Slashdot Mirror
Enlist, Boot Up, Change Fewer Batteries
BigBragger writes: "Upside has an article declaring that ViA will begin using the Crusoe chip in the wearable computers it currently designs for the US Army. Crusoe will debut in the next version. There's hope for a Transmeta PC yet, but will I have to enlist to get one?"
WillSeattle points to C|Net's story on the same thing and adds harshly:
"Soldier, when was the last time you compiled this kernel! You are a disgrace to the uniform! Give me 10,000 lines of code, pronto!"
It's not a 33% across the board raise, it's more like 2% to 33%. If you're a GS-5 (that's pretty low level, like 20k a year or so) in San Francisco, you get the 33%.
When I started my current job in Seattle three years ago while still in college, I was a Computer Specialist, grade 7. These guys are getting something like 20% raises. Now I'm a Computer Scientist, grade 11 and I'll get something like 4%. If I were a Computer Engineer, which would require nothing more than having a degree in comp E instead of comp sci, I'd have been getting the extra pay all along.
Is anyone besides me saddened by the continuting trend of new and promising technology being immediately conscripted for use in war? I realize a lot of money stands to be made by selling to the military, but it seems as though the computing field as a whole is drifting slowly towards being less a tool for creation and more a tool for destruction, which raises ethical questions.
Precisely because of the versatility of computers which has made them such a success, nearly any work done in many areas of the field such as hardware, security, or even graphics could easily be adapted for military use without the knowledge of the engineers or programmers who did the work. While a machinist can refuse to work in a shop which manufactures guns, it's not so easy for an principled engineer to avoid developing hardware with military applications. Worse still, any GPL software which the military of any country finds useful might easily be adapted into a tool of destruction, completely against the wishes of the people who developed it.
I'm probably taking the issue a bit far, but it is something to think about as you're writing that clone of Scorched Earth.
The *interface* may or may not be a variant of X (an embedded graphics toolkit such as Qt/embedded would probably be more appropriate), and you're certainly not going to bother with a full posix environment, let alone KDE or GNOME, but why reinvent the wheel and write your own kernel? I'm not saying that Linux is the only or even the best choice, but it would probably be a quite viable one.
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
Oh, there was a "toilet seat" bru-ha-ha a few years back. Which ignored that it was "toilet seat" for a combat aircraft; a Home Depot toilet seat wouldn't have fit. And it was a prototype; when production was set up, each was of course much cheaper than the the first hand-tooled one.
Similarly, there was a "hammer" controversey. To be used on reactive metals. Using an ordinary steel hammer from a Home Depot would have been a very good way of causing thousands of dollars of damage to high-performance aircraft and risking serious injury to the person using the hammer.
So, ignorant deficit-reduction organizations horrified by the Reagan defense buildup did an outside examination of a military procurement bugdet and found those items, and then sent press releases about these "$$$$ for hammers and toilet seats" to ignorant reporters who contacted ignorant PR people in the DoD PR offices who couldn't explain things. So the reporters ran their stories without having talked to anyone who knew what they were talking about, and Americans were told that the military pays hundreds of dollars for toilet seats.
There's no "we" in team, only "me"
I can see it now, distro wars between the services.
Navy: Windows. Yeah, unfair, but it's huge, it's a battleship, it's slow as hell.
Army: Embedded Linux. Clean, mean, and green, stripped down and ready to fight.
Marines: Embedded BSD. Secure but limber. Fewer apps.
Air Force: Windows 2000. They crash and burn.
Green Berets: Since they're fazed out, Windows 98. Buggy as hell, but it usually works when it's not drunk.
Will in Seattle
The use of Crusoe pieces in a military (or any government) contract is a guarantee of not only of millions of orders for the product, but also repeat business for years to come. This should go a long way on Wall Street.
because these Crusoe chips have enough horsepower to run minesweeper.
Sargent: This is the chip for me and you.
Troops: This is the chip for me and you.
Sargent: And its software is GNU.
Troops: And its software is GNU.
Sargent: And if the OS code should crash.
Troops: And if the OS code should crash.
Sargent: Recompile and kick some ass.
Troops: Recompile and kick some ass.
Sargent: We all code for Uncle Sam.
Troops: We all code for Uncle Sam.
Sargent: And our army runs a WAN.
Troops: And our army runs a WAN.
Sargent: Sound off.
Troops: Zero, One.
Sargent: Sound off.
Troops: Two, Three.
--GrouchoMarx
--GrouchoMarx
Card-carrying member of the EFF, FSF, and ACLU. Are you?
From a risk management perspective, the Crusoe is very important - it provides a bridge from the old world to the new world, and eventually, in the history of computing will be seen as such.
What Crusoe does is add an intermediate layer, that can adapt to various upper layers and therefore emulate different types of chips. Also, this intermediate layer, can be optimised and adapted to different firmware implementations that sit underneath - and as the marketing blurb tells you, in such a way to reduce power consumption as well.
In an age where computing is moving to a distributed, connected, more homogeneous type of environment, this is brilliant.
What I could expect to see is various different forms of firmware, with tight multiprocessing, and other custom features - and VLIW layers that can adapt. Also, VLIW adaption for intel, motorola and other processors which would be excellent as a migration path for existing software. Perhaps also, custom VLIW instructions can be useful for high performance applications.
It would be good to see Crusoe as the end for Intel, Motorola and others. I think the ARM may still have a price/benefit advantage in embedded systems, what many people don't remember is that ARM is often used as a ASIC core within silicon next to other functionality (e.g. RF, in communications chips) - I don't expect to see Transmetta offering Crusoe as an ASIC core in the immediate future, but who knows! ARM has a different market segment to Crusoe, and designers will know that.
-- Matthew - matthew.gream@pobox.com, http://matthewgream.net
rm -rf enemies
"I have not failed. I've simply found 10,000 ways that won't work." --Thomas Edison
They have already beaten the VLIW problem.
:)
That the transmeta chip has some vliwish elements in it is merely an implementation detail. 95% of software will use the code morphing layer, and not know what chip is underneath.
This is 1) good 2) important.
When you compile natively to a VLIW, you completely lose all binary compatability. If a pipeline length changes, if the number of EU's change, if _anything_ about the implementation of the processor changes - your binaries break. All of them.
Once intel moves to a VLIW architecture, evolutionary changes from 486->pentium, pentium->ppro are no longer possible. Each of these represented new EUs being added or pipeline changes.
The transmeta approach is different. Instead of breaking all the software for a new transmeta chip, you just change hte code morphing layer for the underlying physical processor. THe user-level (and emulated OS level) stuff never really knows the difference - it just runs (presumably) faster.
Of course, stuff running natively against transmeta chips will break, but it is expected this will be a small amount of software. If its just the linux kernel, it wont even matter - people tend to build kernels alot more often than transmeta releases chips
It will be a shame if transmeta doesn't make it in the market. They've got a lot more big names and big brains than linus, and for reasons i mentioned above, they do actually have a solid advantage over intel's strategy going forward.
Time will tell.
My opinions are my own, and do not necessarily represent those of my employer.
I'd much rather get ahold of a wearable. Does anyone know of any such plans? I've seen the PC104 based wearble that Charmed Technolgies makes but it's still a bit clunky. Maybe someday I can get a Microoptical display and set of datagloves, attached to a Transmeta wearable. Check out weracam for some really cool uses of wearables. Steve Mann has a system that let's you take notes on people and if the mounted camera recognizes them it automatically pops up the notes in overlay. He also has a high speed camera that lets him read the writing off the sides of tires on moving cars.
Now, as far as what these chips will be used for, there is a project called the "Land Warrior" project. The goal of this project seems to be to try and load a soldier up with almost every ungodly amount of sensory equipment as one can pack onto an individual human. We're talking like 45 extra pounds of technology, a heavier, albiet, more capable weapon system with things like rangefinders, Laser Target Designators and thermal sights all built into a rifle, with a grenade launcher and 5.56 round capablity, all in a single system.
The soldiers will have to also contend with sensory equipment integrated to thier helmet and a head's up display that can communicate terrain and other types of intel. The commander will be able to see what each soldier sees and intel can be shared from any one element on the battlefield to almost any other point.
This program assumes, in my view as an *infantryman* that the average soldier, who they just lowered the entrance standard (ASVAB - GT score) on so that they could get recruitment numbers up, will have a clue as to operating all this crap while still maintaining eyes on that critical 40-65 feet in front of them in an urban environment.
Uh, just a minute Bob, I have to reboot. Or "my head's up display got smudged from my camoflauge, and while I'm cleaning it so I can see what the heck might be around the next corner, I get whacked from some 3rd world irregular who leaned out the window across the street.
Now, these are bold strides that the military is trying to make in advancing thier state of technology. But technology alone is no *magic bullet* which will enable soldiers on the ground to be more lethal, if they can't move or don't have the intelligence to work the equipment they have been issued. We've barely got enough time to get people to be really effective as regular infantrymen and can't even afford enough fuel and bullets to have soldiers learn their basic jobs. So, now were going to buy all this neato technology that goes beyond the current (and future) warfighting need?
But hey, I hear they sure are going to be wearing neat hats pretty soon. Goes to show you, when someone was thinking "Can we do this", someone else forgot to ask "should we"?
But hey, good for Transmeta and all that. I'm sure those who got in at the IPO will be happy. As for the everyday soldier who has to content with all this stuff and doesn't even have the resources to maintain basic proficencies it won't mean a damn, not now, and surely not after they get loaded up with a lot more crap that their units will let them use about once a quarter.
The machines will almost certainly not run Linux or another other current OS. In a combat situation, it simply doesn't make sense for the computers the troops are using to have more than basic functionality, as the more complicated it is, the more attention will have to be put into using it.
I would imagine the basic idea of a wearable military computer would be to transmit terrain data and specific messages from commanding officers, and as such would be built to do just that with a minimum of user interaction. The machine will probably run a custom bare bones OS/application combination and will function more like an advanced GPS device than a desktop machine.
After all, do we really need our soldiers to be checking their email, buying stocks or talking to their buddies on AIM in the middle of a fight?
Palm: Low cost, low power, low heat, passable speed, huge user and programmer base.
I wonder what would happen if Transmeta and Palm ever hooked up? Could Transmeta's code-morphing answer the question that Palm users have been asking for months about Palm's eventual move to StrongARM chips? (Specifically: How to do it without locking out the millions of existing users and 100,000 existing developers.)
--GrouchoMarx
My other account is CmdrTaco
--GrouchoMarx
Card-carrying member of the EFF, FSF, and ACLU. Are you?
No, there are real uses for information technology of this type. I read about an airborne drop where soldiers transmitted their locations (computed throught GPS) to their platoon leader after they hit ground. (encrypted, of course.) The PL then looked at his map, picked a rally point, and transmitted the RP to all his troops... (encrypted again, of course).
One of the big problems with modern armies is that we don't make big formations anymore - people move in wedges and columns with 10-20 meters between each soldier. Multipy that by 200, and an infantry company will take up a square km of terrain. It's very easy to get lost/separated, wearable computers might solve this.