It doesn't matter that 2's complement is the encoding format used to represent a negative number on certain architectures, or if sign magnitude encoding is used, the MSB is still called a sign bit, if a "1" in that position indicates negative numbers regardless of the encoding format represented by the other bits.
No, atheism can merely be lack of belief in the existence of a supreme being. Hence my point. Atheism is no more a religion than illiteracy is.
And christianity can be merely the lack of belief in the non-existence of a supreme being. Hence my point. Christianity is no more a religion than literacy is. You could base a religion on literacy too, but it would not make literacy a religion.
Obviously if Apple is working on their own version of Google Latitude (or owns the IP rights to this functionality), they'd be hesitant to put an app with the same functionality on their devices from another company."
If this were Microsoft, we'd be talking about how evilly they were using their monopoly power, to quash a competitor.
How interesting that we say Obviously Apple would do this...
In other words, we have already taken for granted that Apple is an even more evil monopolist than MS.
Microsoft tilted the playing field by giving their software an advantage (such as private APIs), but they never (that we know of) "blocked" competing application programs altogether from their platform, for the purpose of ensuring they were the first to market...
The characteristics of "Christianity" and "Atheism" are the same; Christianity is a belief in Christ, and nothing more; Atheism is a belief in the non-existence of a supreme entity.
Christianity is a religion because you define it to be?!
The argument that christianity is the name of a religion is nonsensical.
Perhaps Atheism just happens to be the name of a religion too then.
Feist v. Rural, US Supreme court 1991, telephone companies do not have a copyright on telephone listings.
Copyright covers creativity, not the mere act of collecting existing information.
However, there are things they could do to make their listings as a whole come under copyright : things like adding 'creative' fake entries, inserting jokes, or various other bits of miscellany in the pages of the directory, that involve a creative process.
But the actual information, as in real names, and phone numbers, is not copyrightable.
Additional protections/restrictions on the use of databases.
Databases aren't subject to copyright, so some big companies felt they needed a law to allow them to restrict use of their publicly accessible databases in ways they couldn't otherwise
The day spent resting is still part of the time spent creating the universe, earth, and everything in it.
He got paid for 7 days worth of "creating the universe and everything in it", because that's what was in his contract:)
Since he was up working 144 hours straight, he needed that 24 hours of rest, before he could move on to the next project.
Good luck getting Microsoft operating systems to support any MMU functionality beyond what the operating systems currently provide.
Microsoft currently supports NX (No-eXecute bit) functionality. If customers want it, and they can add it to their "Feature checklist" as a security function, I think MS will implement it.
Hardware IOMMU already something Solaris and Linux have to lord over Microsoft with, it would be a competitive disadvantage for them to ignore it.
Provided the system even is running. If the operating system will not boot, the "Safely remove hard drive" option would have to be in BIOS.
Well, the point is just.. if everything's been loaded onto RAM, the system can suspend I/O to the hard drive/long-term storage, and allow you to remove it physically, while it queues up writes by sending them to the journal instead (provided the journal is large enough... e.g. 32gb SSD; is plenty for most workstation I/O loads, for Database servers, much larger sizes might be needed). Applications don't need to notice any immediate difference, as long as it is not rebooted while the hard drive is out.
If you can force a commit of pending transactions before doing that HDD removal, then you can snapshot your system at any point in time "at the point of commit", for the purpose of having a "backup hard drive", then insert a new blank hard drive, and you just need an option to dump the RAMDISK volume to it, and then resume committing of journal.
Another possibility would be to boot the second system (with identical hardware), and have some type of OS bootup option "transfer the journal, main memory, and complete application state from another system"
Then on the original system, have some command to transmit the journal and main memory contents over say GigE Ethernet to the first system, and at some point of your choosing, reboot the second system, kill the first system as soon as the 2nd one comes up, and execution of the software transfers seamlessly.
I'm thinking along the lines of "backup server" setups, where you had a failed node, you moved to a backup server, and maybe now want to move execution back to the main physical machine, with minimal downtime.
What's your point?
Christianity isn't a religion either, by your definition. Being christian is simply the state of believing in one Jesus Christ. There is no dogma, no canon, no "book of how to behave", no punishment, no reward. Just faith in one thing.
Of course there are many christian churches.
Some of them do have some of those components. The only dogma shared by all of them really have is that 'God exists' and 'Jesus christ existed. Some have a 'canon' some don't.
Some have 'book of how to behave', some don't. Some have punishment/reward (just faith alone VS faith+good works debate), some don't.
None of these things you mentioned define religion.
Apparently Atheists do have some sacraments, such as marriage.
The possibility of congretations also cannot be ruled out.
The possibility of an atheist church existing, that has all the characteristics you describe as "religion" is very real.
Just because they're atheist, doesn't mean they can't have a church with a dogma and a book of how to behave (E.g. bylaws of the organization).
They can have punishment too -- break the rules, and you get sanctions laid upon you by the church.
Religion doesn't define a person's outlook, morals, or ethics either.
There are lots of good christians with very poor morals/ethics, by objective standards. There are lots of bad christians and even atheists with very good morals/ethics, by objective standards.
Aren't "There is a God" "God created the world in 7 days" both grossly abusive and insulting in relation to matters held sacred by the practitioner of atheism?
I have no doubt underground vendors are willing to do questionable things.
But it would at least help to force them to actually go underground, rather than use a public exposed website for anonymous scanning (without sample sharing), make their service harder for novices to access, increase the price.
And reduce the "legitimacy" or "credibility" of the service designed to facilitate malware authors.
The DMCA and various DMCA-inspired laws passed by various countries and the notion of 'takedown letters' really sucks, but maybe it can be put to one good use...
On a proprietary OS platform, it's only appropriate that the antivirus programs contain license restrictions against using them for evil, or using them to circumvent other users' need to buy their own copy and update subscriptions.
These programs already contain very restrictive EULAs. It's logical for them to contain a restriction against this type of abuse.
Otherwise... someone could just write a free "stub AV" everyone installs on their desktop, that uses an outsourced, online scanner to actually do all the file checking.
Then the manufacturer of the AV scanner loses all their business to the "outsourced AV programs"...
This is proprietary software.
They profit by selling copies of the software, not by enabling as much freedom as possible.
Don't fool yourself into thinking you have software freedom, or software freedom principles, somehow apply to someone else's closed source, "pay for use, but restricted" software product.
All commercial AVs are in that category... in general, basically all AVs are in that category (except the likes of ClamAV)
Obstruction of justice is apparently not applicable to civil matters..
The penalty for failure to comply with a subpoena depends on the jurisdiction/court that issued it, and usually some warning of that effect would be on the document.
It also listed people who would be exempted from these screening procedures such as heads of state and their families.
Now the bad guys know is all they have to do is get a family member of a "head of state" (whatever that means), to defect to their cause, or impersonate such a person successfully.:(
What the heck is the security justification for heads of state, or their families to be exempt?
Their luggage (or the item a bad guy secretly planted in their luggage at some point to get it past security) can be just as much a security risk as anyone else's luggage.
AV makers should include a clause in the EULA, that:
the software may not be used to provide a virus scanning service for more than one third party.
You may not scan a file for another person without purchasing an additional license to be permanently assigned to each person.
And then they can send their army of lawyers at any "paid AV scanning website" that doesn't have an agreement with them.
the files collectively are too big to fit in RAM (capacity miss).
I think the premise is RAM becomes cheap..
so you can have enough of it to hold the entire filesystem. E.g. 64gb or 128gb of RAM easily meets the needs of most users, with a couple gbs to spare for the kernel/app working memory partition.
During the boot process, 2-4gb (or user's choice) is reserved for OS and application working memory, and the rest is partitioned as the RAMDISK.
Probably none of the filesystems currently in existence are suitable for this, without modification.
But one problem with journaling writes to data is that the user expects shutdown to be fast, and if the user has accumulated hundreds of megabytes of logs, it could take a while to copy them out to the hard disk.
I would suggest a solution is only to commit out a small amount during shutdown.
There is no need to commit the entire journal at the time of shutdown, as long as the journal is stored on a stable medium, the user will be in the same place when they power back on.
The uncommitted transactions can be simply loaded back at next boot, and committing can resume at the normal pace when the system boots back up.
Yeah, I'm saying 'normal power off' can be treated similarly to system failure case.
Except during a normal power off, application consistency is assured (instead of just crash-consistency).
This should be a tunable. Users should get to control (during shutdown) whether their system fully commits the journal to guarantee they can remove the disk; I assume by default they would not need safe removal of the HD.
Or even while the system is running.. presumably there should be an option like "Safely remove hard drive"
At which point, committing the journal stops, they can move the drive to another system, to boot up a second system which will now be identical, but, they'd better insert a new one in the original system, tell it to "mirror main memory back to HDD", and have the manual mirroring complete, before the journal drive overflows.
And the journal chip would have to be in the same chassis as the hard disk; otherwise, the user won't be able to take out the disk and put it in another machine when repairing the computer.
Wouldn't the answer just be a 'safe removal' procedure? Or move the CHIP and the journal at the same time.
The journal medium should contain metadata about the hard drive, so it can't accidentally be committed to a different volume.
And also, so it can't be accidentally committed to a clone of the original volume that has a different 'version' of the data than the one the journal was made against.
I take it "rarely changing" is relative: it would change every Tuesday when the updates come out.
Possibly, but once a month every patch Tuesday is not that often in the grand scheme of things, I don't expect they'd kill CF, if done efficiently.
I'm thinking more along the lines of UNIX systems such as Linux, BSD, MacOS, that don't get updates every tuesday, though.
Yeah: unreadable. If the operating system stores user documents in the same space as less persistent objects used by applications, a kernel crash may corrupt the file system beyond repair.
The RAMDISK is a portion of memory; I think there should be 2 memory areas: application working memory, and filesystem RAMDISK region.
The 'OS/application working area' should be zero'ed during each boot, and not journaled.
The only real bad new failure modes left there, is the RAM starts accumulating bit errors in the filesystem area (ECC memory might help, just like physical disks use ECC technology to detect errors, this goes back to current desktop RAM not being designed to reliably store bits over long periods of time).
Or the kernel started doing stray writes to the RAMDISK region, e.g. buffer overflow.
Or a buggy driver hit the wrong memory area with a DMA.
Nope, the maximum value for 32-bit time_t is 2147483647.
Increment that by 1, and the time_t value becomes -2147483648.
Although time_t is a 32-bit value, the 1st bit is the sign bit.
Jan 18 21:14:07, 2038
For 64-bit time_t it should be 9223372036854775807. But I don't believe the standard time functions can handle this value...
While it may be a perfectly valid 64-bit time_t value,
if gmtime/localtime/strftime/ctime don't work with the maximum value, it's not a usable value, really
It allows an application to tell the kernel how it expects to use some mapped or shared memory areas, so that the kernel can choose appropriate read-ahead and caching techniques.
Of course... reading from a file (from an app point of view) is really nothing more than accessing data in a mapped memory area.
Oh.. I suppose unless you actually use the POSIX mmap call to map the file into memory for reading, you won't have an easy ability to provide the advise.
And it makes portability a bitch regardless, as not all OSes are POSIX, and not all OSes have mmap().
Nevertheless, it's not fair to say it is impossible for an app to provide hints.
Whether giving the hints or not actually has a useful effect (usually) may be a matter of debate.
Load the system image into RAM at boot from the "image source".
Journal changes to user datafiles.
When a certain number of transactions have occured, commit them back to the main disk.
If the system crashes... load the "boot volume" back up, replay the journal.
No need to journal changes to the "system files" file system (that isn't supposed to change anyways).
If a system update is to be applied, the signed update package gets loaded into the journalling area, and rolled into the main image at system boot.
Another possibility would be to borrow a technology from RAID controller manufacturers...
and have a battery backup for the RAM in the form of a NiMH battery pack.
If power is lost, upon system boot, the RAM image will be restored to the same state it was in as of the unexpected shutdown/crash.
Avoid clearing the RAM region used for file storage at boot also .
The order essentially says MS can't license any new copies of the software with the Custom XML editor.
So, I understand this to mean, if for example, your enterprise has a volume license, and you buy more copies after the effective date of the order... the new copies might not be able to support Custom XML, according to the order..
Similarly, if you upgrade from a trial version of Word via an online license purchase, after the effective date, MS must not convey you the ability to use the Custom XML feature.
It seems like enforcing the patch might be a requirement in those cases for MS to comply...
I wonder what predicament this puts people in who rely on the Custom XML feature to build templates for programmatic population of documents and segmenting the presentation?
One way to make the plaintiff whole is to force Microsoft to stop distributing word, in addition to forcing them to pay the royalties they should have paid to license the technology legally.
Shouldn't they be department-specific?
There are IT managers... then there are managers over other departments...
The CFO Accountants, Auditors should wear a "Finance" badge.
Lawyers should wear a "Lawyer" badge. (Provided it's not too much a hazard to their health)
All the people in Public relations/Marketing/Sales, should have to wear a conspicuous "Salesperson" badge everywhere while working, especially when meeting with customers.
Slashdot Mirror
It doesn't matter that 2's complement is the encoding format used to represent a negative number on certain architectures, or if sign magnitude encoding is used, the MSB is still called a sign bit, if a "1" in that position indicates negative numbers regardless of the encoding format represented by the other bits.
No, atheism can merely be lack of belief in the existence of a supreme being. Hence my point. Atheism is no more a religion than illiteracy is.
And christianity can be merely the lack of belief in the non-existence of a supreme being. Hence my point. Christianity is no more a religion than literacy is. You could base a religion on literacy too, but it would not make literacy a religion.
Obviously if Apple is working on their own version of Google Latitude (or owns the IP rights to this functionality), they'd be hesitant to put an app with the same functionality on their devices from another company."
If this were Microsoft, we'd be talking about how evilly they were using their monopoly power, to quash a competitor.
How interesting that we say Obviously Apple would do this...
In other words, we have already taken for granted that Apple is an even more evil monopolist than MS.
Microsoft tilted the playing field by giving their software an advantage (such as private APIs), but they never (that we know of) "blocked" competing application programs altogether from their platform, for the purpose of ensuring they were the first to market...
You're begging the question.
The characteristics of "Christianity" and "Atheism" are the same; Christianity is a belief in Christ, and nothing more; Atheism is a belief in the non-existence of a supreme entity.
Christianity is a religion because you define it to be?!
The argument that christianity is the name of a religion is nonsensical.
Perhaps Atheism just happens to be the name of a religion too then.
Feist v. Rural, US Supreme court 1991, telephone companies do not have a copyright on telephone listings.
Copyright covers creativity, not the mere act of collecting existing information.
However, there are things they could do to make their listings as a whole come under copyright : things like adding 'creative' fake entries, inserting jokes, or various other bits of miscellany in the pages of the directory, that involve a creative process.
But the actual information, as in real names, and phone numbers, is not copyrightable.
It was the whole point of the Database directive in Europe.
Additional protections/restrictions on the use of databases.
Databases aren't subject to copyright, so some big companies felt they needed a law to allow them to restrict use of their publicly accessible databases in ways they couldn't otherwise
Indeed...
First Party: The original person the information belongs to.
Second Party: An entity the 1st party directly shared the information with.
Third Party: EVERYONE or anyone else who got the information from any source other than directly from the First party.
The day spent resting is still part of the time spent creating the universe, earth, and everything in it. He got paid for 7 days worth of "creating the universe and everything in it", because that's what was in his contract :)
Since he was up working 144 hours straight, he needed that 24 hours of rest, before he could move on to the next project.
Good luck getting Microsoft operating systems to support any MMU functionality beyond what the operating systems currently provide.
Microsoft currently supports NX (No-eXecute bit) functionality. If customers want it, and they can add it to their "Feature checklist" as a security function, I think MS will implement it.
Hardware IOMMU already something Solaris and Linux have to lord over Microsoft with, it would be a competitive disadvantage for them to ignore it.
Provided the system even is running. If the operating system will not boot, the "Safely remove hard drive" option would have to be in BIOS.
Well, the point is just.. if everything's been loaded onto RAM, the system can suspend I/O to the hard drive/long-term storage, and allow you to remove it physically, while it queues up writes by sending them to the journal instead (provided the journal is large enough... e.g. 32gb SSD; is plenty for most workstation I/O loads, for Database servers, much larger sizes might be needed). Applications don't need to notice any immediate difference, as long as it is not rebooted while the hard drive is out.
If you can force a commit of pending transactions before doing that HDD removal, then you can snapshot your system at any point in time "at the point of commit", for the purpose of having a "backup hard drive", then insert a new blank hard drive, and you just need an option to dump the RAMDISK volume to it, and then resume committing of journal.
Another possibility would be to boot the second system (with identical hardware), and have some type of OS bootup option "transfer the journal, main memory, and complete application state from another system"
Then on the original system, have some command to transmit the journal and main memory contents over say GigE Ethernet to the first system, and at some point of your choosing, reboot the second system, kill the first system as soon as the 2nd one comes up, and execution of the software transfers seamlessly.
I'm thinking along the lines of "backup server" setups, where you had a failed node, you moved to a backup server, and maybe now want to move execution back to the main physical machine, with minimal downtime.
What's your point? Christianity isn't a religion either, by your definition. Being christian is simply the state of believing in one Jesus Christ. There is no dogma, no canon, no "book of how to behave", no punishment, no reward. Just faith in one thing.
Of course there are many christian churches. Some of them do have some of those components. The only dogma shared by all of them really have is that 'God exists' and 'Jesus christ existed. Some have a 'canon' some don't. Some have 'book of how to behave', some don't. Some have punishment/reward (just faith alone VS faith+good works debate), some don't.
None of these things you mentioned define religion.
Ever hear of the 1st church of Atheism ?
Apparently Atheists do have some sacraments, such as marriage.
The possibility of congretations also cannot be ruled out.
The possibility of an atheist church existing, that has all the characteristics you describe as "religion" is very real.
Just because they're atheist, doesn't mean they can't have a church with a dogma and a book of how to behave (E.g. bylaws of the organization).
They can have punishment too -- break the rules, and you get sanctions laid upon you by the church.
Religion doesn't define a person's outlook, morals, or ethics either.
There are lots of good christians with very poor morals/ethics, by objective standards. There are lots of bad christians and even atheists with very good morals/ethics, by objective standards.
Aren't "There is a God" "God created the world in 7 days" both grossly abusive and insulting in relation to matters held sacred by the practitioner of atheism?
I have no doubt underground vendors are willing to do questionable things.
But it would at least help to force them to actually go underground, rather than use a public exposed website for anonymous scanning (without sample sharing), make their service harder for novices to access, increase the price.
And reduce the "legitimacy" or "credibility" of the service designed to facilitate malware authors.
The DMCA and various DMCA-inspired laws passed by various countries and the notion of 'takedown letters' really sucks, but maybe it can be put to one good use ...
On a proprietary OS platform, it's only appropriate that the antivirus programs contain license restrictions against using them for evil, or using them to circumvent other users' need to buy their own copy and update subscriptions.
These programs already contain very restrictive EULAs. It's logical for them to contain a restriction against this type of abuse.
Otherwise... someone could just write a free "stub AV" everyone installs on their desktop, that uses an outsourced, online scanner to actually do all the file checking.
Then the manufacturer of the AV scanner loses all their business to the "outsourced AV programs"...
This is proprietary software. They profit by selling copies of the software, not by enabling as much freedom as possible.
Don't fool yourself into thinking you have software freedom, or software freedom principles, somehow apply to someone else's closed source, "pay for use, but restricted" software product.
All commercial AVs are in that category... in general, basically all AVs are in that category (except the likes of ClamAV)
According to the article it was a civil subpoena.
And TSA agents are not police officers.
Obstruction of justice is apparently not applicable to civil matters..
The penalty for failure to comply with a subpoena depends on the jurisdiction/court that issued it, and usually some warning of that effect would be on the document.
Such a subpoena would be illegal prior restraint on free speech.
Thankfully, we still have the constitution and the 1st ammendment. No item in the patriot act or other laws has the legal authority to override that.
Besides, if they can tell their lawyer about it, their lawyer can disclose the information.
Maybe so, but one thing:
Hrm..
It also listed people who would be exempted from these screening procedures such as heads of state and their families.
Now the bad guys know is all they have to do is get a family member of a "head of state" (whatever that means), to defect to their cause, or impersonate such a person successfully. :(
What the heck is the security justification for heads of state, or their families to be exempt?
Their luggage (or the item a bad guy secretly planted in their luggage at some point to get it past security) can be just as much a security risk as anyone else's luggage.
AV makers should include a clause in the EULA, that: the software may not be used to provide a virus scanning service for more than one third party. You may not scan a file for another person without purchasing an additional license to be permanently assigned to each person.
And then they can send their army of lawyers at any "paid AV scanning website" that doesn't have an agreement with them.
the files collectively are too big to fit in RAM (capacity miss).
I think the premise is RAM becomes cheap.. so you can have enough of it to hold the entire filesystem. E.g. 64gb or 128gb of RAM easily meets the needs of most users, with a couple gbs to spare for the kernel/app working memory partition.
During the boot process, 2-4gb (or user's choice) is reserved for OS and application working memory, and the rest is partitioned as the RAMDISK. Probably none of the filesystems currently in existence are suitable for this, without modification.
But one problem with journaling writes to data is that the user expects shutdown to be fast, and if the user has accumulated hundreds of megabytes of logs, it could take a while to copy them out to the hard disk.
I would suggest a solution is only to commit out a small amount during shutdown. There is no need to commit the entire journal at the time of shutdown, as long as the journal is stored on a stable medium, the user will be in the same place when they power back on. The uncommitted transactions can be simply loaded back at next boot, and committing can resume at the normal pace when the system boots back up.
Yeah, I'm saying 'normal power off' can be treated similarly to system failure case. Except during a normal power off, application consistency is assured (instead of just crash-consistency).
This should be a tunable. Users should get to control (during shutdown) whether their system fully commits the journal to guarantee they can remove the disk; I assume by default they would not need safe removal of the HD.
Or even while the system is running.. presumably there should be an option like "Safely remove hard drive"
At which point, committing the journal stops, they can move the drive to another system, to boot up a second system which will now be identical, but, they'd better insert a new one in the original system, tell it to "mirror main memory back to HDD", and have the manual mirroring complete, before the journal drive overflows.
And the journal chip would have to be in the same chassis as the hard disk; otherwise, the user won't be able to take out the disk and put it in another machine when repairing the computer.
Wouldn't the answer just be a 'safe removal' procedure? Or move the CHIP and the journal at the same time.
The journal medium should contain metadata about the hard drive, so it can't accidentally be committed to a different volume.
And also, so it can't be accidentally committed to a clone of the original volume that has a different 'version' of the data than the one the journal was made against.
I take it "rarely changing" is relative: it would change every Tuesday when the updates come out.
Possibly, but once a month every patch Tuesday is not that often in the grand scheme of things, I don't expect they'd kill CF, if done efficiently. I'm thinking more along the lines of UNIX systems such as Linux, BSD, MacOS, that don't get updates every tuesday, though.
Yeah: unreadable. If the operating system stores user documents in the same space as less persistent objects used by applications, a kernel crash may corrupt the file system beyond repair.
The RAMDISK is a portion of memory; I think there should be 2 memory areas: application working memory, and filesystem RAMDISK region. The 'OS/application working area' should be zero'ed during each boot, and not journaled.
The only real bad new failure modes left there, is the RAM starts accumulating bit errors in the filesystem area (ECC memory might help, just like physical disks use ECC technology to detect errors, this goes back to current desktop RAM not being designed to reliably store bits over long periods of time).
Or the kernel started doing stray writes to the RAMDISK region, e.g. buffer overflow. Or a buggy driver hit the wrong memory area with a DMA.
This c
Nope, the maximum value for 32-bit time_t is 2147483647. Increment that by 1, and the time_t value becomes -2147483648.
Although time_t is a 32-bit value, the 1st bit is the sign bit.
Jan 18 21:14:07, 2038
For 64-bit time_t it should be 9223372036854775807. But I don't believe the standard time functions can handle this value...
While it may be a perfectly valid 64-bit time_t value, if gmtime/localtime/strftime/ctime don't work with the maximum value, it's not a usable value, really
What is "a certain number" that won't require the disk to be spun up all the time committing transactions?
Why spun up? use a write-optimized SSD for the journal, and compact flash for the rarely-changing system boot image.
"A certain number", the exact choice is a design/engineering concern, but probably fairly small values should be used, to avoid data loss.
It will be restored to the same state: a crashed state.
Well, of course, the filesystem would be in the same state as at the time of the crash.
That doesn't by any means indicate a second crash will occur.
Even UPSes have fuses that can blow / breakers that can trip. A UPS can overload.
Someone can accidentally hit the EPO, or power-off switch on the UPS.
The UPS battery may be too low to permit a graceful shutdown before power expires.
The PC power supply can fail.
Someone could trip over the power cord running to the PC.
Even with a solid UPS, it doesn't require much imagination at all to recognize how likely a power failure or 'hard down' is to occur eventually.
Losing all your data/changes in such cases, is probably unacceptable.
We have it today. Tfa's on crack.
It's called madvise
In Linux there is also fadvise()
Of course... reading from a file (from an app point of view) is really nothing more than accessing data in a mapped memory area. Oh.. I suppose unless you actually use the POSIX mmap call to map the file into memory for reading, you won't have an easy ability to provide the advise.
And it makes portability a bitch regardless, as not all OSes are POSIX, and not all OSes have mmap().
Nevertheless, it's not fair to say it is impossible for an app to provide hints. Whether giving the hints or not actually has a useful effect (usually) may be a matter of debate.
Power failure happens.
That's what journaling is for.
Load the system image into RAM at boot from the "image source".
Journal changes to user datafiles.
When a certain number of transactions have occured, commit them back to the main disk.
If the system crashes... load the "boot volume" back up, replay the journal.
No need to journal changes to the "system files" file system (that isn't supposed to change anyways). If a system update is to be applied, the signed update package gets loaded into the journalling area, and rolled into the main image at system boot.
Another possibility would be to borrow a technology from RAID controller manufacturers... and have a battery backup for the RAM in the form of a NiMH battery pack. If power is lost, upon system boot, the RAM image will be restored to the same state it was in as of the unexpected shutdown/crash.
Avoid clearing the RAM region used for file storage at boot also .
The order essentially says MS can't license any new copies of the software with the Custom XML editor.
So, I understand this to mean, if for example, your enterprise has a volume license, and you buy more copies after the effective date of the order... the new copies might not be able to support Custom XML, according to the order..
Similarly, if you upgrade from a trial version of Word via an online license purchase, after the effective date, MS must not convey you the ability to use the Custom XML feature. It seems like enforcing the patch might be a requirement in those cases for MS to comply...
I wonder what predicament this puts people in who rely on the Custom XML feature to build templates for programmatic population of documents and segmenting the presentation?
One way to make the plaintiff whole is to force Microsoft to stop distributing word, in addition to forcing them to pay the royalties they should have paid to license the technology legally.
Shouldn't they be department-specific? There are IT managers... then there are managers over other departments...
The CFO Accountants, Auditors should wear a "Finance" badge.
Lawyers should wear a "Lawyer" badge. (Provided it's not too much a hazard to their health)
All the people in Public relations/Marketing/Sales, should have to wear a conspicuous "Salesperson" badge everywhere while working, especially when meeting with customers.