As much as I'd hate to see internet tax, it might be a mechanism to fight SPAM. Introducing a tax of 1 penny for each e-mail sent would set the average user back about $1 - $5 a month.
SPAM houses would pay through the nose... I thin this would be a small investment for all of us to make junk mail less profitable.
As much as I agree with you, I have already had long and detailed debates about these subjects. With your proposal, there are a few "problems".
1. How do you regulate it? I receive 100 messages a day that are logs of virus infections on computers I monitor, each of these computers send it directly to my server over the net.
2. Legimate mailing lists, such as the Debian projects mailing list, the kernel mailing list, etc. How do they pay? If you suggest that mailing lists are exempt, then of course, spam lists are just mailing list.
The next line of thought goes to, a fee for any unsolicitated spam sent, but there are many problems with that. How do you find the actual spammer? Maybe fine the person who paid for the spamming (the seller), but if they are located outside your country, that can make it hard. They can also easily claim they never asked any spammer to send this, and the burden is on your to prove they did. So you are back to finding who the spammer was.
Essentially, in any free system, there will *always* be abuse. Its the way it is, and it will probably always be that way. Remember, there are no country lines on the internet. And US law doesn't apply anywhere but the US.
And by that I mean the coders (if they have any left)? Sure, the sales people are just doing their job, but what about the coders at the company? Surely they can't believe the drivel thats coming from above? Is there actually anyone left in that company that has more than 2 ethical brain cells?
Remember, this company was once Caldera, who produced a linux distro, so is there anyone left from those days?
Still, flying a "piddly single engine" irl beats anything you can do with flight simulator (atleast indoors). It's like comparing pr0n with the real thing (a really really bad comparsion considering this is slashdot...).
Agreed, however, when you're sick of forking over large chunks of your paycheck for said engine, you can return home and sit up front in the cockpit of a 747-400 and fly from New York to London. All for arround the same price as a couple of hours of real flying.
I use to fly, but got too expensive, or rather, my other hobbies got too expensive. So now, I fly from the comfort of my home PC.
This is where, in some cities, these Fligh Sim places can really, er, fly. (excuse the pun).
I'm more than willing to fork over a few dollars an hour to sit in a "real" flighsim and have some fun there... only wish I could, none in my City:(
Why don't all these guys who spend b i g bucks on a flightsim mock-up just get a pilots licence? It's way more fun, 100% real and much cheaper. Heck, some of these guys could actually buy a real plane for less than they spend on the computers alone.
Sheesh!
Cheaper than real flying?
Its betweeen $60-$100/hr to rent a cessna to fly. And thats just a piddly single engine job.
Flight Sims have their definate place. There a a HUGE number of reasons why you would spend this much on a flight sim. If I were to bet, I would say that some of these "mega" flighsim buffs either are private pilots, or have been. And I bet a few were airline pilots.
Oh yeah, and you can't just drive to your local airport and RENT a 747 to have a few fun flights in. They don't really do that.
If you believe that its much cheaper to get a pilots license, then you haven't tried. If you want to go the cheap path, join the airforce, they PAY you to learn to fly.
Back during the great auckland power crises of 97, my ISP was Binary Brothers, a now extinct ISP. They were a great ISP, run by a few guys who knew their stuff.
Turns out the owners were physically located on the coremandle peninsula (about 4 hrs drive from Auckland), while their servers/modem racks, etc were located in the heart of auckland CBD.
The power in auckland blinked out, and as did my net connection (I was located outside the blackout area). I rang 'em up and asked if they were out due the the power crises. They replied that they were currently in the wagon driving from the coremandle with a generator in the back, and the ISP will be back online in a few hours.
Well, they were right, it flicked back online in a few hours. Gotta hand it to them, they really did a good job. I just have this mental picture of these two guys in a wagon, speeding down the windy roads, one yelling, drive faster, we need more power!
Anyway, that was those days, these days I'm lucky if my home connection stayed up during rain storm, let alone a power blackout.
Having just recently configured another Xerox Document Centre on our network, I must say, they are a breeze to configure and print to.
They are a true network printer, they talk LPR, SMB, and, not that I've tested, apple-talk and netware.
I use to have trouble finding drivers, and configurating them, but then I discovered a few settings here and there that made them work right (to do with the hi-cap feeder, fax, and duplex).
The document centres come with various options, but we've got two (a 250 and a 400 model) that have the fax, duplex and high cap feeder options.
Oh, and they talk postscript. *REAL* postscript.
They print, scan, and fax, and I've heard some of the xerox boys at the R&D labs hooked up a coffee maker, so it can even do that *GRIN*.
The only thing I'm not sure about is the network scanner side. There is a windows client that interfaces to it, but I've never tried with linux.
What was said about Auckland, New Zealand in the article was a pretty fair assesment.
Being as I live in Auckland, I was there for the power crisis. Yes, businesses folded, yes, most CBD businesses lost money, but those that folded were most likely going to fold anyhow, and money can be made back.
It hurt the people the most though. Some were fortunate (like the BNZ bank staff), and their companies moved the staff arround to keep them working, but a lot of people couldn't work during those 5 weeks. And 1 month without a pay cheque hurts.
The same obviously applies to the States at the moment. Bussinesses aren't the one to be worring about, its the people.
The power company stiffed Auckland though. Mecury energy is still a force here, they do still run a lot of the power. But then again, lightning never strikes the same place twice... does it?
For the most part, the power grid in most countries performs amazingly well. You try designing a system that can handle an average 20 or 30 lightning strikes a day and still keep on pumping.
How about installing the ROBOCOPY from the Windows 2000 Server resource kit in the NT box, and then using that (with the/SEC switch) to copy the files over. Wouldn't that work?
I tried that, well, tried scopy, didn't have the 2k resource kit. Kept getting, of all things, permission denied errors when trying to copy the file. (I permission to read the source, and permissiong to write to the destination). If I manually copied (eg using xcopy) the src to the dest, and manually applied the acl, it worked..
Samba 3.0 is the first real samba (excluding samba-tng), imho, that can replace a WinNT4 PDC (Primary Domain Controller) *fully*.
(eg: with samba3, the windows usrmgr.exe works for adding/deleting users & groups. (usrmgr.exe communicates over RPC, so I consider it something that should work for a windows primary domain controller).
I have just recently setup for a company:
A samba PDC, with usrmgr.exe working.
With an LDAP backend for authenciation.
With posix ACLs on the file system (to allow *real* permission settings. The perms are still a bit wierd, and I feel better setting them in Linux rather than through the windows gui, but they do work).
With cups printer backend, so printing works great.
Basically, this machine fully replaces their windows NT4 server, and does it pretty damn well.
The move from NT4 to PDC was pretty good. Once everything is setup on the samba side, you can "net vampire" all of the user and group accounts over to the samba server, and the users can login with no problems.
The only missing feature was I needed some way to copy the file system on the NT box to the linux box and keep the ACLs.
* If you have an nForce chipset, make sure to add "mem=nopentium" to your kernel boot parameters, or else your system will be incredibly unstable. Better yet, ditch your nForce chipset (I did) since the Linux support totally blows, at least for now. Give your old nForce chipset to your wife, girlfriend, mother, Windows box, or whatever.
Oh how I agree with that statement.
Recently my motherboard died of bad caps, so decided to splash out on an nvidia motherboard. Damn thing had crap all linux support. I don't know if it was the m/b manufacturer (MSI), or nvidia chipsets in general, but:
USB didn't work (even under windows w/ or w/o nvidia special drivers), actually, the worked, but very rarely.
Onboard Sound didn't work under linux (w/ nvidia special binary only drivers)
I didn't know about the mem=nopentium trick, and would have tried it, but I sent the board back to my supplier with, "tried it, won't work, won't buy, won't recommend". They didn't put up much protest.
I stick with the via chipsets. Sure they aren't supported by via for linux (atleast, not in any real way, they do have drivers, but they are binary only, and don't work). The opensource drivers work damn well. Infact, I suspect, better than the manufacturers binary only (provided you can get them to work).
Then again, thats just for my home pc, which I cheap out on. For any server, I just stick with IBM parts. damn things just plain well WORK!
More MP3 archives of good public radio shows would be most welcome. (I don't suppose there's a good Real Adio --> mp3 converter for Linux?)
Of course there is, this is linux. Setup a named pipe, call it/dev/dsp, and realplayer should happly write to it.
Ok, so maybe you may run into some small blocks, but generally speaking, it can be done. Especially in Linux. Thats why the dvd/etc industry don't like Linux. Its too easy to simply write an audio/video driver that writes whats being played/displayed to a file rather than to the speaker/screen.
Then again, if you wanted to get real low-tech, you could hook the audio out of one sound card to the audio in of another (or if your card is decent enough, back into the same card). Then convert that to mp3.
As one of the helpers for the installfest, I can say that this is pretty much only going to help our cause. We couldn't ask for better advertising (both the NZ Herald, and Slashdot).
We will be ready, Saturday, with plenty of CDs (we hope).
Van Eck's freaking is still possible with an LCD display. Its to do with the rythmic timing of a PC. Its easy to spot the 70Hz (60, 80, whatever) of your monitor. Your LCD also refreshes. There is the writes to the video memory, etc. I suppose you could say its more difficult, but compared to actually doing Van Eck freaking in the first place, its only marginally more difficult. If you can freak VDUs, you can freak LCDs.
As for the initial problem of restricting access. If you want to ensure that nobody can be coereced into authenticating, then you'll need more than one person. preferably a commity of 11 people who hate each other. If you put a gun to my head, I'll give you all the passwords i know. So lets say I'm a good employee, and won't give out the passwords/authenticate under the threat of death, what about threatning someone I know, or love.
If humans are responsible for authenticating, then another human will be able to coerce the first one into giving access. This is why good safes and bank vaults are on a time delay. It doesn't matter how many passwords you know, the vault will still take 30 minutes to open from the time you enter the right code. The assumption being that 30 mins is a) enough time for the cops to get there, and b) that the criminals want to avoid a standoff with the cops.
Want to make a totally secure server? turn it off, bury it in concrete, and under a mountain, make sure you are the only one who knows where it is, then shoot yourself. Seriously, you can't make something that will give access only under certain circumstances, there will always be ways around it. This is why security is made up many levels.
Ok, so lets assume that you just want a system that only YOU can access, and only if you choose (and assuming you can make a valid choice with a gun to your head). Well, first we'll encrypt everything with some pretty good encryptiong. Thats the easy part done. Now you have a key. How to decrypt the data, with the key, without exposing the data, or the key, to any other party (call them "Eve") who are listening in.
You can't.
Take a look behind you... At that wall over there. Yes, that one. See that black spot? No, not that one, the other one... Yes... how do you know its not a camera, recording everything you see and do?
Ok, so lets put the screen inside googles that you wear.. whoops, how do we know that noone has tampered with the googles, or that they are Van Eck'ing the googles.
Its impossible. If you come up with some solution that is very very very good, then come talk to the CIA, they'll have a job for you. If you can come up with a way around it, you'll be able to get two pay checks, the other from the NSA.
In the end its about risk-management. What will it cost me, or you, if whatever method I use, fails? How much money does my adversary have? How much money does the solution cost? Now balance those. If the risk is you goto jail. And you figure you can buy your way out of jail for say $20millon, and the protection will protect you for $10 million, but it'll cost $30 million to by pass, and your adversary only have $5, then you should use the solution. But if you adversary can bypass it, then you may aswell just save $20million for the times you get caught, and stick with something simple.
It looks just like one of those home gyms. And then when the marketing speak said: > The MasterPeace Rocks and Cradles you I thought... "You can rock, rock, rock! yourself to firmer abs!"... damn those infomercials.
The biggest problem with those is there is no place to put good old fashion pen to paper.
Even when I was in R&D I still liked to use pen and paper beside my superfast machine. I like a big desk with lots of room to scribble.
Interesting side note, since we were in R&D, we weren't allowed any "scrap" paper to work on. All our work had to be in our logbooks, that had all our notes (aparently incase of patent dispute). So I started doing my doodles in that. Then I put things like, "if you are reading this, then I must be dead, the treasure is buried....". Of course, after our head of project died, i stopped doing that.
There was a proposal for dynamic lengths of IP addresses. When I thought about it, it sounded pretty neat too, however it becomes pretty unmanagable once you get too far into it.
As for the Y2k thing. Yeah, I really think we will suddenly all say, "This calender sucks, lets move to a different one". The last country to adpot the "new" calander finally switched about 20 years ago. And the calander was switched arround 200 years ago.
The reason for using fixed length is efficency and ease. Its much easier as a programmer to know that hostnames won't be larger than 64 characters (true!). That way I can put a static buffer 64 chars big, and then if I'm a good programmer, check that the data I'm copying into the buffer isn't bigger than 64 chars.
If you have a dynamic address length, everything changes. try writting a program that deals with arbitary strings of any length. sprintf and friends are usually implemeneted with a 1024 character buffer, so you can't use those easily.
If I'm writting a program that takes lines of text of arbitary length, I usually assume that the lines are no longer than 1024 characters. Otherwise I have to start reallocing buffers, or rewinding the stream.
If you have a dynamic size IP address, then there is a limit to how long it is. If the field that specifies the length is 1 byte, then the IP address can be 255 octets long. Now, since a 255 octect address can be represented in hexidecimal by 510 characters (assuming you don't have seperators), do you really think it would be practical to have an address that long?
The ISP doesn't have to prove it. Most (all?) ISPs will have a phrase in their contracts with you stating "The service can be canceled by ISP at any time for any reason" or something very similar.
Yes, they would have to prove it if they wished to take you to court, and they would only take you to court if they levied charges against you for your NATing. (However, in New Zealand, if you use Xtra (which is owned by Telecom), and you don't pay your bill, they just cut your phone line).
ISPs can drop your service for any reason they want. The NAT analysis is for their benifit, if they can detect users with NAT, send 'em a warning, if they continue, maybe investigate further, or just cancel the account.
The printer manufactures need to sell many printers, at low cost, to many users. Now, in order to do that, they need to reduce manufactoring costs (thus lower quality) and reduce profit margins.
Some bright spark[1] decided that once a person buys a printer, they are commited to it, so will have to buy the print cartridges for it. So if we make the cartridges expensive, we can still maintain our profit margins, and have continous profits rather than once off for each customer.
Now enter the business side of things. Our business customers don't want to keep buying the latest bubblejet/inkjet/crapjet every 3 months, so they produce a seperate business line of machines. Mostly these are laser based, however, there are some top-of-the-line inkjet systems that are mostly used in the printing industry (eg signs/cars/etc).
So you either buy a business quality printer, preferably laser based, and you pay good money for it. Or you do what some of my customers do:
They buy a new printer when the old print cartridge runs out. However, they are being thwarted by the print manufactores who are now selling print cartridges half full on new printers, so they buy a new cartridge with the printer (usually at a discount, since they can wrangle one with the printer), and run it till it runs dry, and pick up the next latest and greatest model.
Ok, so thats a bit extreme, but I do have one customer doing that.
Basically, printers are becoming a consumable product.
[1] Reminds me of the quote: May a bright spark grow into a flaming idiot.
First, as some people have said, you Rent standard equipment, not buy. But if you are interested in some equipment you own, then I would suggest looking at contracting a hobbyist/small time electronics tinkerer.
A standard wireless mike from places like DSE (Dick Smith Electronics, equiv to maybe Radio Shack in US) have the problem of drift and that they use the commecial FM band. a) you don't have much room for your signals, especially if you are in a built up area, and b) the signals drift due to not using PLLs.
If you look on the net, you should find some PLL wireless mikes. You won't be able to make them very small, I would estimate you should be able to make them (the transmitters) the size of say a small paperback book.
The receivers are part of the same design, as for recording the audio tracks, that can be done by many different means. If you want post editing possibilities, you'll need either a professional recording device, that can record many tracks onto one media. Or you'll need 15 different recording devices (8 if you only record mono and use a stereo recording device). As for what you use, I can't recommend much more on that, other than try google.
Now, back to the transmitter/receiver. Most commerical units, atleast in NZ (and I believe in AU), use the TV band VHF band, just below standard FM radio. Newer units use the GHz bands, but they are harder to build (less tolerances at higher bands).
However, come to think of it, you can pick up some video/audio GHz transmitters/receivers pretty cheap (I'm not talking about using off-the-shelf, they are still expensive afaik). I'm sure you'll be able to find a company that sells GHz audio transmitters for a cheap price.
You mentioned run time. 2 hours should be reasonable with standard battery packs.
Your biggest problem, if you get this stuff made, is size. You will be hard pushed to match the commercial gear for transmitter size.
Just as one more possibility, audio bugs for spying/PI work could be used. Talk to a PI company if they could rent/sell some reasonable units. The gear made for this stuff is expensive, so look at gear made for other stuff thats similar.:)
My mum had a really expensive watch that did that. It was rather ammusing, but you should have seen the looks she got when she went to the jewelers to get it fixed.
"it runs backwards.. uh huh.. are you sure its not upside down?"
Old answer I'm affraid
on
LCD Price Fixing?
·
· Score: 4, Interesting
A lot of it has to do with the old story of supply and demand. There is a very large call for LCDs for laptops, and the laptop manufacturers get them at almost cost, then intergrate them into the laptops.
However, there isn't much (comparitivly) demand for LCD computer screens, or even worse, TV screens.
When I was in singapore a few years ago, RCA input LCD screens weren't that bad a price, but the problem is that price hasn't drop that much.
It does take some more work to make a LCD screen take VGA or RCA inputs, so there is the cost the LCD is brought at (a lot more than the laptop manufacturers buy them at), and then the intergration of circuits to accept VGA or RCA input.
Cheaters not a major problem for Joe User
on
Deathmatch for Dollars?
·
· Score: 4, Interesting
The PunkBuster (PB) system is pretty good, they've put some serious thought into the design of the system, and provided they keep updating it, it should prove sufficient to stop Joe User from cheating.
However, cheating is still possible, but it will come from dedicated cheaters, rather like it does in casinos. Casinos have an advantage though, first they are very rich, and can afford all sorts of checks and balances, and second, players/cheaters are physically present, and therefore can be ID'd. In the online world, noone knows you're a dog, so it will be harder to stop a dedicated cheater from coming back. If the dedicated cheater does not reveal his/her method, then it will be up to the PB team to try and stop it. The PB team have an advantage when the cheat is released in the wild, but not when its kept secret.
PB has the great advantage of being able to update everyones code at any point. So if someone does do a major number on the current PB system, they can simply change it and bingo, all legit users are now running the new code.
The big problem is punishment for cheating. How do you stop a cheater? Kick him out? So he comes back with a new account and continues. You need a way to identify a user. The problem is any ID system will be open to comprimise aswell. The big casinos come down to using people to recognise other peoples faces, and you can't do that online.
If this becomes a very big venture, with real (i'm talking $10000+wins), then the big boys will come along and try to cheat, you'll find they will succeed sometimes.
The difference between this and normal casinos is you aren't playing against the house, so it will be difficult to play and win $10000, unless you find some pretty stupid people. However, lesser sums of $100 or so will be possible. I don't think that these stakes are high enough for the big boys to play for.
Where real money will come from will be when they do have a "house". Playing against computers (the house). They'll have to do it so the house wins >50% like they do at casinos, but it could become very interesting.
In all I think this will become interesting, but not interesting enough to the big boys, so go ahead, play for a few dollars, if you don't make it worthwhile a user risk cheating against you, then they won't.
The biggest issue with this idea is cross-platform. So far a few suggestions have been raised, and I like the idea of a samba frontend, though it seems a little extreme.
BestCrypt is the only cross-platfrom encrypted drive/volume software I know of, its only free for Linux though.:(
Scramdisk/
e4m are options.
Though Scramdisk doesn't run on w2k or XP,
nor Linux. E4M doesn't run on linux either.
The source for Scramdisk and E4M is available,
but I've forgotten what the license is.
I *think* its GPL, but don't count on it.
DriveCrypt
is made by the same people as ScramDisk,
but DC is closed source. Though they are promising
a Linux release (as well as the current XP/2K/etc
clients).
Developing On-The-Fly encrypted drives for
linux isn't all that hard, afterall, its been
done before a few times. Doing so for Windows 95
though to XP is a lot harder.
As for the Mac side, I have no idea. I think
the most portable option would be the Samba
idea mentioned before. It shows the most promise,
you are esentially piggybacking off a known and
support product.
Slashdot Mirror
As much as I agree with you, I have already had long and detailed debates about these subjects. With your proposal, there are a few "problems".
1. How do you regulate it? I receive 100 messages a day that are logs of virus infections on computers I monitor, each of these computers send it directly to my server over the net.
2. Legimate mailing lists, such as the Debian projects mailing list, the kernel mailing list, etc. How do they pay? If you suggest that mailing lists are exempt, then of course, spam lists are just mailing list.
The next line of thought goes to, a fee for any unsolicitated spam sent, but there are many problems with that. How do you find the actual spammer? Maybe fine the person who paid for the spamming (the seller), but if they are located outside your country, that can make it hard. They can also easily claim they never asked any spammer to send this, and the burden is on your to prove they did. So you are back to finding who the spammer was.
Essentially, in any free system, there will *always* be abuse. Its the way it is, and it will probably always be that way. Remember, there are no country lines on the internet. And US law doesn't apply anywhere but the US.
4. Then we Brought SCO... Oh wait, no, we did that first... why was it we did that?
But what do their employees think?
And by that I mean the coders (if they have any left)?
Sure, the sales people are just doing their job, but what about the coders at the company? Surely they can't believe the drivel thats coming from above?
Is there actually anyone left in that company that has more than 2 ethical brain cells?
Remember, this company was once Caldera, who produced a linux distro, so is there anyone left from those days?
Did you get Linux booting off the array?
Agreed, however, when you're sick of forking over large chunks of your paycheck for said engine, you can return home and sit up front in the cockpit of a 747-400 and fly from New York to London. All for arround the same price as a couple of hours of real flying.
I use to fly, but got too expensive, or rather, my other hobbies got too expensive. So now, I fly from the comfort of my home PC.
This is where, in some cities, these Fligh Sim places can really, er, fly. (excuse the pun).
I'm more than willing to fork over a few dollars an hour to sit in a "real" flighsim and have some fun there... only wish I could, none in my City :(
Sheesh!
Cheaper than real flying?
Its betweeen $60-$100/hr to rent a cessna to fly. And thats just a piddly single engine job.
Flight Sims have their definate place. There a a HUGE number of reasons why you would spend this much on a flight sim. If I were to bet, I would say that some of these "mega" flighsim buffs either are private pilots, or have been. And I bet a few were airline pilots.
Oh yeah, and you can't just drive to your local airport and RENT a 747 to have a few fun flights in. They don't really do that.
If you believe that its much cheaper to get a pilots license, then you haven't tried. If you want to go the cheap path, join the airforce, they PAY you to learn to fly.
But a funny story none-the-less...
Back during the great auckland power crises of 97, my ISP was Binary Brothers, a now extinct ISP. They were a great ISP, run by a few guys who knew their stuff.
Turns out the owners were physically located on the coremandle peninsula (about 4 hrs drive from Auckland), while their servers/modem racks, etc were located in the heart of auckland CBD.
The power in auckland blinked out, and as did my net connection (I was located outside the blackout area). I rang 'em up and asked if they were out due the the power crises. They replied that they were currently in the wagon driving from the coremandle with a generator in the back, and the ISP will be back online in a few hours.
Well, they were right, it flicked back online in a few hours. Gotta hand it to them, they really did a good job. I just have this mental picture of these two guys in a wagon, speeding down the windy roads, one yelling, drive faster, we need more power!
Anyway, that was those days, these days I'm lucky if my home connection stayed up during rain storm, let alone a power blackout.
Having just recently configured another Xerox Document Centre on our network, I must say, they are a breeze to configure and print to.
They are a true network printer, they talk LPR, SMB, and, not that I've tested, apple-talk and netware.
I use to have trouble finding drivers, and configurating them, but then I discovered a few settings here and there that made them work right (to do with the hi-cap feeder, fax, and duplex).
The document centres come with various options, but we've got two (a 250 and a 400 model) that have the fax, duplex and high cap feeder options.
Oh, and they talk postscript. *REAL* postscript.
They print, scan, and fax, and I've heard some of the xerox boys at the R&D labs hooked up a coffee maker, so it can even do that *GRIN*.
The only thing I'm not sure about is the network scanner side. There is a windows client that interfaces to it, but I've never tried with linux.
What was said about Auckland, New Zealand in the article was a pretty fair assesment.
Being as I live in Auckland, I was there for the power crisis. Yes, businesses folded, yes, most CBD businesses lost money, but those that folded were most likely going to fold anyhow, and money can be made back.
It hurt the people the most though. Some were fortunate (like the BNZ bank staff), and their companies moved the staff arround to keep them working, but a lot of people couldn't work during those 5 weeks. And 1 month without a pay cheque hurts.
The same obviously applies to the States at the moment. Bussinesses aren't the one to be worring about, its the people.
The power company stiffed Auckland though. Mecury energy is still a force here, they do still run a lot of the power. But then again, lightning never strikes the same place twice... does it?
For the most part, the power grid in most countries performs amazingly well. You try designing a system that can handle an average 20 or 30 lightning strikes a day and still keep on pumping.
I tried that, well, tried scopy, didn't have the 2k resource kit. Kept getting, of all things, permission denied errors when trying to copy the file. (I permission to read the source, and permissiong to write to the destination). If I manually copied (eg using xcopy) the src to the dest, and manually applied the acl, it worked..
Must have been some strange bug/missing feature.
Samba 3.0 is the first real samba (excluding samba-tng), imho, that can replace a WinNT4 PDC (Primary Domain Controller) *fully*.
(eg: with samba3, the windows usrmgr.exe works for adding/deleting users & groups. (usrmgr.exe communicates over RPC, so I consider it something that should work for a windows primary domain controller). I have just recently setup for a company:
A samba PDC, with usrmgr.exe working.
With an LDAP backend for authenciation.
With posix ACLs on the file system (to allow *real* permission settings. The perms are still a bit wierd, and I feel better setting them in Linux rather than through the windows gui, but they do work).
With cups printer backend, so printing works great.
Basically, this machine fully replaces their windows NT4 server, and does it pretty damn well.
The move from NT4 to PDC was pretty good. Once everything is setup on the samba side, you can "net vampire" all of the user and group accounts over to the samba server, and the users can login with no problems.
The only missing feature was I needed some way to copy the file system on the NT box to the linux box and keep the ACLs.
Anyway, the samba team does a great job
Oh how I agree with that statement.
Recently my motherboard died of bad caps, so decided to splash out on an nvidia motherboard. Damn thing had crap all linux support. I don't know if it was the m/b manufacturer (MSI), or nvidia chipsets in general, but:
USB didn't work (even under windows w/ or w/o nvidia special drivers), actually, the worked, but very rarely.
Onboard Sound didn't work under linux (w/ nvidia special binary only drivers)
Onboard network card, (again needed nvidia binary only drivers), didn't work.
And the damn thing crashed often.
I didn't know about the mem=nopentium trick, and would have tried it, but I sent the board back to my supplier with, "tried it, won't work, won't buy, won't recommend". They didn't put up much protest.
I stick with the via chipsets. Sure they aren't supported by via for linux (atleast, not in any real way, they do have drivers, but they are binary only, and don't work). The opensource drivers work damn well. Infact, I suspect, better than the manufacturers binary only (provided you can get them to work).
Then again, thats just for my home pc, which I cheap out on. For any server, I just stick with IBM parts. damn things just plain well WORK!
Of course there is, this is linux. Setup a named pipe, call it /dev/dsp, and realplayer should happly write to it.
Ok, so maybe you may run into some small blocks, but generally speaking, it can be done. Especially in Linux. Thats why the dvd/etc industry don't like Linux. Its too easy to simply write an audio/video driver that writes whats being played/displayed to a file rather than to the speaker/screen.
Then again, if you wanted to get real low-tech, you could hook the audio out of one sound card to the audio in of another (or if your card is decent enough, back into the same card). Then convert that to mp3.
As one of the helpers for the installfest, I can say that this is pretty much only going to help our cause. We couldn't ask for better advertising (both the NZ Herald, and Slashdot).
We will be ready, Saturday, with plenty of CDs (we hope).
Van Eck's freaking is still possible with an LCD display. Its to do with the rythmic timing of a PC. Its easy to spot the 70Hz (60, 80, whatever) of your monitor. Your LCD also refreshes. There is the writes to the video memory, etc.
I suppose you could say its more difficult, but compared to actually doing Van Eck freaking in the first place, its only marginally more difficult. If you can freak VDUs, you can freak LCDs.
As for the initial problem of restricting access. If you want to ensure that nobody can be coereced into authenticating, then you'll need more than one person. preferably a commity of 11 people who hate each other. If you put a gun to my head, I'll give you all the passwords i know. So lets say I'm a good employee, and won't give out the passwords/authenticate under the threat of death, what about threatning someone I know, or love.
If humans are responsible for authenticating, then another human will be able to coerce the first one into giving access. This is why good safes and bank vaults are on a time delay. It doesn't matter how many passwords you know, the vault will still take 30 minutes to open from the time you enter the right code. The assumption being that 30 mins is a) enough time for the cops to get there, and b) that the criminals want to avoid a standoff with the cops.
Want to make a totally secure server? turn it off, bury it in concrete, and under a mountain, make sure you are the only one who knows where it is, then shoot yourself. Seriously, you can't make something that will give access only under certain circumstances, there will always be ways around it. This is why security is made up many levels.
Ok, so lets assume that you just want a system that only YOU can access, and only if you choose (and assuming you can make a valid choice with a gun to your head). Well, first we'll encrypt everything with some pretty good encryptiong. Thats the easy part done. Now you have a key. How to decrypt the data, with the key, without exposing the data, or the key, to any other party (call them "Eve") who are listening in.
You can't.
Take a look behind you...
At that wall over there.
Yes, that one.
See that black spot?
No, not that one, the other one...
Yes... how do you know its not a camera, recording everything you see and do?
Ok, so lets put the screen inside googles that you wear.. whoops, how do we know that noone has tampered with the googles, or that they are Van Eck'ing the googles.
Its impossible. If you come up with some solution that is very very very good, then come talk to the CIA, they'll have a job for you. If you can come up with a way around it, you'll be able to get two pay checks, the other from the NSA.
In the end its about risk-management.
What will it cost me, or you, if whatever method I use, fails?
How much money does my adversary have?
How much money does the solution cost?
Now balance those.
If the risk is you goto jail. And you figure you can buy your way out of jail for say $20millon, and the protection will protect you for $10 million, but it'll cost $30 million to by pass, and your adversary only have $5, then you should use the solution. But if you adversary can bypass it, then you may aswell just save $20million for the times you get caught, and stick with something simple.
It looks just like one of those home gyms. And then when the marketing speak said:
> The MasterPeace Rocks and Cradles you
I thought... "You can rock, rock, rock! yourself to firmer abs!"... damn those infomercials.
The biggest problem with those is there is no place to put good old fashion pen to paper.
Even when I was in R&D I still liked to use pen and paper beside my superfast machine. I like a big desk with lots of room to scribble.
Interesting side note, since we were in R&D, we weren't allowed any "scrap" paper to work on. All our work had to be in our logbooks, that had all our notes (aparently incase of patent dispute). So I started doing my doodles in that. Then I put things like, "if you are reading this, then I must be dead, the treasure is buried....". Of course, after our head of project died, i stopped doing that.
There was a proposal for dynamic lengths of IP addresses. When I thought about it, it sounded pretty neat too, however it becomes pretty unmanagable once you get too far into it.
As for the Y2k thing. Yeah, I really think we will suddenly all say, "This calender sucks, lets move to a different one". The last country to adpot the "new" calander finally switched about 20 years ago. And the calander was switched arround 200 years ago.
The reason for using fixed length is efficency and ease. Its much easier as a programmer to know that hostnames won't be larger than 64 characters (true!). That way I can put a static buffer 64 chars big, and then if I'm a good programmer, check that the data I'm copying into the buffer isn't bigger than 64 chars.
If you have a dynamic address length, everything changes. try writting a program that deals with arbitary strings of any length.
sprintf and friends are usually implemeneted with a 1024 character buffer, so you can't use those easily.
If I'm writting a program that takes lines of text of arbitary length, I usually assume that the lines are no longer than 1024 characters. Otherwise I have to start reallocing buffers, or rewinding the stream.
If you have a dynamic size IP address, then there is a limit to how long it is. If the field that specifies the length is 1 byte, then the IP address can be 255 octets long. Now, since a 255 octect address can be represented in hexidecimal by 510 characters (assuming you don't have seperators), do you really think it would be practical to have an address that long?
Yes, they would have to prove it if they wished to take you to court, and they would only take you to court if they levied charges against you for your NATing. (However, in New Zealand, if you use Xtra (which is owned by Telecom), and you don't pay your bill, they just cut your phone line).
ISPs can drop your service for any reason they want. The NAT analysis is for their benifit, if they can detect users with NAT, send 'em a warning, if they continue, maybe investigate further, or just cancel the account.
I've heard it as; The longer the title, the less important the job.
Case in point:
President of the United States of America.
Some bright spark[1] decided that once a person buys a printer, they are commited to it, so will have to buy the print cartridges for it. So if we make the cartridges expensive, we can still maintain our profit margins, and have continous profits rather than once off for each customer.
Now enter the business side of things. Our business customers don't want to keep buying the latest bubblejet/inkjet/crapjet every 3 months, so they produce a seperate business line of machines. Mostly these are laser based, however, there are some top-of-the-line inkjet systems that are mostly used in the printing industry (eg signs/cars/etc).
So you either buy a business quality printer, preferably laser based, and you pay good money for it. Or you do what some of my customers do:
They buy a new printer when the old print cartridge runs out. However, they are being thwarted by the print manufactores who are now selling print cartridges half full on new printers, so they buy a new cartridge with the printer (usually at a discount, since they can wrangle one with the printer), and run it till it runs dry, and pick up the next latest and greatest model.
Ok, so thats a bit extreme, but I do have one customer doing that.
Basically, printers are becoming a consumable product.
[1] Reminds me of the quote: May a bright spark grow into a flaming idiot.
A standard wireless mike from places like DSE (Dick Smith Electronics, equiv to maybe Radio Shack in US) have the problem of drift and that they use the commecial FM band. a) you don't have much room for your signals, especially if you are in a built up area, and b) the signals drift due to not using PLLs.
If you look on the net, you should find some PLL wireless mikes. You won't be able to make them very small, I would estimate you should be able to make them (the transmitters) the size of say a small paperback book.
The receivers are part of the same design, as for recording the audio tracks, that can be done by many different means. If you want post editing possibilities, you'll need either a professional recording device, that can record many tracks onto one media. Or you'll need 15 different recording devices (8 if you only record mono and use a stereo recording device). As for what you use, I can't recommend much more on that, other than try google.
Now, back to the transmitter/receiver. Most commerical units, atleast in NZ (and I believe in AU), use the TV band VHF band, just below standard FM radio. Newer units use the GHz bands, but they are harder to build (less tolerances at higher bands).
However, come to think of it, you can pick up some video/audio GHz transmitters/receivers pretty cheap (I'm not talking about using off-the-shelf, they are still expensive afaik). I'm sure you'll be able to find a company that sells GHz audio transmitters for a cheap price.
You mentioned run time. 2 hours should be reasonable with standard battery packs.
Your biggest problem, if you get this stuff made, is size. You will be hard pushed to match the commercial gear for transmitter size.
Just as one more possibility, audio bugs for spying/PI work could be used. Talk to a PI company if they could rent/sell some reasonable units. The gear made for this stuff is expensive, so look at gear made for other stuff thats similar. :)
It was rather ammusing, but you should have seen the looks she got when she went to the jewelers to get it fixed.
"it runs backwards.. uh huh.. are you sure its not upside down?"
A lot of it has to do with the old story of supply and demand. There is a very large call for LCDs for laptops, and the laptop manufacturers get them at almost cost, then intergrate them into the laptops.
However, there isn't much (comparitivly) demand for LCD computer screens, or even worse, TV screens.
When I was in singapore a few years ago, RCA input LCD screens weren't that bad a price, but the problem is that price hasn't drop that much.
It does take some more work to make a LCD screen take VGA or RCA inputs, so there is the cost the LCD is brought at (a lot more than the laptop manufacturers buy them at), and then the intergration of circuits to accept VGA or RCA input.
However, cheating is still possible, but it will come from dedicated cheaters, rather like it does in casinos. Casinos have an advantage though, first they are very rich, and can afford all sorts of checks and balances, and second, players/cheaters are physically present, and therefore can be ID'd. In the online world, noone knows you're a dog, so it will be harder to stop a dedicated cheater from coming back. If the dedicated cheater does not reveal his/her method, then it will be up to the PB team to try and stop it. The PB team have an advantage when the cheat is released in the wild, but not when its kept secret.
PB has the great advantage of being able to update everyones code at any point. So if someone does do a major number on the current PB system, they can simply change it and bingo, all legit users are now running the new code.
The big problem is punishment for cheating. How do you stop a cheater? Kick him out? So he comes back with a new account and continues. You need a way to identify a user. The problem is any ID system will be open to comprimise aswell. The big casinos come down to using people to recognise other peoples faces, and you can't do that online.
If this becomes a very big venture, with real (i'm talking $10000+wins), then the big boys will come along and try to cheat, you'll find they will succeed sometimes.
The difference between this and normal casinos is you aren't playing against the house, so it will be difficult to play and win $10000, unless you find some pretty stupid people. However, lesser sums of $100 or so will be possible. I don't think that these stakes are high enough for the big boys to play for.
Where real money will come from will be when they do have a "house". Playing against computers (the house). They'll have to do it so the house wins >50% like they do at casinos, but it could become very interesting.
In all I think this will become interesting, but not interesting enough to the big boys, so go ahead, play for a few dollars, if you don't make it worthwhile a user risk cheating against you, then they won't.
The biggest issue with this idea is cross-platform. So far a few suggestions have been raised, and I like the idea of a samba frontend, though it seems a little extreme.
BestCrypt is the only cross-platfrom encrypted drive/volume software I know of, its only free for Linux though. :(
Scramdisk/ e4m are options. Though Scramdisk doesn't run on w2k or XP, nor Linux. E4M doesn't run on linux either. The source for Scramdisk and E4M is available, but I've forgotten what the license is. I *think* its GPL, but don't count on it.
DriveCrypt is made by the same people as ScramDisk, but DC is closed source. Though they are promising a Linux release (as well as the current XP/2K/etc clients).
You may also like to try The Linux crypto mailing list to search for answers there.
Developing On-The-Fly encrypted drives for linux isn't all that hard, afterall, its been done before a few times. Doing so for Windows 95 though to XP is a lot harder.
As for the Mac side, I have no idea. I think the most portable option would be the Samba idea mentioned before. It shows the most promise, you are esentially piggybacking off a known and support product.