Any RBL that charges a fee to remove you is not working in the best interest of the Internet. IMNSHO they are just practicing extortion.
Phat -- Remember when that was the noise Bill the Cat made caoughing up a hairball.
So if it only interests 1 in 100,000,000,000,000 then someone should devote any energy to it. Wow, I hope it interests more that an insignificant fraction of 1 human being.
Of course while Ken was saying that Digital was working on reducing the size of the PDP-11 systems so that they fit into monitors, small (12x13x2.5 inches) "shoebox" systems, etc. and two years after was when the first work started on "Single-user" Vax (or SU-Vax, which office politics killed before the hardware team built the prototype Vax for it, it started proto-typed with a PDP-11/34 in the pedestal of the desk). 1980 saw the start of work on the desktop PDP-11 (Dec Professional series) and the Rainbow personal computer and the Decmate II word processing systems); all designed to share peripherials and internal media (floppy and disk drives, not formatted to be compatible of course).
Assumming (and its a big assumption) you could get people to switch to a different email mechanism than the free one currently in use, I'd suggest the following.
Every email sent results in a "email debit" of $.20 and every email read (as in placed in a email inbox and not filtered by the ISP) results in an "email credit" of $.20. At the end of the month and positive balance is reset to zero. Any negative balance is billed to the sender. Normal business or personal email (as opposed to SPAM) would tend to balance out (or be cost competitive with postal rates), and spammers sending millions of SPAM a month (is SPAM its own plural?) would soon find it not paletable due to cost. Take the money gathered and allocate 10% to the local postal system (this gives the government an interest in collecting, and since they have the "guns" so to speak, the SPAMmers would be more inclined to pay if they play), take the other 90% and split it amongst the sending ISP, the recieving ISP and the consummers who receive the email. Thus the ISPs involved get paid, the SPAM recievers get paid, and the governmental agency doing postal service gets paid. If you want to complicate the system then set the fee and credit at the local first class postage level for in country (by destination IP) email and international rate for other email.
While the above would be a fairer system, it would still be woefully inadequate for "moms" or "aunts" or mailing list operators who send more than they receive and are public service oriented or family interest related. So...
It would seem to be best to fine the cr*p out of the SPAMmers where possible, and in the meantime increase the filtering tech and for the love of FLCL please don't allow automatic robots to resend virus laden payloads to the poor b*stards who get their email address put into spoofed headers. Really, how hard is it to make the simple check that the originating email address is in a CIDR associated with the domain of the sender. It is not like you'd need to do it for all email messages, just the bounces to be very effective.
Ranting off, sorry, 300 plus and rising SPAM a day now get past the filters, over 500 get killed by them. Lots of that is because I pre-filter all postmaster and administrative addresses since people quote spam in legitimate complaints and these rules lead to holes for the SPAM to follow.
BTW Anyone else started getting ads that look like complaints but aren't from legitimate users. This is right up there with the SCUM (tm) (Spammers Causing Untold Mayhem) using fake virus problem reports for the Sheep to respond to and install the viruses manually. Dog I love Apple since most all (actually all to date) have been Windoze oriented using that tactic.
I assume that FL and any other states reaching an agreement on this datbase sharing have reached an accord that is blessed by Congress, et al as required in the Constitution? Agreements and treaties between states require Federal approval. If they wanted to, Commerce could probably stop out-of-FL-state data providers from providing data to the state government. Should be interesting.
He purchased the songs in the US. What was not anticipated by the Apple cadre of trained seals was that someone might want to re-authorize from outside the US. So they should store the relevant info to allow one to re-authorize, say a "region code" (he says ducking) in the encoded music that says where it was purchased. Alternately he could have burnt them to CD (maybe even a virtual CD image? don't know if that is currently possible. Maybe the elves will makes something other than shoes tonight). Then if they are in CD form, he could reimport them untagged into iTunes... Anyone know if iTunes will allow transfer such that a CD-RW or/DMG (disk image) can be the output instead of a CD-R?
One could actually use Linux and prepare to pay the SCO license fee extorted to use Linux w/o threat of lawsuit. Then, provided the SCO license does not preclude it, and maybe even if it does and it is considered a contract of adhesion, prepare to sue SCO if their claims are found without merit. Perhaps having SCO be required to put up a bond for the eventual damage suits if their claims are without merit might also slow them down.
Database applications dont have the database running on the client machine. They have it running on the oracle cluster or mainframe in the back room. The client side wouldn't need 2G of memory. And nobody in their right mind is going to run their DB server off of a client box.
It is the hidden databases that are important to the client machines like iTunes, which if they sucked the songs into the DB would be that large , and the registry / preferences (in all its various names and forms). Mostly the current mode is to use the file systems as a non-relational DB, but very soon relational DB based file systems will start being useful (OK a prediction, but one based on current trends). Not to mention the DB built into graphics applications, etc. Apple should consider a MySQL license and bundle it as a generic service (like spell checking is) with a known API to access it (like SQL for example...) That would get people to start considering the use of a generic DB instead of all the proprietary ones built into the applications. And Apple's service concept allows for pretty easy plug-in replacements and for operation in parallel of multiple similar services, so the proprietary-for-ratioanl-reasons DB and legacy DB support could sit along side a "standard / default" DB service.
As to CAD uses, you are a bit narrow minded. it is used for all manner of ancillary uses (like network layout, furniture design, which while able to be in either architecture or engineering still broadens it in the same manner as separating arcitecture from the super class of engineering).
Time to market means nothing. Marketing is the end all. Else the 64 bit Dec Alpha would have won the battle a decade ago. Still IMNSHO a contender for best all around RISC processor and still, unfortunately, horribly undersupported with a dismal future since it directly competes with its current owner's own processor architecture suffering from HUGE not-invented-here problems at HP.
Tungsten would be better. Resonably sure it is the lowest elemental vapor pressure there is. Easy enough to reproduce accurately. Just define the kilogram in terms of a specific size (just for grins suggest inch measurements for the initial draft:-) cube at a specific purity (100%) and measured at STP. Machine the cube with lasers and use optical measures for the finish. Heck IBM even moves tungsten around atom by atom. So you could very accurately fine tune the copies. Use tungsten and get the best of both worlds!
Quantum mechanics allows that our physical constants aren't all that constant. In the future we'll have to move back to physical standards as some of our constants drift or we travel to other spots in our universe where they are just "wrong".
We have started. So much so that I need both metric and english tools to work on my American made car. One or the other would have been much nicer. I have a dashboard on my Mustang that reads in either metric or english (well, or french for some units). I have mixed electronics equipment as well. It is all rather a pain in the !$@#. At least at the individual item level, I'd like it to be one or the other. BTW, the beauty of the "standard" measures is that almost all important measures are one sylable and that applies when the scale uses needs to be changed. inch, foot, yard, rod, chain, mile, etc. ounce pound stone ton... cup pint quart (gallon must be an outlier as are the other larger liquid measures which seem to be two sylables), volt amp watt... Just as simple as the metric system, just a different axis if you have the right point of view. BTW sign me up for one of the obsolete Bridgeport turret lathes, milling machines, shapers, etc. in english measure. Especially any CNC ones. It is not all that tough to change the gearbox over and the dial plates to make it metric which if you are giving 'em away to go metric I'd be happy to do.
For heaveans sake it has a vapor pressure of something on the order of one atom per universe. It is durable and dense. I really like physically achievable in the garage standards and I don't think I can afford to measure the exact number of atoms in anything, nor afford a absolute pure silicon crystal sphere or the like.
If the writers of the benchmarks actually modeled the real world uses then this would not even be an issue as the "tweaked" drivers would be tweaked for real world applications. Since the most common use of the benchmarks is to make purchasing decisions, the writers of the benchmarks are falling down on the job if they don't closely model the real world applications use. As far as recognizing the sequences of operations to determine if it is the benchmark, the benchmark can render the test screens in such a way that statistically the results are valid even if the sequence varies from run to run. (but please keep a standard sequence to help all of us debug the hardware in the first place). When a driver is being optimized they most likely run the benchmarks as a repeatable means of developing call history, etc. while profiling the code. After all you want to spend 80% of your effort optimizing the most frequently use 20% of the code (or whatever you favorite 90/10 or 80/20 like rule is).
Maybe the Iraqi people will get affordable data plans, too.
Then Sprint PCS should have won. PCS Vision with all-you-can-eat data is only $10 over the voice plan you attach it too. And it is a widely deployed (mostly urban/suburban and major highway corridors nationwide) CDMA system. Pretty sweet deal.
One of the issues that is not addressed so far is that the financial requirements for older technical people is significantly higher that that of a new graduate. Insurance premiums are skewed upward, and base salery is higher (the new grad does not yet have a house, two cars a wife, etc usually). Some of this is hidden (like company paid portions), some is more obvious. There is also the bias that can be expressed as "why are you _still_ a programmer after 20+ years" as one is expected to have transitioned long ago into management. And there is the "fresh outlook" presumtion that goes against older technical people. An interview rarely gets to the heart of whether someone is a innovator or an implementor of someone elses innovation. And last there is a disadvantage in that older techno-geeks will have most likely held some position that the folks hiring think is more advanced than the position they are offering (for example I was a director, lead architect and senior research scientist) so they are hesitant to hire (even on a contract basis) those individuals for a simple programnming assignments. The phrase "you seem to be over qualified for this position" is an all to common statement I receive. BTW I am out of work still from the tech bubble bursting (somewhat because it would cost _a lot_ to relocate, and a major factor is that I was too helpful to my former company after being laid off for too long for too little). If the price is right and the job is in the Greater Seattle area and you need an old dog who knows the new tricks (and creates them!) you can hire me.
TiVi just simply frees you from your TV while letting you watch more of what you want to watch. It is a Zen thing. And with the Gen 2 Home Media Option which allows sharing program conent from room to room and playing Mac or PC hosted photo and MP3 content over the TiVo it just keeps getting better. (BTW DirectTv has declined to offer the HMO upgrade even though they'd make $99 for the first receiver and $49 for each addition upgraded receiver in the same household. A petition is available online to allow one to voice their wish for DirectTv to reconsider. It is available here.
Tornado driven telephone poles bounce off 'em. The are made by inflating a elastomer dome, spraying the inside with foam, embedding the rebar ties in the foam, and application of "shotcrete", a sprayed concrete product that cures to a very strong product, to the inside of the shell. If you then bury it, it will last quite some time. Since the interior does not contain load bearing walls to support the outer structure you have quite a lot of flexability in interior layout. Super insulated (and you _could_ bury the wiring ducts in the walls during foaming...) with little to no annual (exterior) maintenance. Essentially just the grounds if you bury it. I always wanted to do one with "tube" skylights to the surface and put "mushrooms" (well mushroom shaped) light gatherers on the surface. Don't want to bury it? Take some stone facing material and cover it instead... You want to keep maintenance low, so don't just paint it. Use a material that is stable and blocks UV. Paranoid? Put a layer of pond-liner clay over it before you bury it. The dome is self supporting and since the concrete is not required to be watertight, minor cracking is not an issue. Use high quality material to avoid any aging problems. And since you'll have most of the equipment on site anyway, "Gunite" the interior surface when complete for a finished plaster surface that is very durable.
I still get 2-3.5 hours off my original G4's battery. I do find the annoying fan noise has gotten louder over the years and the fan seems to run more, but battery life is still pretty close to the day one figures. Most lithium battery tech wants the batttery "topped off" frequently for maximum life (unlike some older NiCd technology). I've only deep discharged (below 20%) maybe 5-6 times. If you deep cycle your batteries, then expect periodic replacements. It is much better today than when 1-2 hours (if you were lucky) was enough to drain a laptop a few years back. Now I hook into the airplanes power when flying, the hotels when I travel, the cars when driving... And obviously the houses when home. As the battery technology gets better look for easily removable ones to disappear. (My cell phone/pda does not have an externally replaceable battery already.)
Since you can already (under NTFS) have multiple "streams" in a file, why not have a generic name for the DLL as the main file, which points to the latest version (for the I don't care users), and then store the different versions as separate file streams. You can mimic this with yet more "special" folders under FAT. Then you can, since NT derivatives can have disk compression, store just the differences between the versions in the streams (requires a bit of a change to the NTFS but not much, and all the APIs are already in place). This maintains a low on disk foot print. And, if the loader where exceptionally clever and the DLL mechanism and format (obviously for future DLLs) for building and linkage were slightly modified, you could cause the loader to map the common readonly segments across all the versions to one space in memory. Very compact and virtually no additional overhead if done correctly. Less space in memory and less space on disk. One of the alternative means would be to actually use the NTFS versioning features (present in the meta-format data but not currently used) for the on disk structure, although that is tougher to mimic on non-NTFS volumes. Best solution though would be to _dump_ FAT now that the "consumer" OS is based on NT. BTW Microsoft, much as I like the Mac platform, feel free to contact me for research or OS dev positions!
Of course, then you'd want to make a few changes to support multiple stream files better in the GUI and some other utilities. The command line already does of course, although you might add some flags to some commands to expose the multiple streams rather than require explicit filenames that identify them. "DIR" would be a good start, so it could display the multiple streams by default, or optionally. The file tree display is another venue for change, and a few more. "CMD" already recognizes them when you specify them, so you can invoke most commands with a particular stream of a multistream file. Then its just a simple matter of doing a bit of compression/versioning and voila, a much saner approach.
Slashdot Mirror
Any RBL that charges a fee to remove you is not working in the best interest of the Internet. IMNSHO they are just practicing extortion. Phat -- Remember when that was the noise Bill the Cat made caoughing up a hairball.
So if it only interests 1 in 100,000,000,000,000 then someone should devote any energy to it. Wow, I hope it interests more that an insignificant fraction of 1 human being.
Of course while Ken was saying that Digital was working on reducing the size of the PDP-11 systems so that they fit into monitors, small (12x13x2.5 inches) "shoebox" systems, etc. and two years after was when the first work started on "Single-user" Vax (or SU-Vax, which office politics killed before the hardware team built the prototype Vax for it, it started proto-typed with a PDP-11/34 in the pedestal of the desk). 1980 saw the start of work on the desktop PDP-11 (Dec Professional series) and the Rainbow personal computer and the Decmate II word processing systems); all designed to share peripherials and internal media (floppy and disk drives, not formatted to be compatible of course).
Assumming (and its a big assumption) you could get people to switch to a different email mechanism than the free one currently in use, I'd suggest the following.
...
Every email sent results in a "email debit" of $.20 and every email read (as in placed in a email inbox and not filtered by the ISP) results in an "email credit" of $.20. At the end of the month and positive balance is reset to zero. Any negative balance is billed to the sender. Normal business or personal email (as opposed to SPAM) would tend to balance out (or be cost competitive with postal rates), and spammers sending millions of SPAM a month (is SPAM its own plural?) would soon find it not paletable due to cost. Take the money gathered and allocate 10% to the local postal system (this gives the government an interest in collecting, and since they have the "guns" so to speak, the SPAMmers would be more inclined to pay if they play), take the other 90% and split it amongst the sending ISP, the recieving ISP and the consummers who receive the email. Thus the ISPs involved get paid, the SPAM recievers get paid, and the governmental agency doing postal service gets paid. If you want to complicate the system then set the fee and credit at the local first class postage level for in country (by destination IP) email and international rate for other email.
While the above would be a fairer system, it would still be woefully inadequate for "moms" or "aunts" or mailing list operators who send more than they receive and are public service oriented or family interest related. So
It would seem to be best to fine the cr*p out of the SPAMmers where possible, and in the meantime increase the filtering tech and for the love of FLCL please don't allow automatic robots to resend virus laden payloads to the poor b*stards who get their email address put into spoofed headers. Really, how hard is it to make the simple check that the originating email address is in a CIDR associated with the domain of the sender. It is not like you'd need to do it for all email messages, just the bounces to be very effective.
Ranting off, sorry, 300 plus and rising SPAM a day now get past the filters, over 500 get killed by them. Lots of that is because I pre-filter all postmaster and administrative addresses since people quote spam in legitimate complaints and these rules lead to holes for the SPAM to follow.
BTW Anyone else started getting ads that look like complaints but aren't from legitimate users. This is right up there with the SCUM (tm) (Spammers Causing Untold Mayhem) using fake virus problem reports for the Sheep to respond to and install the viruses manually. Dog I love Apple since most all (actually all to date) have been Windoze oriented using that tactic.
Ranting off for real this time.
At first glance I thought scientists might have discovered a real-life rust monster.
They have, it is just that this is the microorganism that is also a symbiot with gelatinous cubes in normal atmospheric conditions.
Or
MUlTistate
ANTi
TeRroism
Information
eXchange
In Russia, er... Florida, you don't eat your breakfast cereal, your breakfast cereal eats you.
I assume that FL and any other states reaching an agreement on this datbase sharing have reached an accord that is blessed by Congress, et al as required in the Constitution? Agreements and treaties between states require Federal approval. If they wanted to, Commerce could probably stop out-of-FL-state data providers from providing data to the state government. Should be interesting.
He purchased the songs in the US. What was not anticipated by the Apple cadre of trained seals was that someone might want to re-authorize from outside the US. So they should store the relevant info to allow one to re-authorize, say a "region code" (he says ducking) in the encoded music that says where it was purchased. Alternately he could have burnt them to CD (maybe even a virtual CD image? don't know if that is currently possible. Maybe the elves will makes something other than shoes tonight). Then if they are in CD form, he could reimport them untagged into iTunes ... Anyone know if iTunes will allow transfer such that a CD-RW or /DMG (disk image) can be the output instead of a CD-R?
One could actually use Linux and prepare to pay the SCO license fee extorted to use Linux w/o threat of lawsuit. Then, provided the SCO license does not preclude it, and maybe even if it does and it is considered a contract of adhesion, prepare to sue SCO if their claims are found without merit. Perhaps having SCO be required to put up a bond for the eventual damage suits if their claims are without merit might also slow them down.
Database applications dont have the database running on the client machine. They have it running on the oracle cluster or mainframe in the back room. The client side wouldn't need 2G of memory. And nobody in their right mind is going to run their DB server off of a client box.
...) That would get people to start considering the use of a generic DB instead of all the proprietary ones built into the applications. And Apple's service concept allows for pretty easy plug-in replacements and for operation in parallel of multiple similar services, so the proprietary-for-ratioanl-reasons DB and legacy DB support could sit along side a "standard / default" DB service.
It is the hidden databases that are important to the client machines like iTunes, which if they sucked the songs into the DB would be that large , and the registry / preferences (in all its various names and forms). Mostly the current mode is to use the file systems as a non-relational DB, but very soon relational DB based file systems will start being useful (OK a prediction, but one based on current trends). Not to mention the DB built into graphics applications, etc. Apple should consider a MySQL license and bundle it as a generic service (like spell checking is) with a known API to access it (like SQL for example
As to CAD uses, you are a bit narrow minded. it is used for all manner of ancillary uses (like network layout, furniture design, which while able to be in either architecture or engineering still broadens it in the same manner as separating arcitecture from the super class of engineering).
Time to market means nothing. Marketing is the end all. Else the 64 bit Dec Alpha would have won the battle a decade ago. Still IMNSHO a contender for best all around RISC processor and still, unfortunately, horribly undersupported with a dismal future since it directly competes with its current owner's own processor architecture suffering from HUGE not-invented-here problems at HP.
All your kilogram are belong to us.
Tungsten would be better. Resonably sure it is the lowest elemental vapor pressure there is. Easy enough to reproduce accurately. Just define the kilogram in terms of a specific size (just for grins suggest inch measurements for the initial draft :-) cube at a specific purity (100%) and measured at STP. Machine the cube with lasers and use optical measures for the finish. Heck IBM even moves tungsten around atom by atom. So you could very accurately fine tune the copies. Use tungsten and get the best of both worlds!
Quantum mechanics allows that our physical constants aren't all that constant. In the future we'll have to move back to physical standards as some of our constants drift or we travel to other spots in our universe where they are just "wrong".
It's all fun and games until a 200' robot dinosaur shows up and trashes Neo-Tokyo... Again
Wouldn't that be a 6.096 dekameter robot dinosaur? (or if you want to lessen panic, just say a 0.006096 myriameter robot dinosaur).
We have started. So much so that I need both metric and english tools to work on my American made car. One or the other would have been much nicer. I have a dashboard on my Mustang that reads in either metric or english (well, or french for some units). I have mixed electronics equipment as well. It is all rather a pain in the !$@#. At least at the individual item level, I'd like it to be one or the other. BTW, the beauty of the "standard" measures is that almost all important measures are one sylable and that applies when the scale uses needs to be changed. inch, foot, yard, rod, chain, mile, etc. ounce pound stone ton ... cup pint quart (gallon must be an outlier as are the other larger liquid measures which seem to be two sylables), volt amp watt ... Just as simple as the metric system, just a different axis if you have the right point of view. BTW sign me up for one of the obsolete Bridgeport turret lathes, milling machines, shapers, etc. in english measure. Especially any CNC ones. It is not all that tough to change the gearbox over and the dial plates to make it metric which if you are giving 'em away to go metric I'd be happy to do.
Don't use Moe for a machinist. Besides the shipping from Springfield is enormously expensive.
For heaveans sake it has a vapor pressure of something on the order of one atom per universe. It is durable and dense. I really like physically achievable in the garage standards and I don't think I can afford to measure the exact number of atoms in anything, nor afford a absolute pure silicon crystal sphere or the like.
If the writers of the benchmarks actually modeled the real world uses then this would not even be an issue as the "tweaked" drivers would be tweaked for real world applications. Since the most common use of the benchmarks is to make purchasing decisions, the writers of the benchmarks are falling down on the job if they don't closely model the real world applications use. As far as recognizing the sequences of operations to determine if it is the benchmark, the benchmark can render the test screens in such a way that statistically the results are valid even if the sequence varies from run to run. (but please keep a standard sequence to help all of us debug the hardware in the first place). When a driver is being optimized they most likely run the benchmarks as a repeatable means of developing call history, etc. while profiling the code. After all you want to spend 80% of your effort optimizing the most frequently use 20% of the code (or whatever you favorite 90/10 or 80/20 like rule is).
Maybe the Iraqi people will get affordable data plans, too.
Then Sprint PCS should have won. PCS Vision with all-you-can-eat data is only $10 over the voice plan you attach it too. And it is a widely deployed (mostly urban/suburban and major highway corridors nationwide) CDMA system. Pretty sweet deal.
One of the issues that is not addressed so far is that the financial requirements for older technical people is significantly higher that that of a new graduate. Insurance premiums are skewed upward, and base salery is higher (the new grad does not yet have a house, two cars a wife, etc usually). Some of this is hidden (like company paid portions), some is more obvious. There is also the bias that can be expressed as "why are you _still_ a programmer after 20+ years" as one is expected to have transitioned long ago into management. And there is the "fresh outlook" presumtion that goes against older technical people. An interview rarely gets to the heart of whether someone is a innovator or an implementor of someone elses innovation. And last there is a disadvantage in that older techno-geeks will have most likely held some position that the folks hiring think is more advanced than the position they are offering (for example I was a director, lead architect and senior research scientist) so they are hesitant to hire (even on a contract basis) those individuals for a simple programnming assignments. The phrase "you seem to be over qualified for this position" is an all to common statement I receive. BTW I am out of work still from the tech bubble bursting (somewhat because it would cost _a lot_ to relocate, and a major factor is that I was too helpful to my former company after being laid off for too long for too little). If the price is right and the job is in the Greater Seattle area and you need an old dog who knows the new tricks (and creates them!) you can hire me.
TiVi just simply frees you from your TV while letting you watch more of what you want to watch. It is a Zen thing. And with the Gen 2 Home Media Option which allows sharing program conent from room to room and playing Mac or PC hosted photo and MP3 content over the TiVo it just keeps getting better. (BTW DirectTv has declined to offer the HMO upgrade even though they'd make $99 for the first receiver and $49 for each addition upgraded receiver in the same household. A petition is available online to allow one to voice their wish for DirectTv to reconsider. It is available here.
She also had ordered McDonalds coffee previously, and placed it between her legs (in a crushable styrofoam cup) knowing how hot it would be.
I measured mine from both the drip and the french press. Drip was 187F, french press was 191F.
Tornado driven telephone poles bounce off 'em. The are made by inflating a elastomer dome, spraying the inside with foam, embedding the rebar ties in the foam, and application of "shotcrete", a sprayed concrete product that cures to a very strong product, to the inside of the shell. If you then bury it, it will last quite some time. Since the interior does not contain load bearing walls to support the outer structure you have quite a lot of flexability in interior layout. Super insulated (and you _could_ bury the wiring ducts in the walls during foaming...) with little to no annual (exterior) maintenance. Essentially just the grounds if you bury it. I always wanted to do one with "tube" skylights to the surface and put "mushrooms" (well mushroom shaped) light gatherers on the surface. Don't want to bury it? Take some stone facing material and cover it instead ... You want to keep maintenance low, so don't just paint it. Use a material that is stable and blocks UV. Paranoid? Put a layer of pond-liner clay over it before you bury it. The dome is self supporting and since the concrete is not required to be watertight, minor cracking is not an issue. Use high quality material to avoid any aging problems. And since you'll have most of the equipment on site anyway, "Gunite" the interior surface when complete for a finished plaster surface that is very durable.
I still get 2-3.5 hours off my original G4's battery. I do find the annoying fan noise has gotten louder over the years and the fan seems to run more, but battery life is still pretty close to the day one figures. Most lithium battery tech wants the batttery "topped off" frequently for maximum life (unlike some older NiCd technology). I've only deep discharged (below 20%) maybe 5-6 times. If you deep cycle your batteries, then expect periodic replacements. It is much better today than when 1-2 hours (if you were lucky) was enough to drain a laptop a few years back. Now I hook into the airplanes power when flying, the hotels when I travel, the cars when driving... And obviously the houses when home. As the battery technology gets better look for easily removable ones to disappear. (My cell phone/pda does not have an externally replaceable battery already.)
Since you can already (under NTFS) have multiple "streams" in a file, why not have a generic name for the DLL as the main file, which points to the latest version (for the I don't care users), and then store the different versions as separate file streams. You can mimic this with yet more "special" folders under FAT. Then you can, since NT derivatives can have disk compression, store just the differences between the versions in the streams (requires a bit of a change to the NTFS but not much, and all the APIs are already in place). This maintains a low on disk foot print. And, if the loader where exceptionally clever and the DLL mechanism and format (obviously for future DLLs) for building and linkage were slightly modified, you could cause the loader to map the common readonly segments across all the versions to one space in memory. Very compact and virtually no additional overhead if done correctly. Less space in memory and less space on disk. One of the alternative means would be to actually use the NTFS versioning features (present in the meta-format data but not currently used) for the on disk structure, although that is tougher to mimic on non-NTFS volumes. Best solution though would be to _dump_ FAT now that the "consumer" OS is based on NT. BTW Microsoft, much as I like the Mac platform, feel free to contact me for research or OS dev positions!
Of course, then you'd want to make a few changes to support multiple stream files better in the GUI and some other utilities. The command line already does of course, although you might add some flags to some commands to expose the multiple streams rather than require explicit filenames that identify them. "DIR" would be a good start, so it could display the multiple streams by default, or optionally. The file tree display is another venue for change, and a few more. "CMD" already recognizes them when you specify them, so you can invoke most commands with a particular stream of a multistream file. Then its just a simple matter of doing a bit of compression/versioning and voila, a much saner approach.
Bacteria have evolved to live in nuke plant cooling pipes. (The H2O ones, not the He or Na ones)