Delphi Renaissance

bongo69 writes "The TIOBE Programming Community Index is reporting that Delphi is experiencing a revival, this coincides with Borland recently releasing Delphi 2005 allowing users to target both win32 and .net platforms, which to some, is a welcome alternative for .net developers reluctant to use Microsoft Visual Studio or the opensource alternative SharpDevelop."

Comments

It's not a language

[wiredog]

it's an IDE. Delphi uses Pascal, but the compiler can also handle C++.

Re:Languages die for a reason

[ites]

The reason is only rarely technological. Borland's languages, from their Turbo- series onwards, were always significantly better than Microsoft's, but the market chooses tools based mainly on intertia and marketing. Microsoft advertised their way to dominance. Remember that so-called "Visual C/C++" was simply a wrapper around a few poor tools, with Visual Basic being the only component-based system, producing slow interpreted code, while for yearsBorland were producing fast compiled OO apps with Delphi.

There's a reason why some people dislike using MS tools and adore Borland's tools. Often, though, the developer does not have the say in such choices.

Borland and its IDEs

[Space_Soldier]

One thing that annoys me about Borland is that they have a bunch of IDEs that overlap. They should unite all of them and have a single IDE similar to how Microsoft has Visual Studio that supports many languages. If you were to buy a bunch of this IDEs to support multiple programmers who want to use their religious language, the price will be higher than Visual Studio, which comes with the same languages except Pascal and Sun's Java.

Very trustworthy

[Guillermito]

A survey based on a Google search referred on Slashdot. How trustworthy.

Price and licensing killed Delphi

[Junks Jerzey]

Delphi is some kick-ass technology. It's a solid language, it compiles like *lightning* (essentially instaneous since ~1997), zero link times, and the provided libraries are great. Maybe not greater than .net, mind you, but an excellent alternative that was there many years earlier.

Delphi used to be the darling of the small developer and hobbyist programmer. Not only did you get all of the above benefits, but the standard edition was only $70. An absolutely brilliant alternative to Visual C++ and Visual Basic.

But then Borland quietly upped the price and changed the licensing. It used to be Standard for ~$70, Professional for ~$500, Enterprise for ~$1000. Then they changed it so the cheapest edition you could use in a commercial environment was $1000+. The only other version is Personal, around $100, but it is strictly license-bound to be used for learning the language and writing applications that other people don't use. Borland essentially made a one-line change to the license that forced programmers to jump to a product that costs 10x more. The result? Delphi web-sites and tutorials and hobbyist-written programs in Delphi dropped like a rock. Too bad, Borland.

Re:Price and licensing killed Delphi

[Lucas Membrane]

You may have noticed that all the other vendors of affordable language vendors have also disappeared. Gone. (Watcom, Symantec, JPI, Utah, Marshall, Oregon, Stoney Brook, ...) Used to be that you could stop at BDalton software and pick up the language of the week for $69. My first copy of Borland's Pascal was $49. And they gave free support.

Making money in that business with competition from Microsoft on one side and free software on the other must be so difficult. I never try to second-guess the pricing decisions of these firms. Microsoft can decide to lose money on langauages, because languages make the OS business possible. They give away dotnet to anyone who will commit to develop products for it. Last I heard they had over 100 people creating and maintaining one of their language products.

Do the math. It takes 10 to 200 people to keep one of these full-reatured IDE products in good shape. You need about $500k of revenue each year per employee to make this work. It's a dismal business. If selling to corporations at high prices is the only way Borland can see, I'm not anyone to say that I know better.

Re:Price and licensing killed Delphi

[b1t r0t]

But then Borland quietly upped the price and changed the licensing.

They also never seemed to grasp the concept of bundling trial versions into books. I never saw a book+CD about Delphi that had anything resembling a trial version of Delphi. This meant you already had to have a copy just to try the examples from the book.

The main reason I never used Delphi was that I was pretty much all-Mac at the time, but the #2 reason was that the price of entry was too steep for just trying it out. Pascal was never the problem, because I used that on the Mac for years, and Turbo Pascal used a UCSD dialect similar to what the Mac used.

Language Loyalty

[Undefined Tag]

What is it about some development languages that causes such devout loyalty? I'm a huge fan of Delphi, and have been for some time. If you're selling shrink-wrapped software, Java and VB are way too difficult to support. Ditto for the various "database development systems". C++ is too costly in terms of development headaches (unless you're doing high-end games or some such). If you're developing shrink-wrapped user-targetted software (ala Quicken or some such), Delphi is definitely the way to go. This isn't to say I have anything against other languages/development environments - even VB has its place.

Re:Languages die for a reason

[geg81]

You are naive if you think that C, C++, Java, or C# are "superior languages". Languages used commercially are basically going in circles and are still at the level of 1960's and 1970's technology.

Re:Languages die for a reason

[voodoo1man]

Throughout the years, we've seen many [programming] languages die out. It's a natural progression of technology.

This statement demonstrates not only a supreme ignorance of technological change, but of Darwin's ideas as well. Don't take it personally, though - this attitude is ingrained in most Slashdotters. If you didn't know, "survival of the fittest" and all the associated bullshit was actually an invention of Herbert Spencer, noted opportunist and pseudo-scientist, building on Darwin's idea of evolution (who correctly identified environmental fitness as the only criterion for evolutionary change). Read Stephen Jay Gould's The Lying Stones of Marrakech for an interesting take on the matter.

As far as technology goes, it's been shown time and again that there is no such thing as a deterministic progression of technology. Most technological change is motivated primarily by environmental factors (much like evolution, actually), and most environmental factors are motivated by political and sociological conditions. Several good books on the subject are Albert H. Teich's (ed.) Technology and the Future, and David F. Noble's Forces of Production. Noble makes a convincing argument in favor of re-visiting previously developed and alternative branches of technology, focusing on point-to-point and continuous numerically controlled automatic metalworking machinery as examples. Despite being developed several years later, being more technically complicated and backed by millions of US military dollars, after a decade of modest growth continuous-path N/C machines were still inferior to point-to-point machines in efficiency, and were quickly outsold by point-to-point machines when they were re-introduced to the market in 1960.

Lucky for (good) programmers, judging whether software technology is crap or not comes quite naturally, and such expensive trial-and-error market experiments shouldn't be necessary. As many people have pointed out, by many metrics Delphi is worthwhile technologically, and enables certain productivity advantages. The environmental factor is key here - witness yourself parroting the unfounded assumption that Delphi is somehow an ancient, inferior technology. I don't think you thought this up all by yourself, but rather this seems a more widespread notion in the IT industry. The question to ask is why is this so? I don't have a good answer.