Re: Borlands linux strategy ?

.......
Hardly sounds like the market that *Delphi* has made money from
historically.
I'm not saying that it isn't being used for Web development, but IMO,
it's
bread and butter has been: client-server style apps, rich GUIs running on
Windows. The things people love about Delphi are certainly more than that
(native compilation, pure OO, solid non-visual library, etc); but that's
not necessarily *why* it sells.

IMO, the reason why Delphi can compete at all in the server-side market
has
little to do with Borland's server-side libraries/frameworks, but more to
do with the third party community (notably Intraweb and RemObjects) and
support
of MS's platform (ASP.NET). I can't imagine a Delphi-based SaaS product
would be successful or help expand Delphi's marketshare.

You're taking an existing market and applying it to a new technology which
is intended to be applied to a new market. The specificity you suggest I
just don't see as a limiting factor to go multi-platform. I think it's a
fair question to ask if it would be much of a leap for Delphi to be part of
a product family which serves this new market.

But Borland has decided to go the "process" route rather than trying to
profit
off of dev tools where they have been killed by Eclipse and face extremely
tough competition from Visual Studio.

That's an extension to the opportunity for Borland. SaaS could have as much
or more of a process side than traditional shrinkwrap. It's a complete
toolkit that could include Delphi as part of it.

The desktop os will lose relevance for the majority of business apps.
So will the server for that matter.

LOL. What's left? ;)

Thinking about solutions in the large instead of letting the client os
determine your market.

1) Do you have a product planed for China?
2) Do you know the Chinese market well enough to serve their needs?

Why not, if it makes me money? Work the problem: Find a channel, identify
regional expenditures, specify the application with the channel's help.

3) Will Borland make more money with Delphi if they pursue Linux
development
or if they knock out .NET 2.0 and Win64?

Borland should stop all that dev this instant. Come on, this is not
either/or; I'm just hoping that Borland analyze this new market and come up
with another great tool that I can use.

4) There are no "shoulds" in life; there is only what you do and what you
don't do. ;)

Who are you, Yoda? ;)

Let's be honest here; this is ultimately a project scope question. I
believe
in narrowing scope and focus to succeed well with a less ambitious project
than to gamble with a large/expansive project. I also believe in
targetting
markets where you know the market; hence I'm not going to sell software
into
China.

Thanks. Less comp from a bright developer.

But I'm also not (that) crazy or stupid in my risk taking. I'm not
willing
to advocate risking Delphi's future for SaaS,

Once again, not what I'm suggesting.

MSFT's whole future of software development
efforts depend on the notion that every developer includes Windows as
part of his deployment requirement in some form or fashion.

Yes, but that's MS, not Borland. ;)

It's Delphi.

To some degree sure, but I buy MS products don't you? I buy Windows XP,
I don't buy Office. That's business, you pay for a product because it
gives
you some value. Some of that value comes from network effects. Does that
make MS evil? Not necessarily, it's good business to deliver a product
that
provides value and leverages network effects. (There are other reasons
why
MS is evil)

We're not on the same page, here. Sorry I've been too obtuse, and I don't
think MS is evil at all.

This is the second/third time I've said as much. Please stop saying that
this is what I'm saying. That's a huge leap and it's starting to irritate
me.

Oh, come on, get off your high freakin horse. I should have never included
in my first post the word "mentality," because now it's another tough guy
thing on a newsgroup, which I never intended. How droll.

I won't speak for others on the matter, but the reason why I favor MS is
not at *my* expense. I prefer Windows to *nix. I like .NET over Java.
Companies *pay* me to build stuff on Windows. Companies would pay me to
build stuff in Java too, previously companies have paid me to write C++
code.
I prefer to use Delphi and I prefer to develop on/for Windows.

Look, you're picking at my comments for the sake of argument. I prefer to
use Delphi as well, and would love to apply it to other opportunities ouside
of windows, as well as windows. You prefer Delphi on windows because windows
is productive or Delphi is productive? Why not use java or c++ on windows?
Because you can be more productive/make more money with Delphi, but only
because Delphi is a windows tool.

Follow your bliss. Just don't whine about it to me. ;)

Thanks for that. I didn't think I was whining, but if so, I'm happy to
present my whining to your status quo position anytime .

The reality is that Delphi is a niche product for Windows Development. I
love Delphi, and believe that it is the best Win32 development tool
ever...and
will be for all time.

It's a niche product because it's not produced by MSFT, but runs on windows.
Let's at least consider giving this great tool another chance outside the
windows world. Nothing wrong with maintaining the existing strategy as well.

At
what point do these bright part-timers feel exploited rather than served?

Please continue arguing. Not what I meant at all, which should have been
easy to see. What I'm saying is that I believe Borland could easily
accomplish what FPC has done. I'm not suggesting using the FPC work at all.

Because those are the kinds of questions that matter to me when I need to
earn a living. Those are the questions that Borland must ask when looking
to develop new product or expand existing ones.

That's not to say that new and emerging markets can't be profitable, but
they are riskier. I just don't see Linux, in particular, as gaining
enough
marketshare for *Delphi* to be profitable there. Borland tried it, they
failed. I think it's possible to be profitable on Linux with dev tools,
but obviously not how Borland executed with Kylix. And it might not be
profitable
*enough* for Borland to do so; startups OTOH are different beasts.

Yeah, we definately look at this from different viewpoints. What I'm
suggesting, is move away from the whole OS thing. Just an effort to make
something work on Linux would certainly help, because the extension to that
work is a platform agnostic tool. My egotistical opinion is that your
thinking is close to the metal. Mine isn't.

Not to further the horribly flawed analogy between software and the
automobile
industry, but notice that Toyota's most successful hybrid car is the
Prius.
Not a hybrid version of the Corolla. Why? because it's best to *focus*
on the problem and *target* the market. They can get the highest MPG
(save
the manual transmission Honda Insight, another hybrid-only car), the
lowest
drag coefficient, the best integration of computer controlled electric/gas
balance, etc.

Come on. That argument is weak, and you say my analogy is "horribly flawed,"
because you apply it to your context. The point is, the market had to grow
enough for hybrid cars to be commercially feasible. That market is growing,
and now all auto manufacturers are faced with developing hybrid models. The
gasoline engine as we know it will represent a smaller and smaller portion
of the market as we move into the future.

Now if I were in your shoes, the questions I would ask myself at this
point
are: Am I a Delphi developer in my heart or a SaaS developer? Am I more
of a business man or more of a developer?

Thanks for that advice.

James


.

Quantcast