I find it amusing that in E*Trade's list of upcoming IPO's,
FTD.com Red Hat, Inc. Bamboo.com, Inc. Garden.com GreatFood.com, Inc. ImageX.com, Inc.
, that RedHat is the only one without a ".com" in its name. Let's hope that the world is finally waking up to the fact that a.com address doesn't guarantee a company's success any more than does a Yellow Pages ad. I think that Red Hat has a good chance to be successful as a company, not just as a buzzword.
As for the timing, it probably looked good when they started the IPO process, and I'm sure it's too late for them to back out now.
1. A typical software license restricts more than copying. There are also restrictions on how it may be used, such as the # of concurrent sessions coming into a server. There may also be restrictions on customizing or modifying the software, benchmarking it, or even reading it (by "decompiling"). Last I heard, the government will enforce a license containing such items. "Law", "license", or "contract", Microsoft is able to decide which actions will or will not get you thrown in jail. The baker who sells me a loaf of bread is not similarly empowered; there's no "per slice" toasting license pack, or restriction on making sandwiches made with that bread.
2. "Just don't use MS products". I don't where I can possibly avoid it. However, I still have plenty to worry about (I'll omit that discussion here; search for 'monopoly' elsewhere).
3. "A can do B" does not equal "only A can do B". I didn't say that _only_ large corporations can have their copyrights protected, though it sometimes happens that way for financial reasons aka the "golden rule".
4. I don't remember saying that licenses were evil, just that their strength originates with the government. As I recall, the original idea of the US patent office was that they'd grant a limited monopoly on an invention in exchange for the public disclosure of that invention. "Intellectual Property" by itself is an oxymoron; it is only through *government* that it acquires any substance. With I.P. laws come antitrust laws, and I don't see why a corporation should be able to pick and choose between them.
5. What's that silliness about RH taking over the world? Of course they can't. How did you deduce that from my posting? The only reason I used the example of Debian was to show that the restrictions on the use of software originate with the author rather than the government, even though the government is the enforcer. Maybe I should have said FreeBSD instead. I'm not saying that we can do without these laws, though that may happen some time in the future.
>That's what governments do. Private corporations cannot do that.
Private corporations use the taxpayer-funded government to do exactly that. Let's say I borrow a Windows NT server CD from a friend (with his consent), and install it on 100 machines at work plus my home machine and my mother's. I then return the CD to my friend in its original condition.
Do I go to jail? Yes. Whose jail is it? The government's. Whose LAW did I break? A corporation's.
For reasons both good and bad, private corporations have been granted the power to write laws, which the government will then enforce on behalf of, and at no cost[1] to, that corporation. Thus, corporations like Microsoft do have power traditionally associated with the legislative branch of government.
If I did exactly the same thing with a Debian CD, I would not go to jail. This is because there is no corporate law saying I can't do this with Debian's software. The government does not say that it's a crime to copy software, but they do say it's a crime to disobey any licenses attached to that software (with surprisingly little restriction on what can be contained in those licenses).
[1] not including bribes or campaign contributions
I never advocated the "pure capitalism" scenario. I just wanted to point out that Microsoft Corporation probably wouldn't exist in such an environment. And, if a government structure is necessary for Microsoft to do business, I don't think it's asking too much for them to be subject to a few government controls on HOW they do that business. After all, what are they selling? They sell licenses, not software.
>As much as I hate Microsoft and would like to see it destroyed, I don't think the government should be involved.
Of course, you do realize that Microsoft's business success directly relies on "government involvement" in the form of enforcement of copyright laws and shrink-wrap licenses, right?
In a "pure capitalism" world, a Microsoft CD would only be worth the polycarbonate it's stamped on.
Anyone else get the feeling that the Microsoft team is calling their own tech support?
"Your computer crashed? Well, that's normal, just reboot it. You did? Well, a fresh install of the OS usually works. Your computer might not be powerful enough to run the newest software, of course. Maybe you should add more RAM..."
It's down again, but I've got it in my cache. From the status page:
-------- [...] The Windows 2000 Internet Test Site is so popular we also to got [sic] a new machine to add more capacity! We're now running on a 500Mhz PIII with 256Mb of RAM. Today we installed a recent build that has lots of updates since the RC1 build.
We're back up and running. [...]
-------- Famous last words... (and they're proud of their lack of scalability!)
The server's not cracked, but the guestbook doesn't filter the comments at all. Therefore, it's easy to drop in a bit of javascript, an open BLINK tag, or even a META redirect. The guestbook has been up most of the afternoon, and so far I haven't seen anything actually exploitable in it. He's now clearing the guestbook every few minutes to get rid of the clutter.
I finally got through to w2k, and clipped from the status page:
8/6/99 9:00am - Reset TCP to handle SYN attacks, and rebooted.
8/5/99 1:00pm - Tuned IIS' performance options reset application protection to Medium, and rebooted. 8:54am - Changed IIS' application protection to Low and rebooted, site back up
8/4/99 6:58pm - IIS stopped sending pages. Restarted service. 6:00pm - Morning crash dump due to known bug in Rdr in our build, shutdown Workstation service 9:42am - Crash dump - investigating causes turns out to be Rdr erro
8/3/99 3:22pm - Network connections down due to router failure, possibly related to thunderstorms and power failures in the area
It looks like Micros~1 is reading from a BOFH excuse-of-the-day calendar as they struggle to tune their flagship product toward a state of adequacy. It will be interesting to see how many holes have to be opened up on the LinuxPPC system before it reaches a break-even point with w2k.
I'd guess it's more a bandwidth problem than a computational one. In other words, simple calculations on a huge pile of data, rather than intensive number-crunching on a small packet.
Still, it would be interesting to offer, and I'd gladly drop RC5-64 for this project.
I have a IIIx, and even if I'd been able to get a V for the same price, I wouldn't have wanted one. The Li-ion battery isn't a selling point; I prefer rechargeable alkalines for this kind of application [1]. The only thing the V has going for it is a bit of industrial design, and I'd much rather have the enhanced internals of the IIIx.
I'd still buy a IIIx, though the IIIe is probably a better value for entry-level users.
[1] A little tip from the O'Reilly book - you can set the Palm's battery-monitor thresholds for NiCd, Alkaline, or Rechargeable Alkaline by writing the 'shortcut' symbol, then '.7'
>Pass a magnet through a copper coil. Do you feel any resistance?
Yes, if the coil is part of a circuit which allows current to flow. A neat demonstration is to hold a copper pipe (essentially a 1-turn coil) vertically and drop a rare-earth magnet in the top. The magnet will take several seconds to fall through the pipe, due to the braking forces caused by eddy currents in the copper.
I don't know if anyone's doing it commercially, but another way to get electrical energy from a keyboard would be to use a piezoelectric polymer film such as PVDF (see http://www.measspec.com/piezo_film.htm). This would be non-magnetic, lightweight, and structurally quite similar to existing membrane-switch keyboards.
I don't think any finger-powered transducer can generate enough power to even put a dent in the consumption of a Pentium-II laptop, but it might help with a Palm Pilot or other micro-power device.
>Regarding learning from Amiga, I think at least Microsoft did.
You mean when they finally got around to releasing a 32-bit, pre-emptive multitasking, multimedia-capable OS with long-filename support and auto-configuration of hardware cards?
Yeah, it's pretty funny that on a machine with no swap and 1M of RAM (which includes the "video card" memory), you could still use half of it as a RAM disk.:-)
And the nice part is that even if they don't base the OS on GNU tools, you'll still be able to install gcc and friends if you want to (I had gcc on my A4000, but I used the commercial SAS C package more).
When I switched from my A4000 to a Linux box, I was amazed by how slow and bloated X was. I still am. Using X as the native interface is not necessarily a good idea for an "Amiga" (depending on what that name means to you).
Also, I think they could modify gtk, Qt, lesstif, etc. so that they ran on top of the native Amiga graphics library (whatever that is for the new OS) rather than Xlib [or whatever it is; I've never ventured into those depths on Linux].
However, for compatibility reasons, an X server will be required. One way of dealing with this is to have a native desktop, then an "X" screen behind it. [If you've never seen an Amiga, "screens" are a *much* cooler version of Linux's "virtual consoles"].
On the Amiga, if your mouse pointer ever froze it meant something was seriously crashing and thrashing memory. On X, it's normal. There is definitely something to be said for physical-only memory (no swap) and sensible OS/hardware integration (the Amiga "mouse port" went right into a quadrature counter in a custom chip, and the OS just read the X and Y position registers when it wanted to know where to position the mouse-pointer sprite. No RS-232, no IRQs, no drivers, no nonsense).
Another area where Linux could learn a lot from the Amiga is removable-media handling. If you eject a "mounted" floppy from an Amiga, you get a message asking you to please re-insert volume 'name' in any drive (or if the system was actually accessing a file, a more urgent "You MUST replace volume 'name' in drive DF0:".
Actually, the original Amiga 1000 OS was not built into firmware, it was on the first disk (called "kickstart"). The firmware only had enough smarts to load this disk image into a special section of RAM, which was then write-protected in hardware (they called it "write-once memory", or WOM), so it would survive a warm-reboot. The second disk was the "workbench", which gave you the desktop and a few utilities.
All models after the 1000 had "kickstart" in ROM, though it was possible to modify the function jump table and replace a ROM routine with a patch in RAM.
Slashdot Mirror
$ bc -l
e(l(2) / 12)
1.05946309435929526455 [12th root of 2]
l() is natural log, e() is exponential.
I find it amusing that in E*Trade's list of upcoming IPO's,
.com address doesn't guarantee a company's success any more than does a Yellow Pages ad. I think that Red Hat has a good chance to be successful as a company, not just as a buzzword.
FTD.com
Red Hat, Inc.
Bamboo.com, Inc.
Garden.com
GreatFood.com, Inc.
ImageX.com, Inc.
, that RedHat is the only one without a ".com" in its name. Let's hope that the world is finally waking up to the fact that a
As for the timing, it probably looked good when they started the IPO process, and I'm sure it's too late for them to back out now.
They are evidently using RedHat 6.0 on their Optiplex machines (see here and here), so I imagine it will be the same on the Dimension.
1. A typical software license restricts more than copying. There are also restrictions on how it may be used, such as the # of concurrent sessions coming into a server. There may also be restrictions on customizing or modifying the software, benchmarking it, or even reading it (by "decompiling"). Last I heard, the government will enforce a license containing such items. "Law", "license", or "contract", Microsoft is able to decide which actions will or will not get you thrown in jail. The baker who sells me a loaf of bread is not similarly empowered; there's no "per slice" toasting license pack, or restriction on making sandwiches made with that bread.
2. "Just don't use MS products". I don't where I can possibly avoid it. However, I still have plenty to worry about (I'll omit that discussion here; search for 'monopoly' elsewhere).
3. "A can do B" does not equal "only A can do B". I didn't say that _only_ large corporations can have their copyrights protected, though it sometimes happens that way for financial reasons aka the "golden rule".
4. I don't remember saying that licenses were evil, just that their strength originates with the government. As I recall, the original idea of the US patent office was that they'd grant a limited monopoly on an invention in exchange for the public disclosure of that invention. "Intellectual Property" by itself is an oxymoron; it is only through *government* that it acquires any substance. With I.P. laws come antitrust laws, and I don't see why a corporation should be able to pick and choose between them.
5. What's that silliness about RH taking over the world? Of course they can't. How did you deduce that from my posting? The only reason I used the example of Debian was to show that the restrictions on the use of software originate with the author rather than the government, even though the government is the enforcer. Maybe I should have said FreeBSD instead. I'm not saying that we can do without these laws, though that may happen some time in the future.
>That's what governments do. Private corporations cannot do that.
Private corporations use the taxpayer-funded government to do exactly that. Let's say I borrow a Windows NT server CD from a friend (with his consent), and install it on 100 machines at work plus my home machine and my mother's. I then return the CD to my friend in its original condition.
Do I go to jail? Yes. Whose jail is it? The government's. Whose LAW did I break? A corporation's.
For reasons both good and bad, private corporations have been granted the power to write laws, which the government will then enforce on behalf of, and at no cost[1] to, that corporation. Thus, corporations like Microsoft do have power traditionally associated with the legislative branch of government.
If I did exactly the same thing with a Debian CD, I would not go to jail. This is because there is no corporate law saying I can't do this with Debian's software. The government does not say that it's a crime to copy software, but they do say it's a crime to disobey any licenses attached to that software (with surprisingly little restriction on what can be contained in those licenses).
[1] not including bribes or campaign contributions
I never advocated the "pure capitalism" scenario. I just wanted to point out that Microsoft Corporation probably wouldn't exist in such an environment. And, if a government structure is necessary for Microsoft to do business, I don't think it's asking too much for them to be subject to a few government controls on HOW they do that business. After all, what are they selling? They sell licenses, not software.
>As much as I hate Microsoft and would like to see it destroyed, I don't think the government should be involved.
Of course, you do realize that Microsoft's business success directly relies on "government involvement" in the form of enforcement of copyright laws and shrink-wrap licenses, right?
In a "pure capitalism" world, a Microsoft CD would only be worth the polycarbonate it's stamped on.
Anyone else get the feeling that the Microsoft team is calling their own tech support?
"Your computer crashed? Well, that's normal, just reboot it. You did? Well, a fresh install of the OS usually works. Your computer might not be powerful enough to run the newest software, of course. Maybe you should add more RAM..."
Handcuffs. Jail. Your computer for sale at the next police auction.
It's down again, but I've got it in my cache. From the status page:
--------
[...]
The Windows 2000 Internet Test Site is so popular we also to got [sic] a new machine to add more capacity! We're now running on a 500Mhz PIII with 256Mb of RAM. Today we installed a recent build that has lots of updates since the RC1 build.
We're back up and running.
[...]
--------
Famous last words...
(and they're proud of their lack of scalability!)
The server's not cracked, but the guestbook doesn't filter the comments at all. Therefore, it's easy to drop in a bit of javascript, an open BLINK tag, or even a META redirect. The guestbook has been up most of the afternoon, and so far I haven't seen anything actually exploitable in it. He's now clearing the guestbook every few minutes to get rid of the clutter.
Ya know, whenever I see "DOS", I'm not sure if the person is talking about Microsoft's Disk Operating System, or a Denial Of Service attack.
Hmmm, now that I think about it, maybe it's the same thing after all...
I finally got through to w2k, and clipped from the status page:
8/6/99
9:00am - Reset TCP to handle SYN attacks, and rebooted.
8/5/99
1:00pm - Tuned IIS' performance options reset application protection to Medium, and rebooted.
8:54am - Changed IIS' application protection to Low and rebooted, site back up
8/4/99
6:58pm - IIS stopped sending pages. Restarted service.
6:00pm - Morning crash dump due to known bug in Rdr in our build, shutdown Workstation service
9:42am - Crash dump - investigating causes turns out to be Rdr erro
8/3/99
3:22pm - Network connections down due to router failure, possibly related to thunderstorms and power failures in the area
It looks like Micros~1 is reading from a BOFH excuse-of-the-day calendar as they struggle to tune their flagship product toward a state of adequacy. It will be interesting to see how many holes have to be opened up on the LinuxPPC system before it reaches a break-even point with w2k.
>Aug 6 1999 01:15PM CDT: :)
:-)
>In response to the brute force attempt, we have
>decided to save him the trouble: linuxppc
I guess the flood of ignorant packets got boring.
Didn't the last poll show something like a 98% / 2% split among /. readers? Heck, it'd be a chick fest if they had 5% female attendance.
;-)
Don't worry ladies, I won't be able to attend the party anyway.
I'd guess it's more a bandwidth problem than a computational one. In other words, simple calculations on a huge pile of data, rather than intensive number-crunching on a small packet.
Still, it would be interesting to offer, and I'd gladly drop RC5-64 for this project.
I have a IIIx, and even if I'd been able to get a V for the same price, I wouldn't have wanted one. The Li-ion battery isn't a selling point; I prefer rechargeable alkalines for this kind of application [1]. The only thing the V has going for it is a bit of industrial design, and I'd much rather have the enhanced internals of the IIIx.
I'd still buy a IIIx, though the IIIe is probably a better value for entry-level users.
[1] A little tip from the O'Reilly book - you can set the Palm's battery-monitor thresholds for NiCd, Alkaline, or Rechargeable Alkaline by writing the 'shortcut' symbol, then '.7'
>Pass a magnet through a copper coil. Do you feel any resistance?
Yes, if the coil is part of a circuit which allows current to flow. A neat demonstration is to hold a copper pipe (essentially a 1-turn coil) vertically and drop a rare-earth magnet in the top. The magnet will take several seconds to fall through the pipe, due to the braking forces caused by eddy currents in the copper.
I don't know if anyone's doing it commercially, but another way to get electrical energy from a keyboard would be to use a piezoelectric polymer film such as PVDF (see http://www.measspec.com/piezo_film.htm). This would be non-magnetic, lightweight, and structurally quite similar to existing membrane-switch keyboards.
I don't think any finger-powered transducer can generate enough power to even put a dent in the consumption of a Pentium-II laptop, but it might help with a Palm Pilot or other micro-power device.
>Regarding learning from Amiga, I think at least Microsoft did.
You mean when they finally got around to releasing a 32-bit, pre-emptive multitasking, multimedia-capable OS with long-filename support and auto-configuration of hardware cards?
Yeah, it's pretty funny that on a machine with no swap and 1M of RAM (which includes the "video card" memory), you could still use half of it as a RAM disk. :-)
And the nice part is that even if they don't base the OS on GNU tools, you'll still be able to install gcc and friends if you want to (I had gcc on my A4000, but I used the commercial SAS C package more).
When I switched from my A4000 to a Linux box, I was amazed by how slow and bloated X was. I still am. Using X as the native interface is not necessarily a good idea for an "Amiga" (depending on what that name means to you).
Also, I think they could modify gtk, Qt, lesstif, etc. so that they ran on top of the native Amiga graphics library (whatever that is for the new OS) rather than Xlib [or whatever it is; I've never ventured into those depths on Linux].
However, for compatibility reasons, an X server will be required. One way of dealing with this is to have a native desktop, then an "X" screen behind it. [If you've never seen an Amiga, "screens" are a *much* cooler version of Linux's "virtual consoles"].
On the Amiga, if your mouse pointer ever froze it meant something was seriously crashing and thrashing memory. On X, it's normal. There is definitely something to be said for physical-only memory (no swap) and sensible OS/hardware integration (the Amiga "mouse port" went right into a quadrature counter in a custom chip, and the OS just read the X and Y position registers when it wanted to know where to position the mouse-pointer sprite. No RS-232, no IRQs, no drivers, no nonsense).
Another area where Linux could learn a lot from the Amiga is removable-media handling. If you eject a "mounted" floppy from an Amiga, you get a message asking you to please re-insert volume 'name' in any drive (or if the system was actually accessing a file, a more urgent "You MUST replace volume 'name' in drive DF0:".
------------
My computer history:
Vic20 -> C-128 -> A500 -> A4000 -> Pentium -> K6 , Netwinder(StrongARM) -> ???
Actually, the original Amiga 1000 OS was not built into firmware, it was on the first disk (called "kickstart"). The firmware only had enough smarts to load this disk image into a special section of RAM, which was then write-protected in hardware (they called it "write-once memory", or WOM), so it would survive a warm-reboot. The second disk was the "workbench", which gave you the desktop and a few utilities.
All models after the 1000 had "kickstart" in ROM, though it was possible to modify the function jump table and replace a ROM routine with a patch in RAM.
I'm just wondering why there was no posting of the sale of Slashdot.
There was, June 29, here.