The term "Megabytes" is ambiguous, because some
say 1 Megabyte = 8,000,000 bits.
Other people say
1 Megabyte = 8,388,608
Bits is the preferred method of counting storage used, because the amount specified is unambiguous.
Nobody can dispute how many bits there are in 700 * 10^6 bits.
Maybe they'll just clone them, creating identical com.oracle.* packages. Meanwhile the com.sun.* packages will ship with java, be deprecated and not receive any updates.
Not like anyone will notice their JRE installs growing from 700mb in size to 1000mb in size.
This does not work so well when the feature is present but buggy in various browsers, or for features whose presence or absence cannot easily be detected, the result would be inconsistent rendering if user agent was not checked to redirect the user to the proper page.
It is not a best practice to rely on attempts to "test for the presence" of a feature, to determine whether it should be used.
"Detection" is only possible if using javascript: not all browsers support, and some users might have it turned off.
User agent is commonly used for this purpose, and it is the best practice as defined by the standard to use the user agent field for the purpose of tailoring responses to avoid particular user agent limitations
Because, without a fixed size, the VM would always eventually grow to consume all available memory.
Recall, Java programmers don't manage their memory, they rely on Java's garbage collection.
GC is triggered when the amount of memory available to the VM drops below a specified threshold.
If the VM can grow to use all available memory, that threshold won't ever be reached, until java is taking over all physical memory: even with the limit it tends to do so anyways, but i'm just saying.....
No they're not. After the fiasco regarding blocking access to firmware updates for Sun servers on the new site we stopped considering using any more Sun/Oracle products for new deployments.
Of course old equipment is still badged with the Sun name. Whatever Oracle's doing is getting the opposite of advertising.
So if Microsoft gets bought out by Oracle and the User-Agent string of IE6, IE7, and IE8 are immediately updated to be OIE6 7 or 8, respectively, with a surprise critical update..
You are saying it's the web developers' fault, and not Oracle's, when this poorly conceived change breaks the web and compatibility for IE users?
All they have to do is file a counter-notice, citing their license to use the work under the CC-NC, then the DMCA letter has no benefit.
Contrary to what the author seems to suggest the "No Commercial Use" restriction of CC-BY-NC does not mean that no advertising or thing of commercial value to the site operator can appear on the same site that a work appears on.
Now, someday, the servers will be smart enough to tell the storage array when they're done with a chunk of storage.
Servers are that smart... most commonly this is needed for using SSD drives -- SCSI PUNCH, SATA TRIM, or writing a block of all zeros to a sector.... there are OS configurations that support this, and if you don't have an OS that can handle it -- a simple piece of software can take care of this, but most SANs do not understand/take advantage of the server sending those commands.
The servers aren't dumb, the uber-proprietary ultra-expensive SANs are.
And when the SAN vendors eventually get the feature to understand the servers' SCSI PUNCH commands, it will probably require a few more million in additional licensing costs, in addition to having current support/upgrades agreements in place.
Writing an all zeros sector is probably the most supported. Some SANs have dedup functionality, and an all-zeros sector is easy to dedup:
it just requires special software running on the server.
"Writing to all sectors" in an overcommitted pool doesn't guarantee squat, when the SAN is operating with special features such as snapshots, clones, copy on write foundation, etc.
In some environments, your ability to write or change sectors on your disk may depend on there being free additional space available in the pool, even if you've already written to that sector.
If the pool runs out, which you would in fact be making more likely with server admins pulling such stupid shenanigans, the I/O could easily still get blocked, even though the server had "written to all sectors" previously.
As for a server suddenly trying to use all the thin-provisioned disk space, there's a fix for that too: quotas.
Or, restricting the rate at which a server can consume additional thin-provisioned storage before setting off alarm bells and throttling the server's I/O limit down to forcibly reduce the rate of additional consumption.
And then what? Actually it sounds like they have already done this:
Kind of funny... this is like trying to open up a Wiki and handing down an edict from on high about how people can write and what they can put up on the Wiki.
You know.... stupid and contrary to the whole point of a medium.
Once they have a good picture of the perpetrator, they can incorporate pictures of the person into their ass-recognition software.
Well, i'd say ban based on face recognition; however, people never need to show their face (really) to do what the chatroulette dev doesn't want people to do
Which is interact like they would naturally when they have anonymity and no law enforcement to bust them (as the sick person they are)
To more changing of privacy controls and what can be made private out form under users who had already selected some strong privacy settings or are trying to select new settings?
"If you don't like our changes to the privacy policy, you have 15 days to press the 'Delete Account Pmernanently' button, and then none of your currently private information will be shared."
Two things... the Geneva conventions apply to prisoners of war, not civillians.
Also, people involuntarily transformed into animals are a gray area in regards to the geneva conventions.
And besides, Yubaba would not obey them anyways. The lawyers be lucky to just be catapulted; in all likelihood, they would be roast beast.
The "right to use" is not an exclusive right protected by copyright. You do not need a right to use: you only need to legally possess a work, to have the right to use it under copyright.
If you legally possess but your contract says you cannot use it, and you use it, that's (maybe) a breach of contract, but not copyright infringement, the copies to RAM are still exempt.
Copyright is solely controlled by the federal government; individual states cannot pass laws that control what is copyrightable, or what is exempt from copyright, because the authority is reserved for the feds.
The copy is already copied into memory by the time the dongle is to be verified regardless of whether the dongle-based verification to continue running the program and give the user access will succeed or not
So the copy is already made into memory, the dongle only controls whether the program will perform a useful function or not after it was launched.
What's interesting is the idea that use prevention does not count as copy prevention.
So if this were to become a precedent, then if Joe average makes a copy of a DVD image onto his hard drive, using ordinary DVD imaging technology,
After copying it, he can decrypt in order to view, because CSS encryption is a use prevention technology, that does not prevent or interfere with copying of the media contents.
Only the ability to view them, because the DVDCA standard requires DVD players verify a volume ID code in order to allow viewing the media.
Actually, it seems the software is required by technicians to SERVICE UPSes, because access to calibrations and voltage controls are restricted on the UPS and only available through the software. And the GE company being sued is a company that sells UPS maintenance and repair services.
I would equate this to vehicle manufacturers using encrypted communications with the cars' computers, to ensure only authorized dealers and servicers can possibly have access to diagnostic information.
So their technicians would be using the software while the company receives payments from other organizations for UPS repair/maintenance.
So they are (possibly) competing against the company that manufactures the UPS for repair, management, and maintenance services, as not an "authorized" servicer of their UPSes.
The software is not generally available to the public, and it sounds as if they might have obtained it through an unapproved channel.
There is also a possibilitythey might be providing service for UPSes that are not under service contracts with the manufacturer.
I. MGE manufactures several lines of UPS machines, some of which require the use of MGE's copyrighted software programs Pacret and Muguet during servicing. This software fixes calibration problems more quickly than traditional manual servicing techniques. Without the software, a service technician can still partially service an MGE UPS machine, but a number of critical procedures (including recalibration and adjustment of voltage levels) can only be performed through use of the software, which works only on MGE-manufactured devices.
The software requires connection of an external hardware security key (called a "dongle") to the laptop serial port. Each dongle has an expiration date, a maximum number of uses, and a unique password.... If the protocol exchange is successful, MGE's software proceeds to collect system status information for the technician.
Years after MGE introduced its security technology, a number of software hackers published information on the internet disclosing general instructions on how to defeat the external security features of a hardware key.
...
PMI is a critical power service company servicing a variety of brands of UPS machines, including MGE UPS machines. PMI initially subcontracted MGE to perform software service on MGE UPS machines, but sometime before June 2000, a group of PMI employees obtained at least one copy of MGE's software from an unknown source. GE acquired PMI in 2001.
Yes, and because they suggested the switch to macs and get to take-over the 'support' role, they will look superior to you, and get all the credit.
They will bow down to the Mac evangelizer's "technical prowess" and believe them if there is ever a technical argument and you are in disagreement.
Your opinion on technical matters will get dismissed "the guy who wants us to use PCs + Androids instead of our iPhones + Macs and must therefore be a fossil who does not know all that much about computers really".
Your geek ego, if you have one, should be crushed by something like that, in the long run...
This is more like: when you let a user plug into your network, there is a possibility of them using ARP Hijacking or IP Spoofing to capture traffic intended for another computer on the same subnet.
In many ways, this is just a wireless protocol equivalent of ye' old ARP hijack.
But I don't see anyone around shouting about how the Ethernet protocol is broken and needs this security hole in the protocol to be fixed.
And the GP post is like saying: a bowl of fruit doesn't have apples in it, bowls of fruit can only have pineapples.
Nowhere has the FCC said the amateur service is just a hobby.
Of course YMMV trying to use amateur frequencies to seek help in an emegrency; it's not as if any other amateur is paid to listen, and if there's no local disaster or communications emergency in the area, it's not like a lot of people will necessarily even be thinking of listening on common frequencies.
You might be better off with an emergency radio, to be used only when human life is threatened and you need assistance.
which are not good in mountainous terrain unless you are certain you'd be in range of one or more repeaters during your trips
i don't think he said anything about cost being a factor.
He could build an unmanned amateur repeater station somewhere up in the mountains himself and deploy it before going out. It would be a bit more expensive than carrying a cell phone, though...... depending on whether the cell phone is under contract with one of those plans with exorbitant roaming fees though:)
Slashdot Mirror
700mb? You mean 700 MB?
No I don't. I mean 700 million bits.
The term "Megabytes" is ambiguous, because some say 1 Megabyte = 8,000,000 bits.
Other people say 1 Megabyte = 8,388,608
Bits is the preferred method of counting storage used, because the amount specified is unambiguous. Nobody can dispute how many bits there are in 700 * 10^6 bits.
Maybe they'll just clone them, creating identical com.oracle.* packages. Meanwhile the com.sun.* packages will ship with java, be deprecated and not receive any updates.
Not like anyone will notice their JRE installs growing from 700mb in size to 1000mb in size.
This does not work so well when the feature is present but buggy in various browsers, or for features whose presence or absence cannot easily be detected, the result would be inconsistent rendering if user agent was not checked to redirect the user to the proper page.
It is not a best practice to rely on attempts to "test for the presence" of a feature, to determine whether it should be used.
"Detection" is only possible if using javascript: not all browsers support, and some users might have it turned off.
User agent is commonly used for this purpose, and it is the best practice as defined by the standard to use the user agent field for the purpose of tailoring responses to avoid particular user agent limitations
The definition of 'immediately' depends on who you are. If you're MS patching a security vulnerability, "immediately" may mean 60 to 90 days.
Because, without a fixed size, the VM would always eventually grow to consume all available memory.
Recall, Java programmers don't manage their memory, they rely on Java's garbage collection.
GC is triggered when the amount of memory available to the VM drops below a specified threshold.
If the VM can grow to use all available memory, that threshold won't ever be reached, until java is taking over all physical memory: even with the limit it tends to do so anyways, but i'm just saying.....
Ah, it used to call doWhatIwant.
Unfortunately, after Sun was bought out there was a change to the symbol table to insert the function doWhatOracleWants into the symbol table.
As a result, all the other system calls were shifted down one in the symbol table.
So old source code compiled using doWhatIwant() winds up invoking doWhatIdontWant() instead of what is expected.
No they're not. After the fiasco regarding blocking access to firmware updates for Sun servers on the new site we stopped considering using any more Sun/Oracle products for new deployments.
Of course old equipment is still badged with the Sun name. Whatever Oracle's doing is getting the opposite of advertising.
Running away from them as fast as possible.
So if Microsoft gets bought out by Oracle and the User-Agent string of IE6, IE7, and IE8 are immediately updated to be OIE6 7 or 8, respectively, with a surprise critical update..
You are saying it's the web developers' fault, and not Oracle's, when this poorly conceived change breaks the web and compatibility for IE users?
It's true, but the first two characters are just padding.
it's const char *osname = "UnBreakable Linux" + 2;
All they have to do is file a counter-notice, citing their license to use the work under the CC-NC, then the DMCA letter has no benefit.
Contrary to what the author seems to suggest the "No Commercial Use" restriction of CC-BY-NC does not mean that no advertising or thing of commercial value to the site operator can appear on the same site that a work appears on.
Now, someday, the servers will be smart enough to tell the storage array when they're done with a chunk of storage.
Servers are that smart... most commonly this is needed for using SSD drives -- SCSI PUNCH, SATA TRIM, or writing a block of all zeros to a sector.... there are OS configurations that support this, and if you don't have an OS that can handle it -- a simple piece of software can take care of this, but most SANs do not understand/take advantage of the server sending those commands.
The servers aren't dumb, the uber-proprietary ultra-expensive SANs are. And when the SAN vendors eventually get the feature to understand the servers' SCSI PUNCH commands, it will probably require a few more million in additional licensing costs, in addition to having current support/upgrades agreements in place.
Writing an all zeros sector is probably the most supported. Some SANs have dedup functionality, and an all-zeros sector is easy to dedup: it just requires special software running on the server.
"Writing to all sectors" in an overcommitted pool doesn't guarantee squat, when the SAN is operating with special features such as snapshots, clones, copy on write foundation, etc.
In some environments, your ability to write or change sectors on your disk may depend on there being free additional space available in the pool, even if you've already written to that sector.
If the pool runs out, which you would in fact be making more likely with server admins pulling such stupid shenanigans, the I/O could easily still get blocked, even though the server had "written to all sectors" previously.
As for a server suddenly trying to use all the thin-provisioned disk space, there's a fix for that too: quotas.
Or, restricting the rate at which a server can consume additional thin-provisioned storage before setting off alarm bells and throttling the server's I/O limit down to forcibly reduce the rate of additional consumption.
And then what? Actually it sounds like they have already done this:
Kind of funny... this is like trying to open up a Wiki and handing down an edict from on high about how people can write and what they can put up on the Wiki. You know.... stupid and contrary to the whole point of a medium.
Once they have a good picture of the perpetrator, they can incorporate pictures of the person into their ass-recognition software.
Well, i'd say ban based on face recognition; however, people never need to show their face (really) to do what the chatroulette dev doesn't want people to do
Which is interact like they would naturally when they have anonymity and no law enforcement to bust them (as the sick person they are)
Thankfully, that is impossible for participants of chat roulette to do, the only thing anyone displays anything to is their computer, via webcam.
Last I checked there was no law that you could not display yourself to a webcam, or even post it online.
It's not your responsibility to prevent children from reaching your website or chat roulette's website, that's their parents' responsibility..
I've yet to hear of anyone being sued for failing to prevent underage access to their porn site.
"Android users aren't as disloyal as reported"
That's part 1... still waiting on part 2:
I wonder if it was Apple marketing or ATT marketing that was indirectly responsible for the previous disinformation?
To more changing of privacy controls and what can be made private out form under users who had already selected some strong privacy settings or are trying to select new settings?
"If you don't like our changes to the privacy policy, you have 15 days to press the 'Delete Account Pmernanently' button, and then none of your currently private information will be shared."
Two things... the Geneva conventions apply to prisoners of war, not civillians. Also, people involuntarily transformed into animals are a gray area in regards to the geneva conventions.
And besides, Yubaba would not obey them anyways. The lawyers be lucky to just be catapulted; in all likelihood, they would be roast beast.
The "right to use" is not an exclusive right protected by copyright. You do not need a right to use: you only need to legally possess a work, to have the right to use it under copyright.
If you legally possess but your contract says you cannot use it, and you use it, that's (maybe) a breach of contract, but not copyright infringement, the copies to RAM are still exempt.
Copyright is solely controlled by the federal government; individual states cannot pass laws that control what is copyrightable, or what is exempt from copyright, because the authority is reserved for the feds.
The copy is already copied into memory by the time the dongle is to be verified regardless of whether the dongle-based verification to continue running the program and give the user access will succeed or not
So the copy is already made into memory, the dongle only controls whether the program will perform a useful function or not after it was launched.
What's interesting is the idea that use prevention does not count as copy prevention.
So if this were to become a precedent, then if Joe average makes a copy of a DVD image onto his hard drive, using ordinary DVD imaging technology,
After copying it, he can decrypt in order to view, because CSS encryption is a use prevention technology, that does not prevent or interfere with copying of the media contents.
Only the ability to view them, because the DVDCA standard requires DVD players verify a volume ID code in order to allow viewing the media.
Actually, it seems the software is required by technicians to SERVICE UPSes, because access to calibrations and voltage controls are restricted on the UPS and only available through the software. And the GE company being sued is a company that sells UPS maintenance and repair services.
I would equate this to vehicle manufacturers using encrypted communications with the cars' computers, to ensure only authorized dealers and servicers can possibly have access to diagnostic information.
So their technicians would be using the software while the company receives payments from other organizations for UPS repair/maintenance.
So they are (possibly) competing against the company that manufactures the UPS for repair, management, and maintenance services, as not an "authorized" servicer of their UPSes.
The software is not generally available to the public, and it sounds as if they might have obtained it through an unapproved channel.
There is also a possibilitythey might be providing service for UPSes that are not under service contracts with the manufacturer.
Hm.. http://www.law.com/jsp/tx/LawDecisionTX.jsp?id=1202463809581
Yes, and because they suggested the switch to macs and get to take-over the 'support' role, they will look superior to you, and get all the credit.
They will bow down to the Mac evangelizer's "technical prowess" and believe them if there is ever a technical argument and you are in disagreement.
Your opinion on technical matters will get dismissed "the guy who wants us to use PCs + Androids instead of our iPhones + Macs and must therefore be a fossil who does not know all that much about computers really".
Your geek ego, if you have one, should be crushed by something like that, in the long run...
This is more like: when you let a user plug into your network, there is a possibility of them using ARP Hijacking or IP Spoofing to capture traffic intended for another computer on the same subnet.
In many ways, this is just a wireless protocol equivalent of ye' old ARP hijack.
But I don't see anyone around shouting about how the Ethernet protocol is broken and needs this security hole in the protocol to be fixed.
And the GP post is like saying: a bowl of fruit doesn't have apples in it, bowls of fruit can only have pineapples.
Nowhere has the FCC said the amateur service is just a hobby.
Of course YMMV trying to use amateur frequencies to seek help in an emegrency; it's not as if any other amateur is paid to listen, and if there's no local disaster or communications emergency in the area, it's not like a lot of people will necessarily even be thinking of listening on common frequencies.
You might be better off with an emergency radio, to be used only when human life is threatened and you need assistance.
which are not good in mountainous terrain unless you are certain you'd be in range of one or more repeaters during your trips
i don't think he said anything about cost being a factor.
He could build an unmanned amateur repeater station somewhere up in the mountains himself and deploy it before going out. It would be a bit more expensive than carrying a cell phone, though...... depending on whether the cell phone is under contract with one of those plans with exorbitant roaming fees though :)