Slashdot Mirror
Macs In Space II
MasterOfDisaster writes: "Some nut is planning to make a global wireless network using apple's Power Macintosh G4 Cubes. Here's the story." We ran a story about this guy last year, but this one has a bit more detail and he's progressed further in his plans.
Sounds like well-planned marketing scam.
... no Trip to Tobago this year, I'll buy me self a Mac G4 Cube!"
...why not!
Mac has fallen from the sky (look at the stock price) and now we the executives have dreamed up a funky story and found a geek to talk about it.
And the expected response is?
(after reading this article)
"Sorry honey, change of plans
But hey
Well, NASA seems to have enough faith that it will work that they are going to put it up there and assemble it in space for him. I doubt they would be willing to do that if they didn't think it might work. Im sure they have better things to do than assemble and set into orbit computers that they think will fail.
Time for some tasty Shiner Bock!
Not sure if this is what the "nut" was thinking of when he said it performed better, but apparently the later Intel processors (post-486) have big problems with cosmic rays - flipped bits, missed instructions, etc. This is one of the reasons why NASA relies on older machines still.
I adblock all animated gifs.
Blessed be the prime numbered slashdotters
Geez... I think I would feel better running windows 2000, serously.
Unless they are using OSx, I can't imagine that they really believe it's any more reliable than your average 1-2 week lifespan uptime. Also, whats to protect the vital comunication systems if a single process freaks out and crashes the entire system? Wouldn't be such a big problem with any modern OS.
*sigh*
I am not sure it is that unrealistic to cover most of the United States or greater Europe w/500+ satelites at the right orbit level. (15 minutes of coverage is bullshit, depends on the altitude of the satelite and it's orbital rotation
Anyway, lots of very interesting inovations come from 'wacky people' with 'wacky ideas'
--------------------
Would you like a Python based alternative to PHP/ASP/JSP?
so that's who's buying G4 Cubes. i was wondering which 50 people bought them, as it certainly wasn't anybody i know.
people have been wondering what the target market was for these things: do you think Apple had this in mind all along? i guess this would account for the astronomical pricing ;)
- j
Really the thing is little more then a FlashROM, an off-brand x86 and of course Lucent's Orinoco PC-Card. The code running it is exemplarily, folks have found lots of goodies in it and Apple's put out several revs. of improvements, but not a PPC to be found.
Since the article talks about flying boxes with MacOS on PowerPC's then clearly the existing AirPort base-station technology is NOT the subject. There'd be nothing in common with either the hardware or the code. It'd be easier to start with a BSD underpinning (MacOS X) or something like Sustainable Software's products (MacOS >X.)
Thus, moderators, please bring down those postings that refer to flying AirPorts.
BTW, a good (though dated) AirPort technology link is http://www.msrl.com/airport-gold
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
Just what we need, molten hunks of plastic falling to earth.
Bill - aka taniwha
--
Bill - aka taniwha
--
Leave others their otherness. -- Aratak
Maybe he should look into using Linux as the OS - he definately isn't going to want to have to figure out a way to reboot the Mac when it runs out of memory.
Ground Control, "When you move your hand over the button it should light up."
Project Reboot, "It's lit up."
Ground Control, "Hit the button when it lights up."
Joseph Elwell.
actually cooling in space is a big problem - you have heat input from direct exposure from the sun and from internal sources - but you can only radiate excess heat on the side away from the sun - no cool convective air flows to take heat away - keeping electronics and batteries cool enough to operate safely is a big worry
The air trapped in the hardrive will exert a constant pressure on the seals holding it in, increasing the chance of hard drive failure.
Have you ever noticed a little hole on the top cover of your hard drive with a BO NOT COVER label next to it? That's a vent hole. Air goes in and out through it (heavily filtered, of course). It's the air that allows the heads to float a teensy-weensy distance above the platters while they're spinning (and dragging air around with them). Remember the Landing Zone parameter on old MFM hard drives? That's where the heads would go before the drive stopped spinning so that once they did stop, the heads would stop floating and slide to a halt on the surface.
Where the hell am I going with all this, you ask? Vacuum of space == no air in drive == heads can't float == hard drive doesn't work. Period. That's why satellites don't have hard drives (they also don't need them).
--
Three quick notes:
1) s/BO/DO
2) The concept of heads floating on a thin cushion of air above the platters is the cornerstone of Winchester hard disk design. This is why the surface of the heads is carefully designed with a contour that promotes the correct "ride height" or "float height" (I can't remember which term is used). IBM's technical documents state the proper float height for their drives.
If you wanted a hard-drive-like thing in a satellite, you might get away with a solid-state hard drive (rad-hardened, of course... if such a version exists at all).
--
I wish people looked at their hard drives before they say something incorrect. All hard drives have a filtered breather hole.
Instead of trying to make a disk drive perfectly sealed, disc drive companies make the drive breath from a well filtered breather hole. This is because when drive run and heat up the air expands and wants to push air out, then get turned off and the air wants to come back in. Better to filter air in one spot then get contaminated air from misc. leak points.
Researcher Dennis Wingo says there's a cheaper, simpler way to set up a network of wireless-data satellites: Girdle the globe with Apple's Cubes
What part of that didn't you understand? And no, he didn't say he'd use OS9, but it does say Mac OS. I agree that he'll probably use Darwin (more probably the whole OSX) because if he claims that Windows crashes a lot, he doesn't use classic Mac OSs very often.
MyopicProwls
MyopicProwls
My homepage
The solar radiation will degrade the hard drive and RAM to the point of unusability. Long term solar radiation studies have shown that that extended exposure in orbit subjects electronics to solar storms and other destructive environments that harm non-shielded systems.
There are also the issues of microgravity and vacuum. Metals behave differently in orbit. Several communications satellites have failed because of the growth of zinc solenoids in the spacecraft in orbit. Plastics will outgas, changing their structure. The air trapped in the hardrive will exert a constant pressure on the seals holding it in, increasing the chance of hard drive failure. All the rules are different in space.
The temperature differential between the light and dark sides of the satellite are approx. 200 F. With all that thermal stress, all components will be mechanically stressed.
While I applaud this guys gung-ho spirit, I think he underestimates the harsh environment of space.
-----
nuclear iraq bioweapon encryption cocaine korea terrorist
Besides, the business model is bogus. Iridum couldn't make it doing phones, what makes him think that he can make money with data-only transmissions competing with DSL, CableModems and GSM/PCS technology?
Oh yea, then there's the latency issue since you are sending your signals further, making it unacceptable for playing games... Anyway, there's no real North American or Europe market that can support him. Maybe in more remote places lacking current infrastructure it would make sense, but can they pay the cost of all those mini-sats at 120lb a pop???
- Mike
this is true.
At high altitudes (>10,000 feet), driving cars becomes hazardous, because there isn't enough air to adequately cool normal brakes. This, coupled with steep downhills. . .
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
<em>Rubbish. You cannot "lock a Mac down harder than a UNIX box". They can both be locked down to any degree desired. </em>
The trick is to get it watertight out of the box. OpenBSD's probably the closest, but starts to wander away from the mark when you add functionality. There's a quicker return on investment with with Mac servers if you're security conscious. You know how much a Unux security guru costs?
<em>And sure you can run a Mac with no software and no extensions, but you can't do anything with it. </em>
In my extensions folder, there are 206 items. (Most of them shared libraries Lotus Notes barfed all over my system.) Do I need an audio CD driver on a server? No. Do I need ColorPicker or ColorSync on a server? No. Do I need MacInTalk? No. Do I need finderpop? No. Ditto for control strips, control panels, any font that didn't come with the OS, etc. Five minutes worth of work gets you one stable Mac server.
What makes the Mac unstable has nothing to do with the OS and eveything to do with the applications it runs. Protected memory systems (like Unix) are more forgiving of buggy code...if the Gimp dies a horrible death, it won't take the kernal with it. On the other hand, the Mac will reboot if the program/thread isn't coded to exit gracefully on error.
This means the Mac is only as stable as its applications. This is a problem when you are running enormous and complex applications, like popular web browsers, office productivity suites, or desktop publishing programs.
Running small RealBasic apps and garden variety networking software tied together with a few Applescripts will likely keep running forever without a reboot.
OTOH, cruising slashdot with Netscape while photoshopping Steve Jobs sodomizing Tux while "working" on sales pitch in Microsoft Word will probably have you cursing up a storm and rebooting once every couple days. (If you used iCab and Nisus and Canvas instead of NS and Word and Photoshop, you'd probably think those singing the "Macs are unstable" chorus are a bunch of idiots who don't know what they're talking about.)
So, ironically, where the Mac is supposed to shine the brightest, as a desktop workstation, it is at it's least stable. As a server, it does all right. Better than NT/2000, at any rate, and probably as good as the BSDs. (My personal favorite younickses. Got an OpenBSD box running an intensive PHP-heavy site on an antiquated Sun workstation. I'm impressed with how well it holds up.)
SoupIsGood Food
How would these systems handle the huge swings in temperature? I think that the temperature extremes these things would go through in space are beyond the tested tolerances for most consumer hardware..
--
I still stand by then "no temperature in space", "space has no temperature" is the same thing just phrased differently (though I just saw your point: mine is a little ambiguous). Radiation is not temperature as temperature is a measure of how much thermal energy is stored in a mass.
Solar wind wouldn't give you a vector I think, but it would give you a wind chill effect :) (mind you negligable due to it's sparsity).
Bill - aka taniwha
--
Bill - aka taniwha
--
Leave others their otherness. -- Aratak
Does it? I was under the impression that convection worked by the fact that hot air rises, pulling in cold air from the bottom of the cube. Hot air rises because it is less dense than cold air - less dense and thus not as heavy per square unit. In zero g, the weight of an object would nothing, since Weight=mass*gravity and gravity=0, thus Weight=0.
`ø,,ø!
Free Software: Like love, it grows best when given away.
Another thing that has me concerned is that the guy is proposing just throwing it from the space station. In addition to resulting in a rather imprecise orbit, there's always the chance that it will hit the space station when it returns. While this isn't as bad as hitting an object orbiting in the oppsite direction (with impact speed = 2X orbital speed), an object that size could still do some damage to the space station.
I think this guy has an interesting idea, but as far as implementation goes, he's a kook.
#naabhaprzrag, #sverubfr-000, #agi-fcbafberq, negvpyr[pynff*=' negvpyr-ary-'] { qvfcynl: abar !vzcbegnag; }
RAM is a rather critical and highly-stressed component of computer systems, these days. In afterthought, it's not surprising that the RAM socket is a source of many failures.
`ø,,ø!
Free Software: Like love, it grows best when given away.
Wouldn't that be the developer of the applications fault?
No, it would not be. The App is not responsible for keeping the box up. Thats what the OS is for
ReadThe ReflectionEngine, a cyberpunk style n
500 MHz is fine for quite a long time. Remember, NASA just recently (well in the last year or two) upgraded the Hubble from 10 year old 386's to brand new 486's... It's not like they need to be running Gnome, MS Office, Photoshop or any other resource hungry application...
AFAIR, the original power macs (60 to 80 MHz PPC601's), with their GeoPorts which were basically serial ports controlled by the CPU rather than an extra chip, could flood a T1 on each port, while still remaining relatively responsive for the foreground applications.
500 MHz G4's should be robust enough to tons of bandwidth flying through them...
Why does a satellite costs as much as it does. You can build a device with the same functionality for less.
The problems is that it is a lot harder to replace a part that breaks down. That's why you don't want to use a Mac(or other end(l)user hardware). If a Mac or a PC for that matter was built with the same requirements for the MTBF as a satellite the price of the hardware would skyrocket(pun intended). The test required for each circut and chip would make it impossible to sell today.
So go ahead and launch your Mac into space, just don't me to go and fix it. well maybe I would like to but who's gonna pay.
--------
after READING the article, (wot a concept!) I start to see a method to his madness.
.).
Nowhere does it say that a Mac cube will be used. Nowhere does it say that MacOS 9 will be used.
It's a 120 lb satellite, not engineered to withstand launch stresses, because it will be assembled in orbit. Many of the parts are off the shelf - including probably not more than a motherboard and CPU of a macintosh (much cheaper than designing a controller) - which will probably THINK it's running an Airport - but I'm betting it will be fooled into thinking that it's running airport when it's running some higher-powered device (airport range = 150 feet on disputed radio bandwidth (in France)), (his satellite range, altitude of 120 miles plus azimuth. .
In fact, I bet it will run Darwin, probably without a hard-drive, (probably some sheilded flash RAM device instead), and probably with lots of custom software (like TiVo runs Linux). (I'm guessing Darwin because it would be much easier to run the transceiver as Airport that way than trying to hack together something with Linux - BSD is supposedly more reliable than Linux anyway, but I digress and risk a flamewar).
Cooling will be an issue, and so might radiation, but a PPC chip will give him some pretty good computing power without worring about heat as much as with SOME OTHER solutions.
Of course, part of the 120 lbs will probably be gyros, solar panels, the transmitter and amplifier,
but the main gimmick here, is that he's using off the shelf parts, and assembling them in orbit, in an attempt to reduce costs. (in other words, he probably plans on all devices being launched from ISS or Shuttle, assembled in orbit). Yeah, the labor of assembling in orbit is probably LOTS higher, but you end up reducing the overall weight by bunches, by not having to design solar-panel deployment systems, shrouds, and shock-resistant innards.
If he's planning on spending $10 million on the first device (instead of hundreds of millions for standard communications satellites), it sounds like a worthy project to me (*cough* irridium)
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
This does not make sense, putting stuff up in space is extremely expensive, therefore the lighter the better. It would actually be cheaper buying a smaller lighter PPC controller board and preloading it. DVD drives in space? I don't think so. And are those 7400's able to stand the radiation up in space?
I don't think so. Off the shelf components will not survive the temperature and radiation extremes. A solar storm would kill them all, and shielding them is not too easy.
I am a man, not a toy.
----------------
I am Moldy.
Macs, when used as stone stock simple servers, tend to be fairly reliable, with uptimes on the order of years for some older (Pre-PowerPC) systems I'm personally aware of.
Macs become unstable when you dump a crapload of extensions, control panels, fonts, third party software, plugins for third party software, plugins for plugins for third party software, and, well, you get the drift.
Clear the cruft out of the system folder, use only proven, reliable third party software and damn little of it, and the lowly and much maligned Mac can keep cranking the bits month after month with the younicks big boys. Figure in some scheduled downtime every so often for a pre-emptive reboot, and it will be spiffy for as long as they're up there.
Why bother putting up with three minutes downtime out of a week? Security. You can lock a mac down harder than a Unix or windows box. Simplicity. Configuring these bad boys will be a breeze. Plus, tools like RealBasic and Applescript make coding applications and scripts painless and powerful...without sacrificing reliability.
Put anything that says "Netscape" or "Quark" on there, and the mission is doooooooomed!
Just 'cuz the consensus has it that Macs are inherently unreliable, doesn't mean it's neccesarily true, or that the "unreliability" is a liability to steep to be surmounted. Question consensus, and do a little thinking and research for yourself.
SoupIsGood Food