Anybody knows WinDev?

[ezequiel]

Hi,we use Progress 9.1 and we want to upgrade, but:

01)We haven't paid mainteinance for years (too many time, too much money)
02)Newest version of Progress is awfully expensive.

We do like Progress Database, is very solid and stable. We want to keep it, but we need a more modern programming language

So, looking for alternatives we have found Windev, a french software. It claims to have a connector to Progress Database and a Progress-like programming language called "5GL" (sounds familiar?).

And very cheaper than OpenEdge.


Sounds great, but we really don't know this Windev, so, anybody knows it?

Has somebody work with Windev / Webdev?

Please, share your experience with us.

Thank you!

 

[TomBascom]

I know nothing about "windev".

But I do know that OpenEdge 10 licenses are cheaper than Progress 9 licenses.

What "more modern" features are you looking for? OpenEdge has OO and .NET... And you can combine them with your existing code instead of re-writing it all.

 

[ezequiel]

Thanks for answering, Tom.

Well, I was not comparing Windev with OpenEdge, but with Progress 9.1

We in Systems Dept would like to go with OpenEdge, so you don't need to convince me, but we have to look for alternatives anyway, and Windev doesn't look bad.

My question is still the same: If anybody know WinDev, please tell us about the experience.

 

[GregTomkins]

I sent away for their trial product a few years ago - they sent me a CD that I never got around to loading. Sorry, that's not very helpful.

In my very humble opinion, 90% of the people who are writing Win32 business apps today are crazy. They should be writing browser-based apps.

Of the people who are writing Win32 desktop apps for valid reasons, 90% of them should be using C#. Personally, I would not waste any time on WinDev, or Progress, any of the other niche desktop suppliers.

Of course, if you have a legacy code base, that's a whole different story. And, when it comes to the business logic aspect, it's also a whole different story.

 

[ezequiel]

We have downloaded the trial, too much colors but interesting. Greg is right, we have a lot of code.

Another thing: this product includes interfaces with Exchange, Office, openoffice.org, native export to PDF, its own report generator...

But how is life with WinDev?

Maybe it's very bad...

 

[GregTomkins]

Well, the fact that the answer to your original question would appear to be 'almost nobody' is a massive stroke against using WinDev (same with Progress and many others), in my opinion.

I don't know for sure but I bet there are literally hundreds of (mostly free and/or open-source) C# alternatives for interfacing to Exchange, exporting to PDF, etc.

With WinDev / Progress etc., chances are in many cases it doesn't exist, or if it does it's primitive or costly or slow or all of the above.

PDF is a good example - compare what's available for Progress (1-2 fairly limited libraries) compared to what's available for C# or Java (literally hundreds, eg. iText is awesome). I bet WinDev is similar to Progress.

 

[ezequiel]

Apparently no.

Windev claims to directly relate with those applications, no need to include 3rd party libraries. I don't know about the performance.

As I posted: "this product includes interfaces with Exchange, Office, openoffice.org, native export to PDF, its own report generator".

Sounds like an enhanced Progress 9.1 to me, but I can't be sure.

The answer to my original question is that you said: "Almost nobody" and, so far, nobody here. I'll wait some days, maybe someone know something.

I forgot to say that looks like Windev compilates a java-based executable file, or sort of. I still have to read a lot of manuals.

Thanks.

 

[GregTomkins]

It's a matter of perspective I guess, but I'd take a nearly infinite choice of 3rd party libraries with full source code included (usually) over a single vendor supplied and controlled built-in function (with no source code) any day.

Well, most days !!

Oh, and I think Java on the desktop is the wrong way to go, personally. The GUI never seems to look quite right, you have JVM deployment issues, etc. For the most part, the Java community itself seems to agree that Swing and Applets are not Java's strength. Though personally, I love Swing.

 

[tamhas]

You don't say much about your existing environment. Presumably, if you have a Progress database, you also have a Progress application. Are you suggesting that you might keep the Progress database and redevelop your entire application in something new ... as a cost saving measure? You would keep a database with 10 year old technology and no maintenance? And discard the application? Somehow, this seems like false economy and priorities. How many users do you have? OE10 workgroup database is $200 per named user or registered client in NA prices (I don't have the SA prices for 2010). And OEA, top of the line ABL development tool is only $3600 a seat ... do you have that many developers?

 

[ezequiel]

Thanks, Tamhas. We use Progress 9.1 for both database and programming. About 18-20 users. Windows 2003 server for database, Windows 98 for applications. Windows workstations.

Are you suggesting that you might keep the Progress database and redevelop your entire application in something new ... as a cost saving measure?

Yes. It would be a progressive migration, with both applications (current and new) connected to the same database. No hurry. We did the same thing migrating from Foxpro 2.6 to Progress 9.1, but without the advantage of keeping the database.

At the end of this new process we could migrate the database to WinDev format, and (sadly) say goodbye to Progress.

All of this is just part of a bunch of ideas we are working at. We are also evaluating OE10, SAP (well, that's the most expensive) and many ERP.

 

[GregTomkins]

Just once, once!!!, could we have someone from "outside" talk about doing a conversion *to* Progress? Maybe we should start doing fake posts, just for PR effect

 

[tamhas]

So, 20 users X $200/user for the database = $4000 and you have a brand new up to date license which could be 10.2B. If you have character clients, that's it. If your clients are ABL GUI fat clients, add $110 per user for Client Networking bringing you to a total of $6200. Add $3600 for a copy of OpenEdge Architect for a total of $9800 and you are all set with the latest and the greatest. Gradually, use ABL GUI for .NET to transform the appearance of your application to absolute state of the art with no disruption, no discontinuity, no compromise on the quality of what you can achieve, and maximum reuse of existing logic. Less that $10K is a drop in the bucket compared to the cost to redevelop the application in a new technology and under $1800 a year will keep you on maintenance, entitled to free upgrades forever to the latest and greatest. Figure in the cost of the programming and it looks like a no brainer to me.

 

[ezequiel]

Maybe we should start doing fake posts, just for PR effect

Come on, Greg, don't be cruel....

Thomas knows the subject, and even we don't completely agree , I'm happy to read his opinion.

 

[GregTomkins]

I didn't mean Thomas; he's an insider, and I'm also happy to read his opinions. But what Progress needs is for people nobody on this forum has ever heard of before, "outsiders", coming in here and saying, "Hi everybody! I'm new to Progress and all 5,000 of our developers are ditching C# and Java so we can rewrite our entire 5,000 program application in Progress!".

 

[CarlosO]

We are actually using Workgroup 9.1D with 24 users in a Windows environment and developing with Provision’s AppBuilder. We are very happy with the Database, in 10 years 0 problems. But now, we need something ‘new’ (maybe OpenEdge 10, maybe WinDev, maybe an ERP, maybe another database) because 9.1D does not officially support Windows7 and Windows 2008 Server and there are a plethora of new tools for developers.
We downloaded the OpenEdge Evaluation Kit and tried the OpenEdge Enterprise and it’s very very fast. And when I say fast is fast. Almost 3 times WorkGroup. I don’t know why we didn’t buy Enterprise 10 years ago. I take a look to Architect and almost fall apart. Too complicated for me. Fortunetly AppBuilder is still there.
So now, talking about costs: 25 user OpenEdge Enterprise (yes, Enterprise, not WorkGroup) U$ 14000, 25 Client Networking U$ 2750, 2 Architect U$ 7200 and 1 year maintenance U$ 4300 = U$ 28000.
Our options: Microsoft Dynamics U$ 90000, Progress OpenEdge U$ 28000, PCSoft WinDev U$ 6000 includes Native Access to Progress.
As Tamhas says, with Progress there is “no disruption, no discontinuity, no compromise on the quality of what you can achieve, and maximum reuse of existing logic”.
We are still looking.

 

[TomBascom]

OpenEdge Architect and all the rest of that new stuff isn't really all that complicated. You just need to get over the initial hurdles and the perfect way to do that is to attend the NEPUG meeting April 12; Getting Started with OpenEdge for .NET. You don't even have to show up in person because we're offering it via WebEx and the Amazon Cloud

 

[GregTomkins]

Isn't Microsoft Dynamics an end-user application and therefore an unfair comparison against a mere development environment?

I'm tempted to say "Tomcat or JBoss $0, MySQL or PostGres $0, Java or C# $0, SubVersion or Git $0, Eclipse or SharpDevelop $0, jUnit or TestNG $0, total cost $0" but then I might sound anti-Progress and I'm not really, plus then I'd open myself up to all sorts of claims about how Progress is 100,000x more efficient to develop in and how the platform costs are minimal and worth it in any case.

 

[tamhas]

OEA is an adjustment ... you can always get OES or just a 4GL license if that is what you want to do.

At 24 users, you should be OK with Workgroup unless you are really , really bashing at the DB. One of the obvious reasons to see such a big difference is the issue with multiprocessor machines on workgroup, for which there are some solutions.

In any case, when you start comparing those numbers to programmer salaries and given that you have a working application in ABL, it is hard to imagine actually saving money by changing technologiesl

 

[rstanciu]

>In my very humble opinion, 90% of the people who are writing Win32 business apps today are crazy.
>They should be writing browser-based apps.

agree with GregTomkins today the world is different. Everybody has a mobil-phone in the pocket.
I just add: "They should be writing browser-based and Smart-Phones applications!"

 

[tamhas]

I'm an RIA fan myself, but in the context of the current thread, there is an existing application. Do you think they should dump the whole thing in order to get a browser client on the off chance someone wants to access it from a smart phone?

Picking RIA for a new app is one thing, but for an existing app there needs to be a business reason to make the investment in change.