http://finance.yahoo.com/news/announ...011240408.html
Quote:
In fact Fink announced on Thursday that the company is working on a brand new free and open-source operating system and is inviting universities to help research and build it.
He threw an little dig at Microsoft when announcing the news, saying:
"We want to reignite in all of our universities around the world operating system research which we think has been dormant or stagnant for decades."
On top of that, HP is working on a brand new operating system for The Machine based on Linux. And another one based on Android, Fink continued:
"We are, as part of The Machine, announcing our intent to build a new operating system all open source from the ground up, optimized for non-volatile memory systems.
...
We also have a team that's starting from a Linux environment and stripping out all the bits we don't need. So that way you maintain ... compatibility for apps.
What if we build a version of Android? ... We have a team that's doing that, too."
They are calling their new platform "The Machine". It is (or will be) a "state computer", the opposite of today's stateless computers.
hp-machine.jpg
Quote:
The Machine's claim to fame is that it can process loads of information instantly while using hardly any power. HP wants this computer to replace the servers being used in today's data centers. But it also hopes the tech will become the basis for the next generation of PCs.
...
And The Machine will not use Windows.
...
"We want to reignite in all of our universities around the world operating system research which we think has been dormant or stagnant for decades."
On top of that, HP is working on a brand new operating system for The Machine based on Linux. And another one based on Android, Fink continued:
Here is where I think it will get squirrelly. We've seen this approach before, when Apple asked FOSS FreeBSD programmers to pitch in and help build "Darwin", which Apple plundered without returning any significant modifications back to the FreeBSD community. Apple even took Konqueror and made it the basis of their Safari browser, without returning anything to the KDE community. They did their exploitation by making the hardware interface (i.e., like libc6 library) a proprietary binary that only had the entrance ports public. When coders looked at the FOSS code they created but Apple modified all they saw were software hooks into proprietary binary blobs, blobs which they didn't have access to.
And, IIRC, it was Steve Riley who pointed out that Microsoft is already selling smartphones that are running Android, so proprietary hardware companies using FOSS isn't new.
The concept of a "state" is not new, either. Both machines and software can have "states", but it has been a lot easier to maintain a state in software than in hardware because of past limitations on the amount of memory available to hardware, since there can be only 2^N states in hardware, where "N" is the number of memory bits available. http://en.wikipedia.org/wiki/State_%...ypes_of_states
An aside:
Martin Fink's reference to "operating system research which we think has been dormant or stagnant for decades" recalls a time in the late 1970s and early 1980s, and my first experience with a "state" computer.
In the spring of 1981, I flew myself and two of my Apple clients to a computer show at Springfield, Ill. They were local farmers using their new Apple's to do farm accounting, but were wanting to look at different software, and hardware if necessary, because spreadsheets and text editors didn't fill their needs. We flew there to see if there was new farm software available. Our plan was simple. They'd test the software and I'd test the hardware, if it wasn't an Apple.
The giant building had a couple dozen vendor stations around its perimeter. We went to one after the other. Either the software was bad or the hardware was junk.
Then we came to "DAVID". "DAVID" was one of a family of state computers made by the Logical Machine Corporation (LOMAC). The other members of the family were TINA, ADAM and GOLIATH. DAVID, like its siblings, is what is called a "State Machine", which usually depends on non-volatile memory. There is little on the Internet about this family of computers, but I did find this magazine article from March of 1981, "ADAM and Edlin", in which the marketing VP of LMC responds to criticism of "ADAM".
Ned and Steve found "DAVID"'s software unique but workable. While Ned was playing with the software I asked Buddy Miller III what would happen if the power went out. He said "Like this?" and yanked the power cord from its socket. DAVID immediately blinked out. He then plugged it in and DAVID rebooted and returned to exactly the same place it was at when the cord was pulled. Ned was returned to exactly where he was when the machine suddenly lost power. The "state" of the machine and software had been preserved. It was perfect except for one thing, the price. But, Buddy also told us about "SAVVY", an "artificial intelligence" program, as he mislabelled it, which allowed one to communicate with the machine in ordinary English. And, it was being put onto a peripheral card for both the IBM PC and the Apple PC. That conversation about SAVVY (Excalibur Corporation) led to another adventure in my life, but that is a story for another time.
DAVID died with LOMAC. HP want's to resurrect it! I wish them well, but I doubt it will benefit Linux or FOSS much. IF HP sold a Personal Computer which was a state machine running a modified (but proprietary) version of Linux, would you buy it? If their proprietary model prevented a timely update of security and bugs patches, like such as with Windows, would you still buy "The PC Machine"?
EDIT:
"The Machine" uses (or will use) what are called Memistors (memory +resistors):
Quote:
What is a memristor? Memristors are basically a fourth class of electrical circuit, joining the resistor, the capacitor, and the inductor, that exhibit their unique properties primarily at the nanoscale. Theoretically, Memristors, a concatenation of “memory resistors”, are a type of passive circuit elements that maintain a relationship between the time integrals of current and voltage across a two terminal element. Thus, a memristors resistance varies according to a devices memristance function, allowing, via tiny read charges, access to a “history” of applied voltage. The material implementation of memristive effects can be determined in part by the presence of hysteresis (an accelerating rate of change as an object moves from one state to another) which, like many other non-linear “anomalies” in contemporary circuit theory, turns out to be less an anomaly than a fundamental property of passive circuitry.
Until recently, when HP Labs under Stanley Williams developed the first stable prototype, memristance as a property of a known material was nearly nonexistant. The memristance effect at non-nanoscale distances is dwarfed by other electronic and field effects, until scales and materials that are nanometers in size are utilized. At the nanoscale, such properties have even been observed in action prior to the HP Lab prototypes.
Stanley Williams should get a Nobel (or some other significant) prize for this. First theorized in 1971, it has taken 53 years to create a working prototype.
Attached Images
hp-machine.jpg
(50.5 KB)