Hi:
I am considering writing a new G/L package as I have over ten (10)
years experience in Pick type database programming and I have some
experience with some G/L packages that when the client needs to do
reporting there are issues. I feel that I can write a package with
non of the reporting restrictions that the packages I have seen.

In addition, if I use OpenQM or jBase I am certain that I can ask the
database vendors for a trial version of their database and therefore
allow a potential client to 'try before they buy.' Accuterm has a
similar policy.

The question is what other folks are using right now for new
development: Accuterm GUI or VB DOT NET for the user interface? On
one hand, I am a qualified Pick and Accuterm GUI programmer so I don't
need to learn anything new if I use Accuterm GUI. Besides, Pete is
a great fellow and wonderful to do business with. On the other hand,
maybe if I wrote the client side of the package in VB.NET it would
have all the DOT NET features available.

By the way, I also have experience with Pick 4GL (Like System Builder
+) and would prefer not to use them for a number of reasons.

Anyway, if you could give me some feedback on Accuterm GUI vs. VB DOT
NET for the user interface I would be most appreciative.

Thanks for your advise and time.

Jack wrote:
In my personal experience, no one has asked me for a thick-client GUI
front-end in about 8 years. The AT GUI is great and AT itself is
great and Pete is great. Great great great. I love working with
AccuTerm.

But! The character UI and the thick-client (Windows Forms) UI have
sort of seen their day as far as business apps go. These are simply
not marketable anymore. These days everyone wants a
thin-client/browser UI and the best MV apps around are getting kicked
in the teeth every day by lower quality apps that are prettier and
deployed in the preferred UI. Thin-client means HTML, JavaScript,
CSS, and Ajax on the front, and your choice of back-end tools. If
you're a .NET guy then you can choose from ASP.NET web forms, MVC, or
Silverlight.

To connect into MV, having evaluated all other options for all MV
platforms, my choice remains mv.NET, which I started selling because I
like it so much. mv.NET allows you to move to any MV platform without
changing your connectivity. It's one less tier to worry about and
gives you maximum versatility in terms of DBMS independence.

What about AccuTerm for Internet Explorer? ATIE is a good interim
solution for an intranet but, in addition to being IE specific and
requiring plugin approval, it has a limited scope and simply can't
compete with "real" RIA (Rich Internet Applications).

So to answer your question. These days I write thick-client Windows
applications just for utilities, tray items, and tests. I use ASP.NET
web forms for everything else. I'll get to MVC and Silverlight as
time permits and as our audience comes to understand their benefits.
YMMV For yourself I would encourage you to write your code in a
manner that is as UI independent as possible. Do not put PRINT/INPUT
statements into the same code that files GL data into accounts. Your
UI, whether character or thin or thick, should call to subroutines
that do the work. This is what that term MVC is all about. If you
separate the tiers you can write a character UI now and then easily
add another UI later, whether browser, web service, or mobile device.
The problem most developers have these days is that their UI is so
tightly integrated with the rules that they can't put on a new
front-end without ripping all of their code apart. Those who have
done this are now selling their apps to new audiences - That is the
reward for their effort. For a new app, write it properly in the
first place and you can do whatever you want later.

As a final note. Many people lament the difficulties of web
development but in almost every case after some discussion the reason
for the difficulties is lack of experience or the wrong tools. When
you get a good stack of tools in your kit, beyond the freebie basics,
your coding can move much faster and the results are stunningly
better. Without the right tools you'll find yourself struggling for
months to solve problems that others solved years ago. Without the
right experience you can also get frustrated - experience only comes
with time and lots of reading. If you have a new project and no
timetable, you have the luxury of experimentation, revision, and
refactoring. If you are on a timetable then you might want to
contract with people to do the things that you cannot yet do. And
that's the nature of my business.

HTH

Tony Gravagno
Nebula Research and Development
TG@ remove.pleaseNebula-RnD.com
Nebula R&D sells mv.NET and other Pick/MultiValue products
worldwide, and provides related development services
remove.pleaseNebula-RnD.com/blog
Visit PickWiki.com! Contribute!
http://Twitter.com/TonyGravagno

Hello everyone.

I'm not sure that the Web applications are valid everywhere. I think
in some places a Windows Forms application will be better (better
performance, better access to local resources). A Windows Forms
application can be built using Web services that can access or expose
your PICK subroutines with the rules of business (unless you've mixed
up the rules with the interface). With Web services and using
ClickOnce (if you work with. NET) you can build a "Smart Client" with
all the advantages of Web applications (deployment) and with all the
advantages of Windows Forms applications.

This is an interesting link which may indicate that Windows
applications are still interesting.

Smart Client Soft Factory (SCSF) (VS2005, VS2008, VS2010, WPF)

smarclient.codeplex.com
msdn.microsoft.com/en-us/library/ff648753.aspx


Sorry for my bad English.

Marcos Alonso Vega

To: MAV
MAV wrote:
deployment with all my client applications. I use D3 rule modules to do
all the heavy lifting - including PDF report generation and the like.
Despite the age & difficulties using the D3 class library with .Net, the
performance is excellent in LAN environments. I'm really looking forward
to D3 v9 and its new .Net interface.

g.


--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.simpits.org/geneb - The Me-109F/X Project

ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://www.scarletdme.org - Get it _today_!
--- Synchronet 3.15a-Win32 NewsLink 1.91
The Retro Archive - telnet://bbs.retroarchive.org

On May 12, 11:42*am, Jack <fordguy... (AT) yahoo (DOT) com> wrote:
Hi,

If you would rather not learn .NET or any of the other Web
technologies to build web applications, then you should have a look at
DesignBais.

Please give me a call for a personal Web demo. We will also be
scheduling a few public demos to occur over the next few months.

Rick Weiser
DesignBais International
www.designbais.com
+1 877-889-9777

"Gene Buckle" wrote:
Some years ago I was a proponent of ClickOnce and self-updating
applications here in CDP. This completely negates the need for IT
people to run around and update some number of PCs. As usual the
kick-back I got here was huge. Today it's an expected feature of most
applications. Story of my life.

I agree that desktop apps have value in some environments. All I'm
saying is that from a marketability perspective they're a tough sell -
even when they might be the right solution. If you're writing
something new for sale to a wider audience, go on the path of least
resistance and the one most often followed - in this case the browser
UI (what I call BUI) and RIA.

Another concept here is the Phase2 of any new app, when people like
what you have, and then they say "great, now can we do 'this'?" If
you sell a desktop app to a site and their Phase2 plans include access
for their trading partners into their extranet, you're compelled to
support both thick-client AND thin-client. To minimize the number of
technologies over time, starting with the more versatile option may be
prudent for most developers.

As a happy tradeoff, these days there are two main options for
thick-client apps that also run with almost exactly the same code in
the browser: Flash/Flex/Air and Silverlight. For years people in CDP
have decried the notion of "plugins", but as usual most CDP people are
about 8 years behind the curve, and in this case the rest of the world
is quite happily using these network-enabled platforms that run in the
browser or on the desktop. I should mention that Java also falls into
this category (with desktop updates too), though you rarely see Java
Applets or applications. And Omnis Studio from TigerLogic is an
excellent tool for cross-platform, cross-UI development and deployment
- though Omnis developers are certainly feeling the market pressure of
Flex/Air and Silverlight too.

The bottom line is to understand the needs of your audience, ask lots
of questions of potential prospects before you adopt a technology, and
don't pay attention to any comments you get in CDP.

HTH
T

Tony Gravagno
Nebula Research and Development
TG@ remove.pleaseNebula-RnD.com
Nebula R&D sells mv.NET and other Pick/MultiValue products
worldwide, and provides related development services
remove.pleaseNebula-RnD.com/blog
Visit PickWiki.com! Contribute!
http://Twitter.com/TonyGravagno

On May 28, 1:23*pm, Tony Gravagno
<address.is.in.po... (AT) removethis (DOT) com.invalid> wrote:

Interesting marketing strategy T. I see that it's a two-pronged
approach:

1) Continually "assure" potential customers that you know way more
than they do.

2) Make sure to remind them that they are way behind the rest of the
world.

You've been singing this song for some time now, so how's it working
out?

--
Kevin Powick

Kevin Powick wrote:
If it was working, would I still be singing? Why would I be ashamed
to say in public what everyone says offline anyway? CDP isn't a
forward-looking venue, never was. I was hoping to change that over
the years but it seems obvious that I gave up a couple years ago - and
so have most of our respected colleagues, if you haven't noticed.
This is just entertainment for me now and when my usenet service
terminates my participation here probably will as well.

T

On May 13, 2:16*pm, Tony Gravagno
<address.is.in.po... (AT) removethis (DOT) com.invalid> wrote:
Hi,

You really don't need AccuTerm Internet Edition to run AccuTerm
applications over the internet. That is really just a special purpose
application for sales use, etc. The Standard Edition easily connects
over the internet and gives the full functionality of AccuTerm. I
like the portability and deployment of AccuTerm. I also like the ease
of integration with Pick applications. Additionally, AccuTerm 7 now
has a wallpaper feature for its standard user interface which, when
properly programmed using artwork and photography, gives a beautiful
interface which is not only competitive with .NET but I think is an
improvement. It is faster, and with a little work, you can develop a
GUI application within a text-based paradigm, and make your text
background opaque against the wallpaper background. Along with GUI
context menus, open file dialog boxes, launch of Windows applications,
file transfers, MS Office integration, and screen caching, it can be
made both faster and more attractive than a .NET or AccuTerm GUI
application. I know this sounds like a silly idea to you, but I find
it is actually true. Anyway, integration, portability, and deployment
are much easier with AccuTerm. The client needs only the AccuTerm
Standard Edition client and both text-based and GUI application can be
designed and integrated seamlessly. The clients who see and use my
products prefer the text-based GUI to the VBA GUI of AccuTerm or
of .NET GUI. It is just more efficient, less prone to freesing up or
lock up or shutdown, faster, and yet has the features of mouse
choices, GUI editing, mouse, ctrl-x, ctrl-c, ctrl-v cut, copy, and
paste editing, and much, much more. With .NET, basically, you just
get a UI, not MS Office integration, file transfer, or MultiValue
compatibility.

John Racine

On May 31, 6:28*pm, Paddy <j... (AT) purveyorsoftware (DOT) com> wrote:
P.S. - Most of the people to whom I show my products are more
impressed with the test-based interface (which granted, I have put
more work into than most ever do) than with any, any GUI they've ever
seen. (My text-based tools have mouse selection, mouse cut, copy,
drpop, and paste, ctrl-x, ctrl-c, delete/drop, ctrl-v , and much, much
more; it is a full GUI application within a text-based paradigm). The
immediate comment I get is that they have never seen anything so fast
or trouble free or easy to understand and that they are very sick of
waiting on their GUI apps and their web pages to display and sick of
them freezing and talking up all of their time, which they could be
spending doing other things, rather than fighting with MS Windows .NET
apps. The text-based I have actually sets them apart from both flat
green-screen technology and from convention .NET GUI apps and gives
them something they can actually sell, versus something which merely
looks like other people's applicatons. It is common for developers to
think their .NET applications are fast and trouble-free (though they
all know not as fast and trouble free as a text-based one), but they
forget that end-users systems are rarely so new, up-to-date, and fine-
tuned as their own.