Delphi still fighting the war lost years ago
Embarcadero announced the release of the Delphi XE Starter edition for building native windows applications. I spent many years as a Delphi developer before I made the switch to Java while the Delphi programming language and tools were top notch at the time. However, with the years of abuse at the hands of Borland and then Codegear, it fell way behind. Java IDEs in general are far superior with a lot of attention paid to the actual code aspects of the IDEs while Delphi was very GUI centric and provided superior visual editors. Of course, Delphi is very rigid in how you build applications while Java provides a lot more flexibility to the point of having too much flexibility (the story of Java’s life).
I bought my last version of Delphi a few years ago, I think it was Delphi 10/2006, it cost several hundred dollars and it was a real stinker. I swore I would never buy another version with my hard earned cash. Even at the office we reverted back to Delphi 6 (from 2001 so it was 5 years old) because it was just horrible.
I think at that point, with the release of .net and Visual Studio which was then starting to be given away free, the writing was on the wall. Stuck between the free competitor IDEs , Codegear and then Embracadero were still trying to push a high priced IDE that nobody really wanted in the first place since, albeit rather unjustly, Delphi adoption has never been high.
So now we have a new version, priced at just $199, where you can choose to use a limited version of a dead language (and probably with licensing restrictions). I always find it interesting to see which features they leave out of the basic version, let’s take a look at the Feature Matrix :
- No Code formatting
- No Class Explorer
- No autocompletion for DTD based documents
- No To-Do Lists
- No refactorings
- No diagramming of any kind
- No unit testing
- NO DATABASE SUPPORT (except for Interbase)
- No logging
- No Reports
- No network components
Eh, I could go on but I won’t, to anyone familiar with something like Eclipse most of these items are the basic elements of any IDE that you wouldn’t want to do without. The point is, for $199 today, you can have an IDE from 15-20 years ago. For an extra $600 you can have the professional version which gets you all the basics, but not much more. To create multi-tier datasnap applications, you are looking at spending thousands on an enterprise or architect license. Compare this to the range and scope of the Java APIs that you can get and use for free, whether it be data access, network components, web frameworks or a variety of web servers.
When I moved from Delphi to Java, or even .net, it was like coming out from under a rock and seeing the surface of the Earth for the first time. I have no desire to pay $199 to go back under there.
The Delphi Market
Of course, there is a market for Delphi, there are plenty of places that use it internally for applications, and it really is one of the best ways currently to produce native applications for Windows which after all is still a rather huge market. These people see enough value in upgrading that there is still a market in there, although given the turbulent ownerships of Delphi, probably not a very good one. Meanwhile, Borland was bought by Micro Focus in May 2009.
I tend to object to the calls for something to be made free and open source simply because someone likes it. It simply ignores the masses of time and money that went into making it. However, in this case, I think if Embracadero would like to see any kind of growth, it should look at slashing the prices. Granted, they may be quite happy with the small market of dedicated customers and if they are able to strike a fiscal balance with that, all the best to them.
While cutting prices might make it more attractive, it still might not generate demand (which is probably why they won’t do this). By some standards, the web is now the platform of choice and native apps are in less demand. For corporate customers, there is probably far more demand for web applications for which there are better tools than Delphi. Since code re-usability was never a strong point of Delphi, the ability to leverage existing code is probably minimal, especially given the stateless nature of the web and the stateful nature of a client/server Delphi application.
3 thoughts on “Delphi still fighting the war lost years ago”
Delphi was never for the masses – this happend and will never come back this way. Its complementary have been cracked copies of third party components … I like Java too and I would recommend people for reporting purposes to consider things like Jasper reports if applicable. At least add something that is not totally locked into this world and combine it with open source VCL stuff.
What you describe in this article concerning the underlying assumption was D2006 free edition, the technical description of the Starter XE seems to fit … this was the free version. The starter serves a different purpose – EMB knows that another free IDE will not help. Why should they compete in a market with no money to make?
Things have changed from this time XE is not comparable, but for the “normal” developer fighting on the street the music played today is a different one, here VS is definitely king on Windows, if you write plain C/C++ also more convenient than in the past. Read the XE License and support agreement and tell me if this is the Software for the tiny shop around the corner…
Check the Freepascal/Lazarus project with 0.9.30 … I worked on the betas the whole year. Under Linux too, OS/X I cannot comment but for an app why not.
If one is more focusing the open source corner because of the Linux infrastructure which is the lot more reliable alternative you will be very soon on Java.
What was Delphi native in the past is Delphi Prism today … times change and you can go on UNIX and OS/X too anyway brewing in the labs an Object Pascal compiler for the JVM.
What kind of Windows Software for the desktop is missing today:). Nothing. Except of the scenario you described … and existing Software.
I personally doubt that the prof SKU will stay in the bitter end, they are currently reducing the price systematically but in the RAD Studio Package … independent from what you think …
The target of Delphi in the future are Mac and Linux too … but still more an offering for existing Delphi developers.
So basically, Embarcadero should open source Delphi (despite the product being a profitable business), since no one will want it anyway…?
That said, I assume by ‘dead language’ you meant ‘not massively popular product’. After all, the product’s obviously not literally ‘dead’ since new versions get churned out on a yearly basis. Moreover, the language has acquired some not insubstantial additions since 2006 – C#2-style anonymous methods (inc. variable capturing), and a generics implementation that doesn’t totally suck by design a la Java’s…
I’ve learned long ago that people who don’t like Delphi never will, no matter what Embarcadero or anyone else does. Along the same lines, people who do not like Java (like me) will never like it, no matter how cool or capable an IDE is.
Delphi XE is the best release of the product to date. It is not an outdated 15 year-old IDE. The IDE itself is very comparable to Visual Studio and similar IDEs. The “Starter Edition” is a welcome addition to the product line. Delphi has always been a software product, and its owner has always been in the business of producing revenue. Finding a way to provide a free version of a commercial product without severely hampering your revenue stream is challenging at best, if not impossible.
Borland was turbulent for over a decade. When has Embarcadero ever been turbulent? Borland’s sale to MicroFocus (whose primary product is a COBOL compiler) just emphasizes the excellent decision from Embarcadero to purchase Delphi. Since purchase, the Delphi development team has grown (not shrunk).
Embarcadero is using Delphi internally for several of their core products, including the new AppWave store. And they have a very serious goal of doing cross-platform development with future versions of Delphi.
Doomsayers have said Delphi was dead for 15 years now. They will probably be saying it 15 years from now. For those of us that use Delphi to create optimized, high performance applications, we’ll just laugh at the silliness.
Comments are closed.