The UK data protection act states that your customer's details must be kept secure. If they aren't there are, possibly significant, legal penalties.
A company giving personal details to an organisation that's been PUBLICLY shown to have piss poor security is a recipe for losing court cases, big time.
Sending the list by unencrypted email was just stupid, but generally doesn't result in a real security breach because like the "purloined letter" the gems are swamped in junk. Nevertheless high profile targets (like ACS law and possibly BT) need to be much more careful. Having a policy of sending or accepting such data by email is another matter, however.
In the dos days you COULD just "flip the switch". All the good disk caches would sync the disk before you got back to the dos command prompt or when the application called the idle trap. (int 28 or so?)
Of course, it wasn't safe to turn off while an application was writing to disk (as well as being slow and unreliable to be continually writing to disk) so the idea of working on a copy of the document in memory and "saving" the copy was born.
Yes, Hibernate is a workaround. It's actually a bad hack that's it's "supposed" to be used to improve the boot (and shutdown) speed, but I suppose nobody really cares.
BillG was right! (Oh shit, I didn't say that did I !!!??)
The application the user was running just took minutes to save the file it has stated that the user data is safely on the disk. From the user's point of view THERE IS NOTHING TO SAVE. So from the user's point of view it's safe to turn off, hard, immediately.
It's even pretty easy to do at the technical level. Databases have been doing this sort of thing for decades. Of course for a filesystem it should be slightly different. Ie: any file that gets an fsync() should be considered important enough to commit to disk early. But everything else gets committed strictly on a snapshot basis because there are no "commit transaction" semantics.
So, every few seconds (or minutes) the system is "checkpointed" such that on a hard reset it will come back to the last, completely consistent, checkpoint (especially including user data). If someone uses fsync() on a file just that file is saved away to disk with instructions to 'fix' the filesystem after a crash. You would do it this way so you could reduce power consumption and improve speed on crappy drives compared to committing every single tiny disk operation.
In technical terms I suppose that could be a copy on write filesystem with a rollforward log for the fsync() and related APIs. I guess doing an fsync() on a file would mean that it isn't copied on write any more, that way a DB could stop external fragmentation and LILO and friends could still work.
Microsoft have tried to do it; NTFS has it's log file, but it only stores the least important data on the disk. So you will get in trouble if you frequently hard power off NTFS. (Only the metadata is logged; Think about it, the user does a backup of the files, they get some metadata but if just the metadata is broken they don't care unless it means they lose the file too.)
Where's my Linux version? The closest is ext3 with full logging but that writes everything twice and checkpoints the entire drive if I tell it to fsync() a single file. I suppose I'm okay with that, for now, but I'm still a bit disappointed... except, I hear ext4 is no better...
Personally, I don't like either the GNU Hurd Bootloader or the GPT partition table. The first is far too complex and the second is badly committee botched. Personally I can't remember the last time I actually needed multi-boot capability anyway, I always use virtual machines now. The last time I got close was testing drivers support for various versions of windows during the recent Vista fiasco, I used multiple disk drives for that.
My preferred Linux partitioning tool is LVM with lilo as the booter. Yes, that is LVM directly on the hard drive, no partition table no GPT; I don't need them and don't want them, they're just one more thing to cause bugs. Lilo is similar, though really Lilo still does too much, but at least unlike grub it doesn't get all pissy and chuck it's toys out of the pram when it can't find it's super secret hiding place.
Of course that particular one shows a good reason for downloading even when you have the DVDs, twelve seasons on one little USB stick ready to plug into my DVD player...
Actually, in some places, they do. Except they call it "contention ratio". If you have an "up to" 20Mb at a contention ratio of 50:1 that's 400kb CIR. Of course, they will often bury it too.
What they can't tell you with ADSL is your PIR. For example, the advertised is 20Mb but the reality is whatever your modem syncs to. (14Mb for me)
For BT wholesale connections (in the UK) the rations are 50:1 for a residential service and 20:1 for a 'business' version. But that only covers the BT part; from your phone socket to the ISP's lan not their interconnects.
If it's cleaning the drive you're right. The data storage on a modern hard drive is so close to not being there at all that there's no room for old copies. A modern hard drive has to do the equivalent of MFM just to work properly. (aka PRML)
But if you're scanning for and fixing bad blocks it a different matter. I normally do four write passes (with different patterns) with a read pass after each one. This obviously gives a good test for each block but more importantly it allows the S.M.A.R.T to catch up with reality. I've had drives go both ways, ie dubious before to working perfectly and fine before to hundreds of reallocated sectors and SMART failure.
Flash drives are different though, they contain more blocks than needed for their official capacity. So you must do more than one simple wipe to be able to get them all. Eg: a secure erase.
Except, it can take quite a while to sync two drives of this size. So you will probably find that the second drive spends at least half it's time sitting next to the primary drive.
You actually need THREE drives so one of they is always a safe backup.
The actual rules:
Make a Backup.
Make it safe. eg: offsite
Keep it safe. eg: THE backup must stay offsite, it only comes back when it's not THE backup.
The problem is that a drive of two real terabytes (not marketing terabytes) is one 512byte sector too large for a normal MBR partition table. So you have to switch over to something else; Windows uses a GPT style partition table.
Unfortunately the committee who designed GPT were dumb (Some of the members must have been smart though, you can easily work around every dumb choice I've seen). A current BIOS doesn't know anything about partition tables and it has no problem with drives up to (IIRC) 64 petabytes. The committee decided that the BIOS should contain a boot manager that understands GPT structures. The BIOS makers don't want this problem; their stuff works fine, some people are already using RAID devices with far in access of 3TB and every OS that can handle a 3TB drive can boot off it... except Windows.
So it's SNAFU by Microsoft. The weird thing is this time they did it by following the written standard and ignoring the defacto standard...
It is a huge assumption, back in the nineties the 3.5 inch drives won because they were big enough and had a speed that was noticeably faster than any 5.25 drive could manage.
It's the same reason that SSD drives are so popular for notebooks now. Personally, I'm also now specifying desktop machines with at least two drives (and SSD and a disk) whenever I can.
The closest they'll do is to put FreeDOS on the machine. Which is actually pretty close really as FreeDOS only shows the CPU & Disk work.
Their main problem though for the 'discerning buyer' has always been the website. It's designed to try and segment the market, to try and turn the collection of parts that a technical buyer sees into a smooth line of "the maximum the punter will pay" of marketing theory. While also trying to simplify everything for the non-technical buyer so they are 'theoretical punters' too.
What this means is that the only way you can really map 'FreeDOS' or 'Linux' onto marketing theory is as 'cheap and nasty' because all OSs are exactly the same and the more you pay the better it is...
No you're saying nothing is ever bricked because by improving your ability or paying someone else you can always get it fixed. This is wrong.
In truth the 'normal' way of flashing your WRT54G can still brick the router because you can't fix the problem using the same technique that you got into trouble in the first place. OTOH JTAG flashing of the router does indeed seem to be unable to brick the device because you can always fix the problem.
That just leaves where you draw the line. IMO the requirements for doing a WRT54G JTAG are too far, ie soldering the board and guessing (parts of) the protocol. But (for example) an edge connector with available documentation would be okay. So you see I put the line at about "Could a professional outsider reasonably fix it without being required to damage or permanently change anything?".
Microwaves are at far too low a frequency for a fibre to act as a wave guide and you wouldn't really want to anyway. You see the visible range is just under one octave, that means that just the optical band plus a tiny bit of the UV band has as much bandwidth as every single frequency below it.
This includes the entire of the radio, microwave, toaster (okay, terahertz band) and infrared combined. So how much data do you think can ultimately be pushed down a fibre?
I'm pretty sure I've written this protocol before myself, it was for uploading files to an old mainframe. (well actually a simulation running on a PC, but it worked the same.) All sorts of nasty limitations meant the file was first converted to ASCII then the lines were sent. They'd get miss ordered and misplaced but the protocol would sort them and retransmit the losses.
The Real News is that the Phone company is screwing everybody again.
You know 80000 messages could have been just one small test file, how much should it cost to transmit a small file..?
It was a scarce resource and flooding it could actually prevent anyone making calls (there was a fairly simple DoS attack possible). This was the original reason for SMS being expensive
The original price of SMS messages was pitched so that they were cheaper than a "snailmail" letter. Nobody imagined that even at that ripoff price they'd be, like, popular or anything.
Remember the original phones weren't programmable so a technical problem like DOS would be seen as a complete waste of time even by the "techys".
It's as shriphani says, the phrase is "the dash for the cash" or sometimes "the short-termer view".
In the computer world this became a quest for more megahertz (then Gigahertz) only now that thermodynamic limits have been crashed into are multiple processors being looked at. And only because that's the easiest way to grab more cash.
Current example, the ARM processor. It's been around since the 6502 and z80 were mainstream processors. It's always been a 32 bit processor with no 8 or 16 bit leftovers. It's always had very good power consumption and is quite able to compete head to head against 65C02s and z80s on that score (the x88 and x86 have always been worse). But, it's not had enough money behind it to get anywhere in the megahertz race so it's been stuck (still making a nice profit tho) in the embedded markets, until now.
Dead example, the road is littered with the remains of good machines that were killed by Wintel as they dashed for the cash. The most well known is the DEC Alpha, then comes the Transputer, the Sparc, and many others their lunch was eaten by the shark called Wintel and unlike AMD they couldn't survive on the leftovers when times were bad.
But right now things have changed. The Megahertz race has ended, the victor wasn't any of the publicly predicted ones; nobody said that heat would end it. But what's next? Cores, well no; it's been known for decades that a compiler with a normal programming language and a normal programmer is limited to providing work for maybe 20 cores, usually less than 10. The Intel i7 with it's 8 threads is already half way there. Parallel programming, but that's hard.
This is the kind of crap the IE excels at. Because you can ignore the browser, write a native APP and call it an ActiveX control. Of course, you'll probably be tied to IE because Microsoft are trying to kill ActiveX and IE version N+1 won't support it. Their current flavour is DotNET (silverlite) but that has built in obsolescence so you'll have another big upgrade in a couple of years. You might try Flash if you want native in a browser.
If you want completely native in a browser and you can choose the browser you're better off using the Citrix client (probably in seamless mode) as your browser analogue. You could also roll your own pseudo browser if you have a mind to, it can be far far simpler than a real web browser because it only has to understand one language not the ten to twenty plus variations that are needed for a real web browser.
Assuming you're fixed on a real web browser, the best one for applications is Google Chrome, that is what it was designed for after all (G-Mail, G-maps, G-Docs, G-ood, G-od). Printing may still be an issue as HTML can't do it, you can't even choose a proper CMYK print colour. A good solution would be some form of PDF download, possibly with an enhanced dialog giving the options 'Save, View or Print' so you can get 'single click'.
But many of them are not, they are occasional admins who don't check FD on a daily basis. If their machines get owned it impacts ME. OTOH, I can wait a little while, I have things for general mitigation of all threats that work a lot of the time with any attack. Layers you know.
This means I want the vendor to be told first so they have a chance to fix all those other machines before the exploit is on s-kiddy release by every two bit crook who thinks they can make a penny.
OTOH, if the vendor doesn't move quickly, I need to know the exploit so I can put in specific mitigations.
It appears that Tavis Ormandy has done this correctly, because Microsoft were reportedly ignoring him.
I'm pretty sure you're mis-reading the privacy levels here.
The way I read this is that your 30 foot wall is like WEP, it's a positive indication of the expectation of privacy. In both cases breaking the privacy is a definite indication of trespassing and is likely to be grounds for prosecution. But neither stops a determined invader.
A normal white picket fence is much lower security or privacy; it's something that someone who's even a little bit fit can just jump over but in general they don't, because they're not that impolite, and they don't want any handy agricultural implements to be waved at them.
If there's no fence at all you would be very surprised to be confronted for just walking there.
That last one is the totally open WiFi point; quite simply if no effort has been made to fence something off it's unreasonable to expect it to be considered private property so connecting is so obvious the computer will do it without prompting. I would say the picket fence is about the same level as MAC filtering or turning off the DHCP, easy to hop over but rarely done.
All these rules are the ancient ones that were first invented when people started living in villages rather than just a single family group and there's even a comparison for Google. They might not like the label, but in this time of the Global village they're the Village Gossip. She (usually) will know where everyone is she will be able to tell you happy details about everyone in village so that you feel you know them, even if you haven't exchanged more than two words in the last 6 months. In a real way she knits the village together.
So the question is would you do something you don't want the Village Gossip to see in you front garden? She's not always there; but it sure feels like it sometimes. If she did see something would you want her to do any more than forget what she saw.
Sorry, all those features are available for XP. With the possible exception of support for disks that don't exist yet. Several are free stuff not from Microsoft and all the better for it.
The ONLY reason to start using MSW7 is the upgrade treadmill.
And there are several thousand reasons for using XP in the form of drivers and programs that still cannot work on MSW7.
I mean, it ridiculous; Linux still has better driver support than MSW7.
The 'working week' is a relatively modern concept, when the machines take over everything that can be written down as a list of mechanical instructions people will go back to the old ways. Many will be like the lions in Africa, doing the minimum they need to get by. But some will be thinking, exploring doing the things that make the world change, just like they were doing at the start of the industrial revolution.
Modern communications being what it is it's impossible that everything will slow back to the old ways, but really is Youtube that much different from the gossips of old?
Slashdot Mirror
The UK data protection act states that your customer's details must be kept secure. If they aren't there are, possibly significant, legal penalties.
A company giving personal details to an organisation that's been PUBLICLY shown to have piss poor security is a recipe for losing court cases, big time.
Sending the list by unencrypted email was just stupid, but generally doesn't result in a real security breach because like the "purloined letter" the gems are swamped in junk. Nevertheless high profile targets (like ACS law and possibly BT) need to be much more careful. Having a policy of sending or accepting such data by email is another matter, however.
http://www.righthavenlawsuits.com/
http://www.lasvegassun.com/news/2010/sep/03/sharron-angle-hit-r-j-copyright-lawsuit/
In the dos days you COULD just "flip the switch". All the good disk caches would sync the disk before you got back to the dos command prompt or when the application called the idle trap. (int 28 or so?)
Of course, it wasn't safe to turn off while an application was writing to disk (as well as being slow and unreliable to be continually writing to disk) so the idea of working on a copy of the document in memory and "saving" the copy was born.
Yes, Hibernate is a workaround. It's actually a bad hack that's it's "supposed" to be used to improve the boot (and shutdown) speed, but I suppose nobody really cares.
BillG was right! (Oh shit, I didn't say that did I !!!??)
The application the user was running just took minutes to save the file it has stated that the user data is safely on the disk. From the user's point of view THERE IS NOTHING TO SAVE. So from the user's point of view it's safe to turn off, hard, immediately.
It's even pretty easy to do at the technical level. Databases have been doing this sort of thing for decades. Of course for a filesystem it should be slightly different. Ie: any file that gets an fsync() should be considered important enough to commit to disk early. But everything else gets committed strictly on a snapshot basis because there are no "commit transaction" semantics.
So, every few seconds (or minutes) the system is "checkpointed" such that on a hard reset it will come back to the last, completely consistent, checkpoint (especially including user data). If someone uses fsync() on a file just that file is saved away to disk with instructions to 'fix' the filesystem after a crash. You would do it this way so you could reduce power consumption and improve speed on crappy drives compared to committing every single tiny disk operation.
In technical terms I suppose that could be a copy on write filesystem with a rollforward log for the fsync() and related APIs. I guess doing an fsync() on a file would mean that it isn't copied on write any more, that way a DB could stop external fragmentation and LILO and friends could still work.
Microsoft have tried to do it; NTFS has it's log file, but it only stores the least important data on the disk. So you will get in trouble if you frequently hard power off NTFS. (Only the metadata is logged; Think about it, the user does a backup of the files, they get some metadata but if just the metadata is broken they don't care unless it means they lose the file too.)
Where's my Linux version? The closest is ext3 with full logging but that writes everything twice and checkpoints the entire drive if I tell it to fsync() a single file. I suppose I'm okay with that, for now, but I'm still a bit disappointed... except, I hear ext4 is no better ...
Personally, I don't like either the GNU Hurd Bootloader or the GPT partition table. The first is far too complex and the second is badly committee botched. Personally I can't remember the last time I actually needed multi-boot capability anyway, I always use virtual machines now. The last time I got close was testing drivers support for various versions of windows during the recent Vista fiasco, I used multiple disk drives for that.
My preferred Linux partitioning tool is LVM with lilo as the booter. Yes, that is LVM directly on the hard drive, no partition table no GPT; I don't need them and don't want them, they're just one more thing to cause bugs. Lilo is similar, though really Lilo still does too much, but at least unlike grub it doesn't get all pissy and chuck it's toys out of the pram when it can't find it's super secret hiding place.
How about this link then?
Of course that particular one shows a good reason for downloading even when you have the DVDs, twelve seasons on one little USB stick ready to plug into my DVD player...
Actually, in some places, they do. Except they call it "contention ratio". If you have an "up to" 20Mb at a contention ratio of 50:1 that's 400kb CIR. Of course, they will often bury it too.
What they can't tell you with ADSL is your PIR. For example, the advertised is 20Mb but the reality is whatever your modem syncs to. (14Mb for me)
For BT wholesale connections (in the UK) the rations are 50:1 for a residential service and 20:1 for a 'business' version. But that only covers the BT part; from your phone socket to the ISP's lan not their interconnects.
If it's cleaning the drive you're right. The data storage on a modern hard drive is so close to not being there at all that there's no room for old copies. A modern hard drive has to do the equivalent of MFM just to work properly. (aka PRML)
But if you're scanning for and fixing bad blocks it a different matter. I normally do four write passes (with different patterns) with a read pass after each one. This obviously gives a good test for each block but more importantly it allows the S.M.A.R.T to catch up with reality. I've had drives go both ways, ie dubious before to working perfectly and fine before to hundreds of reallocated sectors and SMART failure.
Flash drives are different though, they contain more blocks than needed for their official capacity. So you must do more than one simple wipe to be able to get them all. Eg: a secure erase.
Except, it can take quite a while to sync two drives of this size. So you will probably find that the second drive spends at least half it's time sitting next to the primary drive.
You actually need THREE drives so one of they is always a safe backup.
The actual rules:
eg: offsite
eg: THE backup must stay offsite, it only comes back when it's not THE backup.
Ooooh. Can of worms time.
The problem is that a drive of two real terabytes (not marketing terabytes) is one 512byte sector too large for a normal MBR partition table. So you have to switch over to something else; Windows uses a GPT style partition table.
Unfortunately the committee who designed GPT were dumb (Some of the members must have been smart though, you can easily work around every dumb choice I've seen). A current BIOS doesn't know anything about partition tables and it has no problem with drives up to (IIRC) 64 petabytes. The committee decided that the BIOS should contain a boot manager that understands GPT structures. The BIOS makers don't want this problem; their stuff works fine, some people are already using RAID devices with far in access of 3TB and every OS that can handle a 3TB drive can boot off it ... except Windows.
So it's SNAFU by Microsoft. The weird thing is this time they did it by following the written standard and ignoring the defacto standard ...
It is a huge assumption, back in the nineties the 3.5 inch drives won because they were big enough and had a speed that was noticeably faster than any 5.25 drive could manage.
It's the same reason that SSD drives are so popular for notebooks now. Personally, I'm also now specifying desktop machines with at least two drives (and SSD and a disk) whenever I can.
By unifying video and audio into a single cable
LOL! WUT?
That was done in 1970 by SCART, every TV has a SCART connector (most multiple) over here.
Forty years behind the times, well done.
The closest they'll do is to put FreeDOS on the machine. Which is actually pretty close really as FreeDOS only shows the CPU & Disk work.
Their main problem though for the 'discerning buyer' has always been the website. It's designed to try and segment the market, to try and turn the collection of parts that a technical buyer sees into a smooth line of "the maximum the punter will pay" of marketing theory. While also trying to simplify everything for the non-technical buyer so they are 'theoretical punters' too.
What this means is that the only way you can really map 'FreeDOS' or 'Linux' onto marketing theory is as 'cheap and nasty' because all OSs are exactly the same and the more you pay the better it is ...
Sigh.
No you're saying nothing is ever bricked because by improving your ability or paying someone else you can always get it fixed. This is wrong.
In truth the 'normal' way of flashing your WRT54G can still brick the router because you can't fix the problem using the same technique that you got into trouble in the first place. OTOH JTAG flashing of the router does indeed seem to be unable to brick the device because you can always fix the problem.
That just leaves where you draw the line. IMO the requirements for doing a WRT54G JTAG are too far, ie soldering the board and guessing (parts of) the protocol. But (for example) an edge connector with available documentation would be okay. So you see I put the line at about "Could a professional outsider reasonably fix it without being required to damage or permanently change anything?".
Microwaves are at far too low a frequency for a fibre to act as a wave guide and you wouldn't really want to anyway. You see the visible range is just under one octave, that means that just the optical band plus a tiny bit of the UV band has as much bandwidth as every single frequency below it.
This includes the entire of the radio, microwave, toaster (okay, terahertz band) and infrared combined. So how much data do you think can ultimately be pushed down a fibre?
I'm pretty sure I've written this protocol before myself, it was for uploading files to an old mainframe. (well actually a simulation running on a PC, but it worked the same.) All sorts of nasty limitations meant the file was first converted to ASCII then the lines were sent. They'd get miss ordered and misplaced but the protocol would sort them and retransmit the losses.
The Real News is that the Phone company is screwing everybody again. ..?
You know 80000 messages could have been just one small test file, how much should it cost to transmit a small file
It was a scarce resource and flooding it could actually prevent anyone making calls (there was a fairly simple DoS attack possible). This was the original reason for SMS being expensive
The original price of SMS messages was pitched so that they were cheaper than a "snailmail" letter. Nobody imagined that even at that ripoff price they'd be, like, popular or anything.
Remember the original phones weren't programmable so a technical problem like DOS would be seen as a complete waste of time even by the "techys".
It's as shriphani says, the phrase is "the dash for the cash" or sometimes "the short-termer view".
In the computer world this became a quest for more megahertz (then Gigahertz) only now that thermodynamic limits have been crashed into are multiple processors being looked at. And only because that's the easiest way to grab more cash.
Current example, the ARM processor. It's been around since the 6502 and z80 were mainstream processors. It's always been a 32 bit processor with no 8 or 16 bit leftovers. It's always had very good power consumption and is quite able to compete head to head against 65C02s and z80s on that score (the x88 and x86 have always been worse). But, it's not had enough money behind it to get anywhere in the megahertz race so it's been stuck (still making a nice profit tho) in the embedded markets, until now.
Dead example, the road is littered with the remains of good machines that were killed by Wintel as they dashed for the cash. The most well known is the DEC Alpha, then comes the Transputer, the Sparc, and many others their lunch was eaten by the shark called Wintel and unlike AMD they couldn't survive on the leftovers when times were bad.
But right now things have changed. The Megahertz race has ended, the victor wasn't any of the publicly predicted ones; nobody said that heat would end it. But what's next? Cores, well no; it's been known for decades that a compiler with a normal programming language and a normal programmer is limited to providing work for maybe 20 cores, usually less than 10. The Intel i7 with it's 8 threads is already half way there. Parallel programming, but that's hard.
Got any good ideas? Now's the time.
This is the kind of crap the IE excels at. Because you can ignore the browser, write a native APP and call it an ActiveX control. Of course, you'll probably be tied to IE because Microsoft are trying to kill ActiveX and IE version N+1 won't support it. Their current flavour is DotNET (silverlite) but that has built in obsolescence so you'll have another big upgrade in a couple of years. You might try Flash if you want native in a browser.
If you want completely native in a browser and you can choose the browser you're better off using the Citrix client (probably in seamless mode) as your browser analogue. You could also roll your own pseudo browser if you have a mind to, it can be far far simpler than a real web browser because it only has to understand one language not the ten to twenty plus variations that are needed for a real web browser.
Assuming you're fixed on a real web browser, the best one for applications is Google Chrome, that is what it was designed for after all (G-Mail, G-maps, G-Docs, G-ood, G-od). Printing may still be an issue as HTML can't do it, you can't even choose a proper CMYK print colour. A good solution would be some form of PDF download, possibly with an enhanced dialog giving the options 'Save, View or Print' so you can get 'single click'.
For a proficient admin you are correct.
But many of them are not, they are occasional admins who don't check FD on a daily basis. If their machines get owned it impacts ME. OTOH, I can wait a little while, I have things for general mitigation of all threats that work a lot of the time with any attack. Layers you know.
This means I want the vendor to be told first so they have a chance to fix all those other machines before the exploit is on s-kiddy release by every two bit crook who thinks they can make a penny.
OTOH, if the vendor doesn't move quickly, I need to know the exploit so I can put in specific mitigations.
It appears that Tavis Ormandy has done this correctly, because Microsoft were reportedly ignoring him.
I'm pretty sure you're mis-reading the privacy levels here.
The way I read this is that your 30 foot wall is like WEP, it's a positive indication of the expectation of privacy. In both cases breaking the privacy is a definite indication of trespassing and is likely to be grounds for prosecution. But neither stops a determined invader.
A normal white picket fence is much lower security or privacy; it's something that someone who's even a little bit fit can just jump over but in general they don't, because they're not that impolite, and they don't want any handy agricultural implements to be waved at them.
If there's no fence at all you would be very surprised to be confronted for just walking there.
That last one is the totally open WiFi point; quite simply if no effort has been made to fence something off it's unreasonable to expect it to be considered private property so connecting is so obvious the computer will do it without prompting. I would say the picket fence is about the same level as MAC filtering or turning off the DHCP, easy to hop over but rarely done.
All these rules are the ancient ones that were first invented when people started living in villages rather than just a single family group and there's even a comparison for Google. They might not like the label, but in this time of the Global village they're the Village Gossip. She (usually) will know where everyone is she will be able to tell you happy details about everyone in village so that you feel you know them, even if you haven't exchanged more than two words in the last 6 months. In a real way she knits the village together.
So the question is would you do something you don't want the Village Gossip to see in you front garden? She's not always there; but it sure feels like it sometimes. If she did see something would you want her to do any more than forget what she saw.
It's this what Google have already done?
There are four answers to this.
Number 2 is the most common and the first is almost unknown. But all of them happen with different database tools; check your documentation.
Sorry, all those features are available for XP. With the possible exception of support for disks that don't exist yet. Several are free stuff not from Microsoft and all the better for it.
The ONLY reason to start using MSW7 is the upgrade treadmill.
And there are several thousand reasons for using XP in the form of drivers and programs that still cannot work on MSW7.
I mean, it ridiculous; Linux still has better driver support than MSW7.
Only Windows has that limitation. Everybody else can format the hard drive with a GPT and boot off it no matter what the BIOS type.
The 'working week' is a relatively modern concept, when the machines take over everything that can be written down as a list of mechanical instructions people will go back to the old ways. Many will be like the lions in Africa, doing the minimum they need to get by. But some will be thinking, exploring doing the things that make the world change, just like they were doing at the start of the industrial revolution.
Modern communications being what it is it's impossible that everything will slow back to the old ways, but really is Youtube that much different from the gossips of old?