Wednesday, July 4, 2007

Reasons for Optimism

I think my earlier pessimism regarding the future of Delphi was unwarranted, and I am abandoning my bets that CodeGear will see 20% staff cuts by July, 2007. I think there are actually some very encouraging signs of life in the Delphi world that indicate optimism may indeed be warranted.

  • New Blood: The Turbos have apparently been pulling in new Delphi enthusiasts. I saw evidence of this first-hand when I recently went to a user group meeting in Schaumburg, Illinois. (It was a long drive from up here in Wisconsin, but it was worth it.) There were a few attendees who said they just got introduced to Delphi by Turbo Pascal, and were attending the meeting to see what it was all about. This included individuals that had in fact been doing C# programming for a living. This is good to see, as it shows that CodeGear has the potential here to do quite well in attracting new users, even though the conventional wisdom (which is seldom wise) says that nobody doing C# would ever want to “look back” to Native Win32 tools. The attendance at the meeting was very good, in fact every seat was taken, and a few chairs had to be brought in. This is in stark contrast to many previous Delphi user group meetings in years past, when sometimes it would be just Mike Leslie and myself.
  • Third-party component and tool makers continue to sell product. For example, Ray Konopka mentioned that this is the best year that Raize Software has had in the past 4 (IIRC). He continues to upgrade his products and he is not alone. The Delphi.third-party.general newsgroup is very busy with new announcements of products for Delphi and C++Builder. DevExpress continues active development in Delphi components and tools, as do TMS, NexusDB, and a whole lot of other vendors too numerous to list here. The Delphi Win32 world is alive and well after all, people.
  • Delphi still remains the best way to create Win32 applications. This week I decided that I need to write an email client in C++, so as part of my investigation I powered up Visual Studio 2005 and started looking into what is required to get an Email client going. It’s doable obviously, and much better in VS2005 than it had been previously, but it is nothing like what it would be in Delphi. This is important because support for native code is still very vital. I did a little investigation, so I went into and did searches on C++ (7820 jobs) and C# (5926 jobs). People may be surprised to learn that there are more job listings that directly reference C++ than directly reference C#, as “everyone knows” that if you want to get a job you need to learn C#, but this shows that native code is still the most vital. If CodeGear targets native code well, then they ought to make more than enough money to keep Delphi going for ages.
  • There are more Delphi jobs listed than there had been for quite some while. I personally have been contacted for more Delphi jobs than I had in several years. And these are no longer just “help us convert Delphi code to C#” jobs either. There are companies that have very large code bases in Delphi that they have no intention of abandoning. These companies will need to update and maintain their products for years to come, and many seem to have now come to the realization that they need Delphi talent no matter what happens with Borland and CodeGear. In addition, there is new development being done in Delphi by myriad startups across the country. Go to any Delphi user’s group meeting and you will meet several people that run their own business either directly selling a product written in Delphi, or using Delphi in some capacity for in-house needs. Delphi is very popular among the small entrepreneurs that often write their own software. As these businesses grow, they too will be offering opportunities for Delphi programmers. And that is the way the U.S. economy has been operating for ages. Large firms offer jobs to maintain their large investments in code, small firms offer jobs to grow their investments in code.
  • There have been plenty of signs that the CodeGear personnel, though obviously caught off guard by the non-sale of CodeGear to a third party, are making all the right moves to understand and adapt to developer needs and wants. They are talking about adding scripting and AJAX to their supported languages, there has been a refreshing openness in regard to those things that they can legally talk about, and they appear to understand the model to emulate to succeed at what they do. No longer is it just an executive feigning to be all about “Developers, developers, developers!” but it is the rank and file and the “chief scientists” that are doing the kind of things that make developers take notice.
  • CodeGear is obviously aware of the flaws in BDS 2006 and are working very vigorously to over come those flaws. The community has long decried the abysmal help system and content in D2005 and D2006. Now CodeGear has open req’s for technical writers and CodeGear personnel have been churning out an impressive series of Camtasia demos of how to use various features of the IDE. In addition they seem keen on adding QA personnel as well, another sign that they have heard the community’s complaints about quality problems and are actually working to solve them.
  • The non-sale of the CodeGear group, for now, means that the major disruption associated with the divestiture is over for now, and there will be no need for an adjustment to a (possibly hostile) acquisition. It will take at least a year or two for CodeGear to gather enough quarterly results as a separate unit to be able to show prospective buyers what it is really worth. This will give the Delphi and C++Builder teams a chance to get back to work on what they do best: Delphi and C++ Builder. This bodes really well for Delphi, because D2006 was the best Delphi yet, but wasn’t perfect. If the help system gets updated, then D2007 may be one that gets the technical press to write headlines like, “Very Strong Delphi Resurgence Surprises Analysts”.

Yes, in the past I had proffered my share of pedantic pessimism and sour skepticism. It is easy to get pessimistic if the newsgroups are your only world, but one needs to step out of one’s safe little enclave once in a while, and that is what I have done in the last couple of weeks. The first, second and fourth items above were new realizations/discoveries that tipped the balance into optimism. For example, it is hard to know how many Delphi jobs are out there until you actually search. Doing a search on dice or monster can be very misleading. Most of the Delphi contacts have been from those who saw my resume online or saw my post in the newsgroup.

I’m excited. If Delphi does indeed see a renaissance, as seems increasingly likely, people like me that have been using it since it came out stand to do quite well. Sure, I can get jobs based on my C++ and SDLC skills if I need to, so I have an insurance policy, so to speak. Most of the top Delphi talent does though. And I just this last week got a great idea for a software product I could write in a couple of weeks while I have the time. If I write it in Delphi, that is.

No comments: