When i was a boy, i had a copy of the 25th Anniversary edition of Playboy. What do i remember most about that issue? A long Shel Silverstein poem called "The Devil and Billy Markham".
And now, in the deepest bowels of Hell, Shel is forcing the Devil to either keep his own promise, or bend over and get buggered. The thought warms my heart.
Someday, my children will be old enough to be allowed to read Uncle Shelby's ABZ Book.
Can i easily drop in my choice of editors in Outlook? I can in pine. Or mutt. Or elm. Can i change pager behavior? I'd like to be able to use spacebar to page, rather than having to move from touch-typing to pagedown. For that matter, i'd like to not have to use a mouse at all, and easily access all commands, menus, and headers with a few keystrokes. I'd like to be able to filter which headers i do and do not see. I'd like to set my Reply-to: address. I'd like to easily access multiple POP mailboxes, and have replies appear to come from the address in the To: header. Mutt and fetchmail do that one just fine.
Can Outlook do ANY of these things? NO!! So don't give me crap about how "powerful" Outlook is.
When Micros~1 delivers ANY product with half the functionality of emacs, i'll use it! Imagine, a nearly complete set of filesystem functionality built with ftp (like ange-ftp), fully programmable language-specific editing capabilities, shell access from within editors... the mind boggles.
First off, memory footprint is a non-issue... it's almost the sort of FUD i've come to expect from IDC and other "analysts". Is 32M enough? That should comfortably run Linux, X, and Netscape - as it has done on PCs for a long time. How much would 32M cost at manufacturing time? $20? Less?
I've been toying with this very idea for a while now, and here are some of my thoughts on architecture...
CPU: StrongARM or PPC. Forget Pentium hogs or quasi-vaporware Java chips.
Ports: VGA, USB, IR, and maybe 1394 (for high-speed peripherals like disk drives). A cheap low-end chipset is enough to work at adequate resolution with any VGA monitor. Keyboard and mouse can work through IR or USB, no need for parallel, old serial, or PS/2. A really stripped-down box needs only USB and VGA.
Networking: Modem or ethernet, use PCMCIA.
OS: Linux, of course.
Drives: Think they could cram a non-development Linux setup into 340M? Use the IBM microdrives. Add a 1394 port, and attach any other drives you need. No need for a floppy (blech!), or even a CD-ROM, but the 1394 could support backup devices, DVD, whatever.
Power supply: Wall-wart. With a StrongARM and a microdrive, power requirements will be trivial.
Now, consider economies of scale on production. I'll bet this box could be built for under $100, sans monitor. At that price point, giving away PCs makes sense!
How about a RAID box built out of IBM's microdrives? Plugs right in, the size of a sandwich, holds more than a gig, can draw power right off the wire... woo hoo!
The reason Mindcraft is calling on Linux and Alan for help is to recover their own lost legitimacy. The mainstream tech media is openly calling the integrity of their work into question. If, as some worry, they try to attach Linus and Alan's names to another rigged test without giving them proper access to the test environment, then they risk being denounced by Linus and Alan themselves... which would almost certainly be reported in ANY press coverage. That would just expose the tests for what they really are - paid FUD.
What makes them different from us?
on
Why Kids Kill
·
· Score: 1
I find the psychology of society's reaction to these massacres more interesting than the psychology that led to the massacres. And after watching such things for a while, here is my conclusion: that when mass murders happen, the first thing society does is to try to differentiate the murderers from the rest of us. Whether they played Doom, or had a web page, or wore black, or listened to different music... whatever excuse we make, it's all about *how they are different from us*. Why? Because if we *don't* find some rationalizable excuse for their difference, then we have to face the potential for mass murder in each one of us.
Robert Anton Wilson, in his short story "Von Neumann's Second Catastrophe", gave an interesting theory about war... war isn't about killing enemy soldiers, it's about killing civilians, defenseless women, children, and old people. Killing someone who is already a direct threat to you isn't a big leap. But gunning down unarmed civilians (or torching them with napalm as we did in Vietnam, or blowing up their water supply as we did in Baghdad) is a real feat - the man who can do it has gone totally beyond morality, and at least for the moment has power beyond most of humanity. (ps: Robert Anton Wilson should be required reading for anyone who doesn't want to be a sheep the rest of their lives. Some of you might recognize the name "Frank Sullivan")
It wasn't Jobs' "infinite wisdom" that scrapped the Intel port... by 1990, Jobs was long gone from Apple, had nothing to do with its decisions, and owned only one share of stock. Those were the Sculley years, when Apple was being run into the ground by a marketing hack from Pepsi.
At that point, Jobs was busy with a little project called NeXT, which was producing almost-affordable workstations with the best GUI ever sold commercially (some of you might use AfterStep for a WM... that's a NeXTStep knockoff), a BSD Unix base, 17+" monitors, CD-quality stereo sound, Display PostScript (true wysiwyg)... in other words, a decade ahead of their time. If the NeXT had ever found a big enough market niche, the world would be very different today.
Actually, though, i don't fault the lack of an Intel port for the Mac's failure then. Intel hardware was not (IS not) sufficiently mature for MacOS. I personally lay the blame at not providing protected memory and preemptive multitasking for System 7, at a time when Apple was abandoning the older hardware that couldn't support it. Of course, Apple dedicated its next-generation resources to Pink (Taligent), which suffered Death By Design.
It's not downsizing that cut Apple's expenses, not at this point. What cut Apple's expenses in the long (current) run was the most sophisticated and effective supply-chain management in the industry. Apple is manufacturing Macs with only *one day* of inventory! That is by far the best performance of any computer manufacturer. And, considering new hardware has roughly the shelf-life of, well, an apple, vendors need to move it before it rots and has to be dumped at below cost. In other words, Apple isn't hemmorhaging money trying to unload last year's models to make warehouse space and working capital for this year's models - unlike the Apple of yesteryear, and unlike most PC manufacturers.
There is more to the new iMac and G3 lines than packaging. By dumping the old product lines, Jobs also dumped inefficient manufacturing techniques. Apple is now making just three different computers, varying only in CPU speeds, disk, and RAM - this also simplifies the manufacturing process. No, the success of Apple today didn't come from downsizing, it came from quality.
That being said, i completely agree that OS/X is desparately needed, and on EVERY new Mac (not just servers). The sad Mac, i think, is seeing such a wonderful interface on such wonderful hardware with such junky OS internals. It pains me to think that NT is significantly superior to MacOS internally. Personally, i think the beginning of the downhill slide for Apple was when they didn't put hardware memory protection and preemptive multitasking into System 7 (at a time when they were almost ready to drop the last of the 68000-based Macs). If they'd fixed the internal problems back then, i think MacOS could have dominated the market.
But that's neither here nor there. I don't think Apple is out of the woods yet, but they've definitely found a trail. In the end, Linux is FAR more likely to kill the Mac than Windows. Why? Because like the Mac, people LOVE Linux.
"We must harness the powerful new forces of technology, and use them to strengthen our oldest values -- to promote freedom, to educate our children, and to lift our families and our nations up."
- Al Gore
Well, Al... i DO use open source, this powerful new force of technology, to promote freedom, and to educate my children. And i do so love to hear my five year old son say he want to boot out of Windows and "play gnome" instead.:}
From a special section for kids in their privacy policy...
8. Be a good online citizen and do not do anything that hurts other people or is against the law.
Using the Open Source trademark without permission to promote something that violates their guidelines is clearly illegal, no?
Laugh at the site while you can, kids... i'm sure the cease-and-desist order is being Fed-Ex'd now.
You know what REALLY sucks, though? I'd *love* to have a candidate who both understands and promotes Open Source ideals! And i think it's a safe bet that as soon as someone gets the remotest clue there, this page will come down and no mention of it will be made again. No public apology, for sure! Instead, how about a nice page apologizing for not understanding what they got into, a link to opensource.org, and a more reasonable promotion policy?
Think Netwinder/Qube, folks. If AOL or anyone else decides to use Linux to compete head-to-head with Windows boxen for Internet appliances, there's no sense using Wintel hardware. After all, you're talking *millions* of boxes. Economies of scale and standardization apply.
Use a StrongARM, or some other dirt-cheap RISC CPU. Add a 15" monitor, one of the new IBM miniature disk drives, and an ethernet port. The whole thing should be no larger than a telephone, except for the monitor and keyboard (infrared makes sense here!), and should cost well under $100 sans monitor to build. At THAT price point, giving boxes away makes a lot of sense.:}
Actually, companies have to learn a lot about "protecting" those investments. All the compiler R&D in the world won't mean diddly squat to Compaq/Digital's bottom line if they can't move boxes. Already, they're going to have to sacrifice the compilers for free, or effectively free. It's not like they're gonna recoup the costs in software - compiler research is a loss leader, which pays for itself by making the boxes faster, which sells more hardware. So are commercial Unix implementations. Why do you think the vendors are all jumping on the Linux bandwagon? To lower the price of entry for their low-end hardware! The reason Unix workstations don't compete with PCs running NT price-wise is because you have to buy $10k worth of operating system and development tools to get full use of your $5k hardware (well, NT ain't really better, but it fools the rubes). Low-end users don't need the high-end features - no sense hot-swapping CPUs in a single-processor box. So give them Linux for free, and sell more hardware. The guys dropping half a million bucks on a server can afford to pay for the extras that come with proprietary Unix.
Remember also that for serious Open Source users, openness is itself a value. People will use gcc even if it's slower, for trust reasons. And keeping secrets breeds resentment, not to mention serious doubts about the quality of the code. And, as open source becomes more accepted in the business world, this will turn into a suit value as well as a geek value.
With this in mind, tell me again what it is i have to learn? If it's that the PHBs will live in a portable bombshelter of denial while running their own business into the ground, rather than actually *understanding* the changing needs of their customers, don't worry - i already understand that. And i can also understand not open-sourcing code that has legal restrictions due to contracts with other companies. But if you have a cogent argument that there is bottom-line business value in keeping their compiler research proprietary when it produces no income on its own and galls customers to boot, please make it. Simply telling me i'm ignorant isn't much of an argument.
Yeah, the DEC compiler is fast, fast, fast. But from a *configuration* point of view, well... it blows. I do cross-platform integration, and porting code from other unices to DEC took longer than all the other platforms put together. And it wasn't endian issues or any other such technical subtleties, either... it was just various bits of brain damage in the configuration. Non ANSI-compliant headers, for example, or the fact that their linker couldn't resolve symbols that didn't cause problems on any other platform.
And then there's D/UX itself, which was apparently written by a bunch of bitter, resentful VMS coders seeking vengeance on Unix (I know, most of them wound up at Micros~1, writing NT!). Little things like the man command using a hardcoded path to nroff, which of course wasn't in the installed base (so no unprocessed man pages worked), and other PATH hardcodings. Or the which command under root only looking at the "official" root PATH, not the actual PATH in the shell. Which dovetailed nicely with the fact that sudo screwed up the environment for child processes...
But i digress. A sensible, user-tested environment like Linux will fix all these gripes. I just wish they would either open up the source to their compiler and libs, or work closely with gcc/egcs to provide their best tricks to the open source compilers. What reasons do they have to do otherwise? I assume they'll be giving the compilers away, or close enough to it, so they won't recoup their cost. The only things that might hold them back are patents (not the case, afaik), or NDAs with other software companies like KAI (who does the C++ front end for some vendor compilers).
If it's just the usual corporate Catholicism of keeping the Mysteries to themselves, then Compaq has a lot to learn.
The generally recognized term is "fair use", Jon, not "Free Use". But you're essentially right... reviewers have a legal right to quote reasonable excerpts of a publicly available work as part of their review. It's the same right that allows students to quote from books as part of term papers.
My point about revokability is that it CAN happen... regardless of circumstances or restrictions. Think of it this way - a revokable license is one where the original license holder can revoke your rights to any or all of the code, even if YOU have not in any way violated the license. An irrevokable license is one where the license holder may only revoke your rights if YOU PERSONALLY violate the agreement.
I agree that revokation clauses are pretty much necessary to protect corporate legal interests, and that they are unlikely to ever actually be used within the near future. Nonetheless, they exist, and as such limit my will, and the will of other developers, to work on the code.
I'm not some Stallmanesque radical insisting that corporations are bad, or that licenses such as the APSL are inadequate. I'm just saying that there is a real, significant distinction over revokability here, one that affects developers' will to work on a project as well as our personal political stances. And i am very, very afraid that if OSI does not explicitly address this, that it will become a wedge issue which forks the "code base" of OSI (the Open Source Definition). This will make the movement look bad in public, cause unnecessary friction, and force us all to choose sides.
Again, i really think the best thing to do would be to have multiple levels of "Open Source"... one for source which is open, but bound by the legal practicalities of corporations; and one for truly "free" software such as the GPL. Will OSI consider this? I hope so.
Time for a distinction between "Open" and "Free"
on
OSI APSL Response
·
· Score: 1
Bickering and animosity aside, we're encountering a fundamental issue here in "open source" licensing... severability clauses. The real distaste for both the Jikes and APSL licenses lies in the fact that the companies can revoke the licenses, and the work of independent developers will be lost. This is a valid concern for developers, who are spending their precious time (how many of us can and often do sell our time for over $100/hour?), and priceless emotional effort on open source projects.
The distinction i see here is between revokable (APSL, Jikes) licenses and irrevokable (GPL, Artistic) licenses. Developers will be much more willing to work on projects when the license guarantees their work will remain free. For terminology's sake, i call the two types "Open" and "Free" (as in free speech!).
I think the concerns about licenses which are merely "open" are valid and important. On the other hand, i don't want to see companies like Apple and IBM bashed for trying to do the right thing, within their limitations. Moreover, i do NOT want to see a split within the open source developer community over this issue - a split which will surely come if things are not changed.
I strongly suggest to the OSI that they consider making a distinction in the Open Source Definition between revokable and irrevokable licenses (open versus free), so developers know clearly where these corporate licenses are. If OSI does NOT do this, they risk losing credibility with developers (who have valid concerns, especially with self-appointed spokespersons), and also with the media and corporations (who will readily point to any infighting). At worst, we risk having a second "Open Source" certification team, one alienated and radicalized.
Someone at ZDNet observed that while porting Office is not impossible, Micros~1 will have problems with GPF and BSOD emulation. But i believe i have a workable solution to the GPF situation... use code profiling libraries to capture segmentation faults and bus errors, and jump from there to the familiar, user-friendly GPF interface that Windows users demand. Besides maintaining the consistent user experience that is Micros~1's trademark, this solution has another advantage... profiling should effectively emulate the interactive performance of Windows NT.
I'm still working on a BSOD emulation mechanism. So far, all i can come up with is a kernel patch, which isn't close enough to a full OS reinstall to satisfy traditional Micros~1 user demands for ease of use. However, it should be able to perform simple Windows functionality like freezing the mouse and disabling Ctrl-Alt-Delete.
Yesterday's "fox in the henhouse" essay, combined with the "bounty" article (Sun and Adobe offering money for open-source implementations of their pet concepts) made clear to me just how little many suits understand Open Source. See, true suits (not hackers in management positions) have a very difficult time understanding a gift culture. To them, things are measured in dollars. How often have suits and journalists (who don't get it either) said that they can't imagine good programmers giving away their work for free? My own take is this... *mediocre* programmers may only work for money. But the *good* programmers *must* hack, even for free, just as a good painter has to paint, even if she never makes a dime off it.
Back to the bounty... by offering a bounty for implementation, they unwittingly undermine the very things that make Open Source work. The bounty encourages code hoarding (so no one else can use your work), and rush work (finish before the deadline). These are the very things that undermine the quality of commercial code. Open Source is a *quality* win because of constant review, and the lack of time pressures. Get all the help you can, and release when it's really finished.
I think a lot of suits see Open Source as a free labor pool. Hence the bounties... paying a prize for an implementation is cheaper than hiring a couple of programmers to implement it in-house. And of course, no classic suits want to pay for the coders to write open source that their competitors can use. Adobe in particular looks bad here... by paying a bounty for tools for their highly proprietary PDF format, they look like they're bottom-feeding for cheap labor. If Adobe wants Open Source work around PDF, they can release a damn RFC documenting the format!
But i must say, it is promising that they're at least realizing the value of Open Source implementations of standards like XSL, and are willing to put up some money to encourage it. Now, we just need them to spend their money effectively. I'd suggest that they follow Red Hat's fine example - pay for a core development team that releases open, early, and often. This has worked *extremely* well for GNOME and Enlightenment - by paying for full-time developers who release GPL, Red Hat has vastly accelerated the already swift Open Source development process, without sacrificing the peer review that makes Open Source so effective. The resulting high-quality software could well lead to GNOME and Enlightenment becoming interface standards for Linux (and Unix in general - in the current climate, commercial vendors may well pick it up), providing the tools for the user-friendly desktop journalists keep telling us we need to compete. This same logic applies to tools like XSL... Sun would benefit if XSL became an industry standard, especially if built around a standardized Open Source implementation that discourages proprietary extensions and platform-dependent bugs. It would be worth Sun's while, i think, to fund Open Source development of implementations for standards like XSL. Hell, i can see an industry Open Source consortium, with business funding full-time Open Source development for standardized tools which benefit the whole industry. This could be what the quote Open unquote Software Foundation (OSF) dreamed of being...
At home, i use a PC Concepts ergo with a built-in trackpad. It's *wonderful*. The regular old split-layout ergo is much easier on the wrists, and the trackpad is even better, if you don't need too much precision. And the 6 key is in the right place for touch typists.:}
At work, i have one of the newer-model MS ergo keyboards. The keys themselves are better than the PC Concepts (as good as soft-touch gets), but the dinky arrow and home/pgup/etc keys blow. And the 6 is on the wrong side. And i'm using a wrist-killing MS mouse. But It's better than nothing.
If you don't like ergo keyboards, look for an ancient gen-u-wine IBM PC keyboard off an AT or RS/6000. *Beautiful* key touch, and they last forever. The only bad thing is they're noisy.
And if some of you would stop bashing Jobs and Disney for a moment, you'd see the logic.
Disney has been looking for a new CEO, since Eisner wants to scale back his job for health reasons. But Disney needs a visionary for a CEO, not the sort of egomaniacal bean-counter that is the mainstream CEO. Jobs is one of the great visionaries of American business (whether or not you agree with his vision), and is an effective fiscal officer as well (see the Apple turnaround).
The Pixar buyout is obvious, the Apple buyout less so. But think on these factors: first, Disney might want to get into the computer industry. It's a diverse company, and has hired some of the greatest living minds of the industry (Alan Kay, Danny Hillis, etc) just to have visions. The only existing computer company with the right cachet for Disney is Apple. Face it, Disney wouldn't just sell beige Wintel boxen. They've been toying with portals, but a unique and friendly PC is more, well, Disney. That's the modern Mac.
Second, and more importantly i think, they would own Quicktime. There's a lot of thought going on in Hollywood now about digital distribution of theater-grade films - digital media that rivals 70mm film in quality. Distributing 70mm reels is a HUGE expense. Digital media would be smaller, much cheaper to manufacture, and effectively immune to wear. And Moore's Law pretty much guarantees that it *will* be technically feasible within a few years. Whoever gets the technology to market first, and has the best backing, stands to make *billions* in distribution.
Apple's Quicktime is the best video compression technology currently available, and barring a surprise technical breakthrough, will remain so for years to come. And it may scale size/resolution well enough for film-quality video, given a good storage mechanism. This technology (and other Apple tech), combined with Disney's industry presence and marketing savvy, could put Disney in control of a key choke point for the entire film industry.
Stop thinking like a bunch of obsessive Anonymous Cowards and start thinking like Disney stockholders. This makes sense.
Cygnus is probably the oldest provider of *Open Source* commercial software. They maintain egcs and release it under GPL. What that $7000 buys is the sort of support that makes PHBs feel all warm and fuzzy inside - better than the support you'd get from most vendors, i might add. The pricing reflects Cygnus' primary market, which is commercial Unix developers. In that rarified world, compilers always cost that much. And, given the effort involved versus the size of the market, it makes sense. Microsoft charges what, two grand for a copy of Visual Studio Enterprise? And they'll move hundreds of thousands of copies.
Don't worry about that $7000. If you're not selling your software for money, you can get the tools for free. And if you want to run quality Open Source development tools at work, it's a lot easier to convince the PHB to pay $7000 than it is to convince him that free tools could possibly be as good as the $7000 stuff!
Slashdot Mirror
When i was a boy, i had a copy of the 25th Anniversary edition of Playboy. What do i remember most about that issue? A long Shel Silverstein poem called "The Devil and Billy Markham".
And now, in the deepest bowels of Hell, Shel is forcing the Devil to either keep his own promise, or bend over and get buggered. The thought warms my heart.
Someday, my children will be old enough to be allowed to read Uncle Shelby's ABZ Book.
Goodbye, Shel.
Can i easily drop in my choice of editors in Outlook? I can in pine. Or mutt. Or elm. Can i change pager behavior? I'd like to be able to use spacebar to page, rather than having to move from touch-typing to pagedown. For that matter, i'd like to not have to use a mouse at all, and easily access all commands, menus, and headers with a few keystrokes. I'd like to be able to filter which headers i do and do not see. I'd like to set my Reply-to: address. I'd like to easily access multiple POP mailboxes, and have replies appear to come from the address in the To: header. Mutt and fetchmail do that one just fine.
Can Outlook do ANY of these things? NO!! So don't give me crap about how "powerful" Outlook is.
*Half* the functionality of emacs?
When Micros~1 delivers ANY product with half the functionality of emacs, i'll use it! Imagine, a nearly complete set of filesystem functionality built with ftp (like ange-ftp), fully programmable language-specific editing capabilities, shell access from within editors... the mind boggles.
First off, memory footprint is a non-issue... it's almost the sort of FUD i've come to expect from IDC and other "analysts". Is 32M enough? That should comfortably run Linux, X, and Netscape - as it has done on PCs for a long time. How much would 32M cost at manufacturing time? $20? Less?
I've been toying with this very idea for a while now, and here are some of my thoughts on architecture...
CPU: StrongARM or PPC. Forget Pentium hogs or quasi-vaporware Java chips.
Ports: VGA, USB, IR, and maybe 1394 (for high-speed peripherals like disk drives). A cheap low-end chipset is enough to work at adequate resolution with any VGA monitor. Keyboard and mouse can work through IR or USB, no need for parallel, old serial, or PS/2. A really stripped-down box needs only USB and VGA.
Networking: Modem or ethernet, use PCMCIA.
OS: Linux, of course.
Drives: Think they could cram a non-development Linux setup into 340M? Use the IBM microdrives. Add a 1394 port, and attach any other drives you need. No need for a floppy (blech!), or even a CD-ROM, but the 1394 could support backup devices, DVD, whatever.
Power supply: Wall-wart. With a StrongARM and a microdrive, power requirements will be trivial.
Now, consider economies of scale on production. I'll bet this box could be built for under $100, sans monitor. At that price point, giving away PCs makes sense!
How about a RAID box built out of IBM's microdrives? Plugs right in, the size of a sandwich, holds more than a gig, can draw power right off the wire... woo hoo!
The reason Mindcraft is calling on Linux and Alan for help is to recover their own lost legitimacy. The mainstream tech media is openly calling the integrity of their work into question. If, as some worry, they try to attach Linus and Alan's names to another rigged test without giving them proper access to the test environment, then they risk being denounced by Linus and Alan themselves... which would almost certainly be reported in ANY press coverage. That would just expose the tests for what they really are - paid FUD.
I find the psychology of society's reaction to these massacres more interesting than the psychology that led to the massacres. And after watching such things for a while, here is my conclusion: that when mass murders happen, the first thing society does is to try to differentiate the murderers from the rest of us. Whether they played Doom, or had a web page, or wore black, or listened to different music... whatever excuse we make, it's all about *how they are different from us*. Why? Because if we *don't* find some rationalizable excuse for their difference, then we have to face the potential for mass murder in each one of us.
Robert Anton Wilson, in his short story "Von Neumann's Second Catastrophe", gave an interesting theory about war... war isn't about killing enemy soldiers, it's about killing civilians, defenseless women, children, and old people. Killing someone who is already a direct threat to you isn't a big leap. But gunning down unarmed civilians (or torching them with napalm as we did in Vietnam, or blowing up their water supply as we did in Baghdad) is a real feat - the man who can do it has gone totally beyond morality, and at least for the moment has power beyond most of humanity.
(ps: Robert Anton Wilson should be required reading for anyone who doesn't want to be a sheep the rest of their lives. Some of you might recognize the name "Frank Sullivan")
It wasn't Jobs' "infinite wisdom" that scrapped the Intel port... by 1990, Jobs was long gone from Apple, had nothing to do with its decisions, and owned only one share of stock. Those were the Sculley years, when Apple was being run into the ground by a marketing hack from Pepsi.
At that point, Jobs was busy with a little project called NeXT, which was producing almost-affordable workstations with the best GUI ever sold commercially (some of you might use AfterStep for a WM... that's a NeXTStep knockoff), a BSD Unix base, 17+" monitors, CD-quality stereo sound, Display PostScript (true wysiwyg)... in other words, a decade ahead of their time. If the NeXT had ever found a big enough market niche, the world would be very different today.
Actually, though, i don't fault the lack of an Intel port for the Mac's failure then. Intel hardware was not (IS not) sufficiently mature for MacOS. I personally lay the blame at not providing protected memory and preemptive multitasking for System 7, at a time when Apple was abandoning the older hardware that couldn't support it. Of course, Apple dedicated its next-generation resources to Pink (Taligent), which suffered Death By Design.
It's not downsizing that cut Apple's expenses, not at this point. What cut Apple's expenses in the long (current) run was the most sophisticated and effective supply-chain management in the industry. Apple is manufacturing Macs with only *one day* of inventory! That is by far the best performance of any computer manufacturer. And, considering new hardware has roughly the shelf-life of, well, an apple, vendors need to move it before it rots and has to be dumped at below cost. In other words, Apple isn't hemmorhaging money trying to unload last year's models to make warehouse space and working capital for this year's models - unlike the Apple of yesteryear, and unlike most PC manufacturers.
There is more to the new iMac and G3 lines than packaging. By dumping the old product lines, Jobs also dumped inefficient manufacturing techniques. Apple is now making just three different computers, varying only in CPU speeds, disk, and RAM - this also simplifies the manufacturing process. No, the success of Apple today didn't come from downsizing, it came from quality.
That being said, i completely agree that OS/X is desparately needed, and on EVERY new Mac (not just servers). The sad Mac, i think, is seeing such a wonderful interface on such wonderful hardware with such junky OS internals. It pains me to think that NT is significantly superior to MacOS internally. Personally, i think the beginning of the downhill slide for Apple was when they didn't put hardware memory protection and preemptive multitasking into System 7 (at a time when they were almost ready to drop the last of the 68000-based Macs). If they'd fixed the internal problems back then, i think MacOS could have dominated the market.
But that's neither here nor there. I don't think Apple is out of the woods yet, but they've definitely found a trail. In the end, Linux is FAR more likely to kill the Mac than Windows. Why? Because like the Mac, people LOVE Linux.
No, the "Open Source" crap is still there. I flushed my cache and restarted Netscape, and still get it.
"We must harness the powerful new
:}
forces of technology, and use them to
strengthen our oldest values -- to
promote freedom, to educate our
children, and to lift our families and our
nations up."
- Al Gore
Well, Al... i DO use open source, this powerful new force of technology, to promote freedom, and to educate my children. And i do so love to hear my five year old son say he want to boot out of Windows and "play gnome" instead.
From a special section for kids in their privacy policy...
8. Be a good online citizen and do not do anything that hurts other people or is against the law.
Using the Open Source trademark without permission to promote something that violates their guidelines is clearly illegal, no?
Laugh at the site while you can, kids... i'm sure the cease-and-desist order is being Fed-Ex'd now.
You know what REALLY sucks, though? I'd *love* to have a candidate who both understands and promotes Open Source ideals! And i think it's a safe bet that as soon as someone gets the remotest clue there, this page will come down and no mention of it will be made again. No public apology, for sure! Instead, how about a nice page apologizing for not understanding what they got into, a link to opensource.org, and a more reasonable promotion policy?
Nah.
Think Netwinder/Qube, folks. If AOL or anyone else decides to use Linux to compete head-to-head with Windows boxen for Internet appliances, there's no sense using Wintel hardware. After all, you're talking *millions* of boxes. Economies of scale and standardization apply.
:}
Use a StrongARM, or some other dirt-cheap RISC CPU. Add a 15" monitor, one of the new IBM miniature disk drives, and an ethernet port. The whole thing should be no larger than a telephone, except for the monitor and keyboard (infrared makes sense here!), and should cost well under $100 sans monitor to build. At THAT price point, giving boxes away makes a lot of sense.
Actually, companies have to learn a lot about "protecting" those investments. All the compiler R&D in the world won't mean diddly squat to Compaq/Digital's bottom line if they can't move boxes. Already, they're going to have to sacrifice the compilers for free, or effectively free. It's not like they're gonna recoup the costs in software - compiler research is a loss leader, which pays for itself by making the boxes faster, which sells more hardware. So are commercial Unix implementations. Why do you think the vendors are all jumping on the Linux bandwagon? To lower the price of entry for their low-end hardware! The reason Unix workstations don't compete with PCs running NT price-wise is because you have to buy $10k worth of operating system and development tools to get full use of your $5k hardware (well, NT ain't really better, but it fools the rubes). Low-end users don't need the high-end features - no sense hot-swapping CPUs in a single-processor box. So give them Linux for free, and sell more hardware. The guys dropping half a million bucks on a server can afford to pay for the extras that come with proprietary Unix.
Remember also that for serious Open Source users, openness is itself a value. People will use gcc even if it's slower, for trust reasons. And keeping secrets breeds resentment, not to mention serious doubts about the quality of the code. And, as open source becomes more accepted in the business world, this will turn into a suit value as well as a geek value.
With this in mind, tell me again what it is i have to learn? If it's that the PHBs will live in a portable bombshelter of denial while running their own business into the ground, rather than actually *understanding* the changing needs of their customers, don't worry - i already understand that. And i can also understand not open-sourcing code that has legal restrictions due to contracts with other companies. But if you have a cogent argument that there is bottom-line business value in keeping their compiler research proprietary when it produces no income on its own and galls customers to boot, please make it. Simply telling me i'm ignorant isn't much of an argument.
Yeah, the DEC compiler is fast, fast, fast. But from a *configuration* point of view, well... it blows. I do cross-platform integration, and porting code from other unices to DEC took longer than all the other platforms put together. And it wasn't endian issues or any other such technical subtleties, either... it was just various bits of brain damage in the configuration. Non ANSI-compliant headers, for example, or the fact that their linker couldn't resolve symbols that didn't cause problems on any other platform.
And then there's D/UX itself, which was apparently written by a bunch of bitter, resentful VMS coders seeking vengeance on Unix (I know, most of them wound up at Micros~1, writing NT!). Little things like the man command using a hardcoded path to nroff, which of course wasn't in the installed base (so no unprocessed man pages worked), and other PATH hardcodings. Or the which command under root only looking at the "official" root PATH, not the actual PATH in the shell. Which dovetailed nicely with the fact that sudo screwed up the environment for child processes...
But i digress. A sensible, user-tested environment like Linux will fix all these gripes. I just wish they would either open up the source to their compiler and libs, or work closely with gcc/egcs to provide their best tricks to the open source compilers. What reasons do they have to do otherwise? I assume they'll be giving the compilers away, or close enough to it, so they won't recoup their cost. The only things that might hold them back are patents (not the case, afaik), or NDAs with other software companies like KAI (who does the C++ front end for some vendor compilers).
If it's just the usual corporate Catholicism of keeping the Mysteries to themselves, then Compaq has a lot to learn.
The generally recognized term is "fair use", Jon, not "Free Use". But you're essentially right... reviewers have a legal right to quote reasonable excerpts of a publicly available work as part of their review. It's the same right that allows students to quote from books as part of term papers.
My point about revokability is that it CAN happen... regardless of circumstances or restrictions. Think of it this way - a revokable license is one where the original license holder can revoke your rights to any or all of the code, even if YOU have not in any way violated the license. An irrevokable license is one where the license holder may only revoke your rights if YOU PERSONALLY violate the agreement.
I agree that revokation clauses are pretty much necessary to protect corporate legal interests, and that they are unlikely to ever actually be used within the near future. Nonetheless, they exist, and as such limit my will, and the will of other developers, to work on the code.
I'm not some Stallmanesque radical insisting that corporations are bad, or that licenses such as the APSL are inadequate. I'm just saying that there is a real, significant distinction over revokability here, one that affects developers' will to work on a project as well as our personal political stances. And i am very, very afraid that if OSI does not explicitly address this, that it will become a wedge issue which forks the "code base" of OSI (the Open Source Definition). This will make the movement look bad in public, cause unnecessary friction, and force us all to choose sides.
Again, i really think the best thing to do would be to have multiple levels of "Open Source"... one for source which is open, but bound by the legal practicalities of corporations; and one for truly "free" software such as the GPL. Will OSI consider this? I hope so.
Bickering and animosity aside, we're encountering a fundamental issue here in "open source" licensing... severability clauses. The real distaste for both the Jikes and APSL licenses lies in the fact that the companies can revoke the licenses, and the work of independent developers will be lost. This is a valid concern for developers, who are spending their precious time (how many of us can and often do sell our time for over $100/hour?), and priceless emotional effort on open source projects.
The distinction i see here is between revokable (APSL, Jikes) licenses and irrevokable (GPL, Artistic) licenses. Developers will be much more willing to work on projects when the license guarantees their work will remain free. For terminology's sake, i call the two types "Open" and "Free" (as in free speech!).
I think the concerns about licenses which are merely "open" are valid and important. On the other hand, i don't want to see companies like Apple and IBM bashed for trying to do the right thing, within their limitations. Moreover, i do NOT want to see a split within the open source developer community over this issue - a split which will surely come if things are not changed.
I strongly suggest to the OSI that they consider making a distinction in the Open Source Definition between revokable and irrevokable licenses (open versus free), so developers know clearly where these corporate licenses are. If OSI does NOT do this, they risk losing credibility with developers (who have valid concerns, especially with self-appointed spokespersons), and also with the media and corporations (who will readily point to any infighting). At worst, we risk having a second "Open Source" certification team, one alienated and radicalized.
Think about it, OSI.
Someone at ZDNet observed that while porting Office is not impossible, Micros~1 will have problems with GPF and BSOD emulation. But i believe i have a workable solution to the GPF situation... use code profiling libraries to capture segmentation faults and bus errors, and jump from there to the familiar, user-friendly GPF interface that Windows users demand. Besides maintaining the consistent user experience that is Micros~1's trademark, this solution has another advantage... profiling should effectively emulate the interactive performance of Windows NT.
I'm still working on a BSOD emulation mechanism. So far, all i can come up with is a kernel patch, which isn't close enough to a full OS reinstall to satisfy traditional Micros~1 user demands for ease of use. However, it should be able to perform simple Windows functionality like freezing the mouse and disabling Ctrl-Alt-Delete.
Yesterday's "fox in the henhouse" essay, combined with the "bounty" article (Sun and Adobe offering money for open-source implementations of their pet concepts) made clear to me just how little many suits understand Open Source. See, true suits (not hackers in management positions) have a very difficult time understanding a gift culture. To them, things are measured in dollars. How often have suits and journalists (who don't get it either) said that they can't imagine good programmers giving away their work for free? My own take is this... *mediocre* programmers may only work for money. But the *good* programmers *must* hack, even for free, just as a good painter has to paint, even if she never makes a dime off it.
Back to the bounty... by offering a bounty for implementation, they unwittingly undermine the very things that make Open Source work. The bounty encourages code hoarding (so no one else can use your work), and rush work (finish before the deadline). These are the very things that undermine the quality of commercial code. Open Source is a *quality* win because of constant review, and the lack of time pressures. Get all the help you can, and release when it's really finished.
I think a lot of suits see Open Source as a free labor pool. Hence the bounties... paying a prize for an implementation is cheaper than hiring a couple of programmers to implement it in-house. And of course, no classic suits want to pay for the coders to write open source that their competitors can use. Adobe in particular looks bad here... by paying a bounty for tools for their highly proprietary PDF format, they look like they're bottom-feeding for cheap labor. If Adobe wants Open Source work around PDF, they can release a damn RFC documenting the format!
But i must say, it is promising that they're at least realizing the value of Open Source implementations of standards like XSL, and are willing to put up some money to encourage it. Now, we just need them to spend their money effectively. I'd suggest that they follow Red Hat's fine example - pay for a core development team that releases open, early, and often. This has worked *extremely* well for GNOME and Enlightenment - by paying for full-time developers who release GPL, Red Hat has vastly accelerated the already swift Open Source development process, without sacrificing the peer review that makes Open Source so effective. The resulting high-quality software could well lead to GNOME and Enlightenment becoming interface standards for Linux (and Unix in general - in the current climate, commercial vendors may well pick it up), providing the tools for the user-friendly desktop journalists keep telling us we need to compete.
This same logic applies to tools like XSL... Sun would benefit if XSL became an industry standard, especially if built around a standardized Open Source implementation that discourages proprietary extensions and platform-dependent bugs. It would be worth Sun's while, i think, to fund Open Source development of implementations for standards like XSL. Hell, i can see an industry Open Source consortium, with business funding full-time Open Source development for standardized tools which benefit the whole industry. This could be what the quote Open unquote Software Foundation (OSF) dreamed of being...
Shouldn't their poll have been, "Will Linux squash Microsoft like a bug?" I'll bet MS would fare worse than Linux has...
At home, i use a PC Concepts ergo with a built-in trackpad. It's *wonderful*. The regular old split-layout ergo is much easier on the wrists, and the trackpad is even better, if you don't need too much precision. And the 6 key is in the right place for touch typists. :}
At work, i have one of the newer-model MS ergo keyboards. The keys themselves are better than the PC Concepts (as good as soft-touch gets), but the dinky arrow and home/pgup/etc keys blow. And the 6 is on the wrong side. And i'm using a wrist-killing MS mouse. But It's better than nothing.
If you don't like ergo keyboards, look for an ancient gen-u-wine IBM PC keyboard off an AT or RS/6000. *Beautiful* key touch, and they last forever. The only bad thing is they're noisy.
And if some of you would stop bashing Jobs and Disney for a moment, you'd see the logic.
Disney has been looking for a new CEO, since Eisner wants to scale back his job for health reasons. But Disney needs a visionary for a CEO, not the sort of egomaniacal bean-counter that is the mainstream CEO. Jobs is one of the great visionaries of American business (whether or not you agree with his vision), and is an effective fiscal officer as well (see the Apple turnaround).
The Pixar buyout is obvious, the Apple buyout less so. But think on these factors: first, Disney might want to get into the computer industry. It's a diverse company, and has hired some of the greatest living minds of the industry (Alan Kay, Danny Hillis, etc) just to have visions. The only existing computer company with the right cachet for Disney is Apple. Face it, Disney wouldn't just sell beige Wintel boxen. They've been toying with portals, but a unique and friendly PC is more, well, Disney. That's the modern Mac.
Second, and more importantly i think, they would own Quicktime. There's a lot of thought going on in Hollywood now about digital distribution of theater-grade films - digital media that rivals 70mm film in quality. Distributing 70mm reels is a HUGE expense. Digital media would be smaller, much cheaper to manufacture, and effectively immune to wear. And Moore's Law pretty much guarantees that it *will* be technically feasible within a few years. Whoever gets the technology to market first, and has the best backing, stands to make *billions* in distribution.
Apple's Quicktime is the best video compression technology currently available, and barring a surprise technical breakthrough, will remain so for years to come. And it may scale size/resolution well enough for film-quality video, given a good storage mechanism. This technology (and other Apple tech), combined with Disney's industry presence and marketing savvy, could put Disney in control of a key choke point for the entire film industry.
Stop thinking like a bunch of obsessive Anonymous Cowards and start thinking like Disney stockholders. This makes sense.
Cygnus is probably the oldest provider of *Open Source* commercial software. They maintain egcs and release it under GPL. What that $7000 buys is the sort of support that makes PHBs feel all warm and fuzzy inside - better than the support you'd get from most vendors, i might add. The pricing reflects Cygnus' primary market, which is commercial Unix developers. In that rarified world, compilers always cost that much. And, given the effort involved versus the size of the market, it makes sense. Microsoft charges what, two grand for a copy of Visual Studio Enterprise? And they'll move hundreds of thousands of copies.
Don't worry about that $7000. If you're not selling your software for money, you can get the tools for free. And if you want to run quality Open Source development tools at work, it's a lot easier to convince the PHB to pay $7000 than it is to convince him that free tools could possibly be as good as the $7000 stuff!