Linux would have been useless with the GNU project. Even if the GNU project had never completed their Free OS kernel, the 386BSD branch was a full Free Software reimplementation of Unix. You might have arguably helped the world of personal computing by removing the Cult of Personality focus from feasibility. After all, Linus is easy to love when compared to most grizzled Unix wizards. I almost wish you could, just so I can find out if Stallman would have insisted on GNU/BSD. I doubt it, but I still am a bit curious. Another benefit is that we would not have the WWW as we know it today! And are you really implying that there aren't any Windows nerds? Most "Linux" nerds are just last year's Windows nerds.
An do you really think you could have defeated Jessie Owens?
Correct, DJBDNS is not Free Software. But then again, neither is Darwin, you theres not really any loss there. If Freedom is your concern, then perhaps you should try a different flavor of BSD. Or MkLinux if you wasnt to feel closer to the substructure of Darwin.
Of course, there will be lots of work to do to port DisplayPostscript to these other platforms. Perhaps GnuStep would be a good starting point for you.
What, you're not a developer? Then why do you care that DJBDNS isn't Free Software? It is still "free" for end users.
This is completely Off topic, but for some reason, this article is really calling out for the Penis Bird Troll.
Where are you when we need you? Really, I almost felt the need to crack some lame wood-pecker joke. That wouldn't have happened if you weren't sleeping on the job. Get to it!
Yeah, Palm won. Be engineers are prolific, and fully unrestricted with the Be IP acuisition as well.
Anyway, much of BeOS was Open Source and being a microkernel, the APIs were of most importance, and most of the APIs were open. That is all that could have helped anything anyway... Linux couldn't use the code used in the core of the BeOS as Linux can only use C code, not C++. BlueOS would not be maing more progress if the BeOS had been Open sourced. OpenBeOS OTOH, could have used more fre BeOS kits. They are progressing at a pretty good clip now anyway, and will be under a less restrictive license than the GPL which Be tended to use.
What did you try to do, that lack of apps prevented? Everything that a standard install of Windows and Mac OS includes came with it, as well as a large amount of GNU utilities. Meaning that anything you get with a Linux distro was also available.
No, Next comes Ce. After that, an improved version should be called Ceplusplus
This will cause much discussion over whether the next iteration should be called Pe or De.
Oh, damn. None of this applies, Because Be is BNU.
Re:Maybe those with hope in YellowTab will wake up
on
Be Throws in the Towel
·
· Score: 1
Kind of like how BSD was in Big Trouble with AT&T? Then 386BSD split up into NetBSD and FreeBSD? Really, OpnBSD and BlueOS are both distinct projects, but they both will share the BeOS APIs. Eventually FreeBSD and NetBSD had to expand and improve their respective projects, because 386BSD wasn't perfect. As long as BlueOS keeps to the order posted in your link, there won't be a problem, as BeOS functionality will be implemented before they try to expand upon the APIs. Just like OpenBeOS will reimplement BeOS before contiuing along the path of their "Glass elevator" BeOS API expansion project. As long as they stay faithful to the core common APIs, there is no problem with thinking about the future. After all, innovation shouldn't always be sacrificed for the sake of backwards compatibility, even Apple has advanced with their Carbon API, while still maintaining reasonable compatibility with classic Mac OS APIs. BeOS classic emulation is even an option, if you can use VMWare, WINE or even BOCHS to host the still available FreeBeOS download. Or you could host Yellowtab, which seems to be nothuing more than a slightly maintained "classic" BeOS.
The irony of these projects, is that the original goal of BeOS was to defiantly break backwards compaibility to free up future pursuits. Well, despite that, I still think Both BlueOS and OpenBeOS are useful projects.
No, They weren't an alternative OS for the Mac unti later.
First they were going to be an OS that could extract high end performance out of low end hardware. They lost when they gave up on their initial raison d'etre, and sold out completely to the Mac platform. The Mac should have been a complement to the BeBox, not the successor. And they should have kept closer ties to Power Computing Corporation. When PCC made their kickass x86 laptop, it should have been the first x86 hardware officially supported by Be. Tying themselves to former antagonist Apple, even without if Jobs wasn't in the picture, was stupid. It would have been like NVidia only making their Graphics cards available for SGI workstations. They should have kept their break from Apple clean. Supporting Mac PPC was a great idea, but only as a side dish, not the main course!
If Be kept their connections with at least PCC and UMAX alive, they might have had some avenue of early official x86 support. Then the Microsoft shenanigans would have been at least a bit less painful. When they were playing the Sony IA games, they should have struck a deal with Sony's desktop (and even laptop) line as well.
Unfortunately, the small fish of Be put thenselves in Apples pond, and were then driven out Intel's ocean. Be should have kept their BeBox alive.
I won't argue your assesment of the Amiga. But I will use it to counter another argument that many made about Be - that it would have been better had Apple bought then out. There is no reason to believe that Apple, which was on at least a ten year streak of mis-management and poor promotion, would have handled Be any better than ho Commodore handled Amiga. Perhaps it was a case of NIH syndrome, but I truly feel that Be was better off without Apple. It is clear that Apple was better off without Be, and I suspect that Commodore would have been better off without Amiga. The purchase of Amiga stressed Commodore's management problems to breaking point, whereas theiy had unprecidented success with their C=64 line. The C=128 provided a smooth technology transition, but was undercut by the announcement of the Amiga.
Note, for the record, that I am completely speaking out of my ass here. I was only a wee lad when all of this was going on, so I really don't know what I talking about. Please, flame away.
As for Be, the one thing they tried to do better than anyone else was to extract cutting edge hardware performance out of less than top of the line hardware. Their road to doing this was with consumer-priced SMP. They strayed from their path by selling out wholesale on their dream and commiting fully to Apple derived hardware, which screwed them in the end. If they kept the BeBox around, perhaps they could have had a safe haven when Jobs NeXTed Apple. It might have even been feasible to introduce an Intel capable BeBox with the funding Intel gave them, and even if Apple shut Be out of the G3 Macs, they could have just used an SMP G3 BeBox, which would have run circles around the Macs which weren't SMP capable. I guess Gassee got in bed with Beelzebub Amelio, and was surprised by Jobs returning with his red-hot poker.
Linux and NetBSD did fulfill a very similar task to the original intention of BeOS, but not quite the same. Linux and NetBSD (and FreeBSD for Intel) could resurrect "obsolete" hardware into usefulness again, but it could not derive high-end performance at low-end prices. Linux and FreeBSD have proven it can readily scale from low-end to high-end hardware, but for high-end performance (especially graphics performance) you need high end hardware (like DRI capable XFree4 supported video). Of course, with the more recent advances in threading and XFree, Linux and eventually BSD have the potential to extract BeOS level performance from lower end hardware, but it still has a ways to go.
Unfortunately, BeOS on x86 could support all of the hardware available, which lost sight of the original reason for BeOS to exist anyway. It's not quite a lesson in timing as you say... BeOS was also ahead of it's time, but it didn't seem feasible that hardware would advance so much that the high end became so high. In the end, commodity computing was so affordable that there was no reason to maximize use of cost-efficient processors. If anything, BeOS was a lesson against keeping all of your eggs in one basket, especially if that basket was weaved by a competetor with its own eggs. Especially if the weaver was a megalomanic like Mr. Jobs. Sure, the bunny left, but he came back again for Easter, with brightly colored iMacs and a cold vengeance.
In short, the best way to introduce a commercial OS seems to be with with the platform that it is dedicated to. If this hardware is specialized for a certain task, then it will have a niche market. If you can maintain the market, do so long enough that you have it cornered. Only when you own the mindshare should you expand into competition based markets, as long as you can avoid driving away your current market. Avoid driving away your current market as long as you can maintain any level of profitability. Of course, if everybody obeyed these rules, computing would never advance. So forget about it, and just gamble on the future. Sure, your company will de or be swallowed up whole, but it can be worth it if you're having fun, and you can leave your mark on history if you do it well.
You knew where to get IOMEGA Zip drives in the '80s? Man, I wish I had your connections!
Now Syquest, they were pretty cool with their removable drives. I never actually got to play with a Bernoulli box, but I'm pretty sure it was quite unlike a Zip drive.
And what's this about lack of hard drives? Mine has a hard drive. Of course, mine is also color, so maybe that HD stuff only came later. But that is most understandable... prices being what they were. In fact, I think that that was the primary NeXT mistake, the price gouging. Which might just be another way of saying that they offered technology that was "ahead of their time". Then again, looking at the history of Macintosh pricing, maybe that awful price tag was just a Steve Jobs feature.
Aye, this "30 yr old OS" is just a poor limited clone of Multics, which supposedly required hardware that was too powerful for its day. So now that our Microcomputers have processors capable of billions of cycles per second, RAM more abundant than most of the off line storage that was available, and hard drive storage capacity that can only be filled by massive copyright violation (or video editing), we are using the restricted clone of Multics. The name Unix (UNIX is not an acronym, the caps was a historical accident) was a reference to Multics, as "One of whatever Multics was many of" or "Multics without balls."
Of course I see your point, though computer science hasn't quite been as linear as your examples would indicate: Spoken language and transportation technology is usually a tangible progression. Multics isn't the only counter-example; consider the history of Lisp, older than most High Level Languages, and yet still more functional than most contemporary languages, which are still playing catch-up.
As for human language, I'm guessing you've never heard of Superl, have you? Now that is a language done right!
Seriously tho, tis a pity to lose another OS. Somesay this was partially Apple's fault?
And they'd be wrong. So Apple didn't share every motherboard spec with them. Big deal. Projects like *BSD did the same thing without Apple's help, Be could have of well. They picked the poor strategy of targeting to replace a 'good enough' solution on a proprietary platform. Then they switched to an open platfrom, and found that the 'good enough' solution there, while less good, was too entrenched. Signed their own death warrant.
Yes, this was definitely Apple's fault, though there is no reason for Apple to shed any tears over this. Jobs be damned, they pulled it off, and are now in a stronger position then they have been in a long time. Be was lured to the Mac platform by its promises of openness, as apple trying to leverage economies of scale like x86 hardware to reduce their prices. This was strongly evidenced by the Mac clones that were finally allowed to exist. In this environment, I actually haped that Be would resist being bought out by Apple, as they could have done better on their own.
Unfortunately, with Apple's purchase of NeXT, they got Jobs, who killed the clones and closed the platform. Without the assistance of Apple (a hardware company, mind you) the Be could no longer support future Macs, and was prematurely forced into the x86 world - quite a contrast to the luxury of fixed hardware in the apple world, which made driver development almost trivial.
Not to say that it wasn't Be's fault for putting all of their eggs in one basket, and trusting Apple enough to discontinue their BeBox line. Be tried to be more cost-effective by depending on another company, and sufferred when the Openness of Gil Amelio's Apple was replaced by the notorious Steve Jobs. As distastefull as I find Jobs, his decimation of the nascent Open-Apple did help Apple in the long run. Too bad I preferred the open Apple better. Now instead of open hardware, we have "Open Source" Darwin, which is definitely not Open Source in spirit (Read: Free Software").
Speaking of Free Software, Yes, the NetBSD project made considerable progress in supporting Macintosh hardware without Apple's help. Of course, there are differences between an academic environment without deadlines, profitability requirements with razor thin margins or a board of directors, and Be inc. No, Be could not have achieved what NetBSD did, and if they had tried, Apple could have easily stopped it with a lawsuit - even a meritless one would have crippled Be beyond recovery. Apple didn't even have to worry about making that trivial effort... closing their specs was enough to send Be packing.
Be did not try to replace anything. They felt that breaking ties with legacy software could make an OS powerful on hardware that is usually considered anything but. They were right, and they proved this with thier original BeBox. Of course, after that proof of concept, they had enough encouragement that they thought they could expand into wider markets. While they were easily supportable on Apple hardware, their eye was more in the direction of Power Computing being the hrdware maker, so tht they could discontinue their own line of hardware to focus on the OS. This could have worked out great if Jobs had not come back into power and killed the clones. After that, Be was just struggling to tread water. While Be could be used to the exclusion of Mac OS or Windows, that was jst incidental,and not their intention. As for making BeOS Free Software, they did have large parts of the OS source code available, as well as most of the Add ons. They could do this while keeping the core "kernel" of the OS proprietary, one of the advantages of the MicroKernel architechure they used. Of course if they had open sourced it, BeOS would still be alive today. But Be Inc. would not be.
Steve sure did kill Be. Not because Apple didn't buy Be. For me, that would have been almost as bad as Be dying. After Jobs was brought back to Apple, Jobs killed off the "openess" of Apple under Gil Amelio. No more giving Macintosh specs to OS developers or licenses to clone makers. With the continued hardware support yanked from under them, BeOS was given the fatal blow before they had enough driver momentum to properly survive in the x86 world.
Steve didn't buy NeXT... he built it. After Apple's board of directors banished him, he started over, making specialized, easy to use, highle overpriced computer systems. This time he used a Unix style foundation. NeXTSTEP became OpenStep. Apple made OpenStep into WebObjects, Rhapsody, Rhapsody into OS X Server, etc.
Unfortunately, most clocks only have one degree of oscillation.
Perhaps my terminology is off, so let me put it another way. If the clock weren't nailed to the wall, but were instead free falling through space, then when it faces you....
Wait a second, that doesn't quite work right. The hands on a clock are independant of the direction the clock-face is facing. Well anyway, why would it be twice-a-year? Does 360 degrees of motion (rotation) include an AM/PM modulation? Wouldn't once a year make more sense based on your assumptions? I would guess that even once-a-year is overly generous, as that would imply that the sattelite was in some kind of (heliosyncronous?) orbit with the earth. But if that orbit were equidistant to the earth's orbit from the Sun, wouldn't the Satelite eventaully interface with the Sun with unfortunate result?
Now completely off topic at this point, can you imagine how crazy it would be be to have a two dimentional analog clock? Or rather a 3 dimentional clock, as standard single plane clocks already use 2 spatial dimentions. Would a 3 dimensional clock utilize 2 planes or 3? No wait, it might be infinite. Oh yeah, and that's not counting any temporal representaion of dimension, e.g. the 4th dimension of time...
Jeez, did you ever notice how sleep deprivation is almost as good as drugs? I wonder if this will make any sense next time I read it?
Anyway, as I was saying, a mechanically viable clock cannot deviate from it's single plane. It's a single axis of variance, as opposed to a free-flying satellite with AFAIK 2 axises of variance.
It sounds almost like it could be a dirty-movie spin-off of the original Star Trek movie (Which was on HBO tonight, incidentlly).
Except this improved version would have an important change. In the original, 7th Heaven's Pastor Camden got to "interface" with the bald "V-ger probe" chick. Now imagine the possibilities of a successor movie, with Pastor Camden's oldest daughter Mary and the "p-neer probe"...
...and instead of Leonard Nimoy, Jolene Blalock - "Your interpretation of 'probe' is most illogical. No, you misunderstand, the mind meld requires my hand on your face, not..."
***ack*** "Status report." "Captain, the poster has suffered neurological damage. Almost as if his head were about to explode!"
That is, if the pen is used in a way to gather massive public support.
Similarly, the microphone is mightier than the handgun.
Again, the microphone at least provides the option of being less obnoxious than the handgun. Distributed systems have a far greater potential power, if harnessed effectively, than the isolated device. While higher tech may allow the isolated device disproportionately more power, it also provides greater potential for communication. The freedom fighter is more powerful than the terrorist.
The salient issue is this : Which is mightier, the Dollar or the microphone? Unfortunately that question is more convoluted than it seems.
64 bits local memory address plus 128 bits of IPv6 address.
So you could have a pointer to memory location X on IP address Y. Distributed memory access over a network. -- Just because it CAN be done, doesn't mean it should!
Would that be considered irony? Did you just damage your own case?
Well, in any case, I'm probably not as knowledgeable as many of the respondents who felt your case needed stronger opposition... but I still think it might be viable.
Large scale SMP computers like the Origin 3000 series can be "partitioned" into multiple system images, so that they act as multiple independant computers. Inversely, multiple discrete machines can be linked with their low latency high thoroughput "Numa-link"(sp) cables so that they can be clustered into one system image.
Now while the current Internet is hardly up to spec for such a system, wouldn't having a large scale distributed machine cluster behaving as a single system image require a single large addressable space for memory? Please excuse the run-on, but wouldn't this be a requirement for a large scale vertical cluster (as opposed to a horizontal, discrete component beowolf-style cluster?)
I value clueful response over clueful moderation. I value ignorant ignorance over ignorant moderation.
I sure hope they ironed all the wrinkles out... or some skiiers might find themselves travelling across some strange stretches of the universe - "Hey, why are there wormholes on the bunny slopes?"
(okay... that sounds like a reference to A Wrinkle in Time. If that was a pun, it wasn't intentional, I swear! I didn't even use the word tesseract!)
While you have a good point, something important to keep in mind is redundancy. A single copy of an important document on a piece of paper is not as certain as 1000 copies on CD-R. Bleeding ink, Mildew and mold, acid deterioration, fire, and wear and tear will all affect the single paper document. Even stone can be subject to ravages of weathering and age. Many historically significant structures are in danger from the effects of pollution and weather, especially so in recent decades with the greater prevalence of threats as Acid Rain.
The best way to ensure longterm storage and useablity is reproduction and redundancy and repetition and redundancy. Sure, if you have an important document on you computer, reproduce it to many forms... and be aware of the limitations of each. Floppies are readily available and compact, but are very volitile. CD-R when properly cared for can be a very viable long term storage, but is still subject to light and heat damage. Even printing out to paper is not guaranteed... the ink of most ink jet printers will fade over a few years.
Despite the weaknesses of "high-tech" solutions, it is not wise to limit yourself to lower tech either. One hand written document is very vulnerable. Mass production through printing-press technology is more resillient, and may be worth the added expense. Mass production through digital means can be much more cost effetive. Volatility in the various forms of storage can be countered by the ease of reproduction.
Also, realize that for any of the REALLY important stuff, maintenance and retransfer will always be required, ad inifinitum. It it is not worth the effort, then perhaps it is not worth keepig. Eventually, everything will die. Even if by some fluke, something is preserved by unintentional means... it may be unearthed by some other culture far off in the distant future, but the original intented meaning will forever be lost. Even with the Rosetta stone, we cannot fully comprehend the subtle nuances of a dead language. Forensics may provide us with many curios, but wil always fall short of comprehension.
Interestingly enough, this is similar to the way most word-processing documents are written on the BeOS.
The document is just plain text, but all of the formatting is stored as metadata.
The upshot, is that unlike most word processing documents, they are clearly readable with simple text editors. Even if you edit the document in a simple text editor, the formatting will remain coherent the next time the document is viewed in a word processor.
Of course this isn't as robust as TeX or such, because it relies on the metadata storing capabilities of the filesystem, and you may be limiting yourself to the BeFS (though there is no reason why NTFS or any filesystem like perhaps XFS or ReiserFS when taking advantage of Linux's VFS couldn't have similar functionality.) Even then , if you were to copy the document to another less enabled filesystem, you would only lose advanced formatting information. The body of text would still be fully useable.
Slashdot Mirror
Linux would have been useless with the GNU project. Even if the GNU project had never completed their Free OS kernel, the 386BSD branch was a full Free Software reimplementation of Unix. You might have arguably helped the world of personal computing by removing the Cult of Personality focus from feasibility. After all, Linus is easy to love when compared to most grizzled Unix wizards. I almost wish you could, just so I can find out if Stallman would have insisted on GNU/BSD. I doubt it, but I still am a bit curious. Another benefit is that we would not have the WWW as we know it today! And are you really implying that there aren't any Windows nerds? Most "Linux" nerds are just last year's Windows nerds.
An do you really think you could have defeated Jessie Owens?
Correct, DJBDNS is not Free Software. But then again, neither is Darwin, you theres not really any loss there. If Freedom is your concern, then perhaps you should try a different flavor of BSD. Or MkLinux if you wasnt to feel closer to the substructure of Darwin.
Of course, there will be lots of work to do to port DisplayPostscript to these other platforms. Perhaps GnuStep would be a good starting point for you.
What, you're not a developer? Then why do you care that DJBDNS isn't Free Software? It is still "free" for end users.
This is completely Off topic, but for some reason, this article is really calling out for the Penis Bird Troll.
Where are you when we need you? Really, I almost felt the need to crack some lame wood-pecker joke. That wouldn't have happened if you weren't sleeping on the job. Get to it!
It seems to me that the Freenet project makes no secret that it is still considered a project in development, and beta release quality at best.
.doc file generated by MS Word.
If this doesn't mean anything to you, then pretend that your "endpoint" will coincide with release 1.0.
After 1.0, the plan should be quite definite, and hopefully at least as static as any given
Yeah, Palm won. Be engineers are prolific, and fully unrestricted with the Be IP acuisition as well.
Anyway, much of BeOS was Open Source and being a microkernel, the APIs were of most importance, and most of the APIs were open. That is all that could have helped anything anyway... Linux couldn't use the code used in the core of the BeOS as Linux can only use C code, not C++. BlueOS would not be maing more progress if the BeOS had been Open sourced. OpenBeOS OTOH, could have used more fre BeOS kits. They are progressing at a pretty good clip now anyway, and will be under a less restrictive license than the GPL which Be tended to use.
Alpha beta gamma, easy as C flat, C, C# ...
What did you try to do, that lack of apps prevented? Everything that a standard install of Windows and Mac OS includes came with it, as well as a large amount of GNU utilities. Meaning that anything you get with a Linux distro was also available.
Or perhaps the lame horse was just projecting.
No, Next comes Ce. After that, an improved version should be called Ceplusplus
This will cause much discussion over whether the next iteration should be called Pe or De.
Oh, damn. None of this applies, Because Be is BNU.
Kind of like how BSD was in Big Trouble with AT&T? Then 386BSD split up into NetBSD and FreeBSD? Really, OpnBSD and BlueOS are both distinct projects, but they both will share the BeOS APIs. Eventually FreeBSD and NetBSD had to expand and improve their respective projects, because 386BSD wasn't perfect. As long as BlueOS keeps to the order posted in your link, there won't be a problem, as BeOS functionality will be implemented before they try to expand upon the APIs. Just like OpenBeOS will reimplement BeOS before contiuing along the path of their "Glass elevator" BeOS API expansion project. As long as they stay faithful to the core common APIs, there is no problem with thinking about the future. After all, innovation shouldn't always be sacrificed for the sake of backwards compatibility, even Apple has advanced with their Carbon API, while still maintaining reasonable compatibility with classic Mac OS APIs. BeOS classic emulation is even an option, if you can use VMWare, WINE or even BOCHS to host the still available FreeBeOS download. Or you could host Yellowtab, which seems to be nothuing more than a slightly maintained "classic" BeOS.
The irony of these projects, is that the original goal of BeOS was to defiantly break backwards compaibility to free up future pursuits. Well, despite that, I still think Both BlueOS and OpenBeOS are useful projects.
No, They weren't an alternative OS for the Mac unti later.
First they were going to be an OS that could extract high end performance out of low end hardware. They lost when they gave up on their initial raison d'etre, and sold out completely to the Mac platform. The Mac should have been a complement to the BeBox, not the successor. And they should have kept closer ties to Power Computing Corporation. When PCC made their kickass x86 laptop, it should have been the first x86 hardware officially supported by Be. Tying themselves to former antagonist Apple, even without if Jobs wasn't in the picture, was stupid. It would have been like NVidia only making their Graphics cards available for SGI workstations. They should have kept their break from Apple clean. Supporting Mac PPC was a great idea, but only as a side dish, not the main course!
If Be kept their connections with at least PCC and UMAX alive, they might have had some avenue of early official x86 support. Then the Microsoft shenanigans would have been at least a bit less painful. When they were playing the Sony IA games, they should have struck a deal with Sony's desktop (and even laptop) line as well.
Unfortunately, the small fish of Be put thenselves in Apples pond, and were then driven out Intel's ocean. Be should have kept their BeBox alive.
I won't argue your assesment of the Amiga. But I will use it to counter another argument that many made about Be - that it would have been better had Apple bought then out. There is no reason to believe that Apple, which was on at least a ten year streak of mis-management and poor promotion, would have handled Be any better than ho Commodore handled Amiga. Perhaps it was a case of NIH syndrome, but I truly feel that Be was better off without Apple. It is clear that Apple was better off without Be, and I suspect that Commodore would have been better off without Amiga. The purchase of Amiga stressed Commodore's management problems to breaking point, whereas theiy had unprecidented success with their C=64 line. The C=128 provided a smooth technology transition, but was undercut by the announcement of the Amiga.
Note, for the record, that I am completely speaking out of my ass here. I was only a wee lad when all of this was going on, so I really don't know what I talking about. Please, flame away.
As for Be, the one thing they tried to do better than anyone else was to extract cutting edge hardware performance out of less than top of the line hardware. Their road to doing this was with consumer-priced SMP. They strayed from their path by selling out wholesale on their dream and commiting fully to Apple derived hardware, which screwed them in the end. If they kept the BeBox around, perhaps they could have had a safe haven when Jobs NeXTed Apple. It might have even been feasible to introduce an Intel capable BeBox with the funding Intel gave them, and even if Apple shut Be out of the G3 Macs, they could have just used an SMP G3 BeBox, which would have run circles around the Macs which weren't SMP capable. I guess Gassee got in bed with Beelzebub Amelio, and was surprised by Jobs returning with his red-hot poker.
Linux and NetBSD did fulfill a very similar task to the original intention of BeOS, but not quite the same. Linux and NetBSD (and FreeBSD for Intel) could resurrect "obsolete" hardware into usefulness again, but it could not derive high-end performance at low-end prices. Linux and FreeBSD have proven it can readily scale from low-end to high-end hardware, but for high-end performance (especially graphics performance) you need high end hardware (like DRI capable XFree4 supported video). Of course, with the more recent advances in threading and XFree, Linux and eventually BSD have the potential to extract BeOS level performance from lower end hardware, but it still has a ways to go.
Unfortunately, BeOS on x86 could support all of the hardware available, which lost sight of the original reason for BeOS to exist anyway. It's not quite a lesson in timing as you say... BeOS was also ahead of it's time, but it didn't seem feasible that hardware would advance so much that the high end became so high. In the end, commodity computing was so affordable that there was no reason to maximize use of cost-efficient processors. If anything, BeOS was a lesson against keeping all of your eggs in one basket, especially if that basket was weaved by a competetor with its own eggs. Especially if the weaver was a megalomanic like Mr. Jobs. Sure, the bunny left, but he came back again for Easter, with brightly colored iMacs and a cold vengeance.
In short, the best way to introduce a commercial OS seems to be with with the platform that it is dedicated to. If this hardware is specialized for a certain task, then it will have a niche market. If you can maintain the market, do so long enough that you have it cornered. Only when you own the mindshare should you expand into competition based markets, as long as you can avoid driving away your current market. Avoid driving away your current market as long as you can maintain any level of profitability. Of course, if everybody obeyed these rules, computing would never advance. So forget about it, and just gamble on the future. Sure, your company will de or be swallowed up whole, but it can be worth it if you're having fun, and you can leave your mark on history if you do it well.
You knew where to get IOMEGA Zip drives in the '80s? Man, I wish I had your connections!
Now Syquest, they were pretty cool with their removable drives. I never actually got to play with a Bernoulli box, but I'm pretty sure it was quite unlike a Zip drive.
And what's this about lack of hard drives? Mine has a hard drive. Of course, mine is also color, so maybe that HD stuff only came later. But that is most understandable... prices being what they were. In fact, I think that that was the primary NeXT mistake, the price gouging. Which might just be another way of saying that they offered technology that was "ahead of their time". Then again, looking at the history of Macintosh pricing, maybe that awful price tag was just a Steve Jobs feature.
Aye, this "30 yr old OS" is just a poor limited clone of Multics, which supposedly required hardware that was too powerful for its day. So now that our Microcomputers have processors capable of billions of cycles per second, RAM more abundant than most of the off line storage that was available, and hard drive storage capacity that can only be filled by massive copyright violation (or video editing), we are using the restricted clone of Multics. The name Unix (UNIX is not an acronym, the caps was a historical accident) was a reference to Multics, as "One of whatever Multics was many of" or "Multics without balls."
Of course I see your point, though computer science hasn't quite been as linear as your examples would indicate: Spoken language and transportation technology is usually a tangible progression. Multics isn't the only counter-example; consider the history of Lisp, older than most High Level Languages, and yet still more functional than most contemporary languages, which are still playing catch-up.
As for human language, I'm guessing you've never heard of Superl, have you? Now that is a language done right!
Seriously tho, tis a pity to lose another OS. Somesay this was partially Apple's fault?
And they'd be wrong. So Apple didn't share every motherboard spec with them. Big deal. Projects like *BSD did the same thing without Apple's help, Be could have of well. They picked the poor strategy of targeting to replace a 'good enough' solution on a proprietary platform. Then they switched to an open platfrom, and found that the 'good enough' solution there, while less good, was too entrenched. Signed their own death warrant.
Yes, this was definitely Apple's fault, though there is no reason for Apple to shed any tears over this. Jobs be damned, they pulled it off, and are now in a stronger position then they have been in a long time. Be was lured to the Mac platform by its promises of openness, as apple trying to leverage economies of scale like x86 hardware to reduce their prices. This was strongly evidenced by the Mac clones that were finally allowed to exist. In this environment, I actually haped that Be would resist being bought out by Apple, as they could have done better on their own.
Unfortunately, with Apple's purchase of NeXT, they got Jobs, who killed the clones and closed the platform. Without the assistance of Apple (a hardware company, mind you) the Be could no longer support future Macs, and was prematurely forced into the x86 world - quite a contrast to the luxury of fixed hardware in the apple world, which made driver development almost trivial.
Not to say that it wasn't Be's fault for putting all of their eggs in one basket, and trusting Apple enough to discontinue their BeBox line. Be tried to be more cost-effective by depending on another company, and sufferred when the Openness of Gil Amelio's Apple was replaced by the notorious Steve Jobs. As distastefull as I find Jobs, his decimation of the nascent Open-Apple did help Apple in the long run. Too bad I preferred the open Apple better. Now instead of open hardware, we have "Open Source" Darwin, which is definitely not Open Source in spirit (Read: Free Software").
Speaking of Free Software, Yes, the NetBSD project made considerable progress in supporting Macintosh hardware without Apple's help. Of course, there are differences between an academic environment without deadlines, profitability requirements with razor thin margins or a board of directors, and Be inc. No, Be could not have achieved what NetBSD did, and if they had tried, Apple could have easily stopped it with a lawsuit - even a meritless one would have crippled Be beyond recovery. Apple didn't even have to worry about making that trivial effort... closing their specs was enough to send Be packing.
Be did not try to replace anything. They felt that breaking ties with legacy software could make an OS powerful on hardware that is usually considered anything but. They were right, and they proved this with thier original BeBox. Of course, after that proof of concept, they had enough encouragement that they thought they could expand into wider markets. While they were easily supportable on Apple hardware, their eye was more in the direction of Power Computing being the hrdware maker, so tht they could discontinue their own line of hardware to focus on the OS. This could have worked out great if Jobs had not come back into power and killed the clones. After that, Be was just struggling to tread water. While Be could be used to the exclusion of Mac OS or Windows, that was jst incidental,and not their intention. As for making BeOS Free Software, they did have large parts of the OS source code available, as well as most of the Add ons. They could do this while keeping the core "kernel" of the OS proprietary, one of the advantages of the MicroKernel architechure they used. Of course if they had open sourced it, BeOS would still be alive today. But Be Inc. would not be.
Steve sure did kill Be. Not because Apple didn't buy Be. For me, that would have been almost as bad as Be dying. After Jobs was brought back to Apple, Jobs killed off the "openess" of Apple under Gil Amelio. No more giving Macintosh specs to OS developers or licenses to clone makers. With the continued hardware support yanked from under them, BeOS was given the fatal blow before they had enough driver momentum to properly survive in the x86 world.
Steve didn't buy NeXT... he built it. After Apple's board of directors banished him, he started over, making specialized, easy to use, highle overpriced computer systems. This time he used a Unix style foundation. NeXTSTEP became OpenStep. Apple made OpenStep into WebObjects, Rhapsody, Rhapsody into OS X Server, etc.
Unfortunately, most clocks only have one degree of oscillation.
Perhaps my terminology is off, so let me put it another way. If the clock weren't nailed to the wall, but were instead free falling through space, then when it faces you....
Wait a second, that doesn't quite work right. The hands on a clock are independant of the direction the clock-face is facing. Well anyway, why would it be twice-a-year? Does 360 degrees of motion (rotation) include an AM/PM modulation? Wouldn't once a year make more sense based on your assumptions? I would guess that even once-a-year is overly generous, as that would imply that the sattelite was in some kind of (heliosyncronous?) orbit with the earth. But if that orbit were equidistant to the earth's orbit from the Sun, wouldn't the Satelite eventaully interface with the Sun with unfortunate result?
Now completely off topic at this point, can you imagine how crazy it would be be to have a two dimentional analog clock? Or rather a 3 dimentional clock, as standard single plane clocks already use 2 spatial dimentions. Would a 3 dimensional clock utilize 2 planes or 3? No wait, it might be infinite. Oh yeah, and that's not counting any temporal representaion of dimension, e.g. the 4th dimension of time...
Jeez, did you ever notice how sleep deprivation is almost as good as drugs? I wonder if this will make any sense next time I read it?
Anyway, as I was saying, a mechanically viable clock cannot deviate from it's single plane. It's a single axis of variance, as opposed to a free-flying satellite with AFAIK 2 axises of variance.
Now that is what I would call a core dump!
Uhh... Gates, William?
Guess I'm shooting for some of your negativekarma,now.
I really do deserve the wrath of a moderator for this one. Sock it to me!
It sounds almost like it could be a dirty-movie spin-off of the original Star Trek movie (Which was on HBO tonight, incidentlly).
Except this improved version would have an important change. In the original, 7th Heaven's Pastor Camden got to "interface" with the bald "V-ger probe" chick. Now imagine the possibilities of a successor movie, with Pastor Camden's oldest daughter Mary and the "p-neer probe"...
...and instead of Leonard Nimoy, Jolene Blalock - "Your interpretation of 'probe' is most illogical. No, you misunderstand, the mind meld requires my hand on your face, not..."
***ack***
"Status report."
"Captain, the poster has suffered neurological damage. Almost as if his head were about to explode!"
The pen is mightier than the sword.
That is, if the pen is used in a way to gather massive public support.
Similarly, the microphone is mightier than the handgun.
Again, the microphone at least provides the option of being less obnoxious than the handgun. Distributed systems have a far greater potential power, if harnessed effectively, than the isolated device. While higher tech may allow the isolated device disproportionately more power, it also provides greater potential for communication. The freedom fighter is more powerful than the terrorist.
The salient issue is this : Which is mightier, the Dollar or the microphone?
Unfortunately that question is more convoluted than it seems.
-castlan
Well, in any case, I'm probably not as knowledgeable as many of the respondents who felt your case needed stronger opposition... but I still think it might be viable.
Large scale SMP computers like the Origin 3000 series can be "partitioned" into multiple system images, so that they act as multiple independant computers. Inversely, multiple discrete machines can be linked with their low latency high thoroughput "Numa-link"(sp) cables so that they can be clustered into one system image.
Now while the current Internet is hardly up to spec for such a system, wouldn't having a large scale distributed machine cluster behaving as a single system image require a single large addressable space for memory? Please excuse the run-on, but wouldn't this be a requirement for a large scale vertical cluster (as opposed to a horizontal, discrete component beowolf-style cluster?)
I value clueful response over clueful moderation.
I value ignorant ignorance over ignorant moderation.
Uh oh....
Is that the mouse they used in the testing of the The Perpetual ski slope?
I sure hope they ironed all the wrinkles out... or some skiiers might find themselves travelling across some strange stretches of the universe - "Hey, why are there wormholes on the bunny slopes?"
(okay... that sounds like a reference to A Wrinkle in Time. If that was a pun, it wasn't intentional, I swear! I didn't even use the word tesseract!)
This is much funnier than its parent! Too bad you didn't log in, AC.
While you have a good point, something important to keep in mind is redundancy. A single copy of an important document on a piece of paper is not as certain as 1000 copies on CD-R. Bleeding ink, Mildew and mold, acid deterioration, fire, and wear and tear will all affect the single paper document. Even stone can be subject to ravages of weathering and age. Many historically significant structures are in danger from the effects of pollution and weather, especially so in recent decades with the greater prevalence of threats as Acid Rain.
The best way to ensure longterm storage and useablity is reproduction and redundancy and repetition and redundancy. Sure, if you have an important document on you computer, reproduce it to many forms... and be aware of the limitations of each. Floppies are readily available and compact, but are very volitile. CD-R when properly cared for can be a very viable long term storage, but is still subject to light and heat damage. Even printing out to paper is not guaranteed... the ink of most ink jet printers will fade over a few years.
Despite the weaknesses of "high-tech" solutions, it is not wise to limit yourself to lower tech either. One hand written document is very vulnerable. Mass production through printing-press technology is more resillient, and may be worth the added expense. Mass production through digital means can be much more cost effetive. Volatility in the various forms of storage can be countered by the ease of reproduction.
Also, realize that for any of the REALLY important stuff, maintenance and retransfer will always be required, ad inifinitum. It it is not worth the effort, then perhaps it is not worth keepig. Eventually, everything will die. Even if by some fluke, something is preserved by unintentional means... it may be unearthed by some other culture far off in the distant future, but the original intented meaning will forever be lost. Even with the Rosetta stone, we cannot fully comprehend the subtle nuances of a dead language. Forensics may provide us with many curios, but wil always fall short of comprehension.
Interestingly enough, this is similar to the way most word-processing documents are written on the BeOS.
The document is just plain text, but all of the formatting is stored as metadata.
The upshot, is that unlike most word processing documents, they are clearly readable with simple text editors. Even if you edit the document in a simple text editor, the formatting will remain coherent the next time the document is viewed in a word processor.
Of course this isn't as robust as TeX or such, because it relies on the metadata storing capabilities of the filesystem, and you may be limiting yourself to the BeFS (though there is no reason why NTFS or any filesystem like perhaps XFS or ReiserFS when taking advantage of Linux's VFS couldn't have similar functionality.) Even then , if you were to copy the document to another less enabled filesystem, you would only lose advanced formatting information. The body of text would still be fully useable.
-castlan