This page contains absolutely incorrect information.
The relevant portions are quoted below:-
"...make sure that the hardware donation includes the original operating system software. Keeping the operating system with the PC is not just a great benefit - it is a legal requirement. "
and
"Q. Why should a donor include the operating system with their PC donation? A. It is a legal requirement that pre-installed operating systems remain with a machine for the life of the machine. If a company or individual donates a machine to your school, it must be donated with the operating system that was installed on the PC."
There is no such legal requirement. The only legal requirement is that the OS on the donated PC at the time of the donation must be a legally licensed copy and that the licence (and any media etc.) are transferred with the PC.
It is perfectly legal to deinstall the pre-installed operating system and replace it prior to donating provided that the donation includes any necessary license for the OS (and other software) included on the PC. Your page is (deliberately?) misleading on this point. I presume that this is to discourage the use of non-Microsoft (since who elses OS currently gets pre-installed by OEMs?) operating systems within schools.
I don't dispute that iButton's are more secure than smartcards, but there is still more scope to break into the iButton than a fully tamper resistant device (the IBM 4758, for example).
You don't have to take our word for how secure this crypto iButton really is. The National Institute of Standards (NIST) and the Communications Security Establishment (CSE) have validated a version of the crypto iButton for protection of sensitive, unclassified information. FIPS 140-1 validation assures government agencies that the products provide a trusted, physically secure module to properly protect secure information.
FIPS 140-1 classification doesn't necessarily imply tamper resistance. It sets out 4 levels, with level 4 being the highest. At time of printing of my source doc (Ross Anderson's "Security Engineering", published 2001) there was only one level 4 device (IBM 4758 - the crypto unit used in e.g. ATM machines). The iButton falls officially into class 3 in FIPS 140-1, but in fact exceeds level three by some way. (Level 3 only requires potting of the components which doesn't rule out any scraping, sandblasting, drilling, EM leakage or memory remanance attacks etc.). FIPS 140-2 (which supercedes 140-1) is available online here .
The iButton falls into an area commonly known as level 3.5 and attacking it would be difficult, but not to the level of difficulty of a 4758 or similar device.
I would be particularly curious of how the iButton intends to detect "Micro-probing the chip" in order to trigger zeroisation. If this is purely based on the mesh layer in the chip then a sophisticated attacker using the "drill through the side" approach may be able to bypass this since the tamper resistant layer doesn't completely enclose the chip.
Not easy by any means, and certainly orders of magnitude better than a smart card, but it doesn't warrant the "You CANT do this to an iButton" position!
In fact, the IBM 4758, (or rather the CCA software supplied with it) can be cracked under certain privileged access conditions as demonstrated by a team in Anderson's group in Cambridge.
Re:iButonsare more secure than a smartcard.
on
Cracking the Smartcards
·
· Score: 2, Informative
The iButton's tamper resistance is not perfect - at least according to Ross Anderson, in "Security Engineering" (Chapter 14, "Physical Tamper Resistance"):-
... one might try drilling in through the side, then either probe the device in operation or disable the tamper-sensing circuitry. Because the iButton has lid switches to detect the can being opened, and its processor is mounted upside-down on the circuitboard (with a mesh in the top metal layer of the chip), this is unlikely to be a trivial exercise. It might well involve building custom jigs and tools. In short, it's a tempting target for the next bright graduate student who wants to win their spurs as a hardware hacker.
i.e. the "no-tamper technology" in the iButton is in the form of lid switches which may be defeatable by drilling in from the side, unlike e.g. the IBM 4758 cryptoprocessor which has a tamper-sensing mesh encasing it.
I'd agree. If the Internet Chess Servers are anything to go by, provided people can get interested enough (i.e. playing with a basic client such as the simple text based one mentioned in the article) then people will write clients.
There are loads of clients for (F)ICS and compatible servers out there, even though the original players had to use plaintext (telnet) sessions to play. Things started out with the likes of xics (now defunct, I think) and xboard, and have spread into other OSes (Winboard and others...), and into other languages (Java for example).
If the protocol is open, people will write clients for many reasons - an urge to use their pet programming language; a desire for features not available in other clients; just plain perversity or any combination thereof[1]
Good luck! I think provided you do get a simple example client that people can use to get hooked and give themselves ideas for what they'd like *their* interface to do, you should be alright.
Cheers,
Mark
[1] I wrote a meta-interface to ICS (ics.el) for a combination of all three of the reasons given - to learn more emacs lisp; for better commandline editing, text colourisation and the perversity of doing yet one more thing from within Emacs
Quantum Mechanics has been known to be a time-trasnlation invariant theory. In layman's term, it means that you can run the clock backwards and everything is fine. There is no "irreversible" process. (For the jargon-empowered, QM does not have a natural "arrow of time").
Whilst this may be true for "plain old QM" as embodied in the Schrodinger equation, for example, the Quantum Field Theories of modern particle physics certainly are not time-reversal invariant.
QFTs such as the standard model are provably invariant under a symmetry known as CPT. This is the combination of three individual symmetry operations:-
C - Charge reversal (roughly translated : swap every particle for its antiparticle)
P - Parity (roughly translated: reverse spatial coordinates - i.e. hold up a mirror)
T - Time reversal (the same as described above - run time backwards)
The decay of neutral K-mesons (or kaons) measurably violates CP
This implies a violation of T if CPT is to be preserved. The Standard Model (Glashow-Weinberg-Salam) incorporates CP violation, albeit in a kludgy manner (imaginary values in the Cabibbo-Kobayashi-Maskawa (CKM) mixing matrix)
Google for "CP-violation" for reference links
Paul Davies wrote a lay-persons book exploring the different "natural arrows of time" in physics including CP violation and 2nd Law of thermodynamics called (duh!) "The Arrow of Time" although I haven't read it (but have read other of his books).
Not totally useless - I used my copy of "Linux Unleashed" or "Slackware Linux Unleashed" or whatever it was called for a monitor stand for ages.
(It was initially bought for quick access to a CDROM in the days before distros came on magazine covers, but I digress.)
Aleph One (Aleph1) is math-speak for an uncountable infinity i.e. one that can't be mapped one-to-one to the set of natural numbers. (Also known as the second transfinite cardinal).
A countable infinity (e.g. the set of natural numbers) is given the name Aleph Null (or Aleph0).
This does not, of course, get us much closer as it still doesn't explain why Elias should choose to be uncountably infinite although Second Transfinite Cardinal has a kind of a cool pseudo-ecclesiastical ring to it.
I ran 8.0 since it came out on my P133 (non-mmx) 72MB box, at runlevel 3 (but used X via startx)
and it was fine. A bit slow in places
(gnome was painful as it painted menus on first use, KDE v. slow to start up but OK after that, WindowMaker was fine). I found 7.1 (or 7.2 I don't remember) a bit faster, particularly KDE1
startup as compared to 8.0's KDE2 startup.
It also works ok on my new 1GHz PIII 512MB box, but as you state above that's not unexpected;-)
"Theatri cal" trailers are those played in theatres, by live actors, not in a cinema (US: theater) or on television!
</REAL_(I.E._NON_US)_ENGLISH_TRANSLATION> ;-)
I once applied for a job in a company that
worked on Ministry of Defence projects.
As part of the application process I had
to fill in a security vetting form which
had that exact question, as well as a few
other yes/no tickbox questions along the
lines of
Have you ever smuggled illegal narcotics
Have you ever tried to overthrow the legitimate government of any country by force?
there were others, but it was a long time ago and I
don't remember that well.
Surely there hasn't been enough time for them to
have read rthille's
comment in the recent "Ask Slashdot" wireless
serial adapters article, much less for the resulting signals to have made it back to us already. --
I just looked at the Google cache of the site (link in somebody's comment above) and I see what you mean - they *do* give a strip 15 pixels wide,as you rightly note.
While the site was slashdotted, I was going on the slashdot story writeup which said "none of the artists get to see what has been created in
the adjacent squares" - I took this to mean zero-knowledge. --
I think it would perhaps make for a more interesting whole if you were given a (small,
maybe 10-20 pixel) slice from the edge of any
adjacent tile(s)that have already been filled
in order to be able to match (or blend or clash!)
as desired.
You could set up the square allocation algorithm
to prefer giving out squares that are adjacent to already filled in parts, and not adjacent to
already allocated, but not yet returned squares.
There wouldn't be a requirement to match/blend
and people could ignore that (or deliberately clash) if they wanted. It would be interesting to see the proportions that chose to do each...
If they have mass, then we must include that mass in all calculations, but for some reason they don't want to:-)
The current standard model does not predict the masses of neutrinos, but its equations are simpler if neutrinos have no mass.
That's like saying calculating the velocity of an object is easier to calcuate if we don't count friction!
Just because physically observed particles have mass, it is not necessarily required that the theory has particle masses in its bare Lagrangian form from which the perturbation theory Feynman rules are determined. (And I'm not talking about the Standard Model's Higgs Mechanism for mass generation by spontaneous symmetry breaking - which is another thing altogether...)
Non-perturbative calculations using the Schwinger-Dyson equations, Ward identities and renormalisability constraints show that masses can be generated dynamically through interactions of massless fields.
Some (8-10 year old) references can be found via this HEPDATA query. Note that this is not talking directly about neutrinos, but rather about generating masses for electrons in a simplified version of QED in which electrons start out massless.
There are almost certainly some newer papers that you could find either at HEPDATA or SPIRES.
(Full Disclosure: Mike Pennington was my Ph.D. Supervisor, although I didn't work in the non-perturbative SD equations field myself except for a short while at the start) --
Slashdot Mirror
FIPS 140-1 classification doesn't necessarily imply tamper resistance. It sets out 4 levels, with level 4 being the highest. At time of printing of my source doc (Ross Anderson's "Security Engineering", published 2001) there was only one level 4 device (IBM 4758 - the crypto unit used in e.g. ATM machines). The iButton falls officially into class 3 in FIPS 140-1, but in fact exceeds level three by some way. (Level 3 only requires potting of the components which doesn't rule out any scraping, sandblasting, drilling, EM leakage or memory remanance attacks etc.). FIPS 140-2 (which supercedes 140-1) is available online here .
The iButton falls into an area commonly known as level 3.5 and attacking it would be difficult, but not to the level of difficulty of a 4758 or similar device.
I would be particularly curious of how the iButton intends to detect "Micro-probing the chip" in order to trigger zeroisation. If this is purely based on the mesh layer in the chip then a sophisticated attacker using the "drill through the side" approach may be able to bypass this since the tamper resistant layer doesn't completely enclose the chip.
Not easy by any means, and certainly orders of magnitude better than a smart card, but it doesn't warrant the "You CANT do this to an iButton" position!
In fact, the IBM 4758, (or rather the CCA software supplied with it) can be cracked under certain privileged access conditions as demonstrated by a team in Anderson's group in Cambridge.
i.e. the "no-tamper technology" in the iButton is in the form of lid switches which may be defeatable by drilling in from the side, unlike e.g. the IBM 4758 cryptoprocessor which has a tamper-sensing mesh encasing it.
They ended up with University of Northumbria at Newcastle.
There are loads of clients for (F)ICS and compatible servers out there, even though the original players had to use plaintext (telnet) sessions to play. Things started out with the likes of xics (now defunct, I think) and xboard, and have spread into other OSes (Winboard and others...), and into other languages (Java for example).
If the protocol is open, people will write clients for many reasons - an urge to use their pet programming language; a desire for features not available in other clients; just plain perversity or any combination thereof[1]
Good luck! I think provided you do get a simple example client that people can use to get hooked and give themselves ideas for what they'd like *their* interface to do, you should be alright.
Cheers,
Mark
[1] I wrote a meta-interface to ICS (ics.el) for a combination of all three of the reasons given - to learn more emacs lisp; for better commandline editing, text colourisation and the perversity of doing yet one more thing from within Emacs
Whilst this may be true for "plain old QM" as embodied in the Schrodinger equation, for example, the Quantum Field Theories of modern particle physics certainly are not time-reversal invariant.
QFTs such as the standard model are provably invariant under a symmetry known as CPT. This is the combination of three individual symmetry operations:-
The decay of neutral K-mesons (or kaons) measurably violates CP This implies a violation of T if CPT is to be preserved. The Standard Model (Glashow-Weinberg-Salam) incorporates CP violation, albeit in a kludgy manner (imaginary values in the Cabibbo-Kobayashi-Maskawa (CKM) mixing matrix)
Google for "CP-violation" for reference links
Paul Davies wrote a lay-persons book exploring the different "natural arrows of time" in physics including CP violation and 2nd Law of thermodynamics called (duh!) "The Arrow of Time" although I haven't read it (but have read other of his books).
OT: Flanders and Swann wrote a song (v. funny) about the first and second laws of thermodynamics.
Not quite - the zSeries is the new name for the S/390 mainframes.
The 'midframe' AS/400 became the iSeries in the same renaming.
zSeries and iSeries scratch different itches and don't, by and large, compete
(Also the Netfinity x86 boxes became xSeries and the RS/6000 became pSeries)
Cheers,
Mark
I hope that the heat at the focus will be enough to melt/damage CDs on a good sunny day.
(Not my idea originally - someone suggested it during the /. coverage of the great AOL CD competition)
Not totally useless - I used my copy of "Linux Unleashed" or "Slackware Linux Unleashed" or whatever it was called for a monitor stand for ages.
(It was initially bought for quick access to a CDROM in the days before distros came on magazine covers, but I digress.)
Not in the distribution, but there is a paperclip.el available, of course.
A countable infinity (e.g. the set of natural numbers) is given the name Aleph Null (or Aleph0).
One reference is at http://mathforum.org/dr.math/problems/simakovsky10 .28.97.html
This does not, of course, get us much closer as it still doesn't explain why Elias should choose to be uncountably infinite although Second Transfinite Cardinal has a kind of a cool pseudo-ecclesiastical ring to it.
It also works ok on my new 1GHz PIII 512MB box, but as you state above that's not unexpected ;-)
"Theatr
</REAL_(I.E._NON_US)_ENGLISH_TRANSLATION>
I just upgraded my box to a 1GHz PIII + GeForce2 and installed quakeII just to see what I'd been missing and the difference is awesome.
I must dig out and install MDK, since that did use to have some difficulties on my old box.
As part of the application process I had to fill in a security vetting form which had that exact question, as well as a few other yes/no tickbox questions along the lines of
- Have you ever smuggled illegal narcotics
- Have you ever tried to overthrow the legitimate government of any country by force?
there were others, but it was a long time ago and I don't remember that well.(+1, Funny), I'd have thought, but then unclosed italic tags are a pet hate of mine
In particular, they advise limiting oneself to 4 cups of coffee over a 6 hour period, reduced to 2 if you are also drinking cola type drinks.
They don't mention caffeinated peppermints or Bawls, though.
Surely there hasn't been enough time for them to have read rthille's comment in the recent "Ask Slashdot" wireless serial adapters article, much less for the resulting signals to have made it back to us already.
--
--
While the site was slashdotted, I was going on the slashdot story writeup which said "none of the artists get to see what has been created in the adjacent squares" - I took this to mean zero-knowledge.
--
You could set up the square allocation algorithm to prefer giving out squares that are adjacent to already filled in parts, and not adjacent to already allocated, but not yet returned squares.
There wouldn't be a requirement to match/blend and people could ignore that (or deliberately clash) if they wanted. It would be interesting to see the proportions that chose to do each...
Just a thought.
--
ours blocks that too... (websense, I think)
--
Just because physically observed particles have mass, it is not necessarily required that the theory has particle masses in its bare Lagrangian form from which the perturbation theory Feynman rules are determined. (And I'm not talking about the Standard Model's Higgs Mechanism for mass generation by spontaneous symmetry breaking - which is another thing altogether...)
Non-perturbative calculations using the Schwinger-Dyson equations, Ward identities and renormalisability constraints show that masses can be generated dynamically through interactions of massless fields.
Some (8-10 year old) references can be found via this HEPDATA query. Note that this is not talking directly about neutrinos, but rather about generating masses for electrons in a simplified version of QED in which electrons start out massless.
There are almost certainly some newer papers that you could find either at HEPDATA or SPIRES.
(Full Disclosure: Mike Pennington was my Ph.D. Supervisor, although I didn't work in the non-perturbative SD equations field myself except for a short while at the start)
--