you start over from scratch.
http://www.rocketsoftware.com/portfolio/applicationtransfer/oracleforms.htm
>>From my experience with .NET ... I am far less than impressed. Some of
>>the worst code I've ever seen has been in .NET. If you are going to work
>>with SQL Server ... .NET may make sense. Against Oracle I'd definitely
>>go with Forms. And if not forms I'd be looking at Delphi or other tools
>>as my second choice.
>
> I cannot speak to what you've seen--just as the reverse is true--but I've seen
> nothing in .NET so far that makes it seem predisposed to writing bad code
> either. As far as Windows apps go, .NET is very powerful, IMHO. By far,
> more so than Forms, and probably on par, at the very least, to any other
> product you could name, including Delphi.
Perhaps. But the development cost is far higher.
> Against Oracle-only databases, Forms 6i is great... if it weren't being
> desupported, I'd be using it now. I've already made my feelings known about
> 9i and it's web-only deployment, and I cannot speak to 10g until I know
> something about it. But note that the OP indicated (I thought) that he might
> be connecting to non-Oracle databases as well... and, in that case especially,
> Forms isn't a very good choice, IMO.
If not connecting to Oracle I'd be inclined to agree with you. I don't
see it as a generic tool though I know it could be used that way. It
just isn't its strong point.
Greg Forestieri
> In article <1074267648.488805@yasure>, Daniel Morgan <damo...@x.washington.edu> wrote:
>
> >1. Everybody is leaning .NET. When pretty much everyone knows something
> >competition for jobs becomes more difficult and the salary goes down.
> >There is just more money is being a Fprms expert.
>
> Basically, number #1 on your list, says go with something that's less
> prevalent and perhaps harder to use just to keep your salary high, nevermind
> what's good for the shop. That's a piss poor reason to choose one over the
> other, IMHO.
[snip]
Jeff, whether your household is a household of 1 or a household of 12
your responsibility is to earn as much over the long run as well as
consistently possible. There is that which is fun (or ideal, or
optimum) and there is that which pays the bills. Sometimes they are
not the same.
Greg
Greg Forestieri
Forms is a wonderful environment for development, but with some
limitations depending upon how your system is architected. .NET
sounds like all-MS, and where I'm sitting all-MS is possibly
short-sighted. Everyone is different but about 11% of the clients in
my world are non-MS or non-IE clients. We use Forms and JSP. JSP
mostly for wider-network apps where an "unconnected" app comes in
handy. There are a ton of issues that come to play in a decision like
this.
Good luck
Greg
Jeff
>I agree. But anytime you want to have a head-to-head competition for ROI
>with me using Forms and you using .NET to build an application I will
>gladly put my reputation on the line.
>
>I'll likely be Beta testing while your still trying to create a stable
>connection.
Bluster, buster. If you can be at beta testing in 5 minutes, then my hats
off to you.
>> That's funny because Oracle hasn't offered any options for 9i Forms
> deployment
>> (that I've seen anyway) other than through 9iAS. Kindly point us towards
> some
>> Oracle-supported solutions for 9i Forms deployment that don't require the
> iAS.
>
>Here is one of many links:
>http://dbforums.com/showthread.php?threadid=534153
>
>I wouldn't expect Oracle sales people to make a big deal out if simply
>because it isn't in their best interest.
Did you even read that thread??? It's NOT supported for production
deployment. In fact, they specifically state that you NEED a 9iAS license.
Yeah, I suppose you can buy the 9iAS license and only do the OC4J thing, but,
hello, that's EXPENSIVE... and not exactly a minor detail to leave out.
>Greatly improved installation that prompts for the configuration
>information. The nightmare seems to be over.
>
>Now I can't guarantee what I've seen is what will be released ... but
>I'd take what I've seen over 9iAS anytime of any day.
9iAS was "challenging" in more ways than simply the installer. If that's the
only improvement, then the nightmare continues, IMHO.
>>>challenging. He also talks about Forms 6i which is meaningless. The
>>>curent version is 9i and the 10g version will be here long before you
>>>could have anything ready to deploy.
>>
>> Forms 6i is meaningless now that it's about to be desupported, but it wasn't
>> impertanent to my response. The OP didn't seem interested in Web apps (I
>> could be wrong), and for many shops this is true as well. 6i was the last
>> Forms that supported the simple client-server model... unless Oracle's
>> reintroduced it in a new version that I'm not aware of.
>
>Either I misunderstood you or you misunderstood me or both. I have long
>said Oracle made a huge mistake dropping client-server capabilities from
>Forms. Basically they ceded the market for client server to Microsoft.
>And there are a lot of Oracle shops that can save money and get what
>they need without deploying the latest alphabet soup.
I was/am well aware of your position on this, which is why I was a little
surprised at your comments in your previous post. I brought up 6i in mine
because, to me, it's the last version that I would even recommend (seeing as
it was the last CS version), so when you called it meaningless, I wondered if
you'd changed your tack.
>>>6. Far more sensible approach if using the database for anything other
>>>than columns and rows.
>>
>> Define this, please.
>
>Getting past VARCHAR, NUMBER, and DATE and into TIMESTAMP, INTERVAL,
>user defined data types, etc. Maybe I missed something but I don't
>recall see a lot of Oracle's newer functionality supported by .NET.
I cannot yet speak to that... I am new to it yet. However, I would be
surprised if there wasn't SOME way to accomodate this... either through SP or
such. There is a M$-provided "data provider" for Oracle, which uses the
native Oracle client (which is why I prefer it over the OLE DB connection).
It's possible that some of that advanced functionality is supported there.
BTW, what's the difference between a TIMESTAMP and a DATETIME?
>True today. But many of these same companies are beginning to realize
>that Linux costs a lot less. If they build a huge pile of MS reliant
>infrastructure it makes the cost of moving from Windows to Linux much
>more expensive.
It's expensive if they transition from Windows to Linux? And if they don't?
As I said, Windows dominates the desktops and not just because you've an app
to deploy. As long as there's all that other software that people need
Windows for, it's not likely that the OS wars will heat up again soon.
>>>8. More stable
>>
>> I doubt this.
>
>I'll put my Linux, Solaris, or HP/UX machines up against any Windows
>machine and let a bunch of university students try to crack them or
>bring them down by any means possible. The Windows machine will be the
>first one to fry.
My original thought was that you were saying that a Forms app is more stable
than a .NET app, running under the same OS. This simply isn't true in my
experience, so far.
I won't argue the stability of Windows over Linux--I'm no fool--but I don't
believe your argument is fair or valid either. You're deploying a Forms app
on a single server (we'll assume Linux) that's probably exposed to the
internet, while I'm deploying a .NET app on clients that aren't. Do you
seriously think that's a fair comparison... or even one that works in favor of
your argument? Only if I decided to deploy to the web (only for internet
customers) would this even become a fair comparison, and I ask you, honestly,
would you seriously deploy Forms to the web for internet customers???
I certainly wouldn't. There are many network topologies that can make a
server or client more or less vulnerable to hacking... this discussion could
degenerate rapidly if we're to bring all of that into what was, I thought, a
comparison between .NET and Forms.
>>>9. More scalable
>>
>> Debatable.
>
>Even Microsoft doesn't use Windows for their own accounting system.
>Don't you think they would if they could?
Do they use Forms??? If so, you got me. I bow to the king.
Again, comparing Apples to Apples (ah, maybe we should leave them out of this
;-)), I think the scalability of Forms versus .NET is very debateable.
>>>10. There are several few companies that sell products that convert
>>>Oracle Forms to Java apps and Forms independence.
>
>> Links? I am admittedly skeptical of such claims, because as far as I've ever
>
>> seen, the conversions are buggy, maintenance nightmares. So much so, that
>> they get thrown out and you start over from scratch.
>
>http://www.rocketsoftware.com/portfolio/applicationtransfer/oracleforms.htm
Have you tried said software? I can find no mention of it outside of that
site, no reviews, nothing, not even in usenet. I've no reason to believe
it'll be any better than what I'd expect. Some firsthand testimony could
change my mind.
>> either. As far as Windows apps go, .NET is very powerful, IMHO. By far,
>> more so than Forms, and probably on par, at the very least, to any other
>> product you could name, including Delphi.
>
>Perhaps. But the development cost is far higher.
Based on what?
>> be connecting to non-Oracle databases as well... and, in that case
> especially,
>> Forms isn't a very good choice, IMO.
>
>If not connecting to Oracle I'd be inclined to agree with you. I don't
>see it as a generic tool though I know it could be used that way. It
>just isn't its strong point.
So then, basically, you are inclined to agree with my original response to the
OP. Thanks! I knew I wasn't wasting my time. :-)
Jeff
>Jeff, whether your household is a household of 1 or a household of 12
>your responsibility is to earn as much over the long run as well as
>consistently possible. There is that which is fun (or ideal, or
>optimum) and there is that which pays the bills. Sometimes they are
>not the same.
And to hell with your employer, eh? If it were only a choice over which to
learn (for the purposes of future employment), the argument would make sense,
but when picking one for the shop in which you work, it's a very poor argument
indeed. In fact, using this argument in justification could easily get you
fired... not exactly good for paying the bills.
Greg Forestieri
Sorry Jeff, missed the part where I mentioned "hell" or "employer". I
believe Daniel's post that started that sidebar mentioned competition
for jobs, not screwing your employer. As you say for future
employment that argument makes sense. I'm glad you agree. By the way
I believe in a number of situations Forms server is the superior
solution. 8^|
Jeff
>Sorry Jeff, missed the part where I mentioned "hell" or "employer". I
>believe Daniel's post that started that sidebar mentioned competition
>for jobs, not screwing your employer.
I guess you missed the OP as well, which I and, one would logically assume,
everyone else here were responding to:
>I need to make decision on which development tool to use: Oracle Forms or
>Visual Studio .Net. I am new to Oracle and hope those who have gone the
>decision path could advise the pros and cons. We have Oracle 8 database in a
>AIX unix box (plan to upgrade to 9i in future), MS SQL Server and MS IIS. We
>do not run any Java web server or have any java programs. Mostly VB and ASP
>stuff.
>If I use Oracle Forms for form-based app and web-based app, then I would
>need the resource and skill to run Oracle AS right?
>Will there be any issues or problems I might faced if we standadised on .NET
>and Visual Studio as platform and development tool and Oracle as the
>database? Are there any company doing this?
>Thanks
>regards
Note the repeated use of the word "we." As in this decision is for the shop,
not the individual for his own growth or gain. Daniel's post was in response
to the OP (and my response to the OP) and gave a rather selfish reason,
personal financial consideration, as his first reason for picking one over the
other. Perhaps he misunderstood the OP (or myself), but that's what my
response objected to when you jumped in.
Daniel Morgan
I never advocate screwing one's employer. I believe professional ethics
must at all times be high. If you don't like your employer ... quit. At
the same time ... if two solutions are equally viable ... the employee
that doesn't choose the one the most enhances the employee's
employability is a fool. And no employer should want a fool for an employee.
Greg Forestieri
Yes, and getting back to the OP, it mentions...
"Will there be any issues or problems I might faced if we standadised
on .NET
and Visual Studio as platform and development tool and Oracle as the
database? Are there any company doing this?"
I think that last question begs an answer that speaks to
possibilities. One of them is that we might be doing something that
either 'many' or 'few' others are doing. That does impact not only
your ability to find a job but conversely the ability to find others
to employ who can also do that job, at least if you are looking to
hire. It works both ways. That would tend to be a factor in a the
decision, at least it would be at my shop. And what motivates me
might be motivating those seeking employment.
Just a thought.
Regards
Greg
Jeff
Greg, you make a very good point in that it goes both ways.
In my opinion, if you have the company's best interests in mind anyway, then
if you choose to do something that few others are doing (vs. many, and where
both options are reasonably viable), you are likely doing your shop a
disservice by choosing the option that few are doing. For one, as you point
out, the shop could have difficulty in finding replacement employees as
current employees leave. Also, there is the difficulty in finding and
maintaining support... new versions and patches (from Oracle or M$, in this
case), books, education/training, 3rd-party utilities, etc. The shop could
soon find their selves essentially "orphaned" and having to redo systems all
over again. And as far as personal interests go, you may find yourself with
skills that are not all that high in demand after all... because they're
mostly all .NET shops (or whatever) out there. (That's not exactly smart
either, Dan.)
Daniel Morgan
>And as far as personal interests go, you may find yourself with
> skills that are not all that high in demand after all... because they're
> mostly all .NET shops (or whatever) out there. (That's not exactly smart
> either, Dan.)
Your point is well made. And in the extreme situation you would be
correct. But in the case of what we are discussing here ... certainly we
are not talking about a skill as widespread as VB or .NET because
Microsoft is, after all, the world's largest software house.
But we are talking about software made by the second largest on the
planet. Not the 10th or 100th ... the 2nd largest. It is core technology
in Oracle's own applications and is extremely widespread in corporate,
non-profit, and government usage. If you can make your argument
successfully against it then one would, by extension, have to assume
that knowing Linux, Solaris, HP/UX, AIX, SAP, PeopleSoft, Siebel, ERwin,
etc. of even lesser import and not even worthy of discussion.
My point being your argument is valid only if taken to the extreme. And
at that extreme I would agree with you.
Niall Litchfield
news:6a8cdd95.04012...@posting.google.com...
> Jeff, whether your household is a household of 1 or a household of 12
> your responsibility is to earn as much over the long run as well as
> consistently possible.
I absolutely disagree. Greed is a terrible character trait and tends to lead
to bad bad decisions. It also has a tendency of going badly wrong.
Jeff
>Your point is well made. And in the extreme situation you would be
>correct. But in the case of what we are discussing here ... certainly we
>are not talking about a skill as widespread as VB or .NET because
>Microsoft is, after all, the world's largest software house.
>But we are talking about software made by the second largest on the
>planet. Not the 10th or 100th ... the 2nd largest. It is core technology
>in Oracle's own applications and is extremely widespread in corporate,
>non-profit, and government usage. If you can make your argument
>My point being your argument is valid only if taken to the extreme. And
>at that extreme I would agree with you.
I'm not sure why my argument must already be at some "extreme" to be valid.
Simply do a Monster.com search for "oracle forms" jobs vs. ".NET" jobs. It's
better than 6 to 1 today in favor of .NET. And just because Forms is not dead
yet doesn't mean that it's not headed that way. Maybe you're right, maybe
not... one can only put oneself in the minds of IT managers and watch the
trends. Right now, IMHO, I feel the trend is moving away from Oracle Forms,
while .NET is gaining momentum. Your opinion is different, I guess.
Greg Forestieri
Someone else brought this up already, that is, that the issue is not
so much Forms vs. .NET. They really don't go head to head that much.
It's more J2EE vs. .NET. Forms is it's own market, which really never
has been "huge" in the sense that .NET could become huge someday (and
is seeming to be on it's way there already in some regards). Don't
get me wrong I love Forms, it's great in a lot of apps, but internet
and large intranets is not one of them.
I believe .NET will be successful but that as the use of non-IE
browsers continues to slowly increase it may be a somewhat risky
proposition, of course depending upon your own company's
circumstances. If you require IE as part of your platform, you're in
the clear. In our case we do both J2EE and .NET, both with Oracle as
the backside. :)
Greg
Jeff
>Someone else brought this up already, that is, that the issue is not
>so much Forms vs. .NET. They really don't go head to head that much.
Well, that WAS THE issue when this thread got started. Someone else may have
had a different mindset, but the OP was asking about Forms vs. .Net (as the
thread title shows). As for whether it's a fair or valid comparison, .NET
can be used for the same type of applications... if you're looking for a
simple C-S solution, both will serve (Forms 6i, anyway... with the proviso
that 6i is about to become unsupported).
I would NOT speak to "J2EE vs. VS.NET," because I'm not qualified to make an
argument there.
>It's more J2EE vs. .NET. Forms is it's own market, which really never
>has been "huge" in the sense that .NET could become huge someday (and
>is seeming to be on it's way there already in some regards). Don't
>get me wrong I love Forms, it's great in a lot of apps, but internet
>and large intranets is not one of them.
I like Forms 6i too... though, it has its faults, as they all do. If it is
agreed that the Forms market is small(ish) to begin with, I believe it is only
going to get smaller as desupported 6i users and customers looking for simple
c-s solutions are given no (acceptable) Forms option and forced to move on to
something else.