The Old Joel on Software Forum: Part 4 (of 5) - Delphi for .net : marketing blunder?

Delphi for .net : marketing blunder?

I own a software company. I have 18 programs written in VB3. They're getting old and in need of a rewrite. For various reasons, Delphi looks like a good choice. (Primarily for producing stanadline static linked EXE's.

Borland has spent a lot of effort on Delphi 8 for .net. This would seem to appeal to the 300k or so Delphi programmers (or at least those that might want to upgrade).

Meanwhile, there's a marketing segment of VB 6 programmers (I think it's well over 3 million) who MUST eventually switch languages.

So Borland COULD have (perhaps) come out with a version that's designed to ease the transition of VB6 programmers to Delphi. Perhaps even a version that uses VB syntax.

That would have given them a chance to lure a market 10x the size of the Delphi market. A market filled with programmers who MUST (eventually) move to a new language. Plus, these would be NEW customers, increasing the total # of Delphi programmers. This would have a tipping effect. More programmers means more programmer resources which means it's easier to program, which lures more programmers.

Instead they're going head to head with the OWNER of .net: Microsoft.

WHAT were they THINKING?

Seems like another kylix type blunder: producing a product thier customers don't want.

Mr. Analogy
Tuesday, April 27, 2004

I agree. But, come longhorn, the line between a .Net app and a standalone executable will blur so it might not be such a bad idea.
Knowing borland though, they will probably give up on .Net just in time for it to become viable for the large markets.

Eric Debois
Tuesday, April 27, 2004

"I agree. But, come longhorn, the line between a .Net app and a standalone executable will blur "

True, but LG is AT LEAST 2 years off.

They could have come out with (for example) Delphi-VB, grabbed up 20% of those 3million (hypothetical) VB programmers who'll switch languages in the next couple of years.

THEN, after they managed to triple the # of delphi users, sell them all a .net upgrade.

Oh... and if .net is a miserable failure (and the jury is still out on that one) they'd avoid the trainwreck and concomitant expense.

Borland needs to start stealing MARKETING people from Microsoft.

Mr. Analogy
Tuesday, April 27, 2004

Mr. Analogy ,

Look at Gupta Team Developer. Its a good choice
They dont have a .NET version but will make one once Longhorn comes out.

And it beats VB6 by a long long way

www.guptaworldwide.com

Karthik
Tuesday, April 27, 2004

Seems like you can always count on language threads with people mentioning Delphi and Gupta.

Brad Wilson (dotnetguy.techieswithcats.com)
Tuesday, April 27, 2004

By the way, any development tool that's going to wait for Longhorn until they get into .NET is going to be way behind the curve.

Brad Wilson (dotnetguy.techieswithcats.com)
Tuesday, April 27, 2004

Brad .

I apologize. Just wanted to tell him that something else was available too.

Karthik
Tuesday, April 27, 2004

I guess I must be missing something. One can rewrite the VB application in the current Win32 Delphi (Delphi 7) and, with very mild forethought, have it compile and run without modification under .Net when it becomes prevalent. Meanwhile you are insulated from the myriad of changes that are likely to happen to .Net during the next few years.

What's to decide?

Larry in Iowa
Tuesday, April 27, 2004

Will somebody please tell me what is so special about Longhorn? Why all the fuss and attention?

Mitch & Murray (from downtown)
Tuesday, April 27, 2004

I think you're making an assumption that's incorrect. If I understand, you're saying that Borland shouldn't have moved Delphi to .NET, instead sticking with win32.

The problem with this assumption is that they are sticking with win32. Delphi 7 is a very nice and very current environment (and is about to get a HUGE service pack). When Delphi 9 comes out towards the end of the year, it will have support for both .NET and win32 in one IDE with one code base.

So .NET or not, Delphi is still up-to-date, and looks like it will stay that way.

Jt
Tuesday, April 27, 2004

"I think you're making an assumption that's incorrect. If I understand, you're saying that Borland shouldn't have moved Delphi to .NET, instead sticking with win32."

No, what I meant to say was that this is a critical time (with LOTS of VB 6 programmers transitioning to __something___). It's a prime time to get more Delphi users. Creating a .net version doesn't offer those vb6 programmers anything that vb.net doesn't already have.

"So .NET or not, Delphi is still up-to-date, and looks like it will stay that way. "

Glad to hear it. I actually really like Delphi. I just think there are better things they could have focused on.

Mr. Analogy
Tuesday, April 27, 2004

"Look at Gupta Team Developer. Its a good choice
They dont have a .NET version but will make one once Longhorn comes out."

Looked a Gupta. Fine for DB intensive apps (as I recall) not nearly as good as Delphi for UI intensive apps.

Thanks for the suggestion. (I looked at dozens of languages before settling on Delphi. I looked at Power Basic, Real Basic, etc. etc.).

I really like Delphi.

Mr. Analogy
Tuesday, April 27, 2004

B++ Builder :-)

http://www.softpae.com/default.asp?p=102

Fred
Tuesday, April 27, 2004

"Will somebody please tell me what is so special about Longhorn? Why all the fuss and attention?"

It's the same nonsense that has happened regarding an oasis Microsoft release that's occurred for years. "When Cairo comes out, boy everything is gonna change!" Already the feature list of Longhorn has seen a pretty big trim, and you can certainly expect more as delivery date nears. Overblown features will be brought down to earth as well.

Classic fire and motion though.

Dennis Forbes
Tuesday, April 27, 2004

If near, appear far.
If big, appear small.
If small, appear big.

And, appropriately for LongHorn:

If distance, appear soon.

I taught Bill G. everything he knows.

Sun Tzu
Tuesday, April 27, 2004

"B++ Builder :-)"

Clever product idea. Now, if BORLAND had done that with Delphi (hmmm... Belphi, Delphib, Delphibe ?)

But, after a 5 minute try, one glaring problem: no autocompletion. That's a HUGE time saver.

Also, I'm leary of investing a thousand hours in my company's next generation of programs using a $129 editor/compiler.

You can maintain a company selling a $129 product.

(The product is, I suspect, well worth $129.)

Sun Tzu
Tuesday, April 27, 2004

Anyway for people who already tried Delphi 8, what is your opinion? How does it compare with VS.Net?

George Wanker Bush
Tuesday, April 27, 2004

Gupta Team Developer is a nice tool but slow

nobody
Tuesday, April 27, 2004

Answering the original question:

In my opinion, yes, Borland is making a marketing mistake by moving Delphi toward the CLR.

Delphi is popular and seems very highly regarded even by those who do not use it.

Yes, yes, eventually we will all be assimilated and all code will run on the CLR. Resistance is futile, yadda, yadda, yadda.

But in the meantime, which by the way will be a time period of several years, Delphi has turned down a chance to win a market position that is still pretty sweet: The best RAD development tool for Win32.

VB6 obviously had this position, and Microsoft abandoned it.

Sure, Gupta sounds cool, but the only reason they have a chance at this position is because Borland chose not to take it.

Don't say that Delphi 7 can retain this position while Delphi 9 moves to the CLR. If that were true, VB6 could retain it. Neither one can. Positioning is about the way a product is perceived by its target market. If Delphi is moving to .NET, then it is perceived as abandoning the Win32 position, regardless of the extent to which the technology remains viable.

It's a shame really. This market position isn't going to exist in 10 years, but for a while, a lot of money remains to be had. How unfortunate that all of the companies capable of winning this position don't seem to want it.

Even more unfortunate is the fact that Borland could have easily had its cake and eaten it too:

1. Keep Delphi in its current form and win the position. Be the best RAD development tool for Win32. Nobody else would stand a chance.

2. Go ahead and build your CLR version of Delphi, but do NOT call it Delphi. Give it a new name, and tell the world that it is "Delphi-compatible".

Eric Sink
Tuesday, April 27, 2004

"Don't say that Delphi 7 can retain this position while Delphi 9 moves to the CLR. If that were true, VB6 could retain it. Neither one can. Positioning is about the way a product is perceived by its target market. If Delphi is moving to .NET, then it is perceived as abandoning the Win32 position, regardless of the extent to which the technology remains viable."

Sorry, I don't follow. Delphi had superseded VB as the best Win32 tool even while VB was being actively developed. VB6 was released in, what, 1997? Even then the current version of Delphi (Delphi 3) was acknowledged as a better tool. Since then, Delphi has been improved upon with 4 or 5 major releases (depending on whether you count Delphi 8).

But more importantly, I just don't see why the move of Delphi to .NET must be perceived as an abandonment of Win32 for Delphi. I don't perceive it that way, and Borland doesn't intend for it to be perceived that way.

Several reasons for this: First, Borland says it hasn't abandoned Win32 Delphi, Borland has a major update of Win32 Delphi 7 planned within next month or so, and Delphi 9 (due out end of 2004) will integrate Delphi for .NET and the next version of Win32 Delphi within the same IDE. Does that sound like they're abandoning Win32?

Second, Borland's major marketing push seems to be to position Delphi as a cross-platform tool. Making use of VCL.NET you can port existing Delphi applications to .NET and/or develop new applications that will run on either .NET or Win32. There is no other product out there that even tries to do this. VB6 and VB.NET are the same in name only; VB6 apps require major rewriting to port to VB.NET since they are essentially different languages. In contrast, Delphi.NET and Win32 Delphi are essentially the same language; there are a few things you can't do in Win32 Delphi if you want the code to run on Delphi for .NET, but not much. Moreover, most all the major third party Delphi tool vendors, and there are many, have either released or are planning to release VCL.NET versions of their tools. So right now you can develop for Win32 using these tools and later recompile to run on .NET. There's even some speculation (for what it's worth) that Win32 for Delphi now and a later port to whatever version of .NET is at in a few years after Longhorn is out will be less painful than developing in .NET now and making changes for the .NET version that is to accompany Longhorn. I don't know what to think of that, but Borland certainly does have a much better track record at maintaining backward compatibility than MS does. (Many Delphi 1 programs will compile without changes on Delphi 7, for example.)

People gripe about Borland's poor marketing and about their poor marketing strategy. Well, I agree that it would be much better if they would do more marketing. But I'm not at all sure that there is a better marketing strategy for them to take. They've clearly positioned Delphi as a cross-platform tool, for both .NET and Win32. It's the only product in that market space. I wish they'd push this message out there more, but their resources are a tiny fraction of Microsoft's, after all. Hard to tell whether this is a strategy that will work. But it seems like a pretty good one to me. The impression I get from time spent in Borland newsgroups is that after quite a big of complaining about the move to .NET and fear that Win32 was being abandoned, the developers who spend time on the newsgroups are becoming more comfortable with Borland's strategy and with their commitment to Win32.

If Delphi fails, I wouldn't feel deprived if I had to move to C#. Not like I would have felt left out in the cold if Delphi had failed with only MS C++ and VB6 as major alternatives. But I really would like to see Borland's Delphi strategy succeed. Don't know if it will. But other than a perceived lack of marketing efforts (which are most likely result of limited resources), seems to me like the strategy itself is not too bad. What else would you have them do? It's got to be pretty tough when MS changes the playing field on them like this.

Herbert Sitz
Tuesday, April 27, 2004

Borland has always wanted to be perceived as a cross-platform dev tools company.

Even in the days of Borland C++, when MS has abandoned Win16, they were making a product (Borland C++) which could compile for Win32, Win16, OS/2 and DOS!

I know that Borland C++ is not a competitive product any longer. I think that this happens because quality C++ compilers are not available for free, and because people are switching to languages such as C# and Java. It's the same reason for which Microsoft is releasing the Visual C++ command line compiler for free.

But what I would like to point out is the fact that Borland has experience with this sort of "cross-platform dev tools" strategy.

I don't have the sales numbers, so I can't be 100% sure, but I think that they keep doing this because it WORKS.

It is probable that Borland is positioning their tools as cross-platform development tools simply because this strategy works.

MX
Wednesday, April 28, 2004

> Borland needs to start stealing MARKETING people from Microsoft

No, Borland need a clue.

Wednesday, April 28, 2004

"2. Go ahead and build your CLR version of Delphi, but do NOT call it Delphi. Give it a new name, and tell the world that it is "Delphi-compatible". "

Excellent point!

Mr. Analogy
Wednesday, April 28, 2004

http://bdn.borland.com/article/0,1410,32050,00.html

Ricardo Antunes da Costa
Wednesday, April 28, 2004

JT
Wednesday, April 28, 2004

"Delphi 9 will be ..."

The problem is the "WILL BE", from a marketing perspective.

If Borland hopes to convert VB6 users to Delphi (as I am in the process of ) they must remember that talk means nothing because we've seen that with Microsoft.

I'm not saying Delphi would be as... devious... as Microsoft. But I'm saying that I, as a MS customer, am jaded.

I believe only what's ACTUALLY available.

Gee, if I believed the MS hype, I'd be using .net NOW.

And I'd already have Longhorn.

Mr. Analogy
Wednesday, April 28, 2004

Good discussion about the future of Delphi:

http://delphizine.com/opinion/2004/05/di200405fn_o/di200405fn_o.asp

MX - not a native English speaker
Wednesday, April 28, 2004