Board index » kylix » Re: How can Borland make from with Kylix?

Re: How can Borland make from with Kylix?


2005-08-17 02:02:11 AM
kylix1
"Marco van de Voort" < XXXX@XXXXX.COM >wrote in message
Quote
/me agrees.
Do you play BZFlag ?
Jonathan
 
 

Re:Re: How can Borland make from with Kylix?

Jeff Overcash (TeamB) wrote:
Quote

You'd still have to show how Borland will make money with this.
Borland makes money from selling the IDE enhancement modules/wizards or
components whatever they may be . As the examples I gave in the original
post.
Quote
I see your advantage, you get a free tool.
Yes, but that is not the real reason (Borland also gains from the
positive marketing and is in a stronger position against the opposition
Surely you also want to attract as many as posible developers to your
technology? - however the main reason is to find a business solution
where Borland makes a profit and us developers can work and invest with
a peace of mind.
..
Quote
revenue and millions in profit, you are proposing that they also deal with
something that at best might make high six figures and might show a
profitability of low six figures. That is not a product Borland is interested
in.
Are you saying that Borland does not care about products that don't
bring in the millions? What about the obligation Borland has towards its
customers who have purchased those products (in this case Kylix)? If
that is so it sends a seriously bad message to all the future and
existing customers :( Maybe the discussions in the Delphi non technical
group regarding a smaller seperate commited development tools only
supplier company looks like a better proposition for the current Borland
customers.
Quote


Once again, you haven't shown how this will increase revenue. It is
questionable how much it will lower expenses (although it will lower it some).
From what has been hinted at it has to lower it a great deal and even more as
there will be less revenue with your plan, not more. I really don't see why
Borland would invest in any plan that reduces an already very small revenue
stream. They need to concentrate on things that a) bring in a lot of revenue
and b) show potential for great revenue growth. Kylix currently and under your
proposal don't seem to fit either category.

Well, I do not think Borland can bank on getting millions initially,
(that's why you need a strong vision for the appropriate business model-
a long term investment at the least), but when they are in the situation
where their customers are demanding a better service they should be
obliged to deliver something or suffer the consiquences. Neglecting
Kylix could hurt Delphi too.
Quote
>'The community' is hypothetical at the moment, but the community will
>develop when Borland makes the appropriate annoucements.


They didn't with FreeClx. That project essentially died due to utter lack of
community involvement over the first 2 years. This came straight from Borland
at BorCon.
What? The problems had nothing to do with the community. It was and
still is the fact that FreeCLX is licensed under the GPL !!! Who on
earth did Borland want to sell Kylix to ???? No one can work with GPL
components unless you want to produce GPL applications - for whixh Kylix
would be one of the very last tools those developers will use, I can
smell a serious misunderstanding by Borland of what OSS or
deploying/development on Linux is. Linux is a viable deploynent platform
for binary only business applicaitons.The only problem here is Borland
not updating the tool to keep the binaries up to date - at least 3 years
too old - hence why we are debating an alternative business model. It is
not only for OSS GPL applications! Look at Kylix like you look at Delphi
- is there a FreeVCL?
Quote
Several years later there is now a revised closed community that is
working on Clx, but that seems to be going at a very slow rate (which does seem
to be more Borland's fault than the dedication of the people in the project).
What? From the last I read was that the community is standing ready and
eager, but the delay is with the Borland legal department who are slow
to create the appropriate license. Apparantly lead developers are not
even allowed to share a central repository for the source of the Kylix
Community project!
..
Quote
If you think Borland is going to OSS their IDE code you are kidding yourself.
They have to show a very strong financial gain to do so. The board of directors
can be held legally liable (IOW jail time) if they mis managed Borland this
way. Borland is a public company and must handle their resources (which the IDE
code is) in a responsible way.

My suggestions were not to OSS the IDE for the sake of OSS, but to find
a solution which benefits both Borland and it's customers. Reasons
discussed already.. The financial implecations need to be modelled by
Borland to investigate a viable business model From what I suggested my
gut feeling tells me Borland would make a profit. Perhaps not millions,
but at least it could enable Borland to continue to be a development
tools provider and best of all it's customers would be happy. Happy
customers are returning customers..
siegs
 

Re:Re: How can Borland make from with Kylix?

siegfriedn wrote:
Quote
What? The problems had nothing to do with the community. It was and
still is the fact that FreeCLX is licensed under the GPL !!! Who on
Borland can change that to LGPL (LesserGPL).
 

{smallsort}

Re:Re: How can Borland make from with Kylix?

siegfriedn wrote:
Quote

Jeff Overcash (TeamB) wrote:
>
>You'd still have to show how Borland will make money with this.

Borland makes money from selling the IDE enhancement modules/wizards or
components whatever they may be . As the examples I gave in the original
post.

No, that is how they would generate a revenue stream. A revenue stream that
would be much lower than it is today. That does not show how they would make
money IOW make a profit. I've already pointed out that building these plugins
clearly are not expense free and the Linux market has already demonstrated it
will not pay Windows like money for comparable technology. How are you making a
profit on a much smaller revenue stream with a small target market compared to
Windows or even MacOS X (Linux overall market is 2.8%, Mac is 3.0%)
Quote
>I see your advantage, you get a free tool.

Yes, but that is not the real reason (Borland also gains from the
positive marketing and is in a stronger position against the opposition
But they gave away free versions of Kylix and did not get a positive reaction
from the slashdotters. They got the exact opposite. History already has shown
that this statement is not right.
Quote
Surely you also want to attract as many as posible developers to your
technology? -
But it is mostly no longer Borland's technology, it is the OSS.
Quote
however the main reason is to find a business solution
where Borland makes a profit and us developers can work and invest with
a peace of mind.
..
But you haven't shown that they can make a profit. They can't sell their plugin
version for the same price they currently sell the closed source IDE (if
StarOffice is any indicator, it went from $250 per unit closed t o$70 per unit
when they OSS and when to plugins like you are advocating). Plus that means
that Borland would lose money on the full versions of StarTeam, CaliberRM etc as
why pay the higher price for that if you can get what you need in a lower priced
plugin? So far everything you have suggested does nothing but take money out of
the revenue stream and doesn't greatly lower expenses.
Quote

>revenue and millions in profit, you are proposing that they also deal with
>something that at best might make high six figures and might show a
>profitability of low six figures. That is not a product Borland is interested
>in.

Are you saying that Borland does not care about products that don't
bring in the millions?
Yes.
Quote
What about the obligation Borland has towards its
customers who have purchased those products (in this case Kylix)?
None. The only legal obligation any publicly traded company is to the
shareholders. Borland is not a charity so the customer should never give it
money unless it feels it is getting good value, but the reverse is also true.
The customers are not charity cases and Borland is not obligated to continue
something if it does not make sound business sense. That is why they have
stopped active Kylix development in the first place. They could get a much
better return on their investments doing other things.
Quote
If
that is so it sends a seriously bad message to all the future and
existing customers :(
Name a single public company that continues with a product that loses money and
has a declining revenue stream.
Quote
Maybe the discussions in the Delphi non technical
group regarding a smaller seperate commited development tools only
supplier company looks like a better proposition for the current Borland
customers.
No, a smaller company has less wiggle room for failing product lines and will
have to cut them loose quicker.
Quote
Neglecting
Kylix could hurt Delphi too.
There is absolutely no evidence of this. Borland has stated that they have
never found any market research that showed a correlation between the two.
Besides that with .Net and Mono Borland still provides a cross platform
possibility with Delphi, but in this case there is a definite revenue stream
measurable on the Windows side to justify this direction and the Linux would be
nothing but gravy.
Quote

>>'The community' is hypothetical at the moment, but the community will
>>develop when Borland makes the appropriate annoucements.
>
>
>They didn't with FreeClx. That project essentially died due to utter lack of
>community involvement over the first 2 years. This came straight from Borland
>at BorCon.

What? The problems had nothing to do with the community. It was and
still is the fact that FreeCLX is licensed under the GPL !!!
Actually it was dual licensed. Linux itself is GPL, why does Linux succeed
(since you think it does).
Quote
Who on
earth did Borland want to sell Kylix to ???? No one can work with GPL
components unless you want to produce GPL applications - for whixh Kylix
would be one of the very last tools those developers will use, I can
smell a serious misunderstanding by Borland of what OSS or
deploying/development on Linux is.
Very wrong on your part. If you bought Kylix you were not under the GPL -
period.
Quote
Linux is a viable deploynent platform
for binary only business applicaitons. The only problem here is Borland
not updating the tool to keep the binaries up to date - at least 3 years
too old - hence why we are debating an alternative business model. It is
not only for OSS GPL applications! Look at Kylix like you look at Delphi
- is there a FreeVCL?
No, there is no FreeVCL. Never will be. Clx and the VCL are licensed exactly
the same for anyone who purchased Kylix or Delphi. Clx also has a secondary
license for people who want to do OSS project - the GPL. As a purchaser of
Kylix you have the choice of what license to use at any time.
Quote

>Several years later there is now a revised closed community that is
>working on Clx, but that seems to be going at a very slow rate (which does seem
>to be more Borland's fault than the dedication of the people in the project).

What? From the last I read was that the community is standing ready and
eager, but the delay is with the Borland legal department who are slow
to create the appropriate license. Apparantly lead developers are not
even allowed to share a central repository for the source of the Kylix
Community project!
Which is exactly what I said. Read it carefully again, I placed the blame in
the right spot.
Quote

..
>If you think Borland is going to OSS their IDE code you are kidding yourself.
>They have to show a very strong financial gain to do so. The board of directors
>can be held legally liable (IOW jail time) if they mis managed Borland this
>way. Borland is a public company and must handle their resources (which the IDE
>code is) in a responsible way.
>

My suggestions were not to OSS the IDE for the sake of OSS, but to find
a solution which benefits both Borland and it's customers.
No, you discussed things that benefit the customer. You've yet defined how the
revenue stream would be and what expenses Borland would do.
Quote
Reasons
discussed already.. The financial implecations need to be modelled by
Borland to investigate a viable business model From what I suggested my
gut feeling tells me Borland would make a profit.
Business models are not done on gut feelings. Borland needs a lot more than
that. I've shown you how revenues (on an already very low revenue generating
product) are going to drop further with your idea. You've yet to show any
reason why they would increase.
Quote
Perhaps not millions,
but at least it could enable Borland to continue to be a development
tools provider and best of all it's customers would be happy. Happy
customers are returning customers..
So if they made their Kylix customers happy while losing money they are
guaranteed to have them come back and let Borland enjoy losing money on the next
release too :). Please don't forget that Kylix was dropping in units shipped.
It wasn't showing any growth at all.
--
Jeff Overcash (TeamB) On waves of silver I dreamed of gold
(Please do not email 'Till I lost the peace that dreaming gives
me directly unless I dreamed of the moment of my own death
asked. Thank You) That no one ever dreams and lives (Marillion)
 

Re:Re: How can Borland make from with Kylix?

Marco van de Voort wrote:
Quote
You can submit patches and bugs now too. And get them released. Just give
them to Andreas. So while the handling will simplify somewhat, I don't
think a GPL model will be significantly different from the current
situation.
It would be significanty different because I would no more contribute to a
GPL'ed CLX. That's the reason why I do not support FreeCLX with my
patches. And the members of the 'Kylix Community Project' are also not
interested in a GPL'ed CLX. They want to make money with closed code.
--
Regards,
Andreas Hausladen
(www.kylix-patch.de.vu - unofficial Kylix 3 patches)
(andy.jgknet.de/blog)
 

Re:Re: How can Borland make from with Kylix?

Marco van de Voort wrote:
Quote

I don't know how much Borland is gaining or loosing on Kylix either. While
Kylix might not have been the blast that was expected, it could still be
a moneymaker. IIRC a few years ago, even BP7 was one.
Kylix was losing money big time. It has been stated that Kylix did not make
money.
Quote

>Once again, you haven't shown how this will increase revenue.

One never can without access to internal data. Positively or negatively (show
that it is not possible this will work)

But his plan is to open source Kylix which will have the effect of some people
who are currently buying the IDE will just use the free version and if previous
commercial products that have gone in this direction, Borland will have to
greatly lower the price of their bundled plugins (StarOffice went from $250 pre
OSS to $70 under this model). So Borland still has R&D expenses (to write the
plugins), new management expenses (to over see the OSS project), full QA
expenses of today which would also mean they'd have to have some percentage of
the R&D team to respond to bugs found by QA because they can't predict when the
OSS team would get around to fixing things and if they are going to ship a
product they will have to deal with this on their own, still have t he same
expenses for marketing people, same expenses for packaging and shipping. All
that has happened on the expense side of his is a few less R&D people are
needed, but have added on R&D for these plugins that do not exists today under
Linux. This plan has managed to take a small revenue stream and make it smaller
though.
Quote
>It is questionable how much it will lower expenses (although it will lower
>it some).

Personally I think that there are two distinct possibilities. One that does
more to maintain the current project in a community-assisted way, and one
that leads to K4.

K4 will need a major investment, with all uncertainties attached that always
are attached to a project with that kind of magnitude.
I don't see that happening unless Linux suddenly starts growing in the overall
market IOW finally makes it into the desktop arena. We've heard for more than a
decade how that is going to happen real soon now. Meanwhile Linux has gone from
an overall market share of 2.5 in '99 to 2.8 in '04. It is still behind Mac.
Quote

However I think that Borland could do a lot better on the K3 front.

>From what has been hinted at it has to lower it a great deal
>and even more as there will be less revenue with your plan, not more. I
>really don't see why Borland would invest in any plan that reduces an
>already very small revenue stream.

IMHO opening CLX sources wouldn't decrease the revenue source. Worse,
_Borland Partners_ have been asking this for a while now, just to simplify
post-mortem Kylix maintainance. No decision.
The Clx sources are already dual licensed. I don't understand why people keep
ignoring this? The GPL only affects those who want to use the free version, if
you want to spend money with Borland you are not constrained by using the GPL
license you can use the normal license. If you are using the free version then
Borland didn't get any revenue on it so why shouldn't they restrict how you can
use their code.
Quote

>They need to concentrate on things that a) bring in a lot of revenue and
>b) show potential for great revenue growth. Kylix currently and under
>your proposal don't seem to fit either category.

It is not just money. That is the K4 branch of the story, there is still a
support burden for K3. People payed for that, and IMHO Borland should do an
best effort to support it. If this includes open sourcing CLX, soit.
The GPL license is an open source license. So they have open sourced Clx.
Quote

>>'The community' is hypothetical at the moment, but the community will
>>develop when Borland makes the appropriate annoucements.
>
>They didn't with FreeClx.

That was Borlands own choice. The GPL _on purpose_ excluded some of Pascal's
best programmers (e.g. on the existing Pascal related open source projects,
most of which earn a living during the day using Delphi). IMHO a missed
opportunity. Borland hoped to ride the hype, and keep 100% control, but it
blew up in their face.

Rightly so IMHO.

>That project essentially died due to utter lack of
>community involvement over the first 2 years.

... which was pretty much obvious from the start. No reason to put the blame
solely on the community.
Then why are so many OSS projects GPL projects (like oh say ... Linux?) If the
GPL is so bad for OSS projects why is it so widely used in OSS projects?
Quote

>This came straight from Borland at BorCon. Several years later there is
>now a revised closed community that is working on Clx, but that seems to
>be going at a very slow rate (which does seem to be more Borland's fault
>than the dedication of the people in the project)

Indeed. And keep in mind that we also have been keeping a close watch on the
developments. We even submitted lots of bugs, usually with fixes, to the
freeclx SF site.
But that site is basically no longer maintained. This was due, like I said, due
to the utter lack of community contributions over the first couple of years.
Borland did not get a return on the investment of having people manage that so
they basically don't anymore.
Quote

>What you are proposing is almost exactly what was released with K1. The
>Clx Library was dual licensed, there was a FreeClx repository opened up on
>SourceForge and the community never really showed up.

Only as testers.
No, for everyone. Everything I stated above was for everyone. Every bought
version of Kylix was dual licensed. The Source forge repository was open to the
whole community, not just testers.
--
Jeff Overcash (TeamB) On waves of silver I dreamed of gold
(Please do not email 'Till I lost the peace that dreaming gives
me directly unless I dreamed of the moment of my own death
asked. Thank You) That no one ever dreams and lives (Marillion)
 

Re:Re: How can Borland make from with Kylix?

Max wrote:
Quote
Borland can change that to LGPL (LesserGPL).
Better MPL1.1 or they come up with an own open source license.
--
Regards,
Andreas Hausladen
(www.kylix-patch.de.vu - unofficial Kylix 3 patches)
(andy.jgknet.de/blog)
 

Re:Re: How can Borland make from with Kylix?

On 2005-08-16, Andreas Hausladen < XXXX@XXXXX.COM >wrote:
Quote
Max wrote:

>Borland can change that to LGPL (LesserGPL).

Better MPL1.1 or they come up with an own open source license.
Better LGPL/MPL dual license.
 

Re:Re: How can Borland make from with Kylix?

Jeff,
while I do not necessarily agree to the proposals of the OP, I'd like to comment
on a few things.
Quote
None. The only legal obligation any publicly traded company is to the
shareholders. Borland is not a charity so the customer should never give it
money unless it feels it is getting good value, but the reverse is also true.
The customers are not charity cases and Borland is not obligated to continue
something if it does not make sound business sense. That is why they have
stopped active Kylix development in the first place. They could get a much
better return on their investments doing other things.
This is one thing I completely fail to understand. Why should you drop something
because you can make more money on something else? Why not do BOTH?
Looking at Kylix, a single full-time developer would be enough of an investment
for Borland to keep Kylix alive and the revenue stream coming. As small as the
current Kylix customer base might be, they'd still invest more than enough money
to pay for that developer.
Quote
Name a single public company that continues with a product that loses money and
has a declining revenue stream.
Kylix having a declining revenue stream is enterily Borlands fault. The market has
improved a lot since Kylix was released. Usage of Linux servers has grown rapidly
during the last years, and Linux on the Desktop these days also is much more common
than it was a few years ago.
I know one thing for sure: There currently are>500 developers using CrossKylix
alone, and there a lots of Fortune500 companies in that list. The number of "plain"
Kylix users will be much higher than that.
Quote
>Neglecting
>Kylix could hurt Delphi too.

There is absolutely no evidence of this.
Actually there is lots of evidence for this. Go ask the fellows at AToZed, as an
example. There are a lot of enterprises out there who depend on the ability to
deploy to Linux. And naturally, I personally quite a few customers that will
say Good-bye to Borland as soon as it is clear that they won't offer a future
path for Kylix. This includes myself. And oh well, of course Borland also
has already lost quite a few customers to the FreePascal and Lazarus camps.
Oh well, and then there is Borland Brazil, who stated that they have tons
of Kylix customers and get asked about a Kylix 4 all the time...
Quote
Borland has stated that they have
never found any market research that showed a correlation between the two.
Besides that with .Net and Mono Borland still provides a cross platform
possibility with Delphi, but in this case there is a definite revenue stream
measurable on the Windows side to justify this direction and the Linux would be
nothing but gravy.
Borland doesn't officially support Mono. Also Mono isn't a real alternative for
many of Delphi/Kylix current usage scenarios. With hacks it's possible to get
Delphi.NET console and Winforms applications to run under Mono. But VCL.NET and
each and every advantage Delphi has over competing offerings simply won't work.
Quote
>>They didn't with FreeClx. That project essentially died due to utter lack of
>>community involvement over the first 2 years. This came straight from Borland
>>at BorCon.
Sorry, but that's bull. The project didn't die due to lack of community involvement,
but for the following reasons:
1.) FreeCLX didn't get updated by Borland. None of the fixes for Kylix 1 ever got
merged. FreeCLX was never updated to the Kylix 2 and 3 CLX releases.
2.) FreeCLX was crippled. The complete database stuff was missing, for example. This
made it impossible to use FreeCLX in real-life projects.
3.) Changes done to FreeCLX didn't get merged back to CLX
This can easily be verified by todays situation. The community is highly active, and
tons of fixes for CLX are done by the community for a LONG time now, and they never got
merged to either CLX or FreeCLX.
Quote
>>Several years later there is now a revised closed community that is
>>working on Clx, but that seems to be going at a very slow rate (which does seem
>>to be more Borland's fault than the dedication of the people in the project).
>
>What? From the last I read was that the community is standing ready and
>eager, but the delay is with the Borland legal department who are slow
>to create the appropriate license. Apparantly lead developers are not
>even allowed to share a central repository for the source of the Kylix
>Community project!

Which is exactly what I said. Read it carefully again, I placed the blame in
the right spot.
No, what you did is a huge understatement of the reality. It's not "to be more Borland's fault",
it's completely and ONLY Borland's fault. The commmunity team is waiting for 10
months now to get access to the source and the license to be finally be able
to merge the fixes. The community and some of its most prominent members (which
include Borland Technology Partners) have tried everything they could to get
Borland to move. Without any success.
Quote
Business models are not done on gut feelings. Borland needs a lot more than
that. I've shown you how revenues (on an already very low revenue generating
product) are going to drop further with your idea. You've yet to show any
reason why they would increase.
IMHO Borlands business model is flawed big time. See the "cash cow" discussions
over in delphi.non-technical. Or have a look at customer satisfaction regarding
Delphi 8, Delphi 2005 and Kylix. Borland would have needed to invest into their
cash cow a long time ago already. Instead we saw more and more core Delphi and Kylix
developers say good-bye to Borland, without those missing human resources getting
filled up again.
Quote
So if they made their Kylix customers happy while losing money they are
guaranteed to have them come back and let Borland enjoy losing money on the next
release too :). Please don't forget that Kylix was dropping in units shipped.
It wasn't showing any growth at all.
Not surprising as Borland has made it next to impossible to actually buy Kylix.
And by the way: AFAIK Borland has never stated that Kylix was losing money.
Borland is using customers in a lot of areas for years now. Unhappy Delphi customers,
people in need for Unicode support, people who invested into Kylix, C++Builder
customers. During the past 3 years the biggest third-party component suppliers have
said good-bye to Borland. The VCL component community which IMHO made up a huge
chunk of Delphi's success also is dying. The .net route might safe a big amount
of current Delphi users to stay with Borland a little longer, but this alone
won't be enough to get to levels of the "good old times".
Simon
 

Re:Re: How can Borland make from with Kylix?

(I guess I need to get some sleep)
Quote
deploy to Linux. And naturally, I personally quite a few customers that will
+know
Quote
Borland is using customers in a lot of areas for years now. Unhappy Delphi customers,
-using +losing
Simon
 

Re:Re: How can Borland make from with Kylix?

On 2005-08-16, Jeff Overcash (TeamB) < XXXX@XXXXX.COM >wrote:
Quote


Marco van de Voort wrote:
>

>I don't know how much Borland is gaining or loosing on Kylix either. While
>Kylix might not have been the blast that was expected, it could still be
>a moneymaker. IIRC a few years ago, even BP7 was one.

Kylix was losing money big time. It has been stated that Kylix did not make
money.
Possible. But that could have been a result of the half hearted approach. And
what do you count as Kylix investments?
Compare:
Did D8 (the initial for pay .NET project) btw make money overall, if you
include everything .NET related, so including the development of the D7
cmdline compiler? I doubt it. Otherwise there shouldn't have been a .NET in
D2005 according to your reasoning, since .NET would be moving money.
However it is logical even if D8 was not highly profitable that D2005 would
have .NET support. Why? Strategic.
All cleanups and experience from Kylix probably significantly helped with the
.NET adventure anyway.
Quote
>>Once again, you haven't shown how this will increase revenue.
>
>One never can without access to internal data. Positively or negatively (show
>that it is not possible this will work)

But his plan is to open source Kylix which will have the effect of some
people who are currently buying the IDE will just use the free version and
if previous commercial products that have gone in this direction, Borland
will have to greatly lower the price of their bundled plugins (StarOffice
went from $250 pre OSS to $70 under this model).
If that is the way to bulk sales, it could still work. While I don't subscribe
to the opening up everything view, I still think it could work. However it
will be risky. OTOH the project is dead already.
Note that Borland software is not typical end-user software like StarOffice,
so comparing pricing to it is IMHO absurd.
Quote
So Borland still has R&D expenses (to write the plugins), new management
expenses (to over see the OSS project)
full QA expenses of today which would also mean they'd have to have some
percentage of the R&D team to respond to bugs found by QA because they
can't predict when the OSS team would get around to fixing things and if
they are going to ship a product they will have to deal with this on their
own, still have t he same expenses for marketing people, same expenses for
packaging and shipping.
All that has happened on the expense side of his
is a few less R&D people are needed, but have added on R&D for these
plugins that do not exists today under Linux. This plan has managed to
take a small revenue stream and make it smaller though.
Funny however, it is pretty much the same as the plugin model to Eclipse for
which customer interest was tested in the last Borland poll.
Quote
>that leads to K4.
>
>K4 will need a major investment, with all uncertainties attached that always
>are attached to a project with that kind of magnitude.

I don't see that happening unless Linux suddenly starts growing in the
overall market IOW finally makes it into the desktop arena. We've heard
for more than a decade how that is going to happen real soon now.
Meanwhile Linux has gone from an overall market share of 2.5 in '99 to 2.8
in '04. It is still behind Mac.
Server is what brings the money on Linux. But I want to be able to write a
GUI server administration app together with my server stuff of course. That
is not linux on desktop, but still GUI.
Quote
>However I think that Borland could do a lot better on the K3 front.
>
>>From what has been hinted at it has to lower it a great deal
>>and even more as there will be less revenue with your plan, not more. I
>>really don't see why Borland would invest in any plan that reduces an
>>already very small revenue stream.
>
>IMHO opening CLX sources wouldn't decrease the revenue source. Worse,
>_Borland Partners_ have been asking this for a while now, just to simplify
>post-mortem Kylix maintainance. No decision.

The Clx sources are already dual licensed.
I don't understand why people keep
ignoring this?
Because GPL is not open. Period.
Quote
license you can use the normal license. If you are using the free version
then Borland didn't get any revenue on it so why shouldn't they restrict
how you can use their code.
Give some get some. Two way.
Quote
>>They need to concentrate on things that a) bring in a lot of revenue and
>>b) show potential for great revenue growth. Kylix currently and under
>>your proposal don't seem to fit either category.
>
>It is not just money. That is the K4 branch of the story, there is still a
>support burden for K3. People payed for that, and IMHO Borland should do an
>best effort to support it. If this includes open sourcing CLX, soit.

The GPL license is an open source license. So they have open sourced Clx.
Theoretically: yes. Practically:no. Hence the freeclx failure.
Quote
>>community involvement over the first 2 years.
>
>... which was pretty much obvious from the start. No reason to put the blame
>solely on the community.

Then why are so many OSS projects GPL projects (like oh say ... Linux?)
Two reasons:
Main reason: Simply because THOSE ARE NOT LIBRARIES, and thus not linked
against. You do know what the GPL means don't you? Riffle through *nix
libraries and they are _ALL_ LGPL or MPL Even the GNU
gcc libc runtime is LGPL. FPC/Lazarus is GPL, but the libs are LGPL.
The only exception I know for major libs is QT. (see earlier discussion
about this in this group)
This means the free version is non-commercial, which rules out most of the
serious contributors. Which is what happened with freeclx.
Quote
If the GPL is so bad for OSS projects why is it so widely used in OSS
projects?
Because you can use a GPL binary in a commercial environment, but a GPL
lib is pretty useless.
Quote
>developments. We even submitted lots of bugs, usually with fixes, to the
>freeclx SF site.

But that site is basically no longer maintained. This was due, like I said, due
to the utter lack of community contributions over the first couple of years.
Borland did not get a return on the investment of having people manage that so
they basically don't anymore.
Which was written on the wall already when they chose to GPL and not LGPL or
MPL it.
 

Re:Re: How can Borland make from with Kylix?

On 2005-08-16, Jeff Overcash (TeamB) < XXXX@XXXXX.COM >wrote:
Quote
>Neglecting
>Kylix could hurt Delphi too.

There is absolutely no evidence of this.
Not entirely the evidence for the Kylix case, but here inhouse we already
skipped Delphi for several projects due to the lack of a native 64-bit
compiler in the Delphi product pallet, or any prospect. .NET is not a
solution.
Quote
Borland has stated that they have never found any market research that
showed a correlation between the two.
That's also no evidence to the contrary either.
Quote
Besides that with .Net and Mono Borland still provides a cross platform
possibility with Delphi, but in this case there is a definite revenue
stream measurable on the Windows side to justify this direction and the
Linux would be nothing but gravy.
3/4 of my native OS indepandant code does not run on .NET. Most of them are
internal db systems heavily using pointers. (these largely compile on a Mac
with FPC btw, no strange x86 stuff or so, simply the native <>.NET thing).
Delphi.NET is therefore for my purposes no migration solution for these apps
to get cross platform support.
 

Re:Re: How can Borland make from with Kylix?

Jeff Overcash (TeamB) wrote:
Quote

>Neglecting Kylix could hurt Delphi too.

There is absolutely no evidence of this. Borland has stated that they have
never found any market research that showed a correlation between the two.
They didn't ask me, obviously :-)
I want to do cross-platform development. I have Delphi 7 and Kylix 3. If
they don't update Kylix, then I have little reason to upgrade Delphi. I
don't see Mono as an option at this point.
--Rob.
 

Re:Re: How can Borland make from with Kylix?

Jeff Overcash (TeamB) wrote:
...
Quote

But it is mostly no longer Borland's technology, it is the OSS.

Huh? If a system is OSS does not mean you relinquesh all rights. You
still own the software. The important thing is the license agreement.
All OSS in general has a license agreement. See also Marco's reply to
your earlier post regarding the problems with GPL 'compile-into' source
libraries (like FreeCLX) which is bad application of the GPL. Also a
good example where the GPL could be considered a 'cancer'
Contributors to the Linux kernel still have ownership of their
contributions, but the kernel is licensed under the GPL - a proper
application of the GPL.
IMO only end user applications, operating systems, desktop environments
and other core end user and server utilities should be GPL. Never for
Compiler linked libraries!
Quote

But you haven't shown that they can make a profit. They can't sell their plugin
version for the same price they currently sell the closed source IDE (if
StarOffice is any indicator, it went from $250 per unit closed t o$70 per unit
when they OSS and when to plugins like you are advocating). Plus that means
that Borland would lose money on the full versions of StarTeam, CaliberRM etc as
why pay the higher price for that if you can get what you need in a lower priced
plugin? So far everything you have suggested does nothing but take money out of
the revenue stream and doesn't greatly lower expenses.

What is better, sell a million of copies at $10 or 1000 copies at $1000?
For a dev tool I would say a million copies at $10 is by far the better
option. - hyphothetical example of course.
Quote
>Are you saying that Borland does not care about products that don't
>bring in the millions?


Yes.

That stinks. I hope it's an official policy of Borland.
Quote

>What about the obligation Borland has towards its
>customers who have purchased those products (in this case Kylix)?


None. The only legal obligation any publicly traded company is to the
shareholders. Borland is not a charity so the customer should never give it
money unless it feels it is getting good value, but the reverse is also true.
The customers are not charity cases and Borland is not obligated to continue
something if it does not make sound business sense. That is why they have
stopped active Kylix development in the first place. They could get a much
better return on their investments doing other things.

..a little bit shocked to reead this, but...
Well, in a legal sense perhaps not, but the Borland we know is not
considered a 'cowboy' outfit is it? It's just bad business to neglect
your paying customers, period.
Quote

Name a single public company that continues with a product that loses money and
has a declining revenue stream.

I can't think of an example, but the whole point was to make a
suggestion to help Borland generate a positive revenue stream, but
'positive' is a BIG $$$ word for Borland according to you :(
Quote
>Maybe the discussions in the Delphi non technical
>group regarding a smaller seperate commited development tools only
>supplier company looks like a better proposition for the current Borland
>customers.


No, a smaller company has less wiggle room for failing product lines and will
have to cut them loose quicker.

I agree that a smaller company is not the way to go - it was merely a
response to the millions of dollars point.
Quote

Actually it was dual licensed. Linux itself is GPL, why does Linux succeed
(since you think it does).

I did not know that, but the GPL license still wrong wrong wrong for a
compile time linked library.
Quote

>Who on
>earth did Borland want to sell Kylix to ???? No one can work with GPL
>components unless you want to produce GPL applications - for whixh Kylix
>would be one of the very last tools those developers will use, I can
>smell a serious misunderstanding by Borland of what OSS or
>deploying/development on Linux is.


Very wrong on your part. If you bought Kylix you were not under the GPL -
period.

obviously! my misunderstanding.. my resonse was to the fact that the
FreeCLX is under the GPL and no binary deployment only developer would
touch it in that case.
..
Quote
>- is there a FreeVCL?


No, there is no FreeVCL. Never will be.
That's the whole point. Treat Kylix like you treat Delphi or the VCL in
this case. What's with the FreeCLX GPL stuff?
..
Quote
Clx and the VCL are licensed exactly the same for anyone who purchased
Kylix or Delphi.
I was not aware of that, but it is still a mistake to use the GPL as the
second or first license.
..>
Quote

So if they made their Kylix customers happy while losing money they are
guaranteed to have them come back and let Borland enjoy losing money on the next
release too :). Please don't forget that Kylix was dropping in units shipped.
It wasn't showing any growth at all.

Kylix sales were dropping in response to the lack of
commitment/communication from Borland. The message Borland is sending
out is that Kylix has been 'discontinued', so who would by such a
product in these circumstances - a topic discussed to death already <g>
siegs
 

Re:Re: How can Borland make from with Kylix?

Jeff Overcash (TeamB) wrote:
Quote
Kylix was losing money big time.
Is that your opinion or you are stating it as fact? What Borland
statement from 2002 would support your conclusions? Where are the
financial from Kylix operations which shows Kylix as "losing money big
time"? How do you separate the R&D development between Kylix and
Delphi? How do you separate the sales of Delphi with cross-platform
capabilities? Why is that Kylix revenue dropped in the same period of
time when Delphi 7 revenues went up ( Delphi 7 contain free Kylix)?
As far I know all Borlands product at Kylix times where profitable,
perhaps Kylix was not the most profitable but it starting to gain
momentum after the third release. The expensive R&D for Kylix was
back-ported to Delphi and placed Delphi as unique mature product with
cross-platform capabilities. Why the support for Kylix was suspended
for so long? Why the cross-platform feature was removed from Delphi
2005? Why the considerable investment in this new technology is
continuously wasted and discouraged, was that adequately explained to
shareholders, investors and those who standarised on Kylix/Delphi?
Why Kylix is not updated ? Why is not maintained at the minimum
level? Kylix sparked a considerable interest around the world and
revenues from its sales surely could pay for it support. Besides the
community repeatedly offered to help Borland to fix Kylix, so far
nothing is happening in that regard and the investment is decaying.
The reasons for staling cross-platform development is appearing to me
not to be financial at all, but of some other nature.
juliusz