Switching away from OpenEdge
Abe Voelker drafted a beautiful write-up about how to switch away from OpenEdge.
Cure for the Plague: A Theoretical Progress OpenEdge ABL Migration Plan
He suggests to use JRuby (or maybe Groovy, or Java), because they are OS and data-store independent, modern (i.e. OO), allow easy CRUD data-access, and support Web application development.
I admire his energy, the extended effort to research, and the kindness to share his insights with the world.
I am a bit surprised, though, by his dislike of the ABL. I L-O-V-E it!
Anything I ever wanted to do, I was able to do with PROGRESS. It is a beautiful, easy to understand language... Frankly, he is the first SW engineer using the ABL that I encountered for as long as I can remember, who doesn't like it! Aside of the folks that know nothing but OO, and have a hard time understanding procedural approach and relational DB models...
However, almost every single ABL programmer I encountered over the last 5 to 10 years is thinking about moving away from progress! Amazing discrepancy, isn't it?
Their reasons - as well as mine - is the shrinking market for Progress developers, PSCs business approach impacting their attitude towards developers, and the incapability of PSC to market the product.
Another interesting observation as consultant: all the managers complain about the licensing cost and even more about how difficult it is to do business with PSC. As a consequence, almost every company I went to over the last many years wants to, has tried to, or is currently trying to move away from Progress.
The licensing cost argument, I believe, is just the normal lamenting, typical for people who have to make financial ends meet. Of course, charging individuals 3000$ is ridiculous. However, some of that lament is just an excuse! Because, the same managers are happy to accept even higher licensing costs for Oracle, or happily pay licensing for Citrix farms instead of using thin-client AppServer. (See Employing AppServer Made Easy)
The real serious problem of the two is, when PSCs customers feel they are being bullied, peed on, and not appreciated by their supplier!
Hey PSC:Just like switching to another tool won't solve structural issues in the culture of your clients, buying up other companies won't fix structural problems in your business approach! And not every tool is as wonderful as PROGRESS, and can survive despite your attitude!
I think, the reluctance of using AppServer is a symptom pointing at the real problem with the ABL: the programmers and the culture in the ABL development shops!
There is absolutely no reason to write a top-down program today! However, many companies are still requiring programmers to do just that!
Then they find themselves in deep doodoo – the application is so fragile you can't even look at it without something breaking. They can't find developers – who wants to do something that is outdated, and proven an unsatisfactory approach for 25 years already!
Switching to another tool doesn't fix this issue - except, of course, if the other tool doesn't allow for a top-down approach.
To me, the ABL is still the best thing since sliced bread (except, of course, not sliced bread – but that is a different story...). I don't see any real reason to switch away from it, that couldn't be remedied by PSC, if they'd finally get their act together.
- hutegger's blog
- Login to post comments
Comments
One of the curious things
One of the curious things about this sort of thread, which comes up on one forum or another regularly, is that there is an apparent body of shared "knowledge" which lots of people seems to have, but which is a bit questionable when examined more closely. Fact: Progress is an over $500M company, so it must be selling something. Fact: PRGS share price is higher than it has ever been so the market seems to think it is doing something right. Fact: PSC revenue continues to be dominated by OpenEdge. While other products have certainly helped swell the coffers, the OpenEdge license stream is at least holding and seems to actually be growing, so empirically the OpenEdge product is far from dying. Fact, maintenance revenues continue to climb, so although some people are leaving (any product), it is few enough that new license sales continue to swell the maintenance pool. And, while I don't have any numbers, I think the myth of no marketing is particularly dubious now and that is exactly why the share price is up since the market realizes much better what is here.
Yes, people do get in fights over licenses and maintenance and yes I do have personal experience in which a salesperson's greed blew a deal that would have kept someone in the Progress family. But, as you say, it is often true that people end up paying more to go somewhere else, so where is the sense in that. And, if they go open source or whatever, it is likely that what they pay extra in staff time for care, feeding, and development will more than pay for the OE licenses and maintenance. Clearly, the unlimited license program created a time bomb which explodes periodically, but those have to be very few any more. I think part of it is that people get spoiled not having to buy new versions and not having to spend tons of development time accommodating to version changes and they get cheap and don't want to spend any money rather than being grateful for how little it costs.
Similarity of OpenEdge and Mac
Thank you, Thomas, for your pep-talk. It is good to hear something positive every now and then.
The main message of my original post was intended to be that we often blame the tool for our own mistakes. And if there is a culture that not just allows, but supports and demands meritocracy and old-fashioned outdated approaches, switching tools doesn't fix the problem...
Not to start (or continue?) another long winded discussion about PSC's marketing or lack thereof. I do feel the tremendous urge to add one more thought to this: a comparison of Apple vs PSC, Mac vs OpenEdge.
I started using progress while still at University, and after 25 years am still doing it - and I love it!
What kills me is this: There is no other tool out there that allows me to do what I can do with Progress. No other tool allows me to do it as easily. And the programs created with none of the other tools are as easy to understand and maintain as those written in the ABL.
So why the heck am I not seeing this fact plastered all over the place?
And, if they really believe that word-of-mouth is the best way to go, why are they not making a better effort to get it represented at universities, so people would actually hear about it?
I find, that PSC's leadership never really believed in the product (except when they started it, maybe?). Apple, for example, always believed their product was the best. They would just pitifully smile at anybody who mentioned the word PC. Just like we have pity on somebody who has to use SQL, or any flavor of C to write and maintain complex business-applications.
However, Apple nurtured and lead this culture. PSC never stood up and said anything like this - it is always their users, who do!
I believe strongly, that if PSC's Leadership would really believe in OpenEdge - and not just see it as a cow to be milked until it's dead - they would start doing something about it. No. They would start supporting their users in the message and make it public...
Apple was declared dead for 20 years - today they are bigger than MS. Progress and the ABL is declared dead for many years too. But it is still the best thing around! The difference again: Apple never believed the doomsday predictions themselves, PSC did and still does.
It is so sad to watch a good thing remain a secret and have to worry about it surviving. It is frustrating not to be able to do much about it. And it is scary, when one's livelihood depends on it...
PS: The stock price is 29$, and has been in the 20s to low 30s for the last 10+ years. I believe, if their lead horse were marketed better, the stock price would reflect that.
As for the stock price, take
As for the stock price, take a look
http://finance.yahoo.com/echarts?s=PRGS+Interactive#chart3:symbol=prgs;r...
It is frequent when people are complaining about Progress marketing that they make comparisons like those to Apple ... but these are two companies who are not even vaguely in the same business. The market cap of Apple isn't even based on Mac any more, but rather on iPod, iPhone, and iPad. These are all consumer products whereas Progress from the beginning is targeted on enterprise business.
To be sure, I agree that Progress was not particularly well marketed during much of its life. In particular, there was a great story in the late 80s when other 4GLs were still popular that Progress was the only 4GL where one could actually write the whole application in the 4GL. With everyone else, 5-15% of the hard stuff had to be written in C.
But, I think the marketing is much better now. The issue is not whether the general public gets the message because they aren't selling to the general public. The issue is whether enterprise business software people and people who develop products for enterprise business get the message. We aren't selling to game developers and we sure aren't selling to consumers.
ABL Market is shrinking...
There's something you mentioned that I believe should be highlighted:
"I started using progress while still at University"
The point here is that people usually start loving languages in the University and today people are not learning ABL there, they are learning Java, .Net, Ruby and most of the languages you may find on any leave-progress-today article. They believe 4GL is a bad step IT world has taken and any code remaining should be annihilated, just like most of we thought about COBOL code some years ago.
ABL lovers are 80s or early 90s programmers. We're getting old and retiring (don't get me wrong, I'm just making a market characteristics evaluation :o) ) and you won't find a 20-year-old programmer that don't think Progress smells like a dinosaur. And they all believe that if want to improve your career you should not waste time learning old things.
Of course, you can't expect all these new programmers being able to comprehend things like TCO, or learning curve and how wonderful is a platform when you can have someone productive in a matter of weeks, not months. I love Progress, although I know it does not fit any problem (and then I use Java) just because I know that when I need to add a new programmer to my team it won't take a life before he/she starts being (really) able to work.
Another point: Java (and some other platforms) has something Progress just don't have (and probably never will): open source libraries and a huge community working on it. If I have any kind of problem I shall be able to find at least 2 or 3 different libraries available, just pick one and be happy. It took years before we could call or develop webservices in Progress.. That's the sort of thing that can really kill a platform because you'll have a problem and you'll find a solution (using Java, probably ?).
The way it is, ABL community won't last long or, at least, won't be bigger than the COBOL community today. The solution ? Maybe PSC should consider open OpenEdge source code and try to create a community around it, involving universities. What about the profits ? Well, I believe they may profit higher and take even more milk from this good old cow, even if it were free and open source.
No it isn't
Where are your numbers? What is there about 250,000 new users on Progress-based applications that qualifies for shrinking. What is there about growing OE license revenue which makes for shrinking.
Perhaps you should check out http://blogs.forrester.com/mike_gualtieri/10-11-23-java_is_a_dead_end_fo...
By the way...
I loved the blog post you mentioned... But that's the very first time I have seen something like that. I'm not in the USA so I can't say anything about the market there but I'd love to hear people saying "this is right.. people are considering going back to 4GL again".
Look at the slide deck under
Look at the slide deck under the Explained blog post which comes a bit later. You have to register, but it is quite interesting. And no, Forrester isn't the only one saying such things.
The answer isn't just "go back to 4GL". Part of the answer lies in using technologies like BPM and CEP where very little programming can produce lots of functionality. But, anything that enhances productivity and nimble response matters.
All right, you caught me...
... I really don't have such numbers. My comment is based only on what I have seen around, what new companies are using as platforms, the new applications our customers are buying, etc.
But how much of this growth you mentioned is only vegetative (new users for old applications) ? How many new ABL applications you have seen lately ? I have seen more "how to leave progress" initiatives than "how to adopt Progress"...
I work for a company that has thousands of customers (and users) using applications developed on OpenEdge. We have more users today than we had 10 years ago. We have more code written too. But these applications today are considered legacy, something we keep alive because is profitable and it would cost too much to be replaced. They are good and stable applications with low maintenance costs for both we and our customers. But take a closer look and you'll see that some parts were already replaced by things created using other platforms, Java and .Net mostly.
Developers working with those technologies feel sorry about those selected (or cursed) to spend their days working in the old ABL code.
I feel Progress sees OpenEdge just like that. Something good, stable and profitable that they must keep alive. Put some enhancements (ABLOO, WebServices, .Net Widgets) to keep the platform alive but focus on the future: Sonic, Apama, Object Store.
The problem with anecdotal
The problem with anecdotal evidence is that it is, well, anecdotal. There is a lot happening in the Progress world which isn't very visible in many ways. Yes, there are new APs and new applications. There are companies like Skyward which had an old COBOL application for schools that have converted to Progress and are selling left and right. Unless you work for one of those schools, this is invisible to the Progress development community because most Skyward customers don't even have their own machines, the app runs on a Skyward machine. So, no local programmers.
And, coming soon, the worlds only BPM enabled development environment ... OpenEdge. The enhancements in OpenEdge are hardly just keeping it alive!
Nothing More to Say but PSC - Sell It!
You have said so elegantly what many of us have thought and believed for such a long time. I have managed development efforts with both Progress and Java developers. In fact, the experience led to my wanting to end my "traditional career" working for a (very) large contractor to the federal government. The Java developers take 3 times as long to accomplish the same task, tend to know very little about the Oracle database that their code hits, including effective index usage, and want to defer everything to the Oracle DBAs when there are problems. They are the industry equivalent to the ultra-specialized medical field. You know the type... "I work on the left foot, you need a right foot podiatrist."
Meanwhile, all the time, the customer bashes Progress and embraces Java, simply because it is the latest and hottest technology. They don't correlate all of the runtime issues created by a source application that is not bound to the database with which it works with their development platform of choice. Nor do they realize all of the SQL that fails in production simply because the schema didn't match the SQL, and could not be known until execution time. Yes, I know, that should be caught in testing. Too bad 90% of the Java developers that I know are too young to know their craft as systems engineers. They rely on junit and other testing tools to do their work for them. They are the IT industry's version of the kids that can't do math because they rely on calculators.
So, now I transition to my own business, creating and selling powerful Progress apps that meet my customers' needs - with a minimum of defects and a maximum of reliability.
For me, it is too late for the W-2 based corporate world. Give me a customer with a need. I will fill it.