Yes, and notice that it's only the desktops that include Mandrake and TurboLinux in their certification matrix. Workstations and laptops are limited to being certified with RedHat and Suse, at least according to the link you provided.
From the example pictures it seems that this still won't be able to detect items that have been put in some of the body's natural "hiding places." I really doubt someone that is willing to die to blow up an airliner full of passengers is going to have any scruples about doing something like that.
On the other hand, it's also possible to do that when you're just being checked out with metal detectors.
Additionally, many firmware implementations for 802.11 products have been updated to not use weak initialization vectores (IVs), which are used in ultimately decyphering a WEP key.
It's silly to compare these kits to a brand new Mac when Apple themselves sell refurbished products.Keep in mind that the refurb'ed PowerMacs already come with ram, an hd, graphics, a superdrive, an os, etc. Oh yeah, they also have a one year warranty from Apple and are still eligible for their AppleCare Protection Plan. Two things that I doubt these kits have.
I don't have experience with this myself, but if I were in your boat I would make a system for classifying types of disaster and the appropriate recovery methods for each. For instance at the top level you would have either a disaster resulting in either physical damage or non-physical damage. From there you could classify disaster types according to how much and what physical damage occurred.
So, your meteorite example would probably fall into something between a horrible fire and earthquake, as the kind of damage inflicted on your facility would be similar in such events.
The OpenLDAP library provides a mechanism for local client side caching of results from queries against an LDAP directory. For more info check out the ldap_cache man page.
A connection (A) is made from a machine and then another connection (B) is made from that very machine. In an ideal situation IPid(B) == IPid(A)+1.
However, even if this ideal case were not to occur (i.e, other connections were made in the time between A and B) IPid(A) Gaplen. This Gaplen constant reflects the number of connections the machine could possibly make during the time between connection A and B and is chosen based on the bandwidth and latency of the connection.
So, if connections A and B came from different machines, IPid(B) would most likely lie outside the range of IPid(A) and IPid(A)+Gaplen, because they're not starting with the same IPid at the beginning of the observed timeframe.
(That is if they're not using PRNGs for IPid generation )
If you initiate a connection (Connection A)from a machine and then immediately initiate another (B) ideally you would have sequential IPids (i.e., IPid(B) == IPid(A)+1). Even under non-ideal circumstances (several TCP sessions are created in between A and B) if you take into the account the characteristics of the transmission medium (i.e., bandwidth and latency) IPid(A) IPid(B) = IPid(A) + Gaplen. This Gaplen is basically a range of IPids that the host that initiated connection A could possibly generate in the time between A and B.
If A and B came from different machines there IPids would ideally be generated from a different base number for the the timeframe and thus IPid(B) would fall outside of the range of IPid(A) and IPid(A) + Gaplen.
"...Engelbart and a group of young computer scientists and electrical engineers he assembled in the Augmentation Research Center at SRI were able to stage a 90-minute public multimedia demonstration of a networked computer system. This was the world debut of the computer mouse, 2-dimensional display editing, hypermedia--including in-file object addressing and linking, multiple windows with flexible view control, and on-screen video teleconferencing."
Basically Engelbart came up with the concept of the modern GUI and the means by which most people interact with it. While not strictly a software innovation I would consider this as falling under your criteria as its affect has been widespread.
Related /. Article
on
Sun vs. OpenBSD?
·
· Score: 2, Informative
First of all you weren't very clear about whether you knew that PCI already did this or not. Saying:
PCI goes a long way to do this, and most new devices have some sort of identification. Basically every model of every device from every manufacturer needs a unique ModelID which is easily retrievable according to the basic protocols of the bus in question.
makes it sound more like you were unaware of these capabilities with PCI and were proposing a peripheral bus that had these capabilities.
A "whole lot" of new manufacturers? Doing a quick grep of this reveals there to be ~6243 different vendor prefixes. A mere 0.0003721117 of the vendor prefix address space of ~16mil is being used. Now perhaps 16mil devices per vendor is a bit low in this day and age of pervasive networking, but they seemed to alleviate this problem quite well by allocating multiple prefixes to a vendor. If you would like to provide information to backup your claim of MAC addressing problems I would honestly be very interested in seeing it.
Comparing this to IP addressing is like comparing apples to oranges. Yes theoretically speaking there could be infinite companies producing infinite different hardware products. But in the real world it takes capital to start a company, and even more to bring a product to market. It doesn't seem to take so much of that hard earned cash to plug in another host to the internet though, now does it?
While I do agree with you about the size of the IP address space and the y2k problem being quite shortsighted, I really do not think this is relevent in this case? Yes IPv6 has a 128bit address, but that's because in the wonderful future eveything imaginable will have an IP address.
The only thing I was calling into question was you pulling the figure of 128bits out of thin air. Now this discsussion has just become totally irrelevent to the original question.
The PCI bus already uses vendor and device ids. Here is a list of PCI vendor ids. If you click on the Vendor/ID you will get a list of the known device ids held by that vendor.
And don't you think 64bits for vendor and device is a bit much? That would allow for 18446744073709551616 vendors with an allotment of the same number of devices for each one.
The problem with using rc5 as a benchmark in your comparison is distributed.net's use of an AltiVec optimized core for their client when run on G4 processors. The only valid use of rc5 as a benchmark is when comparing between processors with the same instruction set.
Re:Pricing problem
on
Itanium Problems
·
· Score: 3, Informative
You know, you can download the compiler for evaluation purposes to actually see if there is a speedup in your application. The linux version is even free for non-commercial use.
Get off your sedentary ass, jog and lift weights. Swim too. Why the hell is this an Ask/. question? Don't you think this is the wrong place to ask about fitness? Sure you may be a so called "geek", but as far as fitness that probably means that you're some sedentary lardass like the rest of the country. So, I think that if you are indeed a homo-sapien you should be able to apply the ample advice given to non-geeks.
I'm sorry, I'm just disgusted with all these Ask/. questions with the word "geek" in them.
Will they sue the Internet Archive too? I'm not sure if this contains links to the offending rail sabotage information (I'm a stupid USian), but I would just guess that it does.
Slashdot Mirror
Yes, and notice that it's only the desktops that include Mandrake and TurboLinux in their certification matrix. Workstations and laptops are limited to being certified with RedHat and Suse, at least according to the link you provided.
Here's a previous Wired article about Tanya Grotter from last November.
From the example pictures it seems that this still won't be able to detect items that have been put in some of the body's natural "hiding places." I really doubt someone that is willing to die to blow up an airliner full of passengers is going to have any scruples about doing something like that.
On the other hand, it's also possible to do that when you're just being checked out with metal detectors.
I fear if IT unions were to become common we'd see an acceleration of offshore white-collar job trend.
Additionally, many firmware implementations for 802.11 products have been updated to not use weak initialization vectores (IVs), which are used in ultimately decyphering a WEP key.
Eerily this Gary Morse guy reminds me of John Vranesevich.
It's silly to compare these kits to a brand new Mac when Apple themselves sell refurbished products.Keep in mind that the refurb'ed PowerMacs already come with ram, an hd, graphics, a superdrive, an os, etc. Oh yeah, they also have a one year warranty from Apple and are still eligible for their AppleCare Protection Plan. Two things that I doubt these kits have.
A 2.4Ghz video sniffer built with *GASP* a 2.4Ghz video receiver!
NAGIOS is an opensource network monitoring system, they have some links on their site for ethernet connected temperature probe devices.
All encompassing Digital Rights Management?
I don't have experience with this myself, but if I were in your boat I would make a system for classifying types of disaster and the appropriate recovery methods for each. For instance at the top level you would have either a disaster resulting in either physical damage or non-physical damage. From there you could classify disaster types according to how much and what physical damage occurred.
So, your meteorite example would probably fall into something between a horrible fire and earthquake, as the kind of damage inflicted on your facility would be similar in such events.
This question was asked quite awhile ago here.
There was a tutorial mentioned in the comments.
The OpenLDAP library provides a mechanism for local client side caching of results from queries against an LDAP directory. For more info check out the ldap_cache man page.
And that would stop the president from signing an executive order saying otherwise?
A connection (A) is made from a machine and then another connection (B) is made from that very machine. In an ideal situation IPid(B) == IPid(A)+1.
However, even if this ideal case were not to occur (i.e, other connections were made in the time between A and B) IPid(A) Gaplen. This Gaplen constant reflects the number of connections the machine could possibly make during the time between connection A and B and is chosen based on the bandwidth and latency of the connection.
So, if connections A and B came from different machines, IPid(B) would most likely lie outside the range of IPid(A) and IPid(A)+Gaplen, because they're not starting with the same IPid at the beginning of the observed timeframe.
(That is if they're not using PRNGs for IPid generation )
If you initiate a connection (Connection A)from a machine and then immediately initiate another (B) ideally you would have sequential IPids (i.e., IPid(B) == IPid(A)+1). Even under non-ideal circumstances (several TCP sessions are created in between A and B) if you take into the account the characteristics of the transmission medium (i.e., bandwidth and latency) IPid(A) IPid(B) = IPid(A) + Gaplen. This Gaplen is basically a range of IPids that the host that initiated connection A could possibly generate in the time between A and B.
If A and B came from different machines there IPids would ideally be generated from a different base number for the the timeframe and thus IPid(B) would fall outside of the range of IPid(A) and IPid(A) + Gaplen.
"...Engelbart and a group of young computer scientists and electrical engineers he assembled in the Augmentation Research Center at SRI were able to stage a 90-minute public multimedia demonstration of a networked computer system. This was the world debut of the computer mouse, 2-dimensional display editing, hypermedia--including in-file object addressing and linking, multiple windows with flexible view control, and on-screen video teleconferencing."
Basically Engelbart came up with the concept of the modern GUI and the means by which most people interact with it. While not strictly a software innovation I would consider this as falling under your criteria as its affect has been widespread.
For more info see this previous /. article from the BSD section that didn't make it to the main page here:9 13211&mode=flat&tid=122
http://bsd.slashdot.org/article.pl?sid=02/11/26/1
First of all you weren't very clear about whether you knew that PCI already did this or not. Saying:
PCI goes a long way to do this, and most new devices have some sort of identification. Basically every model of every device from every manufacturer needs a unique ModelID which is easily retrievable according to the basic protocols of the bus in question.
makes it sound more like you were unaware of these capabilities with PCI and were proposing a peripheral bus that had these capabilities.
A "whole lot" of new manufacturers? Doing a quick grep of this reveals there to be ~6243 different vendor prefixes. A mere 0.0003721117 of the vendor prefix address space of ~16mil is being used. Now perhaps 16mil devices per vendor is a bit low in this day and age of pervasive networking, but they seemed to alleviate this problem quite well by allocating multiple prefixes to a vendor. If you would like to provide information to backup your claim of MAC addressing problems I would honestly be very interested in seeing it.
Comparing this to IP addressing is like comparing apples to oranges. Yes theoretically speaking there could be infinite companies producing infinite different hardware products. But in the real world it takes capital to start a company, and even more to bring a product to market. It doesn't seem to take so much of that hard earned cash to plug in another host to the internet though, now does it?
While I do agree with you about the size of the IP address space and the y2k problem being quite shortsighted, I really do not think this is relevent in this case? Yes IPv6 has a 128bit address, but that's because in the wonderful future eveything imaginable will have an IP address.
The only thing I was calling into question was you pulling the figure of 128bits out of thin air. Now this discsussion has just become totally irrelevent to the original question.
The PCI bus already uses vendor and device ids. Here is a list of PCI vendor ids. If you click on the Vendor/ID you will get a list of the known device ids held by that vendor.
And don't you think 64bits for vendor and device is a bit much? That would allow for 18446744073709551616 vendors with an allotment of the same number of devices for each one.
Commands like rm use "--" to mark the end of the command line options. So to use your example, you could delete "-p" by running rm -- -p.
Try it:
touch -- -p; ls; rm -- -p; ls
The problem with using rc5 as a benchmark in your comparison is distributed.net's use of an AltiVec optimized core for their client when run on G4 processors. The only valid use of rc5 as a benchmark is when comparing between processors with the same instruction set.
You know, you can download the compiler for evaluation purposes to actually see if there is a speedup in your application. The linux version is even free for non-commercial use.
Get off your sedentary ass, jog and lift weights. Swim too. Why the hell is this an Ask /. question? Don't you think this is the wrong place to ask about fitness? Sure you may be a so called "geek", but as far as fitness that probably means that you're some sedentary lardass like the rest of the country. So, I think that if you are indeed a homo-sapien you should be able to apply the ample advice given to non-geeks.
/. questions with the word "geek" in them.
I'm sorry, I'm just disgusted with all these Ask
Will they sue the Internet Archive too? I'm not sure if this contains links to the offending rail sabotage information (I'm a stupid USian), but I would just guess that it does.