Why has the Metrowerks sign been taken down?
Peter Lui
they are moving, Freescale has taken down the large Metrowerks sign at
its Parmer HQ.
The Metrowerks website now seems to indicate that Metrowerks has been
absorbed into Freescale as a "team" and no longer exists as a separate
company:
http://www.metrowerks.com/MW/About/default.htm
> Freescale Semiconductor’s Metrowerks organization is a silicon
> enablement team that helps customers experience and fully leverage the
> performance of Freescale products. The organization’s embedded
> development leadership, technology and talent are focused on driving
> success for Freescale and its customers.
Silicon enablement team? That sounds like the company has folded to me
and they are simply keeping the name around until they can sell of the
rest of the product lines. I'm starting to wonder if Metrowerks is
simply MWRon and some lonely guy burning CDs.
I also noticed that Metrowerks press releases are now making no mention
of Metrowerks as a company at the bottom.
Arthur Langereis
> Silicon enablement team? That sounds like the company has folded to
> me and they are simply keeping the name around until they can sell of
> the rest of the product lines.
Along the same lines, I guess that programmers could be called Computer
Enablers then. I'll have someone modify my business cards for me.
> I'm starting to wonder if Metrowerks is simply MWRon and some lonely
> guy burning CDs.
Hmm, if a new programmer tools company was formed by Isaac,
Howard and Ron (along with some usability geeks (and the CD burner guy))
I'd be first in line to buy when they ship something. :)
Maybe if they'd team up with BareBones we'd have a full IDE in .. Fall 2006!
You read it here first ;)
- Arthur
Alwyn
Peter Lui <pelui_...@comcast.net> wrote:
> It appears that Freescale has other intentions for Metrowerks. Unless
> they are moving, Freescale has taken down the large Metrowerks sign at
> its Parmer HQ.
In the words of Bruce Chizen, why did it take them so long. :-)
> The Metrowerks website now seems to indicate that Metrowerks has been
> absorbed into Freescale as a "team" and no longer exists as a separate
> company:
>
> http://www.metrowerks.com/MW/About/default.htm
>
> > Freescale Semiconductor’s Metrowerks organization is a silicon
> > enablement team that helps customers experience and fully leverage the
> > performance of Freescale products. The organization’s embedded
> > development leadership, technology and talent are focused on driving
> > success for Freescale and its customers.
>
> Silicon enablement team?
Yes, Freescale seems to like the term 'silicon enablement'. See:
<http://investors.freescale.com/phoenix.zhtml?c=175261&p=irol-newsArticle
&ID=682162&highlight=>
> That sounds like the company has folded to me
> and they are simply keeping the name around until they can sell of the
> rest of the product lines. I'm starting to wonder if Metrowerks is
> simply MWRon and some lonely guy burning CDs.
For members of this group, the second paragraph of the 'About' page is
just as interesting:
> Metrowerks'award-winning line of CodeWarrior products and close
> collaboration with key Freescale Ecosystem players help maintain
> a continuum of world-class embedded development technologies for
> Freescale customers and prospects.
The Mac business is so insignificant these days, it's not worth a
mention when Freescale tell the world what Metrowerks is about. It
obviously doesn't fit in with their way of thinking.
> I also noticed that Metrowerks press releases are now making no mention
> of Metrowerks as a company at the bottom.
Yes, I believe that applies to all press releases emitted this year.
Alwyn
jonh...@mac.com
>> mention when Freescale tell the world what Metrowerks is about.
Gee, I wonder who's fault that is. It's a shame, a mere 2-3 years ago,
CodeWarrior was a virtual monopoly on the Mac. ProjectBuilder was a
blip on the radar screen. But Metrowerks really screwed the pooch
these past two years, and so Xcode has essentially taken over.
You can't really blame CodeWarrior supporters, as we have pleaded with
Metrowerks no to proceed with a market-killing move, like refusing to
show up at WWDC and dropping x86 support. We watched as they commit
the business equivalent to hari-kari, all the time them saying, "We
know what we're doing."
A company that started out so cool, is now run by a bunch of complete
and utter morons. I wish we had Greg Galanos back at the helm.
Jonathan Hoyle
Eastman Kodak
Alwyn
"jonh...@mac.com" <jonh...@mac.com> wrote:
> I wish we had Greg Galanos back at the helm.
It's my understanding that Mr Galanos & Co. were having difficulty
keeping the company profitable. I've no doubt they were mightily
relieved when it was taken over by Motorola.
But I believe you're right: the fate of CodeWarrior for Macintosh was
sealed in August 1999. It's been in a long, slow decline since then.
Unfortunately, it is also a technology they could not easily have sold,
as Freescale did with the x86 stuff. So it's my guess that they will
hold on to it as long as they can derive revenue from it, but without
making significant investments in it. When the revenue dries up
completely, of course they will let it go.
Alwyn
MW Ron
Peter Lui <pelui_...@comcast.net> wrote:
>It appears that Freescale has other intentions for Metrowerks. Unless
>they are moving, Freescale has taken down the large Metrowerks sign at
>its Parmer HQ.
Metrowerks was seen as a competitor to other compiler and tool vendors
that supplied Freescale, instead we wish to enable Freescale silicon
sales as CodeWarrior tools and in partnership with 3rd parties.
Some products, Games, Desktop and Palm will continue to be sold under
the Metrowerks name
>The Metrowerks website now seems to indicate that Metrowerks has been
>absorbed into Freescale as a "team" and no longer exists as a separate
>company:
>
>http://www.metrowerks.com/MW/About/default.htm
look at http://www.codewarrior.com
>Silicon enablement team? That sounds like the company has folded to me
>and they are simply keeping the name around until they can sell of the
>rest of the product lines. I'm starting to wonder if Metrowerks is
>simply MWRon and some lonely guy burning CDs.
No, it is a healthy organization.
>I also noticed that Metrowerks press releases are now making no mention
>of Metrowerks as a company at the bottom.
My understand is that some will and some won't
Ron
--
CodeWarrior Community Forum is a free online resource for developers
to discuss CodeWarrior topics with other users and our staff
-- http://www.codewarrior.com/community --
Ron Liechty - MW...@metrowerks.com - http://www.codewarrior.com
Peter Lui
> Metrowerks was seen as a competitor to other compiler and tool vendors
> that supplied Freescale, instead we wish to enable Freescale silicon
> sales as CodeWarrior tools and in partnership with 3rd parties.
>
So in 1998, MW was a competitor to those vendors, then Motorola bought
MW and it was still a competitior, then Motorola spun off the entire
semiconductor business and it was still a competitor and now 6 long
years later MW is being replaced with a Freescale branded version that
will not compete?
If Freescale can't make fast PowerPC processors I guess it's too much
to hope for that they have any idea what software development tools are
about and how to sell them. Forrest Gump must have been made CEO of
Freescale a long time ago.
Some products, Games, Desktop and Palm will continue to be sold under
the Metrowerks name
>
I can't beleive anyone buys Palm tools from Metrowerks anymore. This is
another prime example of how a cozy relationship with the OS vendor
went sour and the technology transition to ARM left Metrowerks behind.
Metrowerks seems to be unwilling to invest in themselves and instead
prefers to milk the efforts of the previous decade.
look at http://www.codewarrior.com
>
So Metrowerks as a company is now dead and exists only as a brand
within Freescale.
I think this sums it up pretty well:
Peter Lui
> A company that started out so cool, is now run by a bunch of complete
> and utter morons. I wish we had Greg Galanos back at the helm.
Galanos and Belanger weren't so smart becaue their biggest mistake was
hiring those idiots in the first place and leaving them in charge.
These are the boobs who were left to ruin the company:
http://web.archive.org/web/20010124062100/www.metrowerks.com/about/officers/
Look at their backgrounds: their only business experience is Metrowerks
itself and they only have bachelor degrees. Now fast forward to 2004:
http://web.archive.org/web/20021017223744/www.metrowerks.com/MW/About/bios.htm?cookie_test=1
Same
people, same experience except a couple of them don't even seem to have
an education background worth mentioning. I believe 8 of those 10
people are no longer with Metrowerks perhaps all of them. What company
can survive losing it's entire management team?
Maybe for Metrowerks this is for the best.
la...@skytag.com
is, if they have a sign up, or if CW for Palm is a thriving product. My
only interest is in CW for Mac development and all indications are that
it's a dead product (as opposed to a discontinued product).
I think Motorola/Freescale figured out a long time ago that there was
an expiration date on selling processors for use in Macs, and after
that the death of CW for Mac was inevitable. I remember when CW *was*
CW for Mac development. Now you have to hunt for that product on a page
full of products for embedded systems I don't even recognize. They
didn't sell off their x86 technology because they were stupid, they did
it because they have no use for it in their long-term product plan.
As always, the only *bad* thing in all this is the degree to which
MW/FS keeps spinning out marketing jargon in lieu of providing any real
information. They are clearly going to milk us as long as enough of us
pay them to keep CW limping along for PPC development. Once that is no
longer true and the current release of CW won't work with the current
release of Mac OS X, they'll make some announcement about how in spite
of grand efforts to keep the product alive, further development became
impractical, blah blah blah, goodbye and thanks for all the fish.
Larry
MW Ron
Peter Lui <pelui_...@comcast.net> wrote:
>On 2005-06-23 08:44:52 -0700, MW Ron <mw...@metrowerks.com> said:
>
>> Metrowerks was seen as a competitor to other compiler and tool vendors
>> that supplied Freescale, instead we wish to enable Freescale silicon
>> sales as CodeWarrior tools and in partnership with 3rd parties.
>>
>So in 1998, MW was a competitor to those vendors, then Motorola bought
>MW and it was still a competitior, then Motorola spun off the entire
>semiconductor business and it was still a competitor and now 6 long
>years later MW is being replaced with a Freescale branded version that
>will not compete?
Yeah that is pretty much the problem. We tried to do all brands of
chips we were competing with other 3rd parties with FSL chips
>If Freescale can't make fast PowerPC processors I guess it's too much
>to hope for that they have any idea what software development tools are
>about and how to sell them. Forrest Gump must have been made CEO of
>Freescale a long time ago.
PC Chips were 3% of Freescale's business but you should get quotes
directly from their press releases.
>So Metrowerks as a company is now dead and exists only as a brand
>within Freescale.
I don't think so not legally yet. Actually there is much more to what
is going on and I think it will be good news for Mac developers.
jonh...@mac.com
>> and I think it will be good news for Mac developers.
Just when I am depressed thinking it's over and it's time to throw in
the towel, MWRon makes a remark like that to give us hope. Well,
here's hoping for great news this summer. :-)
Ron, do you think that you will be able to reveal these plans when you
are at MacHack? I would urge you to remind the higher ups that MacHack
would be the ideal place to make such announcements (as WWDC 2006 would
be a bit late in the game for it). MacHack in particular tends to
appeal to the old guard, who are much more likely to be CodeWarrior
users than newbies. An inspiring presentation on what's to come would
be helpful to keep us on between now and when you release CW 10.
Thanks,
Jonathan Hoyle
Eastman Kodak
MW Ron
"jonh...@mac.com" <jonh...@mac.com> wrote:
I go on Vacation for 2 weeks July first and I'm hoping for a roadmap
before I go on vacation. I imagine if that isn't ready they may wait
to make a splash at AdHoc.
Peter Lui
> What's the point of this thread? I couldn't care less what their name
> is, if they have a sign up, or if CW for Palm is a thriving product. My
> only interest is in CW for Mac development and all indications are that
> it's a dead product (as opposed to a discontinued product).
The point is simply that the writing has been on the wall for sometime
and now the wall is finally being taken down.
I think Motorola/Freescale figured out a long time ago that there was
> an expiration date on selling processors for use in Macs,
>
After the PowerPC 970 was announced and Motorola's competitive roadmap
was blank I would agree but as the chip vendor isn't that up to you?
Have you ever seen this poster:
http://web.archive.org/web/20000815081701/www.macedition.com/images/wanted/wantedcolor.jpg
Freescale
set it's own expiration date. Motorola threw it away. Apple made other plans.
and after
> that the death of CW for Mac was inevitable.
>
Inevitable because they were managed by people cutting their teeth for
the first time and used a broken magic-8 ball whose only advice when
shaken was to run as large a company as possible for as long as you can
so your resume looks good. The real reason that the product was doomed:
a FAILED merger. It happens.
Check out the exec bios I found after much websearching:
http://www.wirelessvalley.com/News/PressReleases/PressRelease012604a.asp
http://www.freescale.com/webapp/sps/site/overview.jsp?nodeId=02VS0llCc5pzMPYZjg19507898
http://www.mobiusvc.com/pages.php?pn=team&sub=ggalanos-bio
http://www.reddwerks.com/r_compmanag.htm
Notice
how all of them make the same claim: they helped Motorola buy MW and
they grew the company. None of the Metrowerks leaders ever mention
Apple or Macintosh. They are embarrassed by it although it's what made
them wealthy.
Over 600 people worked at Metrowerks and they couldn't keep pace with
Apple? How big is the XCode team? I bet it's not 600 people. How big
was the original team that wrote CodeWarrior. Maybe 15 people?
Metrowerks management didn't want to make the product, it's that simple.
They
> didn't sell off their x86 technology because they were stupid, they did
> it because they have no use for it in their long-term product plan.
>
Stupid no, bad timing yes.
As always, the only *bad* thing in all this is the degree to which
> MW/FS keeps spinning out marketing jargon in lieu of providing any real
> information. They are clearly going to milk us as long as enough of us
> pay them to keep CW limping along for PPC development. Once that is no
> longer true and the current release of CW won't work with the current
> release of Mac OS X, they'll make some announcement about how in spite
> of grand efforts to keep the product alive, further development became
> impractical, blah blah blah, goodbye and thanks for all the fish.
>
I don't think they are that honorable. MW will simply keep pressing CDs
until there is nobody there who remembers how to make one that installs
on the current OS.
Unfortunately history repeats itself all the time:
la...@skytag.com
>> and I think it will be good news for Mac developers.
> Just when I am depressed thinking it's over and it's time to throw in
> the towel, MWRon makes a remark like that to give us hope. Well,
> here's hoping for great news this summer. :-)
Ron's been making remarks like this for as long as I can remember. I
stopped finding hope in them a long time ago. When they actually
deliver something, then maybe I'll have hope.
Larry
Don Bruder
la...@skytag.com wrote:
Where I come from, hope - and a buck and a quarter - will buy you a cup
of coffee.
With the way Ron's been waffling, I'm closing the book on
Metrowerks/CodeWarrior. It's been a fun ride, but it's becoming more
obvious with every "Don't touch that dial, folks! It'll be just another
couple commercials before we show you the way things end. Honest! No,
this time we really mean it!" message Ron posts that what's going on is
the credits are rolling, but they're hoping somebody buys another tub of
popcorn before they head for home.
Rather sad, but... <shrug> Hardly unexpected.
Sorry, Ron, but you (and Metrowerks/CodeWarrior) have lost all
credibility in my eyes, and with every "I promise I'll get you a
roadmap" post, you sink even deeper. It actually would have been better,
at least to my point of view, if you'd've just shut up and let CW die
while it still had a shred or two of dignity left to wrap itself in,
instead of trying to bullshit the troops the way you've been doing. As
it is, you've made yourself a laughingstock by trying to shill for an
outfit that's obviously in the final stages of sucking the pipe.
Been nice knowin' ya - I'm outta here. Looks like it's Xcode for me,
whether I like it or not.
--
Don Bruder - dak...@sonic.net - New Email policy in effect as of Feb. 21, 2004.
Short form: I'm trashing EVERY E-mail that doesn't contain a password in the
subject unless it comes from a "whitelisted" (pre-approved by me) address.
See <http://www.sonic.net/~dakidd/main/contact.html> for full details.
Christopher J. Henrich
Let's put it this way: wheb Ron actually gets one of those roadmaps,
that's when we can really be hopeful.
--
Chris Henrich
http://www.mathinteract.com
God just doesn't fit inside a single religion.
Bruce Wheaton
1119555402.4...@g44g2000cwa.googlegroups.com, "la...@skytag.com"
<la...@skytag.com> wrote:
> I think Motorola/Freescale figured out a long time ago that there was
> an expiration date on selling processors for use in Macs, and after
> that the death of CW for Mac was inevitable. I remember when CW *was*
> CW for Mac development. Now you have to hunt for that product on a page
> full of products for embedded systems I don't even recognize. They
> didn't sell off their x86 technology because they were stupid, they did
> it because they have no use for it in their long-term product plan.
It occurs to me that Nokia also saw the writing on the wall, and presumably
being Codewarrior users themselves, bought the x86 compilation stuff so that
it wouldn't sink along with Metrowerks (or at least with the MW desktop
tools dept).
That's the real writing on the wall. After all, Nokia aren't going to start
selling IDEs, are they? And if they believed CW for desktops was going to
keep going, surely they'd just buy site licenses. It smells like a safety
move on their part.
Bruce Wheaton
Alwyn
Bruce Wheaton <br...@spearmorgan.com> wrote:
>
> After all, Nokia aren't going to start selling IDEs, are they?
Nokia have been selling Codewarrior Development Studio for Symbian OS at
least since February this year.
<http://press.nokia.com/PR/200502/978604_5.html>
<http://www.forum.nokia.com/main/1,6566,1_74,00.html>
Alwyn
Alwyn
Peter Lui <pelui_...@comcast.net> wrote:
>
> Notice
>
> how all of them make the same claim: they helped Motorola buy MW and
> they grew the company. None of the Metrowerks leaders ever mention
> Apple or Macintosh. They are embarrassed by it although it's what made
> them wealthy.
This is particularly striking in the case of Mr Bélanger, I think:
> Previously, he was Chairman/CEO of Metrowerks (NASDAQ) and helped grow the
> company from three people in 1991 to 250 in 1999. Metrowerks CodeWarrior won
> Software Product of the Year/1994 over [sic] 3,500 applications.
<http://www.reddwerks.com/r_compmanag.htm>
As everybody knows, the Apple Macintosh is a toy system, so he does not
care to have his involvement with it mentioned on his serious business
profile. Yet, as you say, it's what made his career and his fortune.
> He won a World Class Outstanding Achievement Award/1997 and was finalist for
> the Ernst & Young Entrepreneur-of-the-Year Award-Austin/96. Motorola SPS
> purchased Metrowerks in 1999 for a total of $100M...
at the very height of the dot-com boom. Lucky, lucky Mr Bélanger!
Alwyn
Peter Lui
> It occurs to me that Nokia also saw the writing on the wall, and presumably
> being Codewarrior users themselves, bought the x86 compilation stuff so that
> it wouldn't sink along with Metrowerks (or at least with the MW desktop
> tools dept).
No doubt.
That's the real writing on the wall. After all, Nokia aren't going to start
> selling IDEs, are they? And if they believed CW for desktops was going to
> keep going, surely they'd just buy site licenses. It smells like a safety
> move on their part.
>
Metrowerks was essentially a large software contractor. Rarely did they
develop anything without the financial backing of a large company. Be
it Apple, Sun, Sony, Sega, Nintendo, Symbian (Nokia et al), Motorola,
Be, Palm and the list goes on.
In every case, those companies bankrolled or subsidized the development
and when they didn't like what was produced, they pulled out but
Metrowerks continued to mislead the customer there was a product and
support to squeeze every last nickel out of them.
In Nokia's case, there was probably a meeting between the two companies
where Freescale offered their condolences about future development but
Nokia probably had a clause in their contract that allowed them to
acquire the assets and buy time for the developers. I have no idea what
they're actually going to do with it long term.
I'm curious to see what "roadmap" comes out. To date it's been a
Metrowerks vaporware offering. But even if one does appear, who is
going to make it? All of the developers and managers have either, quit,
been fired or moved to other projects. The best they can do at this
point, is open source, have a fire sale or leave up the website of the
dead company for several years like Gibson did when it acquired Opcode
Systems.
la...@skytag.com
> Metrowerks vaporware offering. But even if one does appear, who is
> going to make it? All of the developers and managers have either, quit,
> been fired or moved to other projects.
Yes, that's a huge question. I know at least some of their talent has
moved on, plus they have no working relationship with Apple. Even if
they really wanted to get the Mac IDE back on track, how could they
possibly make that happen? And how can anyone find any hope in anything
Ron says under the circumstances? Who's willing to bet money we'll see
that roadmap Ron has asked for before his vacation? ;-)
Larry
MW Ron
"Christopher J. Henrich" <chen...@monmouth.com> wrote:
>Let's put it this way: wheb Ron actually gets one of those roadmaps,
>that's when we can really be hopeful.
Not only that if it isn't a roadmap with deliverables that are
absolutely reasonable I'll let you know. I will not take anything that
is not acheivable.
MW Ron
>It occurs to me that Nokia also saw the writing on the wall, and presumably
>being Codewarrior users themselves, bought the x86 compilation stuff so that
>it wouldn't sink along with Metrowerks (or at least with the MW desktop
>tools dept).
Nokia has invested a lot into the Symbian OS an open sourced operating
system in the same way that Apple has invested into Mac OS X.
In the same way Apple wanted to control the look and fee and development
of the operating system by having control over the tools Nokia also
wanted to have this control. it was smart of Apple to do so, it was
smart of Nokia to do so.
>That's the real writing on the wall. After all, Nokia aren't going to start
>selling IDEs, are they? And if they believed CW for desktops was going to
>keep going, surely they'd just buy site licenses. It smells like a safety
>move on their part.
Nokia does indeed develop and produce their own Symiban compilers and
they currently license the CodeWarrior IDE. The Symbian OS like Mac
OS X is agnostic about the chip that it runs on and it does run on intel
chips as well as others.
jollyprez
Love that shirt. Still have it. Still wear it on special occasions.
I switched my last two current projects from MW to XCode in the last
two months. Got rid of the last vestiges of PP and support code (though
I'd already gone over to Carbon / NIB in 2003), and use CF for
everything. It's not that bad - at least not for me. And Cocoa rocks.
I'll keep MW to keep some of my legacy apps going (some as old as
1994!), and mull over the future.
'Jolly
Peter Lui
> Yes, that's a huge question. I know at least some of their talent has
> moved on, plus they have no working relationship with Apple. Even if
> they really wanted to get the Mac IDE back on track, how could they
> possibly make that happen? And how can anyone find any hope in anything
> Ron says under the circumstances?
Actually after WWDC the possibility of a solid roadmap has greatly
increased because the road leads to obsolescence.
Metrowerks has ported their technology so many times to stave off death
they are simply unable to make the original version anymore. That's why
MW spent a lot of last year in this newsgroup promising a new IDE that
will never materialize on Mac OS X.
Metrowerks had great tools that kept up with Apple technology even when
there were potholes in the road like Copland (gigantic sinkhole sized
potholes).
Then they fell behind as they snubbed their noses at Mach-O and
modernizing the look/feel of their IDE.
Now they are being left behind by the very same hardware transition
that made them famous.
Ironic.
Alwyn
MW Ron <mw...@metrowerks.com> wrote:
>
> Nokia has invested a lot into the Symbian OS an open sourced operating
> system in the same way that Apple has invested into Mac OS X.
There are certainly open-source projects associated with Symbian, but I
believe the source for the OS itself is closed.
Incidentally, followers of Apple will be aware that Nokia intend to use
the KHTML- and KJS-based WebCore and JavaScriptCore for their
forthcoming browser for their Series 60 mobile phones.
> In the same way Apple wanted to control the look and fee and development
> of the operating system by having control over the tools Nokia also
> wanted to have this control. it was smart of Apple to do so, it was
> smart of Nokia to do so.
So the implication is that it would be smart of Apple to buy CodeWarrior
for Macintosh? Hmm...
> >That's the real writing on the wall. After all, Nokia aren't going to start
> >selling IDEs, are they? And if they believed CW for desktops was going to
> >keep going, surely they'd just buy site licenses. It smells like a safety
> >move on their part.
>
> Nokia does indeed develop and produce their own Symiban compilers and
> they currently license the CodeWarrior IDE. The Symbian OS like Mac
> OS X is agnostic about the chip that it runs on and it does run on intel
> chips as well as others.
The problem here is that I can't see what Metrowerks have that Apple
could possibly want to pay for. They already have Xcode and a complete
set of tools, together with the promise of Intel's x86 compilers, which
are deemed the best in the market. I doubt if Metrowerks' PowerPC
compilers would interest them either, since, unlike IBM's XL range, they
have no PPC970 (G5) support.
Alwyn
Ben Artin
Alwyn <dt015...@mac.com> wrote:
> The problem here is that I can't see what Metrowerks have that Apple
> could possibly want to pay for.
The plc compiler, the STL knowledge (aka Howard) and the C++ compiler knowledge.
MW C++ compiler is still several times faster than gcc.
Ben
--
If this message helped you, consider buying an item
from my wish list: <http://artins.org/ben/wishlist>
I changed my name: <http://periodic-kingdom.org/People/NameChange.php>
Alwyn
Ben Artin <mac...@artins.org> wrote:
> In article <dt015a1979-B5427...@text.news.blueyonder.co.uk>,
> Alwyn <dt015...@mac.com> wrote:
>
> > The problem here is that I can't see what Metrowerks have that Apple
> > could possibly want to pay for.
>
> The plc compiler, the STL knowledge (aka Howard) and the C++ compiler
> knowledge.
> MW C++ compiler is still several times faster than gcc.
That Metrowerks have these assets is indisputable. Whether Apple would
be in a position to acquire them in such a way that they could put them
to good use is another matter.
The message from MW Ron to which I responded is a curious one, coming
from him. Whatever other implications it may have, I find it hard to
interpret it as anything other than a farewell to the Apple platform.
I suppose Metrowerks' departure from the Macintosh development scene
sooner or later is inevitable, but I still find it sad. However, it
would be a nice thought if a genuine transfer of technology could be
effected.
Alwyn
sdfi...@spamcop.net
> Ron's been making remarks like this for as long as I can remember. I
> stopped finding hope in them a long time ago. When they actually
> deliver something, then maybe I'll have hope.
I remember quite a few personal attacks on me when I mentioned this
back in October 2003. Looks like a lot more people are coming around to
this way of thinking. Good.
Right now, I think the only decent product Metrowerks has is their Palm
OS Garnet compilers -- compilers based on the old Mac 680x0 compilers
and hosted only on Windows. It seems roughly half the technology in the
last version is licensed from other companies, and the command line
version of the compilers don't work at all, but since PalmSource is too
busy making Cobalt compilers (that nobody but them cares about),
Metrowerks will probably be able to sell the Palm OS product for a
while.
Peter Lui
> problem here is that I can't see what Metrowerks have that Apple could
> possibly want to pay for.
Now it's way too late but back in 1999 when the future for Mac OS X was
uncertain and CodeWarrior didn't suck ass like it does today, it made
sense.
If Apple bought Power Computing for $400M they could have spared a few
peanuts to buy MW but guess what happened? The greedy management wasn't
going to get rich off any offer from Apple so they roped in Motorola
for $100M then split.
MW could have easily signed a technology transfer agreement anytime
after the buyout but instead they chose to snub Apple and in doing so
all of their customers. Now the tables have turned: MW is desperate to
sell but nobody wants to buy.
la...@skytag.com
>> In the same way Apple wanted to control the look and fee and development
>> of the operating system by having control over the tools Nokia also
>> wanted to have this control. it was smart of Apple to do so, it was
>> smart of Nokia to do so.
>
> So the implication is that it would be smart of Apple to buy CodeWarrior
> for Macintosh? Hmm...
No, no such implication. He's only saying it was smart of Apple to
assume a proactive role in providing a development environment instead
of leaving it to another company. Nothing more and nothing less.
>> The problem here is that I can't see what Metrowerks have that Apple
>> could possibly want to pay for.
>
> The plc compiler, the STL knowledge (aka Howard) and the C++ compiler
> knowledge.
> MW C++ compiler is still several times faster than gcc.
Mac OS 9 is was several times faster than Mac OS X. Apple's solution to
this problem is the same as it was then: They'll work on the speed, and
in the meantime you can buy a faster Mac. Maybe even a cluster you can
use as a server farm. :-)
> That Metrowerks have these assets is indisputable. Whether Apple would
> be in a position to acquire them in such a way that they could put them
> to good use is another matter.
I doubt Apple has any interest in this. They don't have a good
relationship with MW, and what they would get in such an acquisition
would be a bunch of code they didn't write (though some of the people
who wrote it may now be at Apple). When you acquire code from another
company, you have two options:
- Spend a lot of time learning it.
- Use it as is until you junk it because you don't have anyone who
understands it well enough to maintain or improve it.
Of course, I have no inside information about Apple's plans and
intentions, but IMO Apple has too many Unix-heads and gcc is the
open-source, Unix, blah blah blah way, so I just don't see this
happening.
> MW could have easily signed a technology transfer agreement anytime
> after the buyout but instead they chose to snub Apple and in doing so
> all of their customers. Now the tables have turned: MW is desperate to
> sell but nobody wants to buy.
Why do you think they're desperate? I think they don't care. The fact
is that we don't know anything about what MW wants. They may well be
more than happy to let the Mac market go. Everything may very well be
going according to their plans. *We* tend to think of things
differently, but then Mac development is very significant to us.
MW-like Freescale-is in a very different boat. Freescale stated
that the G4 they sell to Apple represents 3% of their business. No
large corporation is going to lose sleep over a niche product that
represents 3% of its market. At this point I doubt CW for Mac
development represents over 3% of MW's business.
Larry
la...@skytag.com
> they are simply unable to make the original version anymore. That's why
> MW spent a lot of last year in this newsgroup promising a new IDE that
> will never materialize on Mac OS X.
I suspect that's true. I suspect that many of their better engineers
are gone now. After all, if you were really good, why would you hang
around MW waiting for the inevitable pink slip?
>
> Metrowerks had great tools that kept up with Apple technology even when
> there were potholes in the road like Copland (gigantic sinkhole sized
> potholes).
>
> Then they fell behind as they snubbed their noses at Mach-O
I don't know all the politics behind this. I'm under the impression
that Apple did some snubbing early on, and it may well be that we're
seeing the inevitable consequence of Apple's actions. If I ran MW and
Apple started making it difficult to do CW, I'm pretty sure I'd think
about abandoning it at some point. I just don't see how any company
could compete with a free alternative when the free product is done by
the people who control the technology and information needed to do
those products. There may have also been other reasons MW didn't see a
long-term future for CW. Does anyone contributing here really know the
whole story?
> and modernizing the look/feel of their IDE.
You mean I'm not the only one who wanted them to update the look and
feel? See, Ron, I told ya. LOL
Larry
Alwyn
la...@skytag.com wrote:
>
> Of course, I have no inside information about Apple's plans and
> intentions, but IMO Apple has too many Unix-heads and gcc is the
> open-source, Unix, blah blah blah way, so I just don't see this
> happening.
I have no doubt that Apple's decision to rely on open-source tools (most
especially GCC) is a business decision supported at the very highest
level. It makes perfect business sense to have a lot of people do most
of the heavy lifting for free.
And then of course there are IBM's and Intel's compilers for the more
demanding customers. I'm sure these firms lose money on their developer
tool operations like almost everyone else in the business, but being
important processor manufacturers, they can hope to offset their losses
on software against increased chip sales.
Freescale too would want to produce software that promotes their chips -
silicon enablement is, I believe, the term they currently like to use.
As a profit centre, I'm sure Metrowerks was lacklustre at best, but it
may have better prospects as a 'silicon-enabling' cost centre. I have
little doubt that this is what it's really about.
<snip>
> Why do you think they're desperate?
I think Metrowerks as a company has been in crisis because their
business model didn't work, hence the dropping of various projects, the
sale of assets to Nokia and the current 'restructuring'.
> I think they don't care. The fact
> is that we don't know anything about what MW wants.
More relevant at this point would be what Freescale want.
> They may well be
> more than happy to let the Mac market go. Everything may very well be
> going according to their plans. *We* tend to think of things
> differently, but then Mac development is very significant to us.
> MW-like Freescale-is in a very different boat. Freescale stated
> that the G4 they sell to Apple represents 3% of their business. No
> large corporation is going to lose sleep over a niche product that
> represents 3% of its market. At this point I doubt CW for Mac
> development represents over 3% of MW's business.
I don't know about the percentages, but the one thing we can be sure of
is that the number of Macintosh CodeWarrior customers is diminishing
quite steeply and will never recover. What sane business person would
want to keep the product alive?
Apart from that, I suspect that the relationship between Freescale and
Apple executives has also cooled lately and is now in the vicinity of
zero degrees Kelvin. In such an atmosphere, any generosity on
Freescale's part would be out of the question.
Alwyn
Peter Lui
> MW-like Freescale-is in a very different boat. Freescale stated
> that the G4 they sell to Apple represents 3% of their business. No
> large corporation is going to lose sleep over a niche product that
> represents 3% of its market.
I suppose Intel has no interest in 3% of the Freescale business then?
3% of $6B is $200M. No company wants to loose that amount of revenue
and it's simplistic thinking that companies enjoy downsizing, paying
severance, benefits, upsetting the employees, losing revenue, paying
for vacant buildings, etc.
At one time the precentage of revenue was obviously much higher when
there was no G5 but it's been an ever shrinking percentage. They threw
away the market for high performance high margin processors to make
slower low power versions. At one time they built both high and low end
chips but Motorola deliberately chose to follow this path that sent
Apple to IBM and now to Intel.
Apple was a high profile customer. Losing them is a huge blow to their
semiconductor business in more ways than just short term revenue.
> At this point I doubt CW for Mac
> development represents over 3% of MW's business.
Who can say. Ever since the buyout, they've buried their financials
from public view. I imagine they are quite embarassing.
Alwyn
Peter Lui <pelui_...@comcast.net> wrote:
>
> I suppose Intel has no interest in 3% of the Freescale business then?
Gaining Apple was a major PR coup for Intel. The numbers don't really
matter. Besides, Intel is unlikely to lose any money by catering to
Apple's needs.
> 3% of $6B is $200M. No company wants to loose that amount of revenue
> and it's simplistic thinking that companies enjoy downsizing, paying
> severance, benefits, upsetting the employees, losing revenue, paying
> for vacant buildings, etc.
>
> At one time the precentage of revenue was obviously much higher when
> there was no G5 but it's been an ever shrinking percentage. They threw
> away the market for high performance high margin processors to make
> slower low power versions. At one time they built both high and low end
> chips but Motorola deliberately chose to follow this path that sent
> Apple to IBM and now to Intel.
>
> Apple was a high profile customer. Losing them is a huge blow to their
> semiconductor business in more ways than just short term revenue.
I think the real loser here is IBM with their Power Everywhere
initiative. Freescale's ambitions are, as you note above, comparatively
modest.
As I recall, however, IBM were complaining that the terms on which they
supplied chips to Apple were far from advantageous to them. I can't
imagine Freescale did much better. I wouldn't be surprised if people at
both companies breathed a sigh of relief when they got wind of the
Apple-Intel partnership.
> > At this point I doubt CW for Mac
> > development represents over 3% of MW's business.
>
> Who can say. Ever since the buyout, they've buried their financials
> from public view. I imagine they are quite embarassing.
Have they ever been anything other than embarrassing? Very, very few
companies are able to earn real money in that market, and Motorola
should have been well aware of this when they were unwise enough to take
over Metrowerks as a going concern.
Alwyn
la...@skytag.com
>> that the G4 they sell to Apple represents 3% of their business. No
>> large corporation is going to lose sleep over a niche product that
>> represents 3% of its market.
>
> I suppose Intel has no interest in 3% of the Freescale business then?
Intel is a different animal. Whereas Apple was Freescale's only
customer for the G4, I doubt Intel will be producing a special chip
just for Apple. For Intel this is a marketing/PR coup and additional
sales of some chip they'll almost certainly be selling to other vendors
as well. I see this as the biggest win from Apple going with Intel.
They'll have a chip supplier who is motivated to bring out better and
better chips without having a "partnership" with Apple in which Apple
has to cajole them into providing what they want. Even if Apple does
want some special feature on a processor from Intel, it's probably
going to be a tweak for Intel, more so than it would be for Freescale.
> 3% of $6B is $200M. No company wants to loose that amount of revenue
> and it's simplistic thinking that companies enjoy downsizing, paying
> severance, benefits, upsetting the employees, losing revenue, paying
> for vacant buildings, etc.
- They still have 97% of their existing business and as soon as they
stop supplying G4s to Apple they can turn the resources they've been
devoting to the G4 to other products which are a better fit with their
current business model. There's no reason to believe they'll have to
shut down anything. A 3% drop is not a killer. If the market for cell
phones dips, it would probably hurt them a lot more.
- Businesses adjust to changes in the market place all the time. From
what I've heard (and I've heard from pretty high up), Apple has a long
history of hiring and laying people off on a fairly regular basis.
- Revenue per se is meaningless. It's profit that counts. The G4 market
is relatively small compared to their mainstay of embedded chips, and
even if the profit margins are higher on the G4, their other chips
probably have volume going for them. The profits were probably not
enough for Freescale to justify investing the resources needed to speed
up the G4s up significantly.
- It would be unrealistic to expect Freescale to continue suppling G4s
to Apple indefinitely. Eventually IBM would have had a low-power G5 or
some other chip to replace it, so Freescale knew this was coming sooner
or later and there really wasn't anything they could do about that
would have made sense to the business.
Larry
Peter Lui
> I can't imagine Freescale did much better. I wouldn't be surprised if
> people at both companies breathed a sigh of relief when they got wind
> of the Apple-Intel partnership.
I'd say the people at Freescale who monopolized the relationships with
Apple and made huge commissions crapped their pants on June 6th.
> Have they ever been anything other than embarrassing? Very, very few
> companies are able to earn real money in that market, and Motorola
> should have been well aware of this when they were unwise enough to
> take over Metrowerks as a going concern.
MW could have made money selling mac tools but since they never broke
ground with any other product (maybe Palm was an exception at one time)
the failure of the other products dragged it down with the rest.
At one time when MW still put out annual reports they had grand visions
of becoming the Microsoft of the embedded world. They saw the billions
of chips sold and the even larger amounts of money made from selling
them but their strategy had a fatal flaw: you only need to buy one set
of tools to write a piece of software that ships on all of those chips,
many that sell for a few dollars or less.
Since they never figured out how to sell services or get in on
royalties they ended up selling shrink wrap software for a loss. The
used the mac business model for all of their products. You can see that
evidence clearly on their website. The difference was that the mac
product had a monopoly on the business while the others did not. A
colossal mistake.
On 2005-06-26 17:09:51 -0700, la...@skytag.com said:
> - They still have 97% of their existing business and as soon as they
> stop supplying G4s to Apple they can turn the resources they've been
> devoting to the G4 to other products which are a better fit with their
> current business model.
What do you think they are going to do with the tooling? Freescale has
been moving their manufacturing offshore to third-parties so they are
probably going to have to write off the Austin fabs and lay off more
people. I'd like to share your optimism but since I'm not a shareholder
I don't care however facts like this remain.
> There's no reason to believe they'll have to
> shut down anything. A 3% drop is not a killer. If the market for cell
> phones dips, it would probably hurt them a lot more.
Motorola cells phones are not based on Freescale semiconductors.
Company divisions are often disfunctional. This is one of the reasons
they were able to dump that business and spin it off without
consequence.
> - Revenue per se is meaningless. It's profit that counts. The G4 market
> is relatively small compared to their mainstay of embedded chips, and
> even if the profit margins are higher on the G4, their other chips
> probably have volume going for them.
Lets stick to comparing PowerPC based chip sales.
> The profits were probably not
> enough for Freescale to justify investing the resources needed to speed
> up the G4s up significantly.
Advances in low power chips are often handed down from the designs in
high power ones. This is not the only factor but Freescale has
endangered their entire PowerPC lineup as a result of losing Apple
which was a huge customer, a source of chip R&D as well as the guinea
pig for many of their other customers interested in similar designs.
> - It would be unrealistic to expect Freescale to continue suppling G4s
> to Apple indefinitely. Eventually IBM would have had a low-power G5 or
> some other chip to replace it, so Freescale knew this was coming sooner
> or later and there really wasn't anything they could do about that
> would have made sense to the business.
Motorola has a license to produce chips following and extending the
PowerPC architecture. G5 is simply an IBM implementation and Freescale
could have done one but they a) chose not to b) can't afford to do it
c) can't figure out how to do it d) all of the above. The answer is D.
la...@skytag.com
> of becoming the Microsoft of the embedded world. They saw the billions
> of chips sold and the even larger amounts of money made from selling
> them but their strategy had a fatal flaw: you only need to buy one set
> of tools to write a piece of software that ships on all of those chips,
> many that sell for a few dollars or less.
I've always wondered about that. Sure, a lot more embedded chips are
sold, but how many people write software for embedded chips? Seems like
a much smaller market than Mac software that anyone could write and run
on his Mac.
> What do you think they are going to do with the tooling? Freescale has
> been moving their manufacturing offshore to third-parties so they are
> probably going to have to write off the Austin fabs and lay off more
> people.
Okay, companies do that all the time. I'm sure Freescale stops
production of some chips and starts producing others on a regular
basis.
> I'd like to share your optimism but since I'm not a shareholder
> I don't care however facts like this remain.
Somehow I just don't see it being a big deal to them. If it had been,
seems like they'd have done something different. It's really hard for
me to imagine that anyone who's used computers for a few years, much
less running a company like that would be naive enough to think any
chip would bring in money forever. CPUs evolve or die, and I'm sure
they know that. I'm sure Apple has had plenty of discussions with them
about that Apple wants and needs and they had to have some idea that
Apple was not happy with the way things were going. I don't know their
entire strategy in all this, I'm just saying I don't think they viewed
it as the big surprise and tragedy some people think they did.
>> - Revenue per se is meaningless. It's profit that counts. The G4 market
>> is relatively small compared to their mainstay of embedded chips, and
>> even if the profit margins are higher on the G4, their other chips
>> probably have volume going for them.
>
> Lets stick to comparing PowerPC based chip sales.
Why? My point is that Freescale doesn't seem to consider PowerPC chip
sales as a cornerstone of their business. Maybe they wanted out of the
desktop CPU business. I don't know, I'm just speculating. Can you say
for certain that it was incompetence and not an intentional business
decision?
> Motorola has a license to produce chips following and extending the
> PowerPC architecture. G5 is simply an IBM implementation and Freescale
> could have done one but they a) chose not to b) can't afford to do it
> c) can't figure out how to do it d) all of the above. The answer is D.
Maybe Freescale didn't want to invest the money to compete with IBM for
Apple's business. The Mac CPU market is not huge to begin with, and if
you split it between two vendors who don't sell those particular
processors to anyone else, maybe it fails to offer enough incentive to
pursue development. When Motorola was supplying all of Apple's CPU,
their research was rewarded every time Apple sold any Mac. That stopped
when IBM started supplying the G5. Freescale lost the desktop sales and
then the iMac sales. Maybe the market just shrank too much to interest
them.
Larry
Alwyn
Peter Lui <pelui_...@comcast.net> wrote:
>
> I'd say the people at Freescale who monopolized the relationships with
> Apple and made huge commissions crapped their pants on June 6th.
I doubt very much that there were ever huge sums to be made from selling
CPUs to Apple. The one thing that Apple is unable to provide is the
volumes to satisfy big business; this applies to software as well as
hardware. The iPod is a different story, but I digress.
> > Have they ever been anything other than embarrassing? Very, very few
> > companies are able to earn real money in that market, and Motorola
> > should have been well aware of this when they were unwise enough to
> > take over Metrowerks as a going concern.
>
> MW could have made money selling mac tools but since they never broke
> ground with any other product (maybe Palm was an exception at one time)
> the failure of the other products dragged it down with the rest.
I think MW did better with those products while they were receiving help
from Apple and Palm respectively. Left to their own devices, they tended
to flounder.
> At one time when MW still put out annual reports they had grand visions
> of becoming the Microsoft of the embedded world.
Well yes, successful businesses want to grow...
> They saw the billions
> of chips sold and the even larger amounts of money made from selling
> them but their strategy had a fatal flaw: you only need to buy one set
> of tools to write a piece of software that ships on all of those chips,
> many that sell for a few dollars or less.
but to do that, they need a viable business model.
> Since they never figured out how to sell services or get in on
> royalties they ended up selling shrink wrap software for a loss. The
> used the mac business model for all of their products. You can see that
> evidence clearly on their website. The difference was that the mac
> product had a monopoly on the business while the others did not. A
> colossal mistake.
I largely agree with your analysis, but any monopoly MW might have had
on the Macintosh software development market was short-lived. One of the
great things about having a monopoly is that it allows you to raise your
prices whenever you feel like it without losing market share. Seldom
were Metrowerks able to do this.
I think Apple decided at some point that they no longer needed
Metrowerks (this is classic Jobsian behaviour, incidentally) and
Metrowerks were left high and dry in the Mac market. Without help from
Apple, it is difficult to see how they could have made a go of it.
This leaves Apple with a near monopoly in the Mac development tools
market. (Yes, I know about Absoft and Real Software - long may they
prosper!) Although Apple are not profiting financially from this
monopoly, I do not believe it is good for developers, or good for the
company either, in the long run. The best thing that Apple could do in
the circumstances would be to open-source their development tools as
they have open-sourced Darwin and other technologies. It is natural for
companies to want to hold on to their intellectual property, but I
believe that the time has come for Apple to make this gesture.
Alwyn
la...@skytag.com
> on the Macintosh software development market was short-lived. One of the
> great things about having a monopoly is that it allows you to raise your
> prices whenever you feel like it without losing market share. Seldom
> were Metrowerks able to do this.
They didn't need a monopoly, just fair competition. Once Apple started
giving away a competing product and not cooperating with MW, CW's days
were numbered, IMO.
>
> I think Apple decided at some point that they no longer needed
> Metrowerks (this is classic Jobsian behaviour, incidentally) and
> Metrowerks were left high and dry in the Mac market. Without help from
> Apple, it is difficult to see how they could have made a go of it.
Exactly, especially when Apple is putting out the competition for free.
I find it interesting that what Apple's done with MW smells like what
Microsoft's been accused of doing: crippling the competition by taking
advantage of inside knowledge of the OS. The Xcode team has full access
to everything it needs, but MW has to wait around for public
announcements.
>
> This leaves Apple with a near monopoly in the Mac development tools
> market. (Yes, I know about Absoft and Real Software - long may they
> prosper!) Although Apple are not profiting financially from this
> monopoly, I do not believe it is good for developers, or good for the
> company either, in the long run.
I agree. The lack of competition and choices in the marketplace is one
of my biggest concerns. Another is that I don't trust Apple. I've seen
what they did with Interface Builder, and that doesn't speak well for
how willing Apple is to provide top-notch tools when they have a
monopoly.
> The best thing that Apple could do in
> the circumstances would be to open-source their development tools as
> they have open-sourced Darwin and other technologies. It is natural for
> companies to want to hold on to their intellectual property, but I
> believe that the time has come for Apple to make this gesture.
I don't see how this would be helpful. My experience with open source
software is that the projects are often behind on their timetable
because they can't get enough volunteers to work on it, they're always
looking for people to work on it, and all too often the end result has
a distinctly non-Mac feel to it (probably because a lot of open source
fans are Unix junkies who don't fully understand Mac software design).
I've never tried an open source application I really liked.
Larry
Alwyn
la...@skytag.com wrote:
>
> > The best thing that Apple could do in
> > the circumstances would be to open-source their development tools as
> > they have open-sourced Darwin and other technologies. It is natural for
> > companies to want to hold on to their intellectual property, but I
> > believe that the time has come for Apple to make this gesture.
>
> I don't see how this would be helpful.
It's the only way that I can see in which Apple can realistically cease
to have a monopoly over the tools and allow people like ourselves to
have more influence over how they develop.
> My experience with open source
> software is that the projects are often behind on their timetable
> because they can't get enough volunteers to work on it, they're always
> looking for people to work on it,
Well, it varies. There were times when GCC stagnated horribly, but now
it is making very impressive progress, with the help of Red Hat and
other companies who stand to benefit from it. In fact, I would say that,
with certain exceptions, the sponsored open-source model is currently
the only viable one for such products.
Remember that open source is now a crucial part of Apple's strategy -
witness Darwin, Bonjour, Webcore etc. I have no doubt at all that they
could hardly survive without it. Above all, it is quite evident that
they have not lost control over the direction in which the software
evolves.
> and all too often the end result has
> a distinctly non-Mac feel to it (probably because a lot of open source
> fans are Unix junkies who don't fully understand Mac software design).
It is a consequence of the Mac's small market share that open-source
software that runs on it tends to be cross-platform, having been
conceived initially for Linux, Windows or some other system. In the case
of Apache or MySQL, this does not matter much, but the same cannot be
said of the GIMP or OpenOffice.org.
> I've never tried an open source application I really liked.
You don't use Adium or VLC maybe? Both are extremely popular open-source
consumer-type applications on the Mac that compete very favourably with
proprietary equivalents.
Alwyn
ward mcfarland
> I've always wondered about that. Sure, a lot more embedded chips are
> sold, but how many people write software for embedded chips? Seems like
> a much smaller market than Mac software that anyone could write and run
> on his Mac.
It is a smaller market by far and many of those doing the coding want a
product that can support many different families of CPU (x86, 68xxx,
ARM, PPC, etc). The development system should support umbilical control
and debugging. Typically, lots of customer support is required for a
myriad of possible hardware configurations.
bols...@hotmail.com
>I think Apple decided at some point that they no longer needed
>Metrowerks (this is classic Jobsian behaviour, incidentally) and
>Metrowerks were left high and dry in the Mac market. Without help from
>Apple, it is difficult to see how they could have made a go of it.
I see it as Apple could see that it was no longer the most
important target of Metrowerks. They didn't want their destiny
tied to a company who was obviously losing interest in their
platform. IMHO, this was as obvious in 1997 as it is today.
If Apple did not have their own toolset, they wouldn't have the G5.
They wouldn't be able to switch to Intel (and as we were told in the
keynote, they kept x86 builds of Mac OS X since the inception of
Mac OS X). They wouldn't also be able to continually bring new
features to the OS that are greatly enhanced by tools support (CoreData, Obj-C
improvements, etc). It's funny to me that people on this group
support MW, which has added no new features for years but catching
up with Mac OS X basics (Mach-O, plists, bundles, packages) and removing
features (x86, pascal, java, etc), while Xcode has many new features
each release, which get released more often than MW!
Apple had to develop Xcode to keep their business viable. Period.
Because of this it keeps the Mac OS X platform alive and viable.
This means that 3rd party developers can continue to make great
Mac OS X applications. If 3rd parties could only rely on MW, I
believe the Mac platform would be much worse for wear.
bols...@hotmail.com
>fans are Unix junkies who don't fully understand Mac software design).
>I've never tried an open source application I really liked.
You're kidding, right? The entire OS you develop for has a basis on
open source. Therefore every app you like on Mac OS X relies on
open source. And I assume you like some apps, right? ;)
Alwyn
bols...@hotmail.com wrote:
> Alwyn wrote:
>
> >I think Apple decided at some point that they no longer needed
> >Metrowerks (this is classic Jobsian behaviour, incidentally) and
> >Metrowerks were left high and dry in the Mac market. Without help from
> >Apple, it is difficult to see how they could have made a go of it.
>
> I see it as Apple could see that it was no longer the most
> important target of Metrowerks. They didn't want their destiny
> tied to a company who was obviously losing interest in their
> platform. IMHO, this was as obvious in 1997 as it is today.
And I was thinking the rot started in 1999 while the company was being
groomed for takeover. (There was a terrible brouhaha over the dropping
of Pascal and amendments to terms and conditions, which was handled in
typically cack-handed fashion. MW Ron, as usual, took the brunt of the
punishment, but it was a public-relations disaster for the whole
company.)
MW Ron posted in December 1997:
> Metrowerks is not a Macintosh compiler company, We make Be, Palm Pilot,
> Sony Playstation, many RTOS, Set Top compilers, Magic Link, Windows NT,
> Windows 95, and soon Apple Rhapsody and I'm sure I'm forgetting something.
> So while we are proud of our Macintosh roots and are planning on improving
> Macintosh and other Apple systems well into the futures, we are not a
> Macintosh compiler company.
<news:MWRon-ya02408000...@news.kal.ameritech.net>
Some of the rhetoric sounds strangely familiar, does it not? Then,
diversification and expansion were the order of the day: now, it is
contraction and regrouping. In both cases, however, the Macintosh is
conspicuously de-emphasised.
<snip>
> Apple had to develop Xcode to keep their business viable. Period.
Well, they already had ProjectBuilder and InterfaceBuilder from NeXT,
didn't they? Mac OS X is really a continuation of NeXTstep.
> Because of this it keeps the Mac OS X platform alive and viable.
> This means that 3rd party developers can continue to make great
> Mac OS X applications. If 3rd parties could only rely on MW, I
> believe the Mac platform would be much worse for wear.
That may be a rather one-sided way of looking at it. It takes two to get
married and two to divorce. Divorces are by nature messy and miserable
events, and more than two can get hurt in the process.
Alwyn
Jøhnny Fävòrítê
> You don't use Adium or VLC maybe? Both are extremely popular open-
> source consumer-type applications on the Mac that compete very
> favourably with proprietary equivalents.
sez you! i don't use adium. i do use vlc, but only as a last resort, when i
come across a movie that won't play in quicktime. it is the very definition
of a sloppy open-source app that doesn't have a good user experience. and
it's not just a mac thing, it's not so hot-looking on any of the other
platforms it runs on, either.
in all the apps i've really liked, you can practically feel the personalities
of the people involved. if it's a small program, it might be the coders
themselves you're experiencing. in a large program, it's the people who did
the design work, i suppose, and laid out a clear path for the coders to
follow.
there are things the cat-herding open-source development model is good at,
obviously. compilers, apache, protocols, almost any server-based software.
guis, however, can't be designed-by-committee.
Jøhnny Fävòrítê
> la...@skytag.com wrote:
>> I've never tried an open source application I really liked.
>
> You're kidding, right?
i don't know if he is, but i'm sure not.
> The entire OS you develop for has a basis on open source. Therefore
> every app you like on Mac OS X relies on open source. And I assume
> you like some apps, right? ;)
not surprisingly, all the open-source stuff in macosx is in the plumbing that
makes the innards work. also not surprisingly, all the gui elements of macosx
are *not* open source. the sourceforge crowd is not good at guis. never has
been, never will be.
MW Ron
Peter Lui <pelui_...@comcast.net> wrote:
>Who can say. Ever since the buyout, they've buried their financials
>from public view. I imagine they are quite embarassing.
In the Freescale reports we are part of "other".
Brian
> Alwyn wrote:
>> You don't use Adium or VLC maybe? Both are extremely popular open-
>> source consumer-type applications on the Mac that compete very
>> favourably with proprietary equivalents.
>
> sez you! i don't use adium. i do use vlc, but only as a last resort, when i
> come across a movie that won't play in quicktime. it is the very definition
> of a sloppy open-source app that doesn't have a good user experience. and
> it's not just a mac thing, it's not so hot-looking on any of the other
> platforms it runs on, either.
But you do use VLC. Correct? As I do. Which I'm very thankful for,
or else a number of AVIs and MPGs I have would not play. Why? Because
there is no other convenient player with all the codecs VLC supports.
And given it's free I would check again to see if you can find any bit
of gratitude.
in all the apps i've really liked, you can practically feel the personalities
> of the people involved. if it's a small program, it might be the coders
> themselves you're experiencing. in a large program, it's the people who did
> the design work, i suppose, and laid out a clear path for the coders to
> follow.
>
I've used many pieces of software -- shareware and commercial -- on
MacOS over the years where I can "feel the personalities of the people
involved". As for being saved by "large programs" don't count on it.
Ever work on a big program with quantities of people? Many many times
I've seen the UI designed by committee, or at least "reviewed" by
committee. You also get disconnect among groups working on a system.
Or various other problems. It cuts both ways. For example, would
someone please give me a very good reason for Apple Mail's new GUI in
Tiger?
How you build the GUI is not inherently dictated by whether or not you
have the words "OpenSource" hiding in your product's license. Until
there exists some a priori deterministic method which can turn a set of
UI requirements into an actual UI you will always "feel the
personalities of the people involved". Regardless of how large or how
free your product is.
--- Brian
jonh...@mac.com
Well, that certainly explains a lot. I mentally partition the company
so that "Metrowerks" is nostalgicly still that cool company that did
such great things in the '90s, and "FreeScale" as the evil villian that
sucked its soul away. Maybe that's too simplistic a viewpoint, but it
works for me. :-)
Jøhnny Fävòrítê
> But you do use VLC. Correct?
once every month or so. three out of five times it turns out it can't play
the file that quicktime couldn't play, so it was a wash.
> As I do. Which I'm very thankful for, or else a number of AVIs and
> MPGs I have would not play. Why? Because there is no other
> convenient player with all the codecs VLC supports. And given it's
> free I would check again to see if you can find any bit
> of gratitude.
"gratitude" does not include telling the authors it's great when it isn't.
that's called "lying." if there was a nicer app that played as many types of
files as vlc does, but cost 50 bucks, i'd buy it in a heartbeat.
> As for being saved by "large programs" don't count on it.
i don't recall saying anything for or against the size of a program, relative
to its merit. i think it's you who brought that up, just now.
> How you build the GUI is not inherently dictated by whether or not
> you have the words "OpenSource" hiding in your product's license.
i'd love to believe that. if i saw two or three open source apps with decent
guis, then i would. so far i haven't seen even one, after decades of trying.
that leads me to believe open source projects will never have decent guis.
duh.
Alwyn
Jøhnny Fävòrítê (it means "A Device Which Is Meowing")
>
> if i saw two or three open source apps with decent
> guis, then i would. so far i haven't seen even one, after decades of trying.
> that leads me to believe open source projects will never have decent guis.
Not even Camino? Some people like Firefox and Thunderbird too. These
things are mainly subjective really - what one person likes, another
hates.
Alwyn
la...@skytag.com
>> that leads me to believe open source projects will never have decent guis.
>
> Not even Camino? Some people like Firefox and Thunderbird too. These
> things are mainly subjective really - what one person likes, another
> hates.
Some people like Emacs and Pine. Heck, some people like Windows better
than Mac OS X. Better is certainly relative, but my position is that
when it comes to Mac software, the standard is what *Mac* software
should be, including adherance to Apple user interface guidelines.
Every open source application I've tried had at least one annoyingly
non-standard aspect to its interface.
Larry
Brian
>> As I do. Which I'm very thankful for, or else a number of AVIs and
>> MPGs I have would not play. Why? Because there is no other
>> convenient player with all the codecs VLC supports. And given it's
>> free I would check again to see if you can find any bit
>> of gratitude.
>
> "gratitude" does not include telling the authors it's great when it isn't.
> that's called "lying." if there was a nicer app that played as many types of
> files as vlc does, but cost 50 bucks, i'd buy it in a heartbeat.
I said "gratitude" not "love". So I take by the comments you made the
attitude is "I'll use your app, thanks for helping me out for free and,
by the way, it's garbage." Yeah, it is a shame those developers
working on VLC went out of their way to let you play media files you
normally couldn't, at no charge to you. I guess it's better to focus
on how sloppy the app UI feels instead of at least giving a nod to what
they have accomplished.
>> How you build the GUI is not inherently dictated by whether or not
>> you have the words "OpenSource" hiding in your product's license.
>
> i'd love to believe that. if i saw two or three open source apps with decent
> guis, then i would. so far i haven't seen even one, after decades of trying.
> that leads me to believe open source projects will never have decent guis.
> duh.
Well, given how UIs aren't some strict science, I guess the "feelings"
you have for UIs you love must be somewhat subjective. Which means you
don't personally like open source application UIs, not that all open
source UIs are bad for everyone. Duh.
-- Brian
la...@skytag.com
> support MW, which has added no new features for years but catching
> up with Mac OS X basics (Mach-O, plists, bundles, packages) and removing
> features (x86, pascal, java, etc), while Xcode has many new features
> each release, which get released more often than MW!
I guess it depends on how you define "support CW." I've continued to
use CW because it has continued to supply my needs as an application
developer without making me port my project to Xcode, learn another IDE
which I don't like as well, compiles at a fraction of the speed of CW,
has inferior STL libraries, and so on. Xcode has never offered me a
single advantage other than price and I don't really care about
$200/year to use what I like. The Intel switch will finally mean Xcode
offers something I need and can't get from CW. Frankly, I hear the
process of converting projects to Xcode continues to get easier and
Xcode gets generally better with every release, so I have no regrets
about waiting as long as I have to make the switch.
Larry
la...@skytag.com
>> files as vlc does, but cost 50 bucks, i'd buy it in a heartbeat.
>
> I said "gratitude" not "love". So I take by the comments you made the
> attitude is "I'll use your app, thanks for helping me out for free and,
> by the way, it's garbage." Yeah, it is a shame those developers
> working on VLC went out of their way to let you play media files you
> normally couldn't, at no charge to you.
I can't speak for anyone else, but I spend a lot of time and energy
ensuring I know the interface guidelines and that my application
follows them. So it bugs me when other developers don't.
> I guess it's better to focus on how sloppy the app UI feels instead of at
> least giving a nod to what they have accomplished.
If I don't care about the UI in applications I use, why do I use a Mac?
I'd be better off using Windows and never seeing a third-party product
I can't use because it only works with Windows.
>>> How you build the GUI is not inherently dictated by whether or not
>>> you have the words "OpenSource" hiding in your product's license.
>
>> i'd love to believe that. if i saw two or three open source apps with decent
>> guis, then i would.
I believe it, but in practice open source applications rarely seem to
follow Apple's guidelines.
>> so far i haven't seen even one, after decades of trying.
>> that leads me to believe open source projects will never have decent guis.
>> duh.
>
> Well, given how UIs aren't some strict science, I guess the "feelings"
> you have for UIs you love must be somewhat subjective.
Not really. Apple publishes the Apple Human Interface Guidelines, and
many commercial applications provide excellent examples of how an
interface should look and feel in a Mac application. I've yet to see an
open source application that doesn't have at least a few aspects of its
UI that vary from the guidelines or implement their own non-standard
behavior in place of system-provided behaviors.
> Which means you don't personally like open source application UIs, not
> that all open source UIs are bad for everyone. Duh.
Bad in the sense that they're non-standard, yes, all that I've tried
were bad.
Larry
bols...@hotmail.com
>I can't speak for anyone else, but I spend a lot of time and energy
>ensuring I know the interface guidelines and that my application
>follows them. So it bugs me when other developers don't.
Hehe. I gotta tease now. You don't need to know the HIG if you use
Interface Builder, as it has the HIG built-in! ;)
-Greg
Brian
>>> that's called "lying." if there was a nicer app that played as many types of
>>> files as vlc does, but cost 50 bucks, i'd buy it in a heartbeat.
>>
>> I said "gratitude" not "love". So I take by the comments you made the
>> attitude is "I'll use your app, thanks for helping me out for free and,
>> by the way, it's garbage." Yeah, it is a shame those developers
>> working on VLC went out of their way to let you play media files you
>> normally couldn't, at no charge to you.
>
> I can't speak for anyone else, but I spend a lot of time and energy
> ensuring I know the interface guidelines and that my application
> follows them. So it bugs me when other developers don't.
VLC is free and fills a gap no other MacOS application does. It's fine
if you don't happen to be crazy about the UI. That's different than a
flame about how sloppy it is -- supposedly like ALL open source -- and
how you wish you could dump VLC in favor of another app, while at the
same time using it. Doesn't sound like constructive criticism to me,
but ingrateful complaining. So I guess we're both ranting.
>
>> I guess it's better to focus on how sloppy the app UI feels instead of at
>> least giving a nod to what they have accomplished.
>
> If I don't care about the UI in applications I use, why do I use a Mac?
> I'd be better off using Windows and never seeing a third-party product
> I can't use because it only works with Windows.
I never said, or would say, people should not care about an
application's UI. All I'm saying is some -- any -- gratitude for the
free -- no cost what-so-ever -- application you are using would be a
nice change, in addition to the complaints. Hence the "FOCUS on how
sloppy the app UI feels" as opposed to "at LEAST giving a nod" part.
>
>>>> How you build the GUI is not inherently dictated by whether or not
>>>> you have the words "OpenSource" hiding in your product's license.
>>
>>> i'd love to believe that. if i saw two or three open source apps with decent
>>> guis, then i would.
>
> I believe it, but in practice open source applications rarely seem to
> follow Apple's guidelines.
>
>>> so far i haven't seen even one, after decades of trying.
>>> that leads me to believe open source projects will never have decent guis.
>>> duh.
>>
>> Well, given how UIs aren't some strict science, I guess the "feelings"
>> you have for UIs you love must be somewhat subjective.
>
> Not really. Apple publishes the Apple Human Interface Guidelines, and
> many commercial applications provide excellent examples of how an
> interface should look and feel in a Mac application. I've yet to see an
> open source application that doesn't have at least a few aspects of its
> UI that vary from the guidelines or implement their own non-standard
> behavior in place of system-provided behaviors.
In his original post he mentioned "in all the apps i've really liked,
you can practically feel the personalities of the people involved" and
how open source apps lack it: "guis, however, can't be
designed-by-committee". Nothing at all about the HIG, just personal
preference and "feeling". That was the issue. And I agree: taste is
taste, not objective comparison.
As for the HIG, I'm fully aware of them. And I agree, the more
Mac-like an application is then hopefully easier for a user to
understand. But what a user has a preference for, personally, that is
different. Someone may freak out because a dialog should be a sheet
when it isn't. Or the toolbar doesn't look like an NSToolbar. Then
again, other users might not really care, or even prefer it in that
particular app.
I agree with the spirit of what you're saying; the more HIG compliant
the better and more consistent everything is. But that's different than
a user's taste or loving an application's UI because "you can
practically feel the personalities of the people involved".
--- Brian
Chris Baum
<la...@skytag.com> wrote:
> so I have no regrets
> about waiting as long as I have to make the switch
If your company has shipping apps, aren't any of them asking your
status vis a vis x86?
Your competitor's response might be "Yes we use XCode and have already
built and tested a Universal Binary and expect a final x86 version
immediately after the release of the first Intel Mac."
Your response: "We use CW now but everybody tells me it's getting
easier to port! Check back next year."
Andy Dent
<dt015a1979-DBBE1...@text.news.blueyonder.co.uk>,
uhh, the analogy is a good one but your math is wrong - it only takes
one to divorce as the other has no choice (sounds like you've never been
divorced!).
Am I misremembering things or did Apple royally screw MW with gdb
issues, continual header changes etc. in the OS/X early year(s)? I was
under the impression that a LOT of the MW development effort was spent
catching up to Apple (especially rather capricious changes) rather than
working in tandem.
I suspect there's a lot of Next-hangover in Apple engineering with the
arrogance that they don't *need* any other dev tools and this resulted
in a lack of cooperation, along with failing Mac sales, that made MW Mac
development unreasonably expensive.
Andy
--
Andy Dent BSc MACS
OOFILE - Cross-Platform Database, Reports, Graphs, GUI in C++
PP2MFC - PowerPlant->MFC portability
http://www.oofile.com.au/
bols...@hotmail.com
>I suspect there's a lot of Next-hangover in Apple engineering with the
>arrogance that they don't *need* any other dev tools and this resulted
>in a lack of cooperation, along with failing Mac sales, that made MW Mac
>development unreasonably expensive.
MW welcomed the competition with the initial Apple Tools release with 10.0.
Apple continued to innovate.
Since Mac OS X came out, MW just played catch up to 10.0 (Mach-O, plist,
Obj-C, etc)! It seems to me that all of that should have been there in the
initial release!
If Apple didn't believe they 'needed' their own Developer tools, they wouldn't
have G5 or x86 support. Remember, Jobs mentioned in the WWDC keynote
that Mac OS X was universal ppc and x86 from the start. This wouldn't have
been possible with the MW toolchain at that time. MW's compiler has never
supported G5 or 64 bit on Mac OS X.
IMHO, having their own Dev Tools allowed Apple to innovate, which is what
enables Apple to survive as a business. Relying upon a 3rd party tools
company whose primary interests for all intents and purposes seemed to
lie anywhere but Apple doesn't seem like a good way for Apple to stay in
business.
Now if Apple wanting to stay in business is perceived as arrogance...
Peter Lui
> My point is that Freescale doesn't seem to consider PowerPC chip
> sales as a cornerstone of their business. Maybe they wanted out of the
> desktop CPU business. I don't know, I'm just speculating. Can you say
> for certain that it was incompetence and not an intentional business
> decision?
Fabs are $B investments and they take decades to pay off so they don't
throw away the business lightly.
However in this case, some fool at Motorola decided that Apple was not
worth catering to as their eyes bulged at the thought of acquiring even
larger customers during the dot-com era. After the smoke cleared, the
dot-bombs had evaporated along with all the potential new revenue and
they were left in a position of nothing to offer Apple except the same
old G4s. They gambled and lost.
Peter Lui
> MW welcomed the competition with the initial Apple Tools release with 10.0.
Yeah, because there was no competiton. PB was a pathetic joke that only
a zealot could love and which customer was going to switch and waste
hours building their projects when it could be done in minutes with
CodeWarrior? MW knew that customers couldn't switch.
> Since Mac OS X came out, MW just played catch up to 10.0 (Mach-O, plist,
> Obj-C, etc)! It seems to me that all of that should have been there in the
> initial release!
Let's take a step back for a second and not try to rewrite history.
Apple's tools let alone OS did not provide a smooth transition path for
customers. Where was their Mac OS and CFM/PEF tools support? Either
non-existant or held back from developers and kept internal.
CodeWarrior had to bootstrap itself the hard way. Even Apple admitted
during the keynote at WWDC that the carbonizing effort was greatly
underestimated.
It's unfair to suggest that MW should have dropped further development
of legacy support in order to support only the new evolving standards.
Customers weren't ready to switch. By the time they were, MW had been
run into the ground by Motorola and it was too late.
When Apple told everyone to switch to Display Postscript, customers
balked and laughed. Should MW have rewritten their IDE in DPS as soon
as Apple commanded? Well, they started to and what a disaster and waste
of time/money that was. MW got burned by Apple so many times (Copland,
Rhapsody) that it was right to take a conservative approach but
unfortunately they failed to recognize when changes were here to stay
and integrate them into the default workflow.
IIRC, MW shipped mach-o compilers in Pro4 in 1999 but they didn't work
until Pro7 in 2001. Apple had no choice but to rely on their own effort
based on that snails pace of adoption.
> IMHO, having their own Dev Tools allowed Apple to innovate, which is
> what enables Apple to survive as a business. Relying upon a 3rd party
> tools company whose primary interests for all intents and purposes
> seemed to lie anywhere but Apple doesn't seem like a good way for Apple
> to stay in business.
Innovate? Is that the Microsoft definition? Wasn't the NeXT acquisition
in Dec 1996? Has it really taken 10 years for Apple's developer tools
to get this far?
la...@skytag.com
>> about waiting as long as I have to make the switch
>
> If your company has shipping apps, aren't any of them asking your
> status vis a vis x86?
Our applications never ask me anything. ;-) Nor have I gotten a single
e-mail asking about this. I recently added a web page explaining that
we'd be following Apple to Intel, but even before that, not a peep. I
suspect the average user isn't as focused on it as developers are.
> Your competitor's response might be "Yes we use XCode and have already
> built and tested a Universal Binary and expect a final x86 version
> immediately after the release of the first Intel Mac."
Who gives a rat's behind if you have a universal binary now?
> Your response: "We use CW now but everybody tells me it's getting
> easier to port! Check back next year."
Thanks for the sarcasm, but believe it or not, you don't have to be
building universal binaries today to be ready for the new Macs. I
believe all of our stuff will work under Rosetta, and right now I'm
focused on product development because I am planning a major version
release before the Intel Macs ship. I still use Panther as my primary
OS, and will at least until 10.4.2 since I'm kind of weary of being on
the Mac OS X bleeding edge. It shouldn't take six months or even one
month for me to convert to Xcode, so what's the big deal? Have I done
something professionally shameful because I wasn't in a hurry to switch
to an immature development environment with a compiler that's
universally acknowledged to be significantly slower than the one I use?
Just how brainwashed are you? Exactly how would I have been better off
to switch to Xcode a year ago? So what if I don't have a universal
binary now? I couldn't sell it even if I did. Not today and not in five
months.
I've been doing Mac development since around the time System 7.1 came
out. I always have products to cover a range of OS versions. I made the
transition from 68k to PPC, from THINK Pascal to MW and C++, from Mac
OS 9 to Mac OS X and I'll make this one too. I have a long history of
supporting what I sell and improving it with every release. Maybe your
customers are in a panic over this, but mine have no reason to worry.
Larry
la...@skytag.com
Okay, since you're teasing I'm assuming you don't believe that. LOL But
for the benefit of our readers:
Not true. First of all, if you're using Cocoa, Size to Fit does not
size push buttons correctly in Cocoa nibs.
Far more important though, is the fact that IB only knows about layout
and sizing guidelines. Window and general application design is still
up to you and there's a lot more to it than getting buttons the right
distance from the edge of the window. ;-) Unfortunately, a lot of
developers--open source and commercial--don't seem to realize this.
For example, I tried Thunderbird once a while back, and its Preferences
window was implemented as a sheet on whatever window happened to be
active at the time I used the Preferences command. This was clearly
wrong. Even worst, I'd click some button to set options and the prefs
sheet would slide up and the extra options sheet would slide down. When
I dismissed it, it would slide up and the prefs sheet would slide back
down (this is also wrong).
This is the kind of stuff I've come to expect in open source
applications. (FWIW I filed a bug report on this on at the end of Sept
of last year and got an e-mail a few days ago saying they'd rewritten
the prefs window and that this shouldn't be an issue anymore, but I
haven't had a chance to try the latest version yet.)
I tried another popular open-source application (I think it was
Foxfire) which rolled it's own toolbar implementation instead of using
the one provided by the system. It worked, but everything about it was
a little different than the system's toolbar. For example, its
configuration window was not a sheet and there were other differences
as well that made it feel awkward.
Larry
la...@skytag.com
>> sales as a cornerstone of their business. Maybe they wanted out of the
>> desktop CPU business. I don't know, I'm just speculating. Can you say
>> for certain that it was incompetence and not an intentional business
>> decision?
>
> Fabs are $B investments and they take decades to pay off so they don't
> throw away the business lightly.
This is kind of my point. To keep Apple as a customer Freescale would
have had to invest a lot of money eventually, and perhaps they were not
willing to do that for a percentage of the Mac market. So they milked
the current product line as long as they could without having to spend
a lot on it.
Larry
la...@skytag.com
> the circumstances would be to open-source their development tools as
> they have open-sourced Darwin and other technologies. It is natural
for
> companies to want to hold on to their intellectual property, but I
> believe that the time has come for Apple to make this gesture.
Darwin is open source. Mac OS X is not and Xcode is a Mac OS X
development tool tied to specific versions of Mac OS X. If Xcode were
open source, it could limit what Apple could do to it in advance of new
releases of the OS.
More generally, though, it seems to me that for any development tool to
be a viable option for Mac OS X development, the developer would need
access to information about upcoming versions of Mac OS X, information
only available under NDAs from Apple. How would that work with open
source tools? Otherwise it would always be playing catch up with Xcode.
Larry
Alwyn
la...@skytag.com wrote:
>
> Darwin is open source. Mac OS X is not and Xcode is a Mac OS X
> development tool tied to specific versions of Mac OS X. If Xcode were
> open source, it could limit what Apple could do to it in advance of new
> releases of the OS.
>
> More generally, though, it seems to me that for any development tool to
> be a viable option for Mac OS X development, the developer would need
> access to information about upcoming versions of Mac OS X, information
> only available under NDAs from Apple. How would that work with open
> source tools? Otherwise it would always be playing catch up with Xcode.
I can't see how this should differ substantially from what is already
happening with Darwin and other open-source technologies which Mac OS X
depends on. Apple make a private fork, which they work on until they are
ready to release it into the community.
Alwyn
Paul
<la...@skytag.com> wrote in message
news:1119911162.4...@f14g2000cwb.googlegroups.com...
True. I tried "importing" a fair sized project into XCode when it first came
out and gave up in disgust after a few hours with hundreds and hundreds of
cryptic errors. I have no real motivation to try again for a while, since
"legacy" software will still be runnable for several years to come and XCode
will hopefully only get better. I haven't given much thought to x86 porting,
probably won't until commercial x86 Mac's are available and seem to be
selling. CW will take me that far, at least, and maybe further on Windows
where the alternative (Visual Studio) sucks even worse and customer
expectations are lower.
Boy, I have a lousy attitude these days. Forgive me, I'm having to do MFC's
8 hours a day lately at the office.
la...@skytag.com
I'm just saying that in some ways Xcode is more like Mac OS X than
Darwin.
In any case, do you really want to rely on an Apple-specific open
source development tool like this? Here's what I see happening: Apple
adds a nice feature to Xcode, but it relies on a new feature of Mac OS
X 10.5, so they keep it private. Because any open source enhancements
are not covered under NDA, Apple doesn't release the Xcode change until
10.5 ships. By the time they release it, the feature works great in
Xcode, and the open source version doesn't have it at all. Now the
people doing the open source version have to incorporate it, merge it
with their stuff, whatever, and what if it's one of those projects that
can't find enough people to work on it? Who would want to rely on such
a tool in a highly competitive industry such as this, especially when
Xcode is free? And, what problem would it solve?
Larry
Alwyn
la...@skytag.com wrote:
>
> I'm just saying that in some ways Xcode is more like Mac OS X than
> Darwin.
Yes, I get your point, at least as much as Xcode and InterfaceBuilder
etc. use the Mac OS X user interface and Darwin doesn't. So we can say
that these tools are not part of Darwin.
> In any case, do you really want to rely on an Apple-specific open
> source development tool like this? Here's what I see happening: Apple
> adds a nice feature to Xcode, but it relies on a new feature of Mac OS
> X 10.5, so they keep it private. Because any open source enhancements
> are not covered under NDA, Apple doesn't release the Xcode change until
> 10.5 ships. By the time they release it, the feature works great in
> Xcode, and the open source version doesn't have it at all.
This is, I think, where you go wrong. Changes committed to the
open-source Xcode should be incorporated by Apple into the next version
as far as possible. It is Apple's responsibility to keep their fork up
to date.
> Now the
> people doing the open source version have to incorporate it, merge it
> with their stuff, whatever, and what if it's one of those projects that
> can't find enough people to work on it? Who would want to rely on such
> a tool in a highly competitive industry such as this, especially when
> Xcode is free? And, what problem would it solve?
Open-source management is hard and requires constant vigilance. So far
as I can tell, however, Apple have been fairly conscientious in living
up to their duties, even to the extent of taking on board the legitimate
criticisms of the KHTML community.
Apple have said that the Mac OS X APIs have been stabilised as of the
release of Tiger. With the death of CodeWarrior, I suggest that this is
the right time for them to open-source their development tools in order
to enhance their goodwill within the developer community.
Alwyn
Paul
"Alwyn" <dt015...@mac.com> wrote in message
news:dt015a1979-1230B...@text.news.blueyonder.co.uk...
Apple could stand to improve their good will within the developer community.
At the risk of a flaming by those who define good and bad w/ respect to
Steve & Co, MW's (admittedly/apparently) shabby behavior didn't happen all
by itself: they were royally shafted by Apple (whom they flat out saved from
an ignominious doom during the PPC transition); they chose to keep quiet and
pass the shafting along, but that's another story.
I'd love to be able to play with the internals of XCode / IB if I'm going to
be stuck with them in another year or two any way. I could at least fix the
stuff I hate the worst, and I think I might like doing custom mods. It has
potential: in all modesty, in my fulltime shareware days I had a much better
reputation for listening to customers than Apple has ever come close to, and
I'm not an OSS purist, which is quite the competitive advantage in this
case.
MW Ron
"jonh...@mac.com" <jonh...@mac.com> wrote:
see you at MacHack ooh AdHoc
Alwyn
"Paul" <remove_this_...@patchdance.com> wrote:
>
> Apple could stand to improve their good will within the developer community.
They could indeed. Developer relations (and for that matter, user
relations) are not as good as they could be.
> At the risk of a flaming by those who define good and bad w/ respect to
> Steve & Co, MW's (admittedly/apparently) shabby behavior didn't happen all
> by itself: they were royally shafted by Apple (whom they flat out saved from
> an ignominious doom during the PPC transition); they chose to keep quiet and
> pass the shafting along, but that's another story.
Well yes, I think MW got a lot of help from Apple when they were needed.
Now that they are not needed, they get no help and don't know what to do.
> I'd love to be able to play with the internals of XCode / IB if I'm going to
> be stuck with them in another year or two any way. I could at least fix the
> stuff I hate the worst, and I think I might like doing custom mods. It has
> potential: in all modesty, in my fulltime shareware days I had a much better
> reputation for listening to customers than Apple has ever come close to, and
> I'm not an OSS purist, which is quite the competitive advantage in this
> case.
Or there's the other possibility: CodeWarrior for Mac will become open
source. It seems unlikely, but in times of flux, you never know!
Alwyn
la...@skytag.com
>> Darwin.
>
> Yes, I get your point, at least as much as Xcode and InterfaceBuilder
> etc. use the Mac OS X user interface and Darwin doesn't. So we can say
> that these tools are not part of Darwin.
>> In any case, do you really want to rely on an Apple-specific open
>> source development tool like this? Here's what I see happening: Apple
>> adds a nice feature to Xcode, but it relies on a new feature of Mac OS
>> X 10.5, so they keep it private. Because any open source enhancements
>> are not covered under NDA, Apple doesn't release the Xcode change until
>> 10.5 ships. By the time they release it, the feature works great in
>> Xcode, and the open source version doesn't have it at all.
>
> This is, I think, where you go wrong. Changes committed to the
> open-source Xcode should be incorporated by Apple into the next version
> as far as possible. It is Apple's responsibility to keep their fork up
> to date.
Hmm. Okay, let's go with your model for the sake of discussion. I am
not comforted, for multiple reasons:
- I prefer a good UI on the applications I use. I've been using and
writing Mac software for more than 15 years. I'm am not a Unix geek or
Windows programmer who has recently joined the fold with the
introduction of Mac OS X. I want tools that, to the greatest extent
possible, have good UIs. That's one of the reasons I prefer CW.
Discounting for a moment the state of things over the past few years,
I've always found CW easier to learn and use than Project Builder and
Xcode. Xcode is making rapid progress, but I would be concerned that
becoming open source would not be in the interest of Xcode in this
area.
- I believe we should have a development tool available that's easy to
learn and use, not only for me, but to attract more people to Mac
software development. I don't see that as a goal of open source
software. It simply isn't, in my experience, software that attracts
newbies. There's no technical reason it couldn't, but for whatever
reasons, open source software never sees wide adoption. I think I can
safely say that if my only option was open source software when I
started writing Mac software, I would never have started.
- Part of the problem with wide adoption may be the lack of support. CW
has historically had MWRon and e-mail support. Xcode has the xcode list
at Apple. Who's going to support the open source version? Open source
software has always struck me as leaving you on your own much more than
products like CW and Xcode, where the developers have a vested interest
in seeing the product widely adopted.
>> Now the
>> people doing the open source version have to incorporate it, merge it
>> with their stuff, whatever, and what if it's one of those projects that
>> can't find enough people to work on it? Who would want to rely on such
>> a tool in a highly competitive industry such as this, especially when
>> Xcode is free? And, what problem would it solve?
>
> Open-source management is hard and requires constant vigilance. So far
> as I can tell, however, Apple have
BTW, "Apple" is singular, so it's "Apple has."
> been fairly conscientious in living up to their duties, even to the extent of
> taking on board the legitimate criticisms of the KHTML community.
Yes, but I strongly suspect these parts of Mac OS X are largely ignored
by much of the developer community, in that most Mac developers never
deal directly with anything in Darwin. For them (okay, us ;-) it's just
another part of the OS. If Mac OS X development were suddenly done left
to people who spend time pouring over Darwin, making improvements, and
taking advantage of the fact that it's open source, the platform would
surely die. We definitely need those people, but they aren't enough to
make the platform a strong presence in a mass market of the
non-technical people who're buying all those iBooks and iMacs.
> Apple have said that the Mac OS X APIs have been stabilised as of the
> release of Tiger.
Really? That would be great news if it's true, but I hadn't heard that.
Where did you see that? FWIW, I was told a long time ago by someone at
Apple that Tiger was going to have fewer changes in it and would focus
more on fixes and addressing problems than previous releases, but I
don't think it worked out that way.
> With the death of CodeWarrior, I suggest that this is
> the right time for them to open-source their development tools in order
> to enhance their goodwill within the developer community.
I certainly agree they need to treat their developers better and with
more respect, but I have other ideas for that than making Xcode open
source. ;-)
Larry
bols...@hotmail.com
>> Apple could stand to improve their good will within the developer community.
>
>They could indeed. Developer relations (and for that matter, user
>relations) are not as good as they could be.
At the WWDC Keynote SJ noted there were more developers than in some time,
like 10 years. I can't recall exactly what he said.. Saying developers don't like
Apple doesn't make sense when there are more of them... Sure the number of
complaints go up when there are more developers (fact of life), but you're not
basing these statements on any facts I can find.
Alex Curylo
la...@skytag.com at la...@skytag.com wrote on 6/28/05 4:14 PM:
> [open source software] simply isn't, in my experience, software that attracts
> newbies. There's no technical reason it couldn't, but for whatever
> reasons, open source software never sees wide adoption.
Hmmmmm-mmmmm ... Fire and Adium are both open source programs, and they seem
to be more widely adopted than proprietary IM programs, from everything I've
seen.
la...@skytag.com
>
>> They could indeed. Developer relations (and for that matter, user
>> relations) are not as good as they could be.
>
> At the WWDC Keynote SJ noted there were more developers than in some time,
> like 10 years. I can't recall exactly what he said..
I do, but frankly, so what? There are a lot more people using Windows
than Mac OS X. Obviously big numbers don't mean squat.
> Saying developers don't like Apple doesn't make sense when there are more of
> them...
Indeed, if that's what he'd said it wouldn't make sense, but that isn't
what he said. It's not a matter of "liking" Apple. It's a matter of how
people feel they're being treated. People don't develop Mac software
because they like Apple, they do it because they think there's a market
there worth pursuing and/or because the Mac is the platform they
prefer. This is analogous to the fact that I use CW, but not because I
like MW.
> Sure the number of complaints go up when there are more developers (fact of
> life), but you're not basing these statements on any facts I can find.
Each of us has his own take on this issue, but I for one have been very
frustrated since the introduction of Mac OS X on multiple fronts:
- Documentation: Apple dumped a major new OS on us virtually
undocumented. The state of documentation is getting better as time goes
on, but it is still far from ideal and it used to be all but
non-existent for much of what was new in Mac OS X. (Case in point: Take
a look at Tech Note 2078:
http://developer.apple.com/technotes/tn2002/tn2078.html
This tech note, dated May 2003, didn't exist when I needed the
information it contains. Of course, this should be obvious since I
wrote TN2078.)
Even now, some five years after the introduction of Mac OS X, new
developers still struggle with the documentation. Some basic APIs still
have no documentation, and documentation is spread out over web pages,
PDF files, header files, tech notes, Q&As, and sample code, often
without cross referencing. Inadequate and difficult-to-access
documentation wastes our time and increases our frustration.
- Stability. I started with Mac OS X in 10.1 and I have consistently
found Mac OS X to have more problems than I believe it should have.
(Thank goodness I didn't bother trying to write for the second public
beta, aka Mac OS X 10.0.) Not that I expect perfection, but more often
than I care to remember I've written code to use a technology or
feature I haven't used before, only to find out it didn't work. I've
lost count of all the times someone has posted a question to Apple's
carbon-dev list only to find out that some API or feature just plain
doesn't work. For example, if you read the HIToolbox release notes for
10.3.5, you'll see that there were two Nav Services APIs that were
nothing but stub functions from 10.0 until they were finally written
for 10.3.5 (and this primarily because I had brought them up on the
carbon-dev list). It's obvious to me (and confirmed to me by people
that have been at Apple) that Apple does not adequately test Mac OS X.
How many versions of the Mac OS prior to Mac OS X had eight and nine
bug-fix releases? This has been a huge time sink hole for me and other
developers I know, and it's largely the result of Apple's unwillingness
to hire adequate QA engineering staff. "Write lots of code fast and let
developers tell us when stuff doesn't work" seems to be Apple's policy
since Mac OS X came out, and I don't appreciate it because it has made
me consistently less productive over the past four years, and reduced
productivity translates directly into reduced revenues. Thank you
Apple. Glad iPod sales are going so well.
- Interface Builder. Ugh. A great tool in concept, and it's useable,
but Apple has really dropped the ball with IB. I didn't start using IB
until Panther, but when I did I found lots of consistently reproducible
bugs and odd limitations (for example, it can't import menus from
'MENU' resources). I filed Radar after Radar (well over a dozen)
against the bugs and glaring limitations and Apple never released a
single update for the Panther version of IB. More of my time wasted,
this time because some bean counter at Apple decided IB didn't warrant
the resources needed to address its many issues.
- Tech support. Technical support incidents used to cost $50. Maybe
Apple could see DTS being overwhelmed due to the above problems, but
for whatever reason, technical support incidents now cost $195. Skimp
on documentation, give us more bugs to work around, and quadruple the
costs of getting serious help. Thank you Apple.
- Transitions. I ported to Carbon. I ported to Mac OS X (which is,
sorry to say, not the same thing). I converted to .strings files and
nibs, to FSRefs and Unicode, from Pascal strings to CFStrings. Now I'll
have to switch to Xcode and do whatever is necessary to support Intel
processors. Some of us are tired of transitions. They require time and
resources that more often than not don't translate to readily apparent
improvements in products. For example, I switched from 'STR#' resources
to .strings files. Do you really think my English users (the majority
of my customers) really care?
The biggest single thing I'd like to see Apple do is start respecting
our time by giving us more documentation, an OS that works more
reliably when we write code for it, and well thought out tools that
work reliably. As long as they're too cheap to do those things, then
no, I'm not going to be happy with them.
Larry
la...@skytag.com
>> newbies. There's no technical reason it couldn't, but for whatever
>> reasons, open source software never sees wide adoption.
>
> Hmmmmm-mmmmm ... Fire and Adium are both open source programs, and they seem
> to be more widely adopted than proprietary IM programs, from everything I've
> seen.
- Maybe it's just me, but from where I sit, two applications hardly
constitute a trend or form the basis for a generalization. ;-)
- These are clearly targeted at non-technical users. IMO, an open
source development tool would be far more likely to suffer from the
misconception that developers don't care about interface or ease of
use.
Larry
Scott Moore
> Each of us has his own take on this issue, but I for one have been very
> frustrated since the introduction of Mac OS X on multiple fronts:
>
> - Documentation: Apple dumped a major new OS on us virtually
> undocumented. The state of documentation is getting better as time goes
> on, but it is still far from ideal and it used to be all but
> non-existent for much of what was new in Mac OS X. (Case in point: Take
> a look at Tech Note 2078:
Thanks for saying that, I thought it was just me. I had/have a full
collection of the "inside Mac" series. When OS 9 happened, I got the
"inside mac" series off line, .PDF, and printed it. Now every
damm thing I need for OS X seems to be either in a short, and very
non-comprehensive .PDF, or WORSE, in some damm web page that looks
like crap printed out. So up to OS 9, I got thick binders, now
I get lots of little "articletts" which look crappy even when put
together in a binder. Apple seems to be following Microsofts
example in putting all their APIs on line, but in a non-printer
friendly format. Its not good. Books are not just about printed
and bound matter, they are (or were) about organizing information
coherently.
I suppose there is nothing to really complain about. My best source
for Mac docs is not the online stuff, but the books that are printed
now for OS X, which is the same situation as Windows (ooopss, just
revealed myself as one of those nasty cross developers). However,
Apple used to ROCK for documenation, and its a real comedown. I have
a copy of Inside Mac with color illustrations. What a highpoint !
What happened ?
Scott Moore
Scott Ribe
la...@skytag.com at la...@skytag.com wrote on 6/28/05 5:14 PM:
>> Open-source management is hard and requires constant vigilance. So far
>> as I can tell, however, Apple have
>
> BTW, "Apple" is singular, so it's "Apple has."
British usage is have when the subject is a corporation.
bols...@hotmail.com
>- Documentation: Apple dumped a major new OS on us virtually
>undocumented. The state of documentation is getting better as time goes
>on, but it is still far from ideal and it used to be all but
>non-existent for much of what was new in Mac OS X. (Case in point: Take
I don't know; I worked on Mac OS X since DP4 came out. I could find the
documentation I needed. I asked questions on email lists when I needed.
So our experiences don't seem to match. And I used Carbon, Mach-O, java,
Obj-C, C++, AppKit, and PB / Xcode. I even used makefiles, otool, and nm.
Even perl... Ugh.
I've also coded from Mac OS 7 on, and I find the docs for Mac OS X to actually
be better, as they aren't all in pascal code I have to convert to C or C++ in
my head! ;)
Alwyn
la...@skytag.com wrote:
>
> - I believe we should have a development tool available that's easy to
> learn and use, not only for me, but to attract more people to Mac
> software development. I don't see that as a goal of open source
> software. It simply isn't, in my experience, software that attracts
> newbies. There's no technical reason it couldn't, but for whatever
> reasons, open source software never sees wide adoption. I think I can
> safely say that if my only option was open source software when I
> started writing Mac software, I would never have started.
This just isn't true. Microsoft Internet Explorer is still by far the
dominant Web browser, yet use of Firefox is steadily growing. On the Mac
side, the related Camino has received a lot of praise for its well
designed user interface.
> - Part of the problem with wide adoption may be the lack of support. CW
> has historically had MWRon and e-mail support. Xcode has the xcode list
> at Apple. Who's going to support the open source version? Open source
> software has always struck me as leaving you on your own much more than
> products like CW and Xcode, where the developers have a vested interest
> in seeing the product widely adopted.
Support is very important, but I see no reason to think that an
open-source Xcode would get worse support than the current one. The
open-source Fink, for instance, is outstandingly well supported, in my
opinion.
<http://fink.sourceforge.net/>
<snip>
> BTW, "Apple" is singular, so it's "Apple has."
I think it depends on where you went to school. It's like telling me
that 'colour' doesn't have a 'u' in it.
<snip>
> > Apple have said that the Mac OS X APIs have been stabilised as of the
> > release of Tiger.
>
> Really? That would be great news if it's true, but I hadn't heard that.
> Where did you see that?
I wish I could recall exactly and give you a link. The origin is Philip
Schiller, Avadis Tevanian or someone of equal authority. However, I can
quote John Siracusa as a fairly reliable secondary source:
> It also draws a line in the sand, vowing to maintain API stability from this
> point on. (No snickering, please.)
<http://siracusa.home.mindspring.com/john/articles/ars/>
> FWIW, I was told a long time ago by someone at
> Apple that Tiger was going to have fewer changes in it and would focus
> more on fixes and addressing problems than previous releases, but I
> don't think it worked out that way.
Well, what I think is true is that the most notable changes in Tiger
have been 'under the bonnet', so to speak (you would say 'under the
hood'). This is a point that Siracusa also makes in the review quoted
above, which is well worth reading in its entirety.
> > With the death of CodeWarrior, I suggest that this is
> > the right time for them to open-source their development tools in order
> > to enhance their goodwill within the developer community.
>
> I certainly agree they need to treat their developers better and with
> more respect, but I have other ideas for that than making Xcode open
> source. ;-)
There are indeed many other things Apple could do to make developers'
work easier. (I'm told Microsoft do a far better job in this respect
with their Developer Network.) One of my own biggest complaints is that
their sample code is often of deplorable quality and almost useless for
instructional purposes.
Alwyn
la...@skytag.com
>> undocumented. The state of documentation is getting better as time goes
>> on, but it is still far from ideal and it used to be all but
>> non-existent for much of what was new in Mac OS X. (Case in point: Take
>
> I don't know; I worked on Mac OS X since DP4 came out. I could find the
> documentation I needed. I asked questions on email lists when I needed.
The documentation issue is probably the single most universal complaint
I've seen on Apple's carbon-dev list over the past four years. The
issue is not whether or not you can find information or get code
written. The issue is documentation, and it hasn't been done in a
thorough and cohesive manner for Mac OS X by any stretch of the
imagination. E-mail lists are not documentation and they are not a
substitute for documentation. They're a great source for getting
questions answered, but they are not well suited for gaining a broad
understanding of technologies.
Too many people today, IMO, fail to study technologies to get a good
understanding before they write code. They grab some sample code or
find a likely looking API, learn as little as possible to get it to
compile, and then if they don't get the expected results they post a
question on some mailing list asking why it doesn't work or what API
they *should* be using. Someone answers, they change a few lines of
code and they're off to the next task. I've seen a steady stream of
such questions on the carbon-dev list for years now. I am convinced
this is not the best way to write code. It may get you working code
eventually, but working code is not synonymous with efficient,
well-designed code, or robust code.
On the contrary, it is my belief that code written without an
understanding beyond the minimum to needed to get it to work is a
significant source of bugs. All too often the programmer writes code
that works for him and he has no idea that said code *isn't* going to
work correctly if the user is using a different primary language, a
dual-processor Mac, a scroll wheel, unmounts and remounts a volume,
uses text input other than the keyboard, or any of a hundred other
things which could be different with another user. Robust, correct code
requires understanding more than what parameters you should pass to a
function and what that one specific function does. You need to
understand the big picture and for that you need real documentation.
> So our experiences don't seem to match.
Our experiences aside, the fact remains that Apple is taking *years* to
get real documentation out there, and one of the biggest single reasons
is that they won't hire enough tech pub writers. This is why so many of
the headers now contain documentation. It was taking so long for tech
pubs to get documentation out there, the engineers started documenting
directly in the headers. It was a stopgap measure implemented out of
practical necessity. Of course, the quality of said header
documentation varies widely from header to header, with some of them
providing excellent overviews and detailed remarks, and others
providing nothing.
Two or three years ago people were crediting the carbon-dev list and
Eric Schlegel in particular for the fact that they ever got a product
shipped on Mac OS X. That's just not right. We deal with it, but we
deserve better IMO. If I may indulge in a car analogy, when Chevrolet
releases new models, all of the dealer service departments get service
manuals for maintaining and repairing them. Not five years later or a
year later, but right away. What Apple has done is equivalent to
releasing a new model without service manuals and letting mechanics
figure out how to repair them through online mailing lists with other
mechanics. It's pathetic and shameful.
> And I used Carbon, Mach-O, java, Obj-C, C++, AppKit, and PB / Xcode. I even
> used makefiles, otool, and nm. Even perl... Ugh.
>
> I've also coded from Mac OS 7 on, and I find the docs for Mac OS X to actually
> be better, as they aren't all in pascal code I have to convert to C or C++ in
> my head! ;)
That doesn't make documentation better or worse. If you were using
Pascal today you would have the opposite perspective. When I started
programming == when I started programming the Mac, Pascal was much more
popular and in fact it was the language used in the introductory
programming classes at the local university here. I'd still take
documentation in Pascal over no documentation. No documentation is
essentially what we had in the beginning, and there are still
significant issues with what we have today.
If you're content to spend your time working around the problems in
Apple's documentation and tell yourself there isn't a problem, that's
your choice. But many, many of us acknowledge the problems while
working around them and really wish we didn't have to spend so much
time working around them.
Larry
Alwyn
bols...@hotmail.com wrote:
>
> At the WWDC Keynote SJ noted there were more developers than in some time,
> like 10 years. I can't recall exactly what he said.. Saying developers don't
> like Apple doesn't make sense when there are more of them...
What priceless logic!
Alwyn: Mohammed is a bad husband.
Bolsinga: How can you say that, he's got four wives!
Alwyn
Hendrik Schober
> [...]
> - Transitions. I ported to Carbon. I ported to Mac OS X (which is,
> sorry to say, not the same thing). I converted to .strings files and
> nibs, to FSRefs and Unicode, from Pascal strings to CFStrings. Now I'll
> have to switch to Xcode and do whatever is necessary to support Intel
> processors. Some of us are tired of transitions. They require time and
> resources that more often than not don't translate to readily apparent
> improvements in products. For example, I switched from 'STR#' resources
> to .strings files. Do you really think my English users (the majority
> of my customers) really care?
http://www.joelonsoftware.com/articles/APIWar.html
(See especially "Why Apple and Sun Can't Sell
Computers" about one 4th down the page.)
> [...]
> Larry
Schobi
--
Spam...@gmx.de is never read
I'm Schobi at suespammers dot org
"Coming back to where you started is not the same as never leaving"
Terry Pratchett
Brian
> Too many people today, IMO, fail to study technologies to get a good
> understanding before they write code. They grab some sample code or
> find a likely looking API, learn as little as possible to get it to
> compile, and then if they don't get the expected results they post a
> question on some mailing list asking why it doesn't work or what API
> they *should* be using. Someone answers, they change a few lines of
> code and they're off to the next task. I've seen a steady stream of
> such questions on the carbon-dev list for years now. I am convinced
> this is not the best way to write code. It may get you working code
> eventually, but working code is not synonymous with efficient,
> well-designed code, or robust code.
>
> On the contrary, it is my belief that code written without an
> understanding beyond the minimum to needed to get it to work is a
> significant source of bugs. All too often the programmer writes code
> that works for him and he has no idea that said code *isn't* going to
> work correctly if the user is using a different primary language, a
> dual-processor Mac, a scroll wheel, unmounts and remounts a volume,
> uses text input other than the keyboard, or any of a hundred other
> things which could be different with another user. Robust, correct code
> requires understanding more than what parameters you should pass to a
> function and what that one specific function does. You need to
> understand the big picture and for that you need real documentation.
I'd say the issue of building "robust, correct code" is a software
engineering problem, not an Apple documentation problem. For example,
if you build software where UI language strings are constants in code,
then you don't understand i18n, regardless of the target platform. Or
if you do not properly take threading into account and then are unable
to fully utilize multiple processors, you don't understand
parallelization, regardless of the target platform. So on an so forth.
For example, I had little trouble picking up PowerPlant 10 years ago,
already being familiar with similar UI frameworks. In addition, I was
able to literally understand and start building a Java-Cocoa app in a
day after only examining a few sample apps and using the API reference.
That's because I understand MVC and the other software engineering
principles it is built upon. The same for various other APIs on
various other systems.
Apple isn't a university, they are an OS, software and hardware vendor.
I agree, publishing an API with no documentation is one thing; a
problem Apple should fix. And it can't hurt Apple to publish more
examples, tutorials and further documentation. Sure, I'm all for it.
But Apple's business is not holding a course in software engineering,
and not teaching people with few engineering skills how to build
complex software.
-- Brian
la...@skytag.com
>> learn and use, not only for me, but to attract more people to Mac
>> software development. I don't see that as a goal of open source
>> software. It simply isn't, in my experience, software that attracts
>> newbies. There's no technical reason it couldn't, but for whatever
>> reasons, open source software never sees wide adoption. I think I can
>> safely say that if my only option was open source software when I
>> started writing Mac software, I would never have started.
>
> This just isn't true. Microsoft Internet Explorer is still by far the
> dominant Web browser,
Not on the Mac. That would be Safari. 81.9% of visitors to our web site
are using a Mac, 14.1% are using Windows, and the rest are using
something else or listed as unknown. 58% are using Safari, 19.9% are
using IE (probably a lot of the Windows users), and 17.7% are using
Firefox, which is pretty high for an open source application IMO, but
those may be more Windows people. Only 0.8% are using Camino.
> yet use of Firefox is steadily growing. On the Mac
> side, the related Camino has received a lot of praise for its well
> designed user interface.
Okay, I spoke too unilaterally. Widespread adoption is rarely the goal
of open source software and only a tiny percentage of open source
applications actually achieve any significant level of adoption on the
Mac. Camino's <1% doesn't qualify. ;-)
In any case, I'm making general statements about open source software.
The fact that there are some very rare exceptions doesn't really
invalidate my points.
>> - Part of the problem with wide adoption may be the lack of support. CW
>> has historically had MWRon and e-mail support. Xcode has the xcode list
>> at Apple. Who's going to support the open source version? Open source
>> software has always struck me as leaving you on your own much more than
>> products like CW and Xcode, where the developers have a vested interest
>> in seeing the product widely adopted.
>
> Support is very important, but I see no reason to think that an
> open-source Xcode would get worse support than the current one. The
> open-source Fink, for instance, is outstandingly well supported, in my
> opinion.
> <http://fink.sourceforge.net/>
Fink as another problem. Here's the first thing on that page:
============
The Fink project wants to bring the full world of Unix Open Source
software to Darwin and Mac OS X. We modify Unix software so that it
compiles and runs on Mac OS X ("port" it) and make it available for
download as a coherent distribution. Fink uses Debian tools like dpkg
and apt-get to provide powerful binary package management. You can
choose whether you want to download precompiled binary packages or
build everything from source. Read more...
News
2005-06-09: New Fink releases.
Three new Fink releases are available today: version 0.8.0 (for
10.4), version 0.7.2 (for 10.3), and version 0.6.4 (for 10.2). All
three releases include source files, binary packages, and a Fink
installer for new users.
============
How many non-geeks do you expect to use software described this way?
You just don't describe end-user software this way if you want adoption
among mainstream users. This is software by geeks for geeks.
Understand that I'm not saying open source software can't have a good
interface or good support. But in practice it's typically software by
geeks for geeks to one extent or another, and geeks care less about
interface issues and support.
> I wish I could recall exactly and give you a link. The origin is Philip
> Schiller, Avadis Tevanian or someone of equal authority. However, I can
> quote John Siracusa as a fairly reliable secondary source:
>
>> It also draws a line in the sand, vowing to maintain API stability from this
>> point on.
Of course, what is he really saying here? That the behavior of APIs
will stop changing out from under you? Wow, what an impressive feature,
and it's only taken them five years to give it to us. The whole idea of
APIs changing behavior from one release of the OS to another is just
insane to me and says that Mac OS X was half-baked when it was
initially released. Now, it's done baking and they're finally taking it
out of the oven. LOL
>> (No snickering, please.)
>
> <http://siracusa.home.mindspring.com/john/articles/ars/>
Great article. Thanks.
> with their Developer Network.) One of my own biggest complaints is that
> their sample code is often of deplorable quality and almost useless for
> instructional purposes.
Oh yeah, I forgot about the sample code. I have several issues with it
and use it only as a last resort. My biggest single issue is probably
that the 20 lines of code I need are usually buried in a source file
with a thousand lines of code, the other 980 lines of which are only
there to manage the mundane chores of running the sample application.
Larry
Alwyn
la...@skytag.com wrote:
>
> > This just isn't true. Microsoft Internet Explorer is still by far the
> > dominant Web browser,
>
> Not on the Mac. That would be Safari.
Like IE on Windows, Safari comes with the system. What do you expect?
> 81.9% of visitors to our web site
> are using a Mac, 14.1% are using Windows, and the rest are using
> something else or listed as unknown. 58% are using Safari, 19.9% are
> using IE (probably a lot of the Windows users), and 17.7% are using
> Firefox, which is pretty high for an open source application IMO, but
> those may be more Windows people. Only 0.8% are using Camino.
>
> > yet use of Firefox is steadily growing. On the Mac
> > side, the related Camino has received a lot of praise for its well
> > designed user interface.
>
> Okay, I spoke too unilaterally. Widespread adoption is rarely the goal
> of open source software and only a tiny percentage of open source
> applications actually achieve any significant level of adoption on the
> Mac. Camino's <1% doesn't qualify. ;-)
I believe Camino had a much bigger installed base before Safari was
introduced. In fact, Camino's lead developer was contemplating giving up
after he heard about Safari. Personally, I'm glad he didn't, since
choice is almost always a good thing for users.
Anyway, I mentioned Camino because it is widely acknowledged to have an
excellent, Mac-like user interface, which, you were saying, open-source
software never had.
> In any case, I'm making general statements about open source software.
> The fact that there are some very rare exceptions doesn't really
> invalidate my points.
Your points are well taken, but they do little to support your assertion
that open-sourcing Apple's development tools would be a bad idea. I
cannot seriously entertain the notion that an open-source Xcode would be
used by few, would have a terrible UI and would have inadequate support.
> >> - Part of the problem with wide adoption may be the lack of support. CW
> >> has historically had MWRon and e-mail support. Xcode has the xcode list
> >> at Apple. Who's going to support the open source version? Open source
> >> software has always struck me as leaving you on your own much more than
> >> products like CW and Xcode, where the developers have a vested interest
> >> in seeing the product widely adopted.
> >
> > Support is very important, but I see no reason to think that an
> > open-source Xcode would get worse support than the current one. The
> > open-source Fink, for instance, is outstandingly well supported, in my
> > opinion.
> > <http://fink.sourceforge.net/>
<Fink stuff snipped for brevity>
> How many non-geeks do you expect to use software described this way?
> You just don't describe end-user software this way if you want adoption
> among mainstream users. This is software by geeks for geeks.
>
> Understand that I'm not saying open source software can't have a good
> interface or good support. But in practice it's typically software by
> geeks for geeks to one extent or another, and geeks care less about
> interface issues and support.
You miss the point rather badly here. Xcode is also software by geeks
for geeks. A porting system like Fink requires an awful lot of support -
even the geekiest of us need to have our hand held from time to time.
(Personally, I'm too geeky to use Fink or DarwinPorts in the accepted
manner, but that's another story.)
> > I wish I could recall exactly and give you a link. The origin is Philip
> > Schiller, Avadis Tevanian or someone of equal authority. However, I can
> > quote John Siracusa as a fairly reliable secondary source:
> >
> >> It also draws a line in the sand, vowing to maintain API stability from
> >> this
> >> point on.
>
> Of course, what is he really saying here? That the behavior of APIs
> will stop changing out from under you? Wow, what an impressive feature,
> and it's only taken them five years to give it to us. The whole idea of
> APIs changing behavior from one release of the OS to another is just
> insane to me and says that Mac OS X was half-baked when it was
> initially released. Now, it's done baking and they're finally taking it
> out of the oven. LOL
This is a developer's perspective. From a user's point of view, Mac OS X
has been usable for some time now.
Actually, much the same could be said of Sun's Solaris when it switched
from being BSD-based (SunOS 4 or Solaris 1) to being System V based
(SunOS 5 or Solaris 2). When you change the foundations of an OS, it
takes a long time to stabilise.
Alwyn
la...@skytag.com
> software engineering problem,
>
how the APIs and technologies you're using work and
interact, both within your own software and with the system
as a whole?
> not an Apple documentation problem.
>
I'd say you're just refusing to see what's clearly there
out in the open. Read on.
> For example, if you build software where UI language
> strings are constants in code, then you don't understand
> i18n, regardless of the target platform.
>
I never said or even suggested that not having access to good
documentation is the *only* source of problems in software or even the
#1 source. I only said it was a significant source, as in more than a
tiny percentage.
> Or if you do not properly take threading into account
> and then are unable to fully utilize multiple processors,
> you don't understand parallelization, regardless of the
> target platform.
>
Okay, let's consider this from another angle. I was just
recently converting some operations in my application from
cooperative threads to MPTasks, which are preemptive. I
understand the issues, but I also understand that in order
to ensure I only call thread-safe APIs from a preemptive
thread I have know which APIs are thread safe, and that's a
documentation issue. It really is a documentation issue
because there simply is no documentation describing the
thread safety of each toolbox API. It's addressed for some
APIs in the headers, but not for all. I had to post
questions to the carbon-dev list and get responses from
Apple engineers to get some of the answers I needed. That
shouldn't be necessary, and in practice too many people
just substitute assumptions for accurate information when
they don't know where to find the accurate information and
the boss is wanting this code done when he comes back from
lunch. I'm not going to argue that making assumptions is
good software engineering, but in the real world it's a lot
more likely when real answers aren't readily available.
> So on an so forth.
>
Yes, and so forth. There are lots of ways you can implement
something incorrectly, but still have it appear to work
when you don't understand the big picture. This touches on
another common source of problems in software, the belief
that you can produce robust software by writing lots of bad
code and then clean up the bugs later. This misguided
notion leads people to think it's okay to make assumptions,
because if the assumption is wrong, they'll find out in
testing. The flaw in this logic should be obvious, but it's
an approach that's far too common.
In any case, the fact that there lots of ways to screw up
your code doesn't invalidate what I said.
> Apple isn't a university, they are an OS, software and
> hardware vendor.
>
I never suggested otherwise. GM isn't a university either,
but they publish service manuals in a timely manner, and
those manuals don't teach you how to use your tools.
> I agree, publishing an API with no documentation is one
> thing; a problem Apple should fix.
>
Well gee, isn't that what I've been saying?
> And it can't hurt Apple to publish more examples,
> tutorials and further documentation. Sure, I'm all for
> it. But Apple's business is not holding a course in
> software engineering, and not teaching people with few
> engineering skills how to build complex software.
>
Thank goodness that isn't what I asked for. Now if I just
had a clue why you went off in this direction...
Larry
la...@skytag.com
> introduced. In fact, Camino's lead developer was contemplating giving up
> after he heard about Safari. Personally, I'm glad he didn't, since
> choice is almost always a good thing for users
I agree 100%. The loss of CW means the loss of choice, and that's the
biggest loss of all IMO.
> that open-sourcing Apple's development tools would be a bad idea. I
> cannot seriously entertain the notion that an open-source Xcode would be
> used by few, would have a terrible UI and would have inadequate support.
Oh, I wasn't saying it would be bad for Xcode, although I have no
reason to believe Xcode's interface would improve or that support would
be any better than what Apple provides now. However, I'm not sure what
problems open sourcing Xcode would address.
>> interface or good support. But in practice it's typically software by
>> geeks for geeks to one extent or another, and geeks care less about
>> interface issues and support.
>
> You miss the point rather badly here. Xcode is also software by geeks
> for geeks.
Well yes, I know that. It's one of the reasons I'm so sorry to see CW
go away. Even within the ranks of developers there are differing
degrees of geekness, especially wrt to new developers.
>> will stop changing out from under you? Wow, what an impressive feature,
>> and it's only taken them five years to give it to us. The whole idea of
>> APIs changing behavior from one release of the OS to another is just
>> insane to me and says that Mac OS X was half-baked when it was
>> initially released. Now, it's done baking and they're finally taking it
>> out of the oven. LOL
>
> This is a developer's perspective.
Absolutely.
> From a user's point of view, Mac OS X
> has been usable for some time now.
Only because developers invest a lot of time and energy working around
the problems they encounter in Mac OS X. Apple shifted some of their
share of the work involved in producing robust software onto
developers. Some bean counter at Apple is getting a pat on the back for
not letting someone hire "unnecessary" QA engineers and we're all
working extra as a result. As a developer, the buck stops here. Apple
can ship Mac OS X with lots of bugs because developers will work around
them. We have to. If every time we encountered a bug in Mac OS X we
simply told users "That's a bug in Mac OS X," Mac OS X would have been
in a lot of trouble.
Larry
Brian
>> I'd say the issue of building "robust, correct code" is a
>> software engineering problem,
>>
> Do you think you can engineer software well without knowing
> how the APIs and technologies you're using work and
> interact, both within your own software and with the system
> as a whole?
Nope. But, say, an understanding of threading, parallelization and
deadlock is not the same as an understanding of Apple's -- or anyone
else's -- particular API or implementation.
>> I agree, publishing an API with no documentation is one
>> thing; a problem Apple should fix.
>>
> Well gee, isn't that what I've been saying?
Then why all the complaining about general problems you see in
software? Doing so you appear to be making some vague connection to
the docs Apple puts out. "Robust, correct code requires understanding
more than what parameters you should pass to a function .... You need
to understand the big picture and for that you need real
documentation." And the documentation is necessary so developers will
not "grab some sample code or find a likely looking API, learn as
little as possible to get it to compile, and then if they don't get the
expected results they post a question on some mailing list."
If a developer grabs some sample code and blindly tries to start
creating a complex application without any attempt at understanding the
underlying APIs they are using, it is their problem, not Apple's, or
Apple's documentation.
>> And it can't hurt Apple to publish more examples,
>> tutorials and further documentation. Sure, I'm all for
>> it. But Apple's business is not holding a course in
>> software engineering, and not teaching people with few
>> engineering skills how to build complex software.
>>
> Thank goodness that isn't what I asked for. Now if I just
> had a clue why you went off in this direction...
Hmmmm, how about "too many people today, IMO, fail to study
technologies to get a good understanding before they write code" and
"on the contrary, it is my belief that code written without an
understanding beyond the minimum to needed to get it to work is a
significant source of bugs" and so on. What this has to do with Apple
not documenting a particular API you know about and are having problems
with is beyond me.
Cheers.
-- Brian
Alwyn
My argument is that it would give its users more control over it than
they currently have. You seem to dislike InterfaceBuilder. Wouldn't you
like to fix the parts of it that annoy you most?
<snip>
> > From a user's point of view, Mac OS X has been usable for some
> > time now.
>
> Only because developers invest a lot of time and energy working around
> the problems they encounter in Mac OS X. Apple shifted some of their
> share of the work involved in producing robust software onto
> developers. Some bean counter at Apple is getting a pat on the back for
> not letting someone hire "unnecessary" QA engineers and we're all
> working extra as a result. As a developer, the buck stops here. Apple
> can ship Mac OS X with lots of bugs because developers will work around
> them. We have to. If every time we encountered a bug in Mac OS X we
> simply told users "That's a bug in Mac OS X," Mac OS X would have been
> in a lot of trouble.
This is very true. It's my impression, though, that Apple are
concentrating their efforts on making things easier for Cocoa
developers. (Not that Cocoa doesn't have its share of problems too.)
Unfortunately, a lot of people are compelled to use Carbon, for a
variety of reasons.
Alwyn
Scott Ribe
Alwyn at dt015...@mac.com wrote on 6/29/05 12:37 AM:
>>> Apple have said that the Mac OS X APIs have been stabilised as of the
>>> release of Tiger.
>>
>> Really? That would be great news if it's true, but I hadn't heard that.
>> Where did you see that?
>
> I wish I could recall exactly and give you a link. The origin is Philip
> Schiller, Avadis Tevanian or someone of equal authority. However, I can
> quote John Siracusa as a fairly reliable secondary source:
I think that comment was in reference to kernel-level APIs and driver
development.
la...@skytag.com
>> problems open sourcing Xcode would address.
>
> My argument is that it would give its users more control over it than
> they currently have. You seem to dislike InterfaceBuilder. Wouldn't you
> like to fix the parts of it that annoy you most?
Honestly? No. I'd like Apple to fix them. Unless of course Apple wants
to put an engineer to working on my code while I'm working on theirs or
they want to pay me outright. Apple is a multi-billion dollar
corporation with the best performing S&P500 stock in 2004. Why in the
world would I want to put development of my product-which pays my
bills-on hold so I can volunteer my time to fix Apple's tools? If,
OTOH, you'd like to fix it if it were open source, I'd be okay with
that. ;-) So in the case of IB, open sourcing would be good because
then at least someone could be working on it.
>> can ship Mac OS X with lots of bugs because developers will work around
>> them. We have to. If every time we encountered a bug in Mac OS X we
>> simply told users "That's a bug in Mac OS X," Mac OS X would have been
>> in a lot of trouble.
>
> This is very true. It's my impression, though, that Apple are
> concentrating their efforts on making things easier for Cocoa
> developers. (Not that Cocoa doesn't have its share of problems too.)
> Unfortunately, a lot of people are compelled to use Carbon, for a
> variety of reasons.
I think it's just a matter of degree. The Cocoa people have issues to
deal with as well, even if not as many. I don't use Cocoa myself, but I
get the impression working around problems in Cocoa is no more fun than
working around them in Carbon.
Larry
Alwyn
la...@skytag.com wrote:
> >> However, I'm not sure what
> >> problems open sourcing Xcode would address.
> >
> > My argument is that it would give its users more control over it than
> > they currently have. You seem to dislike InterfaceBuilder. Wouldn't you
> > like to fix the parts of it that annoy you most?
>
> Honestly? No. I'd like Apple to fix them.
Well yes, that's what you'd like. The reality is that IB has been
neglected, especially on the Carbon side. You are not in a position to
dictate Apple's priorities.
> Unless of course Apple wants
> to put an engineer to working on my code while I'm working on theirs or
> they want to pay me outright.
Hehehe, you wish!
> Apple is a multi-billion dollar
> corporation with the best performing S&P500 stock in 2004. Why in the
> world would I want to put development of my product-which pays my
> bills-on hold so I can volunteer my time to fix Apple's tools? If,
> OTOH, you'd like to fix it if it were open source, I'd be okay with
> that. ;-) So in the case of IB, open sourcing would be good because
> then at least someone could be working on it.
Actually, I would do that very gladly. And I bet I wouldn't be the only
one. There are quite a few things in IB that seem to me to require only
a very simple fix.
The point is that - at least in theory, given the community - these
applications would receive more developer time than they do currently.
Alwyn
la...@skytag.com
>>> they currently have. You seem to dislike InterfaceBuilder. Wouldn't you
>>> like to fix the parts of it that annoy you most?
>
>> Honestly? No. I'd like Apple to fix them.
>
> Well yes, that's what you'd like. The reality is that IB has been
> neglected, especially on the Carbon side. You are not in a position to
> dictate Apple's priorities.
Oh gee, thanks for telling me that. It explains a lot. LOL
>> Apple is a multi-billion dollar
>> corporation with the best performing S&P500 stock in 2004. Why in the
>> world would I want to put development of my product-which pays my
>> bills-on hold so I can volunteer my time to fix Apple's tools? If,
>> OTOH, you'd like to fix it if it were open source, I'd be okay with
>> that. ;-) So in the case of IB, open sourcing would be good because
>> then at least someone could be working on it.
>
> Actually, I would do that very gladly. And I bet I wouldn't be the only
> one. There are quite a few things in IB that seem to me to require only
> a very simple fix.
I know. It's been very frustrating.
>
> The point is that - at least in theory, given the community - these
> applications would receive more developer time than they do currently.
Well, given Apple's history with IB it wouldn't take much of a
community to do better than what we seen so far. But on a side note, am
I the only one made nervous by the the thought of reliance on Apple for
developer tools given they way they've handled IB? This is a big reason
I've hoped MW would stick with CW.
Maybe open sourcing would help Apple's tools, but in my mind that would
just be Apple punting on one more thing they should be handling but
aren't. But it's okay, I know in my heart when Steve says they're
selling lots of iPods that everything must truly be right with the
world. ;-)
Larry
bols...@hotmail.com
>> At the WWDC Keynote SJ noted there were more developers than in some time,
>> like 10 years. I can't recall exactly what he said.. Saying developers don't
>> like Apple doesn't make sense when there are more of them...
>
>What priceless logic!
>
>Alwyn: Mohammed is a bad husband.
>Bolsinga: How can you say that, he's got four wives!
Hehe LOL! :)
But you have to admit, no one is forcing these developers to sign up an pony up
$1500 for WWDC!
Alwyn
bols...@hotmail.com wrote:
>
> But you have to admit, no one is forcing these developers to sign up an pony
> up $1500 for WWDC!
Well, it means that 3800 people can justify and afford the expense of
attending WWDC, for whatever reason. This figure could certainly be used
to argue that there is currently a healthy interest in the Macintosh and
that some people believe there is money to be made from writing software
for it. I've never been myself, but I've no doubt that you can learn a
lot by going there.
What you cannot say is that good WWDC attendance means that the
relations between Apple and the developer community at large are
overwhelmingly positive; there is much evidence that points to the
contrary.
Alwyn
Paul
"Alwyn" <dt015...@mac.com> wrote in message
news:dt015a1979-BFE46...@text.news.blueyonder.co.uk...
Most of these people get sent by their companies. It would be interesting to
know how many of the 3800 represent interests that actually help keep the
Mac relevant. It would be really interesting to know how many x86
development systems they ship this year.
In 2002 I thought I might actually get to go.... The company was talking
about it, but it didn't happen and by the next year they'd stopped Mac
development and I was just another Windows guy (and very happy to have a
job!) I'm still all about Mac development on my own time, but the old
fun/excitement/passion about Apple has faded to the point that a pilgrimage
to WWDC would be just another business trip. I haven't been to MacWorld Expo
in years; I'd be more likely go to AdHoc (if I could afford that) than WWDC,
and spend the difference on DTS incidents.