The PowerBuilder Phenomenon – One Perspective


Conceptual Model of PowerBuilder Apps

This is a historical perspective of the application development platform and IDE named PowerBuilder by Sybase Corporation.  The summary was written by Rich Bianco and is accurate to the best of his knowledge.  This summary contains facts and opinions about the history of PowerBuilder, the opinions are Rich’s and do not reflect those of any client he represents.  Rich is an expert level PB developer having extensive experience with Version 3 through Version 11.2 against all major DBMS  (Oracle, MS SQL Server, Sybase SQL Server, Informix).

The Birth and Death of PowerBuilder (1985-87)

The initial prototype of PowerBuilder was presented to management at a company called Cullinet in 1985, however they were facing big problems fending off Computer Associates from a hostile takeover, and were not able to capitalize on the outstanding product.  They lost the battle against CA shortly after in 1986 via a hostile takeover.  The prototype was deemed non-essential to the corporate raiders at CA and shelved along with firing of it’s developers.  It would appear that PowerBuilder was dead before it had a chance.  It is ironic that CA practically gave away the most valuable part of the company they engulfed.

Rebirth – Luck, or Karma?  (1988)

In 1988, PowerSoft had been developing apps for the VAX platform and saw that PC products were about to explode, so they asked the vultures over at CA if they wanted to sell the original code.  CA had looked at the code and determined it “had no future”, so they sold it to PowerSoft for a few dollars and good luck wish.  PowerSoft then assembled the original team and started enhancing the program.  Three years after initial concept the project was brought back to life, either by Luck, or Karma.

“PowerBuilder” was for real (1989)

PB App and Google Maps Services

About a year after getting the code from CA, the team had themselves a product, and christened it with the name “PowerBuilder”.  They were developing VAX applications for customers and started to rewrite some internal applications using the new tool.  It was a brilliant way to test, improve and refine the tool.   However they lacked the funding to undertake the complex projects.  HP was invited for a demo of the product, and apparently was so impressed that they pretty much wrote a blank check to PowerSoft.  HP began developing their in-house PC apps using PowerBuilder, while PowerSoft was rewriting complex VAX applications for the PC and collecting real-world test results and refining the tool along the way.  PowerBuilder had a feature that is not available in any other language to this day, not without purchasing add-ons or sacrificing control or offering database independence.

First global customers (1989-90)

The first few worldwide customers were enough validation that PowerBuilder filled a huge void in a huge untapped market.  Most notable is the second customer.  I don’t believe many people know PB was adopted by Microsoft so early on.  I suppose MS would not prefer to advertise the fact that they weren’t developing with their own tools.

  1. Royal Australian Air ForceFirst global customer
  2. Microsoft Corporation– Second global customer.  Management in Redmond, WA not only purchased licenses and developed in-house applications, but it has been said that employees using PowerBuilder raved about it to friends further fueling the PB fever that was under way.
  3. Canadian Government– PB became the tool of choice for development after a recommendation made to the Canadian Government.  Revenue Canada built GST (Govt. Sales Tax) system that is tracking tax returns to this day.  Most Canadian departments use PB for mission-critical systems ranging from those that scan your license plate when you drive into Canada, scanning of your passport, to critical interfacing with Canadian runway and radar system (a 24×7 operation).  The list of applications is quite large many still in use today.  Other applications are the Old Age Pension (developed in 2002),  Case Logging for the Supreme Court & Tax Court of Canada, Firearms Registration System, UN Troop Deployment, Security Clearance System, and Federal Election support systems.  An independent study compared performance of one of their systems written in PB v.s. Visual Basic and the PB one performed 4000% faster.  Maybe Canadians are smarter than us Americans’ and not so eager to use a tool (e.g. Java, .Net)  just because it is the new buzzword, or the latest fad.  [ courtesy link back was added on 8/18/13 at authors request ]

[adrotate banner=”16″]

Money, Money, Money (1990-96)

PowerBuilder was gaining traction fast, companies couldn’t buy it fast enough, and developers with experience were commanding six figure salaries.  Revenue numbers were amazing growing consistently year after year.  Around 1994 the inevitable happened, PowerSoft merged, or was

Sample uses Flicka Services

acquired by Sybase to the tune of around a billion dollars in a stock deal.  But Sybase was unable to retain the core developers of the product, and it appeared that they could not market the product properly.  Sybase stock took a beating and I’m sure they wished the marriage could have been annulled.  Market share was still very strong however but it seems like 1996 was a distinct turning point when Web mania began taking away some of PB’s thunder.  Sybase wasted several years by making a commitment to Java which gave the founders that left time to make a competing tool called SilverStream (acquired by Novell in ’92)

The new scapegoat for poorly performing applications  (1997-2009)

Through all of the years (over a decade) of using PowerBuilder at numerous companies against every major DBMS there was never a performance problem that could be attributed to PowerBuilder.  I have a very strong theory about why people were starting to complain about how PB applications were too slow, too network intensive, or too hard on the database.  Every single time I worked on one of these so called “slow” applications it was evident that the developer was not trained properly at using PowerBuilder.  The combination of simplicity & power facilitated development of serious applications by persons with little programming experience.   As the name “PowerBuilder” implies, it is extremely powerful tool, and in the hands of a junior level developer with little or no understanding of Object Oriented Programming or weak DBMS Skills a fair number of bad PowerBuilder applications started to make a bad name for the tool.  It is my opinion, that PowerBuilder is still (in 2010) the best 4GL around.  I’ll challenge anyone with their tool of choice and I’ll use PowerBuilder, and have the results independently checked for quality, performance, scalability, maintainability, and whatever other criteria they can dream up.  I wish Sybase would sponsor something like this and prove once again how strong the tool is, but it seems as if they don’t even believe in themselves.  Anyway my theory… almost every company I worked there were PB apps written by novice developers.  PB was far too powerful a tool in the hands of a novice, it is like a civilian shooting a bazooka at the range.  It got the job done but was causing damage either by retrieving the entire database into a datawindow causing database degradation, network congestion, it seemed most of the complaining originated from DBA’s and Network Engineers.   It isn’t practical, but you should be required to earn a license to use PowerBuilder.  It is too easy and forgiving to write bad code giving the tool a bad name.

Sybase Marketing – 1 step forward, 2 steps back? (2009 – 2010)

I can’t speak for most developers, but it will be a cold day in hell when I shell out thousands of dollars for development software to learn with.  Shall I even waste the time installing PB11.5 for 30 days elapsed time, meaning I might open it about 3 or 4 times?   Sybase as usual is doing the exact opposite of everyone else.  We are in the age of Open Source, which is being embraced at the corporate level now, and Sybase decides to do hardware to license marriages.  This means when you install the software you have to log into Sybase, and assign a physical PC with a license. If your hard drive crashes (our corporate laptops averaged every two or three years)  you have to find the original instructions and detach the license and attach it to the new PC.   So much for installing PB on your home PC as the license allows when the company you are working for manages the licenses.

In the past, developers were like a shadow sales team for Sybase, we would check out all the new bells and whistles on new versions and make a case to management about why we need it.  With the new assume everyone is a thief first policy Sybase has effectively taken away the ability for developer to verify new features and feel comfortable recommending an upgrade.  No developer that has been around the block will recommend a product based on marketing hype, we need to see it work.

All software companies have to deal with the cost of piracy, there will always be someone who will crack software it is just one cost of doing business.  This cost is fairly predictable and is probably easy to detect via virtually every PC being connected to the internet.  But I really wonder if Sybase has considered the cost of alienating loyal customers.  How many customers will give it up and move to another tool when they are inconvenienced or are made to feel like a thief.   I think of Wal-Mart which I have boycotted because I feel like a thief with all their cameras, and locked up merchandise, it used to be the smaller expensive items, but now they are locking basic items that you’d have to wear a trench coat to sneak out of the store with.  Forget it, I feel like a valued customer at Target and the prices are competitive.

Who will step in next?  I’m betting that  WaveMaker will be the next “PowerBuilder” for Ajax and Web Applications?  (2010 and beyond…)

With PowerBuilders’ popularity on an apparent decline it is only a matter of time before Sybase decides to pull the cord on it.   Even though the tool offers productivity unmatched by other leading tools it may lose critical mass needed to stay in the market.   Everyone is watching and waiting to see what happens and WaveMaker couldn’t have come a minute too soon.  WaveMaker is the first tool that I’ve been exited about using since PowerBuilder.  I’m committed to riding the next wave (pun) and learning WaveMaker.   Watch my blog in the coming weeks for new WaveMaker articles, tutorials and comparisons between PowerBuilder and WaveMaker.  Here is one article on WaveMaker.

Your thoughts?

Are you a PB developer or ex-PB Developer?  I’d love to hear what you think, what you are doing now.  Please let me know via the comments or email.  If you decide to try WaveMaker look for me on the WaveMaker development forum under username RichBianco.

Misc – About PowerBuilder

Product Manuals
  • Installation Guide InfoMaker 11.5
    This book is for anyone installing InfoMaker 11.5. It addresses installation, product licensing with SySAM, migration information and other topics. For complete topic information, see “Contents”.
  • InfoMaker 11.5 Getting Started
    This book introduces InfoMaker and provides a tutorial for learning to use InfoMaker. The lessons teach basics and how to create forms, reports, queries, and graphs.
  • InfoMaker 11.5 Connecting to Your Database
    This book is for anyone using InfoMaker to connect to a database. It assumes you are familiar with the database you are using and have installed the server and client software required to access the data.
  • PowerBuilder 11.5 Extension Reference
    This book is for programmers who build applications that use built-in PowerBuilder extensions.
  • PowerBuilder 11.5 Connection Reference
    This book describes the database parameters and preferences used to connect to a database in PowerBuilder.
  • PowerBuilder 11.5 PowerScript Reference
    This reference manual describes syntax and usage information for the PowerScript language including variables, expressions, statements, events, and functions .
  • PowerBuilder 11.5 Users Guide
    This book describes the PowerBuilder development environment and the use of PowerBuilder user interface tools in building objects including windows, menus, DataWindow objects, and user-defined objects, creating client/server and multitier applications .
  • PowerBuilder 11.5 Getting Started
    This book provides an overview of the PowerBuilder 11.5 development environment, a tutorial in which you build your first application, create a PowerDynamo Web target, deploy and run a Web site, and more. Please see “About this Book” for more details.
  • Installation Guide PowerBuilder Enterprise
    This book explains how to install the PowerBuilder Enterprise 11.5 product.
  • Installation Guide PowerBuilder Desktop/Professional 11.5
    This book describes the installation of the Desktop or Professional edition of PowerBuilder 11.5.



54 Responses

  1. Do you use PowerBuilder for any of your website? I used to do PowerBuilder and would like to make money online in particular make money blogging by utilizing google adsense, google adwords. I enjoyed the history of PB, agree with it

  2. I love your website! did you create this yourself or did you outsource it? Im looking for a blog design thats similar so thats the only reason I’m asking. Either way keep up the nice work I was impressed with your content really..

  3. I love your website! did you create this yourself or did you outsource it? Im looking for a blog design thats similar so thats the only reason I’m asking. Either way keep up the nice work I was impressed with your content really..

  4. I love your blog! did you create this yourself or did you outsource it? Im looking for a blog design thats similar so thats the only reason I’m asking. Either way keep up the nice work I was impressed with your content really..

  5. This film is definately worth watching! It consists of several disturbing/graphic scenes but absolutely nothing folks couldnt manage. It genuinely offers you an insight into wherever these diamonds come from and how far persons will go to obtain them. If youre after an action packed film having a bit of reality thrown in then it is definately worth a chance and in my opinion, Leonardo DiCaprios attempt at a South African accent is definately really worth hearing 🙂 Get watchingthis now!

  6. I started with PB3 and practically learned everything on my own. I recommended PB to quite a few big corporate and they loved it so much that they sticked to it at least until PB8/9. That’s when web-based applications were so much in demand and PB couldn’t live up to that demand (at least not in this part of the world). The technology has become too sophisticated and discouraged many potential newbies. Worst, all new grads are equipped with either Java/PHP or .Net skills. Later version of PB (glaringly) like PB10 and 11 are really half cooked products and questions in the newsgroup go mostly unanswered or given a lousy workaround. I love PB then but my love is dying with every new release of it. PB12 seems to rock they say but I don’t have the dough to acquire it. Even an eveluation copy has to be asked from the sales team instead of download freely like everything else. Ahhh… so sad!!

  7. It is very good for anyone to make a lot of contribution in knowing about the different types of good things there. This is really a special moment in knowing about the proper things there.

  8. I just wanted to comment and say that I really enjoyed reading your blog post here. It was very informative and I also digg the way you write! Keep it up and I’ll be back to read more in the future

  9. I agree with your assessment that PowerBuilder got a bad rep from all the bad developers out there. I contracted for one large company that had taken a dozen Cobol programmers and given them a weeks training in PowerBuilder. They were building windows with a single very large user defined event (many thousands of lines) that was triggered by almost every action a user could take.

    Not sure if bad developers is what killed it though. I think there are plenty of bad developers using .Net these days.

    Sybase marketing, lack of direction, and clumsy web service integration is what killed PowerBuilder. And although there were other related products that would allow published content, these were never properly integrated into the core product. .Net/IIS and java/apache have dominated this area for the last decade.

  10. Nice post! You truly have a wonderful way of writing which I find captivating! I will definitely be bookmarking you and returning to your blog. In fact, your post reminded me about a strange thing that happened to me the other day. I’ll tell you about that later…

  11. I am fresher and just started my career as a PowerBuilder11.5 developer.
    Is there any future in Powerbuilder language.
    Upto my knowledge day by day companies are converting powerbuilder applications to different languagaes.It looks like in coming time powerbuilder will die and there were no jobs for powerBuilder developers.

    Can anybody tell me is it right to stay in PowerBuilder jobs OR time to look for some other option..?

  12. Hi,

    I was a hard core PowerBuilder developer once upon a time. We had sucessfully developed many back office applications in the finance sector. Somehow in my heart and mind, i am still a hard core fan of PB.

    Then i moved to VB.Net, C# and now elivated in the corporate lader so less of a programming more politics!

  13. Thank you for such a fantastic blog. Where else could one get this kind of info written in such an incite full way? I have a presentation that I am just now working on, and I have been looking for such information.

  14. The datawindow was an excellent solution for the time – the time being client/server apps.

    Back then being able to simply say “select stuff, * from nonsense where sense = none”, drag a few fields around to set the visual form, plonk that on a window with a few lines of code and 5 buttons and have a CRUD app was just great.
    If you knew what you were doing, you could create a base class and inherit – this was before VB even had inheritance.
    Like you said, remains unbeaten.

    we have some random stuff passed to an App Server.
    This Appserver (generally) gets an object template, instantiates it, and then calls of_retrieve(none). The datawindow is retrieved and displayed.
    The trouble is that I seem to need a lot more managing code to cope with CRUD where the object is stateless, setting up the containers etc etc

    So the gain by using PB is swamped by the 3-tier issues that all such tools seem to have, and as no-one uses PB much, it wont be.

    I found this post via google from wavemaker as I was looking for the same thing.

  15. Hello Rich,
    Your Blog was interesting and informative. I have been a PB developer for more than 12 yrs( PB 4.0- PB 11.2) and keen on moving over to more recent technologies(particularly .net), but been putting off as I have been getting jobs in PB. At the same time, I realize the time is past due. WaveMaker seems to be interesting and looks lot more easier to transition into for a PB deveoper. But, how are the job opportunities for WaveMaker ? Also, where are you located currently ? I would like to know this as you had mentioned that the company you work for is looking to hire PB developers.


  16. I really enjoyed reading your article. I have been developing in PB since version 3. I have been with the same company now for almost 10 years. I wrote an application to manage class action lawsuits. Our company handle nearly 200 new projects a year and our application handles all of them dynamically thanks to the power of the datawindow. We store most of the datawindow syntax in the DB to avoid re-building the application. Our company has made two attempts to replace the PB app and have failed miserably. Please send me an email and let me know what you are doing.

    • Hi Steven,

      I hear and see examples of what you described on a regular basis. I can’t mention the name of the companies, but there are some that are literally hand-cuffed to old PowerBuilder applications and having desperate times finding developers to support them. One company I worked planned to retire a PB app within one year of my contract start, I left the contract 8 years later and the application had been enhanced, and new interfaces added to SAP or other systems making the process of moving from PB even more complex.

      I believe this happened because management at many companies made a dire mistake of looking for the next best thing rather than really evaluating what was working and why they wanted to change. Now some companies want to change because they view PB as a risk, developers are hard to find, rates are higher, and us PB developers can command just about anything we want. If you’re a PB developer and not working from home, then you aren’t negotiating strongly enough. If you are GOOD at PB you can be working at a Fortune 500 company from your living room and still pulling six figures.

      I see no solution– we see very few successful Java projects, a few successful Microsoft projects but those tend to be at mid-sized companies. The first tool that I’ve found that MIGHT have what it takes is WaveMaker– I’m currently evaluating it but I believe that I could convert many of the smaller PB applications I’ve worked on in short order, and WaveMaker generates true Java applications so that the management doesn’t need to lose sleep about going with some unheard of technology.

      It is my opinion– and strong one at that– that Java cost the business world a catastrophic amount of money. Too many managers jumped ship and got burned, now they caused PB to lose critical mass in market share further putting them in a really tough spot. Us PB die-hards as they like to call us are hard to convince otherwise because nothing comes close. I recommend keeping an eye on WaveMaker, a cautious eye, because there is NO easy solution to what you are up against.

      I wish I could share specifics — a company I am under contract with now is severely hand-cuffed to MANY PowerBuilder applications and their attempts to rewrite in Java are in year three– the users don’t like using the Java versions so the PB apps just won’t go away. The only reason for the PB demise, is putting PB into the hands of an untrained developer — okay this is purely opinion but there’s tons of anecdotal evidence that maybe PB wasn’t so bad after all.

      IF I still feel as strongly about WaveMaker in a few months I may start marketing myself to lead conversions of PB applications to WaveMaker, which will generate pure Java applications using complex frameworks that would take years to master. Stay tuned…

  17. Good post. I have said for years, since the early-to-mid 90’s that the Bain of PB was its ease of use, just as you said, and whenever I said that, people would look at me like I had 4 heads, basically thought I was nuts. I worked with PB from 1.0a through 8. Was a CPI (Certified PB Instructer), worked for a few years as a consultant, for Sybase in the late 80′ into the 2000’s, so I had a ringside seat and saw it all. Thinking about returning to the PB fold.

  18. PowerBuilder is still my tool of choice with Visual Studio running a close second. It is not that I don’t want to use PowerBuilder. It’s that I have no choice but to switch to java. I am being forced into java by the powers that be and so are many of my colleagues. I will probably never know why. Maybe PowerBuilder out priced themselves? PowerBuilder is by far the superior product.

  19. I don’t know how I found this blog. Regardless I spent some time in PB6 and didn’t care for the restrictiveness that the PowerBuilder Language provides and would not want to be a developer or architect whose main tool is PB. Back when I was using it the primary architecture was client/server now everything is done to the n’th tier of abstraction. Everything now is web services with a relative stupid GUI’s. I find myself in xCode, jQuery, java or .NET – all fun to work in but the power is in the web service – not so much the GUI anymore. Plus now with most companies moving to open source and reducing cost of development tools I really don’t see a future with PB except legacy stuff.
    I’m sorry but – the DataWindow is dead.

  20. in 2012, what would be the expert opinion regarding a power builder developer’s career as it have rapid application development?

  21. Great article. I’m from Mexico. I started over working as a developer analyst and I was trained on Powerbuilder. It was very popular in the top banks between 1997 and 2002, then it started to decline the main raison was cost. The license was very expensive around 3000 dollars while JAVA was free and the Sybase support didn’t exists. I took over around five projects where the goal was to migrate from Powerbuilder to Visual Basic since the applications performance were very poor, after I made a couple of changes, customers realized about the advantages of powerbuilder. Unfortunately, schools and companies didn’t train people because of the high license cost. While between 1997 and 2002 you were able to find at least 4 offers powerbuilder jobs a day, In 2011 I found five in whole year and in 2012 just one has appeared. It is sad to see how people get impressed for what JAVA and NET are able to do nowdays when Powerbuilder was able to make it ten years ago. Even, in your article you talk about Powerbuilder was great in server client but I was able to introduce two solutions with Powerbuilder Web.PB and Distributed Powerbuider on PB7 and there was nothing to ask to the current JSP and ASP solutions. Greetings.

    • yes, PB was a great product. It was fast and easy to develop. Good for Client/Server. However web changed all this. PB applications could not be deployed to net without heavy hardware/software(citrix?) Also they were EAS server was buggy.You could program Oracle forms and deployed to web with one click. PB could not.
      Now PB is me too product in .net with little future. The cost is high. Most of the PB applications are converted to VB,c# or Java. Hard to find jobs if you can only do PB.

  22. I’ve used PB for several months and absolutely hate it. With constant source control issues, frequent corrupt files, SLOWNESS…the list goes on and on. I wish the company I work for would get off it and use another program like C# ASP.NET, but the excuse is that this has more issues than PB which I disagree with. I think it’s more of not wanting to learn a new language and converting. With the frequent headaches encountered, I’m surprised the product hasn’t been pulled yet. I’m guessing in several years at the most SAP will eventually pull it off the market. And when that day comes, I’ll be jumping for joy.

    • If you are stuck using it for a while, you will probably start to appreciate it but if you are using the .NET version than I have to agree it is pretty slow compared to the classic version. If you come across any challenges or curious about how to do things at a design level send me a note and I’ll try to help.

      rich – thanks for the comment!

  23. We use PB for our application and I absolutely hate it. Too many instances where it has trouble with source control and bombs on every situation under the sun. I wish PB would just go away so that the company would migrate the application to something more up-to-date.

  24. Aw, this was a really good post. Spending some time and actual
    effort to produce a top notch article… but what can I say… I hesitate a lot and don’t seem
    to get anything done.

    My personal blog page about engineering: – Kermit

  25. I am a .net developer working in what is primarily a PB shop. In this day and age PB is a second rate product. Whatever it was 15-20 years ago is no more because .net blows it away on several levels.
    The day SyBase pulls the plug IS COMING and that is going to be a very serious time for shops who have a large PB code base. Hackers will salivate for the day support ends and then start attacking PB/Appeon web sites and PB applications. There will be little that can be done to keep them out. And when SyBase kills PB Appeon will do the same.
    Why are all of you trying to hang on to a dying language? The smart move is to start porting NOW. If a large shop waits until the announcement date it WILL be too late to get all of the code ported to a Production standard.
    According to the PB Wiki site in 1998 PB “peaked” at 100,000 users WORLDWIDE. .Net has somewhere in the neighborhood of 10+ million users worldwide and I think that is a very conservative estimate. If you are a PB developer and you want to have a job in the next 5-10 years you would be well advised to start learning .net and prepare yourself to drop PB.

Leave a Reply

Your email address will not be published. Required fields are marked *