Where did Be and Gassee go wrong ?
raj
to look back at the mistakes that Be's management made.
Here is my list of mistakes that contributed to the death by a
thousand cuts of Be Corp.
1. Marketing BeOS as a "media OS"
Sure, it was an interesting niche and it got a bit of free publicity
because of this. Unfortunately, positioning it as a " media os "
turned most consumers and programmers off.
It made people think of high priced ( eg: Silicon Graphics ) computers
that would be used by only a small group.
2. Overpriced Be machines.
Consumers would have bought a Be as their second or third computers.
The prices ( who cared if there were two processors ) were way out of
line with that of commodity Intel PCs.
3. Inadequate development environment.
Some thing like Visual Studio, with a slick IDE and a good GUI
generator was badly needed.
(Hindsight is a wonderful thing, blah, blah... but not learning from
others failures is worse )
Greg Weston
<isra...@optushome.com.au> wrote:
> Now that the Be story is almost over, it interesting and instructive
> to look back at the mistakes that Be's management made.
First, I don't believe "the Be story is almost over." I think we're
headed for a very serious plot twist.
> Here is my list of mistakes that contributed to the death by a
> thousand cuts of Be Corp.
>
> 1. Marketing BeOS as a "media OS"
> Sure, it was an interesting niche and it got a bit of free publicity
> because of this. Unfortunately, positioning it as a " media os "
> turned most consumers and programmers off.
> It made people think of high priced ( eg: Silicon Graphics ) computers
> that would be used by only a small group.
Actually, you missed the real issue: Repeated refocussing of the target
market which confused potential customers and alienated multiple
generations of developers.
> 2. Overpriced Be machines.
> Consumers would have bought a Be as their second or third computers.
> The prices ( who cared if there were two processors ) were way out of
> line with that of commodity Intel PCs.
I disagree. Actually, I think the elimination of Be-branded hardware
was a major mis-step. The price of the 'Box as offered to developers -
the only people who could get one - was _extremely_ good given what it
included. You can't make a fair comparison by putting it next to
contemporary baseline x86 boxes, let alone what's out now.
> 3. Inadequate development environment.
> Some thing like Visual Studio, with a slick IDE and a good GUI
> generator was badly needed.
I disagree here, too. Even without a traditional RAD tool I was getting
shorter development cycles out of the BeIDE than I do out of Delphi -
and I've been using Delphi longer. Very few of the first (or even
second) generation of BeOS developers even made a peep about the lack
of such tools, and those that did were motivated to address said lack -
while they were motivated (see my response to #1).
G
quu...@newsguy.com
<gwesto...@CAPShome.com> wrote:
>In article <rk0iftc97s4lou6e3...@4ax.com>, raj
><isra...@optushome.com.au> wrote:
>
>> Now that the Be story is almost over, it interesting and instructive
>> to look back at the mistakes that Be's management made.
>
>First, I don't believe "the Be story is almost over." I think we're
>headed for a very serious plot twist.
>
Hope springs eternal in the hearts of the faithful....
>
>> Here is my list of mistakes that contributed to the death by a
>> thousand cuts of Be Corp.
>>
>> 1. Marketing BeOS as a "media OS"
>> Sure, it was an interesting niche and it got a bit of free publicity
>> because of this. Unfortunately, positioning it as a " media os "
>> turned most consumers and programmers off.
>> It made people think of high priced ( eg: Silicon Graphics ) computers
>> that would be used by only a small group.
>
>Actually, you missed the real issue: Repeated refocussing of the target
>market which confused potential customers and alienated multiple
>generations of developers.
>
Dead on.
>
>> 2. Overpriced Be machines.
>> Consumers would have bought a Be as their second or third computers.
>> The prices ( who cared if there were two processors ) were way out of
>> line with that of commodity Intel PCs.
>
>I disagree. Actually, I think the elimination of Be-branded hardware
>was a major mis-step. The price of the 'Box as offered to developers -
>the only people who could get one - was _extremely_ good given what it
>included. You can't make a fair comparison by putting it next to
>contemporary baseline x86 boxes, let alone what's out now.
>
...but Be didn't then (and *definitely* doesn't now) have the capital
to be a hardware manufacturer. The BeBox was cool, but it had no
market -- Be couldn't decide who they wanted to sell to: developers,
graphic artists, or musicians. And there was the problem of no
software for *any* of the target markets.
>
>> 3. Inadequate development environment.
>> Some thing like Visual Studio, with a slick IDE and a good GUI
>> generator was badly needed.
>
>I disagree here, too. Even without a traditional RAD tool I was getting
>shorter development cycles out of the BeIDE than I do out of Delphi -
>and I've been using Delphi longer. Very few of the first (or even
>second) generation of BeOS developers even made a peep about the lack
>of such tools, and those that did were motivated to address said lack -
>while they were motivated (see my response to #1).
>
>G
You're in the minority here (even though I agree with your analysis).
Most developers for the BeOS were and continue to be hobbyists, and
these folks want pretty GUI builders and RAD tools. There were a few
attempts to deliver one for BeOS, but AFAIK none ever made it to a 1.0
release.
I have to say that the Be API was very nice from a developer's
perspective; it's one of the things I will miss when Be goes away.
quux111
Greg Weston
wrote:
> On Wed, 09 May 2001 13:04:57 GMT, Greg Weston
> <gwesto...@CAPShome.com> wrote:
>
> >First, I don't believe "the Be story is almost over." I think we're
> >headed for a very serious plot twist.
> >
>
> Hope springs eternal in the hearts of the faithful....
I haven't counted as one of the Be faithful since around R3, actually.
I've been well and thoroughly lambasted for my opinions about the past,
present and future of Be and their products in multiple forae. But
while I _do_ expect the end of Be as an independent entity in the
relatively near future, I _don't_ think it's the end of the legacy,
such as it is, or the products.
> >> 2. Overpriced Be machines.
> >
> >I disagree. Actually, I think the elimination of Be-branded hardware
> >was a major mis-step. The price of the 'Box as offered to developers -
> >the only people who could get one - was _extremely_ good given what it
> >included. You can't make a fair comparison by putting it next to
> >contemporary baseline x86 boxes, let alone what's out now.
>
> ...but Be didn't then (and *definitely* doesn't now) have the capital
> to be a hardware manufacturer.
Be was selling the 'Box for too little in an attempt to build mindshare
among developers. It worked. The first generation of BeOS developers
was active and activIST. We were highly experienced and highly
motivated to contribute what we could to a promising platform. Having
built said mindshare, I'm of the opinion that Be could have sold the
hardware to VARs and end-users at a price sufficient to sustain a
hardware business. That opinion may be wrong in general, but some of us
were successfully selling proposals based on the BeBox.
> And there was the problem of no
> software for *any* of the target markets.
Except for developers, and developers _are_ the first market you have
to sell to.
> >> 3. Inadequate development environment.
> >> Some thing like Visual Studio, with a slick IDE and a good GUI
> >> generator was badly needed.
> >
> >I disagree here, too. Even without a traditional RAD tool I was getting
> >shorter development cycles out of the BeIDE than I do out of Delphi -
> >and I've been using Delphi longer. Very few of the first (or even
> >second) generation of BeOS developers even made a peep about the lack
> >of such tools, and those that did were motivated to address said lack -
> >while they were motivated (see my response to #1).
>
> You're in the minority here (even though I agree with your analysis).
I accept that such opinions put me in the minority. However, to claim
that "a good GUI generator was badly needed" implies that a GUI
generator is somehow necessary for development. That is clearly not the
case. Far too much development has been done on BeOS and other
platforms without such tools - and without the demand for such tools -
for me to take that statement seriously. I understand it...I just don't
think it's realistic.
> Most developers for the BeOS were and continue to be hobbyists, and
> these folks want pretty GUI builders and RAD tools.
I believe that to be true of Mac OS and Windows as well. The follow-up
question is, though, how much the hobbyist demographic contributes to
the commercial well-being of a nascent platform. They can certainly
contribute to growth of an established platform, but I suspect they
don't play much of a role in that establishment (and I've heard that
belief echoed by representatives of 2 major, multi-platform toolchain
vendors).
> I have to say that the Be API was very nice from a developer's
> perspective; it's one of the things I will miss when Be goes away.
The quick response is that I think it's needlessly pessimistic (and for
me to call pessimism needless is just surreal) to think that if Be goes
away BeOS will automatically go away as well. I expect Be's assets will
be bought and that the buying entity will attempt to continue providing
a platform for 3rd party development. Even if I'm wrong, though, it's
not like the installed copies of BeOS will suddenly stop working. I
know of at least one hardware vendor who delivered a new product for
the Apple II (a video card) within the last few years, and somehow the
Amiga still isn't quite dead yet. If you like writing for BeOS, there's
no technological reason to stop in the near future. If, on the other
hand, the presence of BeOS in markets in which you're interested decays
to the point where you personally don't feel like it's worth your
while, you may want to check out (ducks behind a blast shield) Mac OS
X. The API is conceptually similar and a complete RAD toolchain is part
of the standard distribution. On the other hand, you may not want to
since it costs more than twice as much as BeOS and you may be among the
large group of people who are allergic to anything with Apple's name on
it.
G (Who develops for every platform he can get his hands on.)
Marco Nelissen
> Now that the Be story is almost over, it interesting and instructive
> to look back at the mistakes that Be's management made.
20-20 hindsight, eh?
What's the name of your succesful company again?
Lubris
isra...@optushome.com.au says...
> Now that the Be story is almost over, it interesting and instructive
> to look back at the mistakes that Be's management made.
>
> Here is my list of mistakes that contributed to the death by a
> thousand cuts of Be Corp.
>
> 1. Marketing BeOS as a "media OS"
>
> Sure, it was an interesting niche and it got a bit of free publicity
> because of this. Unfortunately, positioning it as a " media os "
> turned most consumers and programmers off.
> It made people think of high priced ( eg: Silicon Graphics ) computers
> that would be used by only a small group.
So this is how the rewriting of history begins? Let's not forget that Be
is a tiny company by computer industry standards. The fact that Be was
savvy enough to get some press in Byte with 2 informative stories back in
95-96 says a lot about their marketing. They used the Web very well in
these early years with a regular (and well archived) newsletter that
informed and excited many a user and programmer. They did a road show in
late '96 unveiling the port to PowerPC Macs. And then made a successful
conversion to Intel. Their task of converting users to a new OS was not
an easy one. They built a better OS, but even good marketing can't get
some people to change. As for turning off users because of expensive
machines, nothing could be further from reality. Any one who read up on
Be knew that it ran faster on uniprocessor machines than most OSes and
took great advantage of dual processors. If major PC makers had made dual
processors more available, the outcome may have been much different.
The blessing and the curse was that BeOS came around with the explosion
of the Internet. It was a blessing because they were able to reach people
and distribute software more easily, a curse because they couldn't catch
up with all the application "advances" that came about.
> 2. Overpriced Be machines.
> Consumers would have bought a Be as their second or third computers.
> The prices ( who cared if there were two processors ) were way out of
> line with that of commodity Intel PCs.
Before the Intel port, this argument was true. I debated buying a
PowerMac, but couldn't justify the extra expense of the hardware and new
software required. Fortunately, my Dell PC that was purchased before the
BeOS port to Intel ran just fine with its popular, mainstream video,
SCSI, etc.
> 3. Inadequate development environment.
> Some thing like Visual Studio, with a slick IDE and a good GUI
> generator was badly needed.
A small company can't do it all. Metrowerks did a pretty good job, but
they were to small to wait out the market. These are not inexpensive
undertakings!
> (Hindsight is a wonderful thing, blah, blah... but not learning from
> others failures is worse )
Be's recent decision to focus on BeIA has been poor in my opinion, but I
must give them a pass because the economic reality didn't give them the
luxury of waiting much longer. They built a fine OS and they did well by
the community in the earlier years.
This trip down memory lane reminded me how much I enjoyed reading those
Be newsletters. I loved William Adams' creative articles with his
anecdotal leads. And a regular article from the CEO showed that they
realized communication was important. Unfortunately, going public
required disclaimers and sealed lips.
Here's how I see their history being written (but it's not over yet!):
A small company that fought the odds and made a damn good product.
--Lubris
Stephen E. Halpin
> In article <rk0iftc97s4lou6e3...@4ax.com>, raj
> <isra...@optushome.com.au> wrote:
>
> > Now that the Be story is almost over, it interesting and instructive
> > to look back at the mistakes that Be's management made.
>
> First, I don't believe "the Be story is almost over." I think we're
> headed for a very serious plot twist.
Does your crystal ball see OpenGL drivers for the ATI Radeon?
>
>
> > Here is my list of mistakes that contributed to the death by a
> > thousand cuts of Be Corp.
> >
> > 1. Marketing BeOS as a "media OS"
> > Sure, it was an interesting niche and it got a bit of free publicity
> > because of this. Unfortunately, positioning it as a " media os "
> > turned most consumers and programmers off.
> > It made people think of high priced ( eg: Silicon Graphics ) computers
> > that would be used by only a small group.
>
> Actually, you missed the real issue: Repeated refocussing of the target
> market which confused potential customers and alienated multiple
> generations of developers.
Erf. The vision was probably wrong from the start. The BeBox had
some novel features, but the software was pretty coarse. I thought
that the Be web site said they sold something like 1,800 boxes, which
wasn't anywhere near the 3 million they had hoped to get from the
old Amiga tinkering crowd. They lost time and money porting to
the Mac, and they lost time and money porting to Intel. They lost
focus, and became the media OS without drivers for anything media
related (I don't count a Sound Blaster as much of a media device.)
I think when it became apparent that people weren't going to pop
$70 a head for a copy of BeOS, they were in trouble. This is a real
shame, as by the time BeOS turned version 5, they had a hell of a
product, but by then they were the web pad company without a web
pad. I'd love to see a successful BeOS 6, and I wouldn't mind paying
a reasonable price for the privilege.
Ironically, NeXT made most of the same mistakes in the other
direction. They dropped out of hardware, wasted many years
porting rather than improving, kept retargetting their market, but
instead of underpricing their software, they overpriced it (what
was the final OpenStep development kit selling for, $7,000??)
For half a billion dollars Apple bought debt and Steve Jobs, and
for five years didn't have the capabilities that A/UX offered
over a decade ago (UNIX, with an API layer that allowed
Mac GUI applications to run in a protected environment, and
a compatability environment to run native MacOS apps in.)
Supporting multiple platforms is the kiss of death in both these
cases and many more - no one has done it successfully in the long
term. NT dropped MIPS, Alpha and PowerPC, NeXT dropped 68K,
HP and SPARC, Sun has half dropped IA32 and never had any real
sales there anyway, the VAX died faster than anyone predicted when
Alpha came out, and yet Be held on to the PowerPC even after
you couldn't buy a new PowerPC system anywhere to run it on.
When all was said and done, the PC was the only system
with an actual architecture, which is why I can run NeXTstep
on a new PC or BeOS 5 on an old one. PowerPC was so fickle
that the only development machine supported by the first beta
of Rhapsody was unsupported by the second beta. SPARC and
HP changed so fast that NeXT only saw one or two releases
for those platforms before dropping the effort entirely.
>
>
> > 2. Overpriced Be machines.
> > Consumers would have bought a Be as their second or third computers.
> > The prices ( who cared if there were two processors ) were way out of
> > line with that of commodity Intel PCs.
Be was never a consumer brand. They wanted the tinkerer crowd, and
had they had more sooner, they may have really taken that world.
>
>
> I disagree. Actually, I think the elimination of Be-branded hardware
> was a major mis-step. The price of the 'Box as offered to developers -
> the only people who could get one - was _extremely_ good given what it
> included. You can't make a fair comparison by putting it next to
> contemporary baseline x86 boxes, let alone what's out now.
>
> > 3. Inadequate development environment.
> > Some thing like Visual Studio, with a slick IDE and a good GUI
> > generator was badly needed.
I think the GUI stuff isn't that big a deal.. The bigger problem had to do
with
the fact that you really needed to do serious multithreading to get the
most
from the system, which made it difficult to bring single threaded stuff
from
other parts of the world and really make it shine. In a competative
environment
a few good apps could have taken them to the mainstream, but that's no
longer
possible. Here's to endless expensive subscriptions where your computer
has
to be tied to the mothership in Redmond before you can use it.
>
>
> I disagree here, too. Even without a traditional RAD tool I was getting
> shorter development cycles out of the BeIDE than I do out of Delphi -
> and I've been using Delphi longer. Very few of the first (or even
> second) generation of BeOS developers even made a peep about the lack
> of such tools, and those that did were motivated to address said lack -
> while they were motivated (see my response to #1).
>
> G
-Steve
Peter Stegemann
> You're in the minority here (even though I agree with your analysis).
> Most developers for the BeOS were and continue to be hobbyists, and
> these folks want pretty GUI builders and RAD tools. There were a few
So you have asked all those developers? Or how
can you know that? Until know, most people I
met who _needed_ such tools for development
spend more time with reading books about
developing software than actually doing it. But
for sure, that's not objective, just my personal
impression.
Peter Stegemann
> > > Now that the Be story is almost over, it interesting and instructive
> > > to look back at the mistakes that Be's management made.
> > First, I don't believe "the Be story is almost over." I think we're
> > headed for a very serious plot twist.
That reminds me on people who still state that BeOS
desktop development is all fine as "BeOS is the
development-platform of BeIA and will benefit a lot
from it". :-)
On the other hand, I say it's dead when it's dead.
We're as close to a dead BeOS as we've been the last
2 years. It's just more obvious nowadays.
> Erf. The vision was probably wrong from the start. The BeBox had
> some novel features, but the software was pretty coarse. I thought
> that the Be web site said they sold something like 1,800 boxes, which
> wasn't anywhere near the 3 million they had hoped to get from the
> old Amiga tinkering crowd. They lost time and money porting to
When Be started selling the BeBox, Amiga-Users
weren't an issue. It was a surprise for Be to
attract so much of us old Amigans.
> the Mac, and they lost time and money porting to Intel. They lost
Porting to Intel was a gain. A problem was just
the way it was done. As always, Be lost great
parts of it's community and this scared a lot
of potential new users and developers. It was
exactly like I said when they switched from the
BeBox to PPC (and btw, end of the year we drop
the support. Bye bye.): Which part of the
community will they drop next? They always failed
to install a trustful environment where people
and companies were willing to invest money in.
raj
<gwesto...@CAPShome.com> wrote:
>First, I don't believe "the Be story is almost over." I think we're
>headed for a very serious plot twist.
Aah !
The eternal optimist.
As the local rednecks say, " Good onya mate !" :-)
>Actually, you missed the real issue: Repeated refocussing of the target
>market which confused potential customers and alienated multiple
>generations of developers.
Yes, refocussing was a very real problem towards the end.
However, that was Gassy's (sic ! ) not-so-bright response to the
already apparent failure of the BeOS.
>> 2. Overpriced Be machines.
>> Consumers would have bought a Be as their second or third computers.
>> The prices ( who cared if there were two processors ) were way out of
>> line with that of commodity Intel PCs.
>
>I disagree.
The point I make is that Joe Normal and Jane Normal do not understand
or care what a PC can do. They just want something to type on, use the
internet on and see the kids play games on.
The Normals are also very price sensitive.
>
>> 3. Inadequate development environment.
>I disagree here, too. Even without a traditional RAD tool I was getting
>shorter development cycles out of the BeIDE than I do out of Delphi -
Well, it certainly did not attract the developers that mattered, did
it ? To not put too fine a point on it , did the software you wrote
cause significant numbers of new customers to adopt BeOS ?
I rather suspect not...
No, the reality is that only a few shareware and freeware authors to
whom programming is more a hobby than a full time job were happy with
the pathetic environment and the limited selection of programming
languages on the Be.
We all know that committed hobbyists will put up any inconvenience.
(Just look at the Amiga users and Triumph riders still around ! )
raj
<gwesto...@CAPShome.com> wrote:
>while I _do_ expect the end of Be as an independent entity in the
>relatively near future, I _don't_ think it's the end of the legacy,
>such as it is, or the products.
Yeah sure, just as the legacy of Minix lives on in Linux and the
legacy of Multics lives on in SCOMP....
But do the great unwashed know or care ?
>Be was selling the 'Box for too little in an attempt to build mindshare
>among developers. It worked. The first generation of BeOS developers
>was active and activIST.
They also failed to produce any software that penetrated the
mainstream markets.When was Be software last sold in Target or KMart ?
( only partially joking )
raj
>Let's not forget that Be
>is a tiny company by computer industry standards.
Be is now a dead company by computer industry standards.
>The fact that Be was
>savvy enough to get some press in Byte with 2 informative stories back in
>95-96 says a lot about their marketing.
They also failed to follow up on this.
Be eventually became classified in the popular mind with the Amiga and
Taligent.
>The blessing and the curse was that BeOS came around with the explosion
>of the Internet. It was a blessing because they were able to reach people
>and distribute software more easily, a curse because they couldn't catch
>up with all the application "advances" that came about.
Translation, they were not savvy enough to use the internet to market
their products successfully.
>> 2. Overpriced Be machines.
>Before the Intel port, this argument was true. I debated buying a
>PowerMac
>> 3. Inadequate development environment.
>A small company can't do it all. Metrowerks did a pretty good job, but
>they were to small to wait out the market.
When I last checked, Metrowerks were still a division of Motorola.
Sure Motrola is hardly IBM, but their revenues last year were in
excess of 30 billion dollars.
>Here's how I see their history being written (but it's not over yet!):
>A small company that fought the odds and made a damn good product.
A small company that tried ................ and died......
A sad requiem.
Greg Weston
<isra...@optushome.com.au> wrote:
> They also failed to produce any software that penetrated the
> mainstream markets.When was Be software last sold in Target or KMart ?
> ( only partially joking )
I don't know. Neither of those stores is around here. I _have_ seen
BeOS software in national retailers, though.
raj
>weren't an issue. It was a surprise for Be to
>attract so much of us old Amigans.
The kiss of death !
That should have warned them ! :-)
Greg Weston
<isra...@optushome.com.au> wrote:
> On Wed, 09 May 2001 13:04:57 GMT, Greg Weston
> <gwesto...@CAPShome.com> wrote:
>
> >First, I don't believe "the Be story is almost over." I think we're
> >headed for a very serious plot twist.
>
> Aah !
> The eternal optimist.
[Snort] Eternal optimist? You don't know me very well, do you?
Rhetorical question, of course, but to illustrate, a friend in college
once asked me how I'd characterize my outlook. I was just about to say
I considered myself a realist when my brain popped up and said: 'So be
realistic then.' I answered: "Pessimism." What I was trying to indicate
is that the end of Be the independent company, when/if it happens (and
I think that's "when"), does not inherently signal the end of Be's
product line. The story is not necessarily almost over; I just expect
the death of a major character.
> >Actually, you missed the real issue: Repeated refocussing of the target
> >market which confused potential customers and alienated multiple
> >generations of developers.
>
> Yes, refocussing was a very real problem towards the end.
> However, that was Gassy's (sic ! ) not-so-bright response to the
> already apparent failure of the BeOS.
Uh-uh. The refocussing I'm discussing happened like clockwork virtually
every 6 months after I formed an association with Be (which was before
the announcement of the 2x133 BeBox). BeOS didn't even exist as a
distinct product (or product name) at that point. Far from impending
failure, they were well into an upswing at the point of the first major
refocussing.
> >> 2. Overpriced Be machines.
> >> Consumers would have bought a Be as their second or third computers.
> >> The prices were way out of line with that of commodity Intel PCs.
> >
> >I disagree.
>
> The point I make is that Joe Normal and Jane Normal do not understand
> or care what a PC can do. They just want something to type on, use the
> internet on and see the kids play games on.
> The Normals are also very price sensitive.
Tough.
a) I never really expected BeOS to be attractive to the mainstream
desktop user.
b) It's completely irrational to categorize as "overpriced" a machine
that was being sold at a loss, regardless of the mainstream's inability
to differentiate between price and value.
> >> 3. Inadequate development environment.
> >I disagree here, too. Even without a traditional RAD tool I was getting
> >shorter development cycles out of the BeIDE than I do out of Delphi -
>
> Well, it certainly did not attract the developers that mattered, did
> it ?
Some, yes. The problem is that Be's focus changed in intervals shorter
than the development cycle of the mainstream tractor apps. You seem to
have ignored the impact of my response to point 1. Be got into the
habit of regularly doing things that happened to beat down the
confidence of a large chunk of their developer base. Given the rapid
changes, there was certainly incentive for the major shops to adopt a
wait and see attitude. (And the emergence of BeOS/x86 immediately
eliminated any chance of any MS product ever shipping for the OS...MS
doesn't support any OS that runs on their principal hardware base,
although they're not above announcing the intent to do so.)
> To not put too fine a point on it, did the software you wrote
> cause significant numbers of new customers to adopt BeOS ?
> I rather suspect not...
Yes, it _is_ rather a dull point. The answer is: I don't know. You
don't either. I _did_ get a lot of feedback and a few positive reviews
... even from media that weren't Be-centric. So, in fact, I may very
well have contributed noticably to what adoption rate Be had.
> No, the reality is that only a few shareware and freeware authors to
> whom programming is more a hobby than a full time job were happy with
> the pathetic environment and the limited selection of programming
> languages on the Be.
A) This is false, and you can't possibly back it up. I can disprove it
right now: I am a full-time programmer. I was completely satisfied with
the IDE included with BeOS.
B) If you consider the BeIDE to be pathetic, that says a lot about how
little you know about judging toolchains. CodeWarrior on Mac OS and
BeOS is one of the best environments I've used on any of the dozen
platforms for which I have written. (Yes: dozen.) Laying out UIs in
code is less of a deterrent for the serious developer than it is for
the hobbyist. (Read that sentence again.) Most of the large shops don't
use RAD tools for production code when they exist, ergo their
nonexistence is not a problem.
G
Lubris
isra...@optushome.com.au says...
> On Wed, 9 May 2001 22:20:58 -0400, lub...@hotmail.com (Lubris) wrote:
>
> >> 3. Inadequate development environment.
> >A small company can't do it all. Metrowerks did a pretty good job, but
> >they were to small to wait out the market.
>
> When I last checked, Metrowerks were still a division of Motorola.
> Sure Motrola is hardly IBM, but their revenues last year were in
> excess of 30 billion dollars.
When did you first check? By the time Motorola acquired Metrowerks in
late '99, Metrowerks had already shifted their focus from BeOS.
> >Here's how I see their history being written (but it's not over yet!):
> >A small company that fought the odds and made a damn good product.
>
> A small company that tried ................ and died......
> A sad requiem.
Did you lose money on their stock or something?
Peter Moore
peter
cdws
http://www.cdws.com.au/
raj
>> A sad requiem.
>
>Did you lose money on their stock or something?
No .
However, if I had let some Be fanatics I know persuade me into buying
their stock, I would have.
I know of at least one person who bought Be at over 20$ a share and
is now sitting on a pile of worthless paper.
Stephen E. Halpin
> "Stephen E. Halpin" wrote:
> > Greg Weston wrote:
> > > <isra...@optushome.com.au> wrote:
>
> > > > Now that the Be story is almost over, it interesting and instructive
> > > > to look back at the mistakes that Be's management made.
> > > First, I don't believe "the Be story is almost over." I think we're
> > > headed for a very serious plot twist.
>
> That reminds me on people who still state that BeOS
> desktop development is all fine as "BeOS is the
> development-platform of BeIA and will benefit a lot
> from it". :-)
>
> On the other hand, I say it's dead when it's dead.
> We're as close to a dead BeOS as we've been the last
> 2 years. It's just more obvious nowadays.
>
> > Erf. The vision was probably wrong from the start. The BeBox had
> > some novel features, but the software was pretty coarse. I thought
> > that the Be web site said they sold something like 1,800 boxes, which
> > wasn't anywhere near the 3 million they had hoped to get from the
> > old Amiga tinkering crowd. They lost time and money porting to
>
> When Be started selling the BeBox, Amiga-Users
> weren't an issue. It was a surprise for Be to
> attract so much of us old Amigans.
I have this recollection of reading one of the early interviews with
JLG where he said he was pitching the product towards the Amiga
crowd. Unfortunately I don't even know where to begin looking
for that article...
>
>
> > the Mac, and they lost time and money porting to Intel. They lost
>
> Porting to Intel was a gain. A problem was just
> the way it was done. As always, Be lost great
> parts of it's community and this scared a lot
> of potential new users and developers. It was
> exactly like I said when they switched from the
> BeBox to PPC (and btw, end of the year we drop
> the support. Bye bye.): Which part of the
> community will they drop next? They always failed
> to install a trustful environment where people
> and companies were willing to invest money in.
The problems were far deeper than the few who couldn't let go
of one bad CPU architecture for another. I still haven't found a
single thing that BeOS can do that I can't do in a Windows
environment, and I had a very long list of things that BeOS
can't do that Windows can. Time spent porting is time that
you stay still while your competition advances.
Going to Intel may have been the only choice at the time as it was
the only way they could stay in the game, but as with some games,
you can make a move in the middle of the game which insures that
all remaining paths to the end will lead to loss. Without something
to clearly differentiate BeOS from other desktop platforms, and
a way to maintain that differentiation over time, the laws of
business say they were highly likely to loose the game.
-Steve
Nate LaCourse
> that I can't do in a Windows environment
It's not what BeOS can do, it's how it does it.
Have fun,
Nate
--
Dig: http://cibo.dhs.org/
Talk: na...@REMOVEcibo.ALLdhs.CAPSorg
--
Stephen E. Halpin
> > I still haven't found a single thing that BeOS can do
> > that I can't do in a Windows environment
>
> It's not what BeOS can do, it's how it does it.
People don't spend thousands of dollars on computers and software
to admire how a computer does it's thing. They spend this money to
get a job done, which for 99% of the desktop users out there is doing
something other than simply using a computer. The continued success
of Windows and MacOS says clearly that protected memory,
preemptive scheduling, etc... is not a primary concern for people.
The lack of drivers and distinguishing user level features make it
difficult for application writers to even match the capabilities of
other platforms, which leads to people continuing to buy and use
Windows and MacOS machines (heck, I like BeOS, but I don't
do anything other than program on it, because I can do more stuff
with less effort on other platforms.)
-Steve
Greg Weston
<s...@quadrizen.com> wrote:
> Nate LaCourse wrote:
>
> > > I still haven't found a single thing that BeOS can do
> > > that I can't do in a Windows environment
> >
> > It's not what BeOS can do, it's how it does it.
>
> People don't spend thousands of dollars on computers and software
> to admire how a computer does it's thing.
Developers do, because for them _how_ a system does its thing can often
directly translate into how well they can get their job done. I am far
more experienced on Windows than on BeOS, just as a reality of how long
I have been working with each (and how long I could have been).
Regardless, how BeOS does some of the things it does generally means
that my development cycles on BeOS are shorter.
> (heck, I like BeOS, but I don't
> do anything other than program on it, because I can do more stuff
> with less effort on other platforms.)
See? The problem I see is that other things that were going on with the
platform caused sufficient concern among the developers who could have
and would have delivered the products that would make BeOS compelling
that they decided to either hold off or abandon the platform entirely.
As a developer, I found Be the OS to be wonderful to deal with, but Be
the company to be frequently frustrating.
G
Peter Stegemann
> I have this recollection of reading one of the early interviews with
> JLG where he said he was pitching the product towards the Amiga
> crowd. Unfortunately I don't even know where to begin looking
> for that article...
Be also states that BeOS was designed as a MediaOS
from the beginning on. Sadly enough, I've lost the
mirrors of the early www.be.com, where they clearly
said that they've just build that cool OS and
machine and let the developers decide where to go.
raj
BeOS Stock closed at 0.53.
Volume 139,681
http://finance.yahoo.com/q?s=BEOS&d=c&k=c1&a=v&p=s&t=3m&l=on&z=m&q=l
Marco Nelissen
> Volume 139,681
Wrong numbers, mate. Thanks for trying though.