According to the article, they can't do it anymore (as of 11/2003), although that may have changed since, I have not researched it heavily.
Regardless of what the FBI legally can or can't do, I'm willing to bet that some bored OnStar employees listen in for entertainment. Even if you believe they don't, this proves the capability to activate your mic and listen to you without your knowledge exists, so I would suggest that from a privacy POV it must be considered a threat.
-speakers -dvi lcd (if the customer is complaining of a DVI problem, you'll need this) -spare PSU(s) -jumpers -Y power splitters -hard drives (pata, sata, various scsi if you got em) -hi-speed usb device (to test usb) -network connections (firewalled into its own DMZ, you don't want the customers wormed out pcs running wild behind your firewall) -cordless drill and charger -solder kit, heatshrink tubing -all the standard various screws computers come with -lots of outlets on a circuit supporting enough amperage to really use them
The correct usage is quickly losing ground, but a real english teacher (or copyeditor) will still find you in error for using the term acronym when you mean initialism, or vice versa.
While Merriam-Webster is good, I suggest you consult the MLA or the Chicago Manual of Style when you have real english usage questions. They (and the OED) are considered the actual standards.
Unless you motherboard is less than a 12-18 months old, don't count on it. It is very rare to see a manufacturer add new features into motherboards older than that.
They will fix serious bugs if they find them, maybe add support for a new type of processor (because it usually requires little effort), but they really don't want to be supporting your board at all any more. They definately don't want to be spending time back-porting code for new features that may very well end up not working or causing problems with your board. Either situation means they have to spend even MORE time bugfixing for this board they just want to be done with.
The idea of needing to verify content over the net before playing it is an AWFUL one.
A DDOS against the verification servers and NO ONE will be able to play anything. A few million people unable to use their players ought to put a quick stop to this kind of DRM, to say nothing of people spoofing the self-destruct update to peoples machines and rendering them useless.
The authors don't understand that both 1024 and 1000 are used, but never (by knowledgeable people), and claim not to understand why 1024?
Actually both are used, but each has its own correct context.
When a hard drive manufactorer describes hard disk capacity, they use the power of 10 series, because they count the actual number of bits that their hd has, if each platter contains 480,000,000 bits, and it has 2 platters, that's 120,000,000 bytes, therefore exactly 120MB by their count.
When networking throughput is described the power of ten series is used, this is because the crystal oscillator governing the clock of the network carrier wave is base 10, if it operates at 100Mhz it really is 100,000,000 Hz. Therefore if you get one bit per Hz you have exactly 100Mbs, it is completely unrelated to powers of 2.
However, when describing file sizes (arguably the most common use), we use the power of 2 series, because that's the way the filesystem clusters work. Clusters are powers of 2 because we describe the number of bytes used in a cluster as some number of bits, we would be wasting some of those bits if we did not make the cluster a full power of 2 size. At that point, for the sake of consistant math, clusters are counted in powers of two as well.
If you didn't, you'd run into problems, i mean you could say a megabyte of storage equals 1000 clusters of 1024 bytes (so 1,024,000 bytes), but what if the clusters are 16384 bytes, then a megabyte is 62.5 clusters, which is contrary to the whole idea of working with integer cluster numbers. By making a megabyte (for instance) 1024 clusters of 1024 bytes (so 1048576 bytes), you can always maintain an integer number of clusters to make a megabyte, and so on to gigabytes, terabytes, etc . ..
As an aside, this difference between filesystem capacity (measured in powers of 2) and raw number of bits on each platter (measured in powers of 10) is why your "120GB" hd formats to less than 111.8GB "free space" (plus a little more loss to filesystem overhead).
RAM and other memories are also measured in powers of 2, although for a different reason. Somewhat simplifying, each bit of information held in these memory technologies represents a stored charge (or lack of one) across a transistor, these transistors are individually addressed using other transistors to select which bit to check. But one transistor can only choose between two sources (two bits), so another is used in parallel to choose between between the next two bits, and then another is used in series with the previous two to choose between those two choosers (so we can now choose between 4 bits), then another group of three is used in parallel with the first group to choose between the next 4 bits, with another transistor in series with those two groups to now choose between 8 bits, and on and on in a continuing power of 2 addressability expansion. Thus it becomes natural to manufacture (and therefore measure) memory in power of 2 capacities.
As a final note: to unambigiously refer to 2^20, one may use the term Mebibyte.
Won't fit on all motherboards in all cases, but it has fit in most of the machines I've built, and the performance is amazing. I suggest a panaflo low or high speed fan (120mm) with it for a super quiet or super cool solution (respectively).
Sirius is losing the car partnership race. Folks that buy new cars have something like a 30-50% conversion rate for whatever ships in the cars. That is becoming, increasingly, XM. Their new deal with Hyundai is going to probably bring in more subscribers than Stern ever would, but it didn't cost half a billion.
Couldn't be more wrong. It is becoming, decreasingly XM. The fact is that most after-market stereos and builtins that had sat radio were XM, that is rapidly changing.
Just look at Alpine, before pretty much all their good decks were XM, now all their newer decks are "sat. ready" and can take XM or Sirius, because their customers were demanding Sirius.
BTW, if you're buying a Hyundai, you probably can't afford sat. radio.
Thank god someone here knows what they're talking about, this use of "reverse firewall" as if it were anything different than an everyday firewall was driving me nuts. Kudos!
Ingress and Egress people; this is what you should have been doing all along. Nothing new to see here, and certainly nothing reversed.
Microsoft is finally doing the right thing and people are still bitching.
Implementing OpenBSD's W^X memory locking, vastly improving the firewall and turning it on by default, and disabling that goddamn messenger POS are all very good things.
If it breaks a few programs so be it: the breaks won't last long, but the vast improvements in security will.
as i look for a network administration job, almost everywhere i see MCSE + CCNA desired, and often required. Nevermind my degree in computer engineering from a top 10 engineering school.
They may have little practical value, but they seem to help get your foot in the door.
Me and another EFF staffer and the Coordinator of the Union for the Public Domain created a heavily editorialized impressionistic transcript of the meeting (EFF mirror, UPD mirror), trying to untie the knots in the negotiation.
I know we're a bunch of tech nerds here, but please, writing like this undermines the importance of what you're saying.
An EFF staffer, the Coordinator of the Union for the Public Domain and I created a heavily editorialized impressionistic transcript of the meeting (EFF mirror, UPD mirror): trying to untie the knots in the negotiation.
Slashdot Mirror
Here's a link to a CNET article about the FBI using OnStar to listen to people: http://news.com.com/2100-1029-5109435.html
According to the article, they can't do it anymore (as of 11/2003), although that may have changed since, I have not researched it heavily.
Regardless of what the FBI legally can or can't do, I'm willing to bet that some bored OnStar employees listen in for entertainment. Even if you believe they don't, this proves the capability to activate your mic and listen to you without your knowledge exists, so I would suggest that from a privacy POV it must be considered a threat.
-speakers
-dvi lcd (if the customer is complaining of a DVI problem, you'll need this)
-spare PSU(s)
-jumpers
-Y power splitters
-hard drives (pata, sata, various scsi if you got em)
-hi-speed usb device (to test usb)
-network connections (firewalled into its own DMZ, you don't want the customers wormed out pcs running wild behind your firewall)
-cordless drill and charger
-solder kit, heatshrink tubing
-all the standard various screws computers come with
-lots of outlets on a circuit supporting enough amperage to really use them
probly lots i'm forgetting, but it's a start.
The correct usage is quickly losing ground, but a real english teacher (or copyeditor) will still find you in error for using the term acronym when you mean initialism, or vice versa.
s m
While Merriam-Webster is good, I suggest you consult the MLA or the Chicago Manual of Style when you have real english usage questions. They (and the OED) are considered the actual standards.
The acronym/initialism situation is discussed on wiki here: http://en.wikipedia.org/wiki/Acronym_and_initiali
Unless you motherboard is less than a 12-18 months old, don't count on it. It is very rare to see a manufacturer add new features into motherboards older than that.
They will fix serious bugs if they find them, maybe add support for a new type of processor (because it usually requires little effort), but they really don't want to be supporting your board at all any more. They definately don't want to be spending time back-porting code for new features that may very well end up not working or causing problems with your board. Either situation means they have to spend even MORE time bugfixing for this board they just want to be done with.
Every new version of windows has beefed up the requirements, and I've always found them usable with less than they say.
With virtual memory disabled, I get over 18 billion terabytes of potential swap file space! Cool!
18 billion terabytes == 18 zettabytes. Coincidence?
The idea of needing to verify content over the net before playing it is an AWFUL one.
A DDOS against the verification servers and NO ONE will be able to play anything. A few million people unable to use their players ought to put a quick stop to this kind of DRM, to say nothing of people spoofing the self-destruct update to peoples machines and rendering them useless.
The authors don't understand that both 1024 and 1000 are used, but never (by knowledgeable people), and claim not to understand why 1024?
.
Actually both are used, but each has its own correct context.
When a hard drive manufactorer describes hard disk capacity, they use the power of 10 series, because they count the actual number of bits that their hd has, if each platter contains 480,000,000 bits, and it has 2 platters, that's 120,000,000 bytes, therefore exactly 120MB by their count.
When networking throughput is described the power of ten series is used, this is because the crystal oscillator governing the clock of the network carrier wave is base 10, if it operates at 100Mhz it really is 100,000,000 Hz. Therefore if you get one bit per Hz you have exactly 100Mbs, it is completely unrelated to powers of 2.
However, when describing file sizes (arguably the most common use), we use the power of 2 series, because that's the way the filesystem clusters work. Clusters are powers of 2 because we describe the number of bytes used in a cluster as some number of bits, we would be wasting some of those bits if we did not make the cluster a full power of 2 size. At that point, for the sake of consistant math, clusters are counted in powers of two as well.
If you didn't, you'd run into problems, i mean you could say a megabyte of storage equals 1000 clusters of 1024 bytes (so 1,024,000 bytes), but what if the clusters are 16384 bytes, then a megabyte is 62.5 clusters, which is contrary to the whole idea of working with integer cluster numbers. By making a megabyte (for instance) 1024 clusters of 1024 bytes (so 1048576 bytes), you can always maintain an integer number of clusters to make a megabyte, and so on to gigabytes, terabytes, etc . .
As an aside, this difference between filesystem capacity (measured in powers of 2) and raw number of bits on each platter (measured in powers of 10) is why your "120GB" hd formats to less than 111.8GB "free space" (plus a little more loss to filesystem overhead).
RAM and other memories are also measured in powers of 2, although for a different reason. Somewhat simplifying, each bit of information held in these memory technologies represents a stored charge (or lack of one) across a transistor, these transistors are individually addressed using other transistors to select which bit to check. But one transistor can only choose between two sources (two bits), so another is used in parallel to choose between between the next two bits, and then another is used in series with the previous two to choose between those two choosers (so we can now choose between 4 bits), then another group of three is used in parallel with the first group to choose between the next 4 bits, with another transistor in series with those two groups to now choose between 8 bits, and on and on in a continuing power of 2 addressability expansion. Thus it becomes natural to manufacture (and therefore measure) memory in power of 2 capacities.
As a final note: to unambigiously refer to 2^20, one may use the term Mebibyte.
Hope at least one person found that useful.
Usually the real server can still be discerned by things like implemented methods and syntax of messages.
But, if its just for the script kiddies, probly work well enough.
Do we really need it revealed? It's not hard to figure out, just send a bad HTTP request:
bbc.co.uk
HTTP/1.1 400 Bad Request
Date: Fri, 03 Jun 2005 19:07:25 GMT
Server: Apache/2.0.51 (Unix)
there is a huge (real) heatsink that is amazing: Thermalright's XP-120
Won't fit on all motherboards in all cases, but it has fit in most of the machines I've built, and the performance is amazing. I suggest a panaflo low or high speed fan (120mm) with it for a super quiet or super cool solution (respectively).
How is this beating "embiggen" and "cromulent" ?
Sirius is losing the car partnership race. Folks that buy new cars have something like a 30-50% conversion rate for whatever ships in the cars. That is becoming, increasingly, XM. Their new deal with Hyundai is going to probably bring in more subscribers than Stern ever would, but it didn't cost half a billion.
Couldn't be more wrong. It is becoming, decreasingly XM. The fact is that most after-market stereos and builtins that had sat radio were XM, that is rapidly changing.
Just look at Alpine, before pretty much all their good decks were XM, now all their newer decks are "sat. ready" and can take XM or Sirius, because their customers were demanding Sirius.
BTW, if you're buying a Hyundai, you probably can't afford sat. radio.
The iRiver players are imo much better than any of the ipods, and yes, they support OGG (and WMA for that matter).
http://www.iriveramerica.com/prod/ultra/
They even offer a firmware that will let's the drive behave like a normal flash drive that you can just drag and drop files to.
I recommend the 800 series, very nice.
Deja vu all over again.
Andrew has it correct, and yes, this does increase security.
The ultra paranoid have been using IDSs that use take both MD5 and SHA-1 on the system files for a while now.
Thank god someone here knows what they're talking about, this use of "reverse firewall" as if it were anything different than an everyday firewall was driving me nuts. Kudos!
Ingress and Egress people; this is what you should have been doing all along. Nothing new to see here, and certainly nothing reversed.
off-by-one-million error.
The only "correct use of a language", by definition, is the way the majority of the populace is using it.
Umm, no, that's simply not true. The majority of the populace mix up ensure, assure, and insure. That doesn't make them interchangable.
There is correct use of language, and there is incorrect; popular misunderstanding doesn't make it right.
"This is my everything's OK alarm! This will sound every 4 seconds, as long as everything is OK!
It can't be turned off, but it does . . . uh, break easily."
I believe i read that one of the points of this was going to be dramatically improving the fail-over abilities of OpenBSD.
Microsoft is finally doing the right thing and people are still bitching.
Implementing OpenBSD's W^X memory locking, vastly improving the firewall and turning it on by default, and disabling that goddamn messenger POS are all very good things.
If it breaks a few programs so be it: the breaks won't last long, but the vast improvements in security will.
as i look for a network administration job, almost everywhere i see MCSE + CCNA desired, and often required. Nevermind my degree in computer engineering from a top 10 engineering school.
They may have little practical value, but they seem to help get your foot in the door.
Me and another EFF staffer and the Coordinator of the Union for the Public Domain created a heavily editorialized impressionistic transcript of the meeting (EFF mirror, UPD mirror), trying to untie the knots in the negotiation.
I know we're a bunch of tech nerds here, but please, writing like this undermines the importance of what you're saying.
An EFF staffer, the Coordinator of the Union for the Public Domain and I created a heavily editorialized impressionistic transcript of the meeting (EFF mirror, UPD mirror): trying to untie the knots in the negotiation.
All Computers is asking for the tidy sum of $500 million USD.
. . . but we'll settle out of court right now for 50 bucks.