For the benefit of idiots who don't know what an.RTF file is, I suppose. Most people are just downright stupid. Although, if you name an.RTF with a.DOC extension, Word will open it anyway. Anyone checked to see if the.RTF and.DOC files are really the same?
But with Blair's "frequent meetings" with Sir
Willam Gates what else can be expected? his knighthood was recommended by Gordon Brown for "services to Industry"
Well, it has to be said, Bill Gates has done many great services to certain industries. If it wasn't for the consequences of having everyone default to almost-root privileges for far too long, there would be no need for the anti-virus and spyware removal industries! If it wasn't for thousands of compromised boxes sitting on ADSL connections ready to spew out mail messages, there would be no dodgy prescription drugs and counterfeit Rolex watches industry! If it wasn't for servers falling over without reason or warning, there would be no need for the "training chimpanzees to push a reset button" industry!
Ever heard of a circular argument, or a self-fulfilling prophecy?
Any line of questioning that begins with "Why don't men wear skirts?" is going to come around to an answer which can best be paraphrased as "Because men don't wear skirts". Similarly any line of questioning that starts "Why are recreational drugs illegal?" will be answered with "Because recreational drugs are illegal" although probably not in those exact words.
It's the same with Linux. Ask "Why do so few people use Linux?" and the answer will boil down to "Because hardly anybody uses Linux".
Let me state this publically now. I will never carry a general-purpose identity card of the type the British Government is proposing, and there is no penalty that would deter me from the crime of going anonymous.
Aside from the fact that it would be, from a security perspective, triple distilled extra virgin snake oil, there is the fact that the Government is supposed to work for me, not the other way around.
If I was forced to carry an identity card, then I would not really be any more free than if I was in prison. So I'm absolutely prepared to risk life imprisonment.
Absolutely. A pound is a pound, or {since we're talking about Australia here} a dollar is a dollar, whoever's money it is; that is the single most important principle on which the entire freaking concept of money as a measure of value rests. Charging different people different prices for exactly the same thing should IMHO be very illegal. I'm surprised it isn't considered at least a form of VAT evasion, if not something similar to cash forgery.
Yes, but if you read the MS EULA, it has a few things to say about Microsoft software:
If it doesn't do what you were expecting, TOUGH TITTY.
If it crashes, TOUGH TITTY.
If it crashes and takes a day's work with it, TOUGH TITTY.
If it has a gaping security hole, and some 5(r!p7 |<!|)|)!3 manages to wipe several terabytes of your entire customer database with a single malformed web query, TOUGH TITTY.
If you use it to design a building, and the building falls down, TOUGH TITTY.
If you use it to control a life support system, and it goes BSOD and kills the patient, TOUGH TITTY.
Spot the pattern, anyone?
At least with Open Source software, you can conduct an independent audit on the source code in order to determine its suitability for a particular application; such an audit can be performed by local programmers, thereby helping the local economy by creating jobs and raising tax revenue. Whereas with Microsoft, you only have the word of a convicted criminal for it that the software will do what you were expecting. And you are taking money out of the local economy to boot.
I just hope the Aussies have the balls to knock back whatever pathetic offer Microsoft are going to come up with to try to keep their bitches working for them. From the over-simplification in the article, it sounds like a no-brain job: just replace the Windows desktops {glorified dumb terminals} with Linux desktops {SSH and Telnet clients already built in}, the Novell file/print servers with Linux file/print servers, and {the hard bit} migrate from Oracle and SAP on Solaris to Postgres on Linux. Total licencing cost: upfront AUS$nil, thereafter annually AUS$nil, this quote valid until forever. The feasibility study and source audit are an upfront cost; but this could be offset by borrowing a sum of money and repaying it over an appropriate timescale using some of the money which would have been spent on Payware.
No you can't. SCO has no authority to distribute the Linux kernel. See, the copyright on the Linux kernel belongs to a whole bunch of people; and if any one of them objects to SCO distributing it, and if SCO's distribution wouldn't be covered by "fair dealing" / "fair use" provisions {which may vary from jurisdiction to jurisdiction}, then SCO are violating copyright law by distributing the Linux kernel. By imposing a "fee" of $699, they are also committing fraud, and quite likely demanding money with menaces.
SCO's only defence against the copyright violation charge is the GPL; but since they have broken the conditions attached to the permissions it grants, then those permissions are withdrawn. However, anyone who has received a copy of the Linux kernel from an unlicenced distributor is licenced to distribute it {unless and until they break the rules}, since it is the copyright holder, not the distributor, who grants the GPL permissions.
The GPL has already been shown to be valid in a Court of Law {albeit a minor one in Germany}; and if the GPL is not legally enforcible then no EULA will ever be legally enforcible either, since most EULAs ask you not to do things the Law of the Land gives you an explicit right to do. {Analogy: If a woman signs a piece of paper stating that she consents to have sex with a man at a certain time on a certain date, then changes her mind, and he has sex with her anyway, it's still rape; because you can't waive your right to say "no" to sex.}
Copyright is one of the greatest misnomers going. It's actually a temporary privilege granted on behalf of Society At Large, in exchange for the promise that a work will eventually enter the Public Domain.
Because sooner or later they would find a way around it. Haven't you looked at your e-mail lately? Why do you suppose spammers are sending strings of nonsense words, and referring to dog-worming tablets {sold under the pretence that they may contain some of the drug Sildenafil Citrate} as "\/i4gr/\" rather than the more-obvious "Viagra" ? Answer: On purpose: to get past filters that were designed to keep them out!
It'd take about 15 minutes to set up an anti-GNAA filter -- and, coincidentally, that's about how long it would work before someone found a way to bypass it. It then becomes a pointless game of cat and mouse, just like e-mail Spam filtering. And meanwhile, the content will become more and more offensive.
The best thing in the long run is to simply ignore it. GNAA and others are just sad, pathetic individuals with no life of their own, so the only way they can get their kicks is by spoiling things for other people. If you respond to their postings, then you will only encourage them. The trolls may claim to hate Slashdot, but we know this is false. Slashcode is Open Source, so they could always set up their own message board -- something they have consistently failed to do. There are also plenty of closed-source message board systems, for those who simply need to feel they are stealing something rather than accepting a gift offered with grace. They have consistently refused this offer because the truth is they need Slashdot, when they are down in their parents' basements with their tweezers and magnifying glasses.
{I once wrote a message board of my own, and I came up with a really good way to keep both trolls and normal people happy. I'm not going to describe it here, though, because that might reduce its effectiveness.}
Once upon a time, a group of cave-dwellers were sitting around a fire, cooking the day's catch. The men were fixing their weapons for the following day's hunt, sharpening their stone axes and spear heads in the flickering firelight; some of the women nursed infants. Already the group's numbers were growing. Thanks to two related discoveries -- that if you hit two of the right kind of rocks together, not only can you shape one of them, but you can also produce a lethal sharp edge; and that dried grass will catch on fire if you hit two of the right kinds of rocks together near enough to it -- they were all better fed than they had been just a few years ago. Of these children, playing, feeding and sleeping in the glow of the flames, more than twice as many would survive as compared to before these discoveries of simple stone weapons and firelighting techniques.
This was the moment when society had become absolutely dependent on technology.
I have also seen it spelt "riduculous" -- by a New Zealander, even, so it was kind of appropriate! This was in response to a complaint of racism made against another employee at my last place of work: he had mis-spelt "Australian" as "Austrilian" and some sad-act -- not even an Aussie, an Aussie would most probably have laughed -- interpreted it as a racial slur {it could be construed as making fun of the Australian accent}.
As for "teh", that seems to be a keyboard rollover thing. When you type fast, you actually strike the next key before releasing the previous one, and the keyboard controller has to work out what you meant. It sounds trivial; but remember, each key "bounces" for about 10ms. on release. Ten whole precious milliseconds, getting dangerously comparable to the duration of a fast typist's keystrokes, and you can't even use a simple R-C circuit for hardware debouncing because of the way the matrix scanning works. So the keyboard controller is having to compensate for key bounce in software, too, as well as rollover. Occasionally it messes up.
I did once write a piece of software that, if it saw the letters "t", "e", "h" typed within a certain timeout of each other, changed it to "the"; but if you typed it slowly and deliberately, one letter at a time, it assumed you really meant "teh" and left it alone. Unfortunately, I thought at the time that it was no more than an amusing novelty. Maybe I should have taken it further?!
I guess other languages must have similar rollover-related problems leading to common typos of their own. Anyone care to comment?
What crack is Ballmer smoking? He is saying computers are obviously too expensive, hence people are paring money off the total cost by "pirating" Windows. Hmm..... That sounds like someone who owns a pay toilet complaining that beer is too expensive. Except Ballmer is keeping a £1-a-leak pay toilet. In the woods. With no roof.
IMHO if anyone is encouraging "piracy" of Windows, it is those manufacturers of cheap and nasty peripherals who provide only Windows drivers and say "f**k you" to Linux / BSD users. It's no skin off their nose if someone has to "pirate" a copy of Windows just to make their brand new, 10mm thick, 96-bit, 76800dpi scanner work. They haven't even got the excuse that "perhaps our customers are running a less expensive but nonetheless legal operating system" if there are no drivers supplied for such systems.
Maybe if purchasers of Windows software -- and that includes Windows drivers for hardware -- were obliged to show proof they are duly licenced to run Windows {and analogously for other OSes; except that very few people indeed are not licenced to run Linux, and nobody at all is not licenced to run BSD} then things would change. In practice, what would happen is that the OSS drivers would be on the same CD as the Windows drivers and thus not physically separable. Just the fact of having Linux and BSD drivers would get the hardware manufacturer off the hook -- it's a far more reasonable supposition that someone is going to use a cheap peripheral under a Free OS if a driver is already provided for that OS than otherwise.
The only thing you can do about stupid people is hope that sooner or later, they will make a mistake that either (a) teaches them not to be so stupid or (b) eliminates them from the respiration-photosynthesis cycle.
In the old days, when you had power tools with big blades and no safety guards, nobody ever made the same mistake twice. Computers have made things easier and safer for idiots, and idiots have made things harder and more dangerous for the rest of us.
To measure the power of a gas boiler, knowing the flow rate V litres per minute and the inlet and outlet temperatures, just multiply the temperature difference by the flow rate by 0.07.
Q = m * c * theta..... (1)
In one minute (=60"), we have V litres of water flowing, or about V kg. So dm/dt = V/60 kg s-1. c is about 4168 J kg-1 K-1. But it'll be easier to handle power in kilowatts, so we shall divide everything by 1000.
This is ripped off from a web application I once wrote. You should be able to modify the time by typing figures into the box or by using up and down arrows. What happens in practice is that adding 1 actually concatenates "1" on the end of the string, so you find that 1 + 1 = 11, and 11 + 1 = 111.
This is what you get when you borrow one idea from Perl about how the computer should be able to work out from context whether or not something is a number or a string; and one from BASIC about re-using operators obviously out-of-context {strings cannot be added} to mean something different {such as concatenation}. The result, as they say, is a mess. For all practical purposes, JavaScript lets you subtract, multiply and divide numbers; but if you want to add, you'd better subtract a negative number.
I mean, it's not freaking rocket science, is it? = is for telling, == is for asking. + is for adding, . is for joining strings. Sheesh!
The Linux driver API *is* stable -- provided that you have the source code available and can recompile anytime. IMHO that assumption is totally reasonable. No matter how many layers of abstraction you provide, something somewhere has to ride the metal.
The problem is that manufacturers are working on the {false} assumption that a compiled binary driver is secure against competitors working out what it does. Decompilation is mathematically similar to shape recognition, and it's a matter of time before a usable decompiler is released under an open source licence. What's going to be worse for some people is that they have released binaries under a BSD-like licence -- which effectively will give anyone who acquires the source code by force, permission to distribute it.
I'd most definitely buy a fully open-specced graphics card, even if it cost a little extra. Better that than a tainted kernel. Although to my mind the question should not be "should someone make an open spec card>", but "shouldn't everyone be obliged to make their hardware open spec?"
If I am the rightful owner of a graphics card {say}, then Common Law says that nothing concerning that graphics card is a secret from me -- and nobody can stop me using whatever techniques are at my disposal for discovering what I have a right to know {which is, basically, every true fact in the universe}. They can bind me to keep secret what I discover; but, as long as I own the card, it is not secret from me. Nor can it possibly be secret from anyone else who owns an identical card.
Proper enforcement of this law is what we need. {It never really mattered up until now, because nobody ever envisaged technology getting as complicated as it has become today; you could figure out how something like a mechanical clock works just by looking at it, hardly needing any instrument more sophisticated than a magnifying glass.} Anyone who owns an nVidia or ATI card already has the right to know how to program it -- they shouldn't have to fight for that.
And spare me the bitching about revealing things to competitors. Are nVidia really so naïve as to think ATI don't reverse-engineer their cards, and vice-versa? Come on, guys.
It still only invalidates the result from one polling station, and that might not even be enough to alter the eventual outcome for the seat {depending upon whether or not you were using transferrable votes, and how close the candidates were} -- but, then again, nobody is likely to try this sort of vote-spoiling tactic in a "safe" seat. It's not such a nightmare logistically. Again, this is a consequence of having many polling stations, each serving a few thousand voters.
Anyway, there are ways an electronic or mechanical voting machine could be tricked into accepting multiple votes from the same person. The problem is with the end, not the means.
In the UK, you cast your vote by making a single, unambiguous mark -- conventionally an X since the intersection of the two lines indicates a precise point on the page -- in a box next to one candidate's name from the pre-printed list. {The lists are randomised; two adjacent papers will have the names in different orders}. It's up to the person counting the votes to use some common sense, obviously; but dubious ones are placed in a separate pile and looked at by everyone at the very end.
In most of the rest of Europe {and in private elections in the UK, e.g. student unions, trade unions &c.}, Single Transferrable Votes are used. You cast your vote by writing a number in a box adjacent to each candidate's name on the pre-printed list. The lists are randomised, but "Re-Open Nominations" {or its local equivalent} is always last. Same procedure applies with "dodgy" papers. Numbers are pretty hard to confuse.
If you change your mind, you have to get another ballot paper from the presiding officer.
I work for an ISP. And knowing what I know, there is no way I would let the fact of who I voted for get out on the Internet.
In a paper ballot, you give every voter an identical ballot paper, and they give it back to you with their vote on it. To emulate this electronically, you would need to issue a token to each voter which is revoked as the vote is cast. But that isn't quite perfect. Paper ballots can't be copied {well, they can; but you aren't allowed to take the ballot paper out of the polling station, nor to bring the necessary wherewithal to make a copy of the paper into the polling station, and let's assume the presiding officer is running a tight ship here}, so you know just from whether or not the paper is in the ballot box, whether or not the token has been revoked; and you know from whether or not the voter's name has been crossed off the list, whether or not the token has been issued in the first place. To be able to know which electronic tokens are valid and which have been revoked, every one would need to be different - and thus identifiable. Furthermore, you have absolutely no idea what is going on in transit between the server and the client and back. If you aren't actually sitting there in a polling station, then you can't supervise voters to make sure they are not misbehaving. They could be "printing their own ballot papers" {generating valid tokens for themselves}, or "stealing other people's ballot papers" {misappropriating valid tokens issued to other people}. How do you know, given the proliferation of Windows PCs, that a client machine has not been infected with a Trojan horse that hijacks its internet connection? You could maybe give out bootable CD-ROMs with a secure OS and client software on them, but then how do you ensure that your special protocol doesn't get reverse-engineered and misapplied between issuing the CD and the vote being cast?
The problem is simply that every time you fix one of the existing failure modes, you introduce some more new ones that you hadn't thought of before. Some of the limitations are not limitations of technology, which can be overcome by invention, but limitations of the universe, which can only be overcome by breaching one or more fundamental laws. At least the failure modes of paper ballots are well understood, and mitigated to the fullest extent possible.
So it's just like an "old-fashioned" modem that you used to plug into a serial port, and with an entry in/dev/ that acts like a ttyS* port? Does that mean I can dial my spare phone line at normal mobile-to-landline rate {which on my tariff is nil in the evening and at weekends for the first 300' a month}, have mgetty auto-answer and gateway myself through my broadband connection? Or am I stuck with using my own telco's dial-up access, or is the call charged extra anyway just because it's a modem call?
But the serial number on the ballot paper is not recorded against your name anywhere. You can always exchange your ballot paper for a different one, which would spoil the records. And nothing is stopping you "swapping votes" with someone you trust -- so you vote for their party, they vote for your party, and everybody's happy.
In the UK, we have hand-counted paper ballots. We have had them since we started having elections. It is a system that everybody can understand, and it's evolved over the years to be surprisingly robust.
Each ballot slip is placed whole into the box. So it's verifiable if necessary, by re-counting. The fact of your voting is recorded, but in such a way as not to be able to link your name to a particular ballot paper. In case the ballot slips are secretly marked or anything like that, you can pick your own if you feel sufficiently paranoid {you aren't forced to accept the one the presiding officer gives you}; so it's secret.
Each polling station takes votes from an area no bigger than the volunteers working there could comfortably count by hand all the votes from. So it's scalable -- if you have more voters, you just add more polling stations. It's also quick -- in each polling station, there are only a few thousand votes to count. All this is going on in parallel, results are initially telephoned through and then the ballot papers are sealed back up in case they need to be re-counted.
The numbers involved mean that to "buy" an election, you would have to pay off a lot of people. So it's actually quite tamper-proof. And if any shenanigans are suspected, a recount can be ordered -- or, in the worst case the ballot repeated -- in just the known affected polling stations.
It is not clear to me how this system could be improved on without introducing new failure modes. Any kind of vote-counting machine is susceptible to tampering. Even if it is absolutely open to public scrutiny for the days when it is not being used for an election, there are stunts that could be pulled on the day. And even if the machine is verified by a hand-count, it will still takes the same number of people to hand-count the ballots after the machine is done, so what have you saved?
If you're going to rely on human honesty, it's best to distribute that reliance as widely as possible, i.e. to trust several thousand people to be just a little bit honest rather than trust a few people or just one person to be very honest indeed. After all, the majority of human beings are generally honest, and more so when the stakes are low. What benefit is there to dishonesty in counting a few thousand votes among tens of millions? On the other hand, if you are the managing director of the company that makes the only officially-approved voting machines, you effectively have every election in your hands -- and that is where the benefits of being dishonest do start to show.
the way it syncs data with a PC seems to be proprietory (it doesnt use the qtopia desktop) and the PC-side software is Windows-only.
Bloody typical.
I am seriously thinking of writing to my MP to try to get it made law that manufacturers must provide full disclosure of driver specs if they want to sell their goods in this country. Placing the driver source code under BSD licence or GPL would of course satisfy the requirements. {I know BSD can be poison; but it isn't actually too bad for situations like drivers, where there is no danger of proprietary extensions taking over from the main code base. The manufacturer can't add proprietary extensions themself, since that would break the requirement for full disclosure.}
Meanwhile, has anyone got any success stories re. getting a Linux laptop on the internet with a mobile phone? Any mobile at all? If I use my own dial-up server, will it just be counted as any old mobile-to-landline call {and therefore covered in the monthly allowance on my current tariff}? I don't need a feature-packed phone with a colour display, camera and integrated self-rolling ashtray, just something that lets me send and receive text messages and answer voice calls. Oh, and go on the internet, obviously:)
It's all very well that the phone itself runs Linux underneath. What's more important from where I'm sitting is, can I use it with my Linux {single-booting and proud of it} notebook? For instance, does it use standard {or at least, well-documented and free for the asking} APIs so I can write my own perl scripts to do cool things with it? Can I use it to get on the Internet? Do Motorola provide the drivers as source.tar.gz files which I'll be able to compile on any system, not just the "commercially viable" ones? Does the phone have a scripting language {I mean, more sophisticated than ash} on-board?
My wishlist would be for something that looked to the host PC as though it was a USB network adapter plugged into some sort of network. You would assign it an IP address in a subnet of your choosing, from the phone's "console". There would be standard servers on standard ports {21 for FTP, 22 for SSH, 80 for web, 3306 for database, &c.; maybe even an Asterisk proxy for hardcore VoIP users} so as to keep the interface clean and simple {I'm implying that they would just be using regular files and databases for storing ringing tones and contact info; there's no reason to assume otherwise}. While you were on the Internet, the phone would do NAT, just like one of those little ADSL gismos.
Slashdot Mirror
For the benefit of idiots who don't know what an .RTF file is, I suppose. Most people are just downright stupid. Although, if you name an .RTF with a .DOC extension, Word will open it anyway. Anyone checked to see if the .RTF and .DOC files are really the same?
Ever heard of a circular argument, or a self-fulfilling prophecy?
Any line of questioning that begins with "Why don't men wear skirts?" is going to come around to an answer which can best be paraphrased as "Because men don't wear skirts". Similarly any line of questioning that starts "Why are recreational drugs illegal?" will be answered with "Because recreational drugs are illegal" although probably not in those exact words.
It's the same with Linux. Ask "Why do so few people use Linux?" and the answer will boil down to "Because hardly anybody uses Linux".
Let me state this publically now. I will never carry a general-purpose identity card of the type the British Government is proposing, and there is no penalty that would deter me from the crime of going anonymous.
Aside from the fact that it would be, from a security perspective, triple distilled extra virgin snake oil, there is the fact that the Government is supposed to work for me, not the other way around.
If I was forced to carry an identity card, then I would not really be any more free than if I was in prison. So I'm absolutely prepared to risk life imprisonment.
Absolutely. A pound is a pound, or {since we're talking about Australia here} a dollar is a dollar, whoever's money it is; that is the single most important principle on which the entire freaking concept of money as a measure of value rests. Charging different people different prices for exactly the same thing should IMHO be very illegal. I'm surprised it isn't considered at least a form of VAT evasion, if not something similar to cash forgery.
- If it doesn't do what you were expecting, TOUGH TITTY.
- If it crashes, TOUGH TITTY.
- If it crashes and takes a day's work with it, TOUGH TITTY.
- If it has a gaping security hole, and some 5(r!p7 |<!|)|)!3 manages to wipe several terabytes of your entire customer database with a single malformed web query, TOUGH TITTY.
- If you use it to design a building, and the building falls down, TOUGH TITTY.
- If you use it to control a life support system, and it goes BSOD and kills the patient, TOUGH TITTY.
Spot the pattern, anyone?At least with Open Source software, you can conduct an independent audit on the source code in order to determine its suitability for a particular application; such an audit can be performed by local programmers, thereby helping the local economy by creating jobs and raising tax revenue. Whereas with Microsoft, you only have the word of a convicted criminal for it that the software will do what you were expecting. And you are taking money out of the local economy to boot.
I just hope the Aussies have the balls to knock back whatever pathetic offer Microsoft are going to come up with to try to keep their bitches working for them. From the over-simplification in the article, it sounds like a no-brain job: just replace the Windows desktops {glorified dumb terminals} with Linux desktops {SSH and Telnet clients already built in}, the Novell file/print servers with Linux file/print servers, and {the hard bit} migrate from Oracle and SAP on Solaris to Postgres on Linux. Total licencing cost: upfront AUS$nil, thereafter annually AUS$nil, this quote valid until forever. The feasibility study and source audit are an upfront cost; but this could be offset by borrowing a sum of money and repaying it over an appropriate timescale using some of the money which would have been spent on Payware.
No you can't. SCO has no authority to distribute the Linux kernel. See, the copyright on the Linux kernel belongs to a whole bunch of people; and if any one of them objects to SCO distributing it, and if SCO's distribution wouldn't be covered by "fair dealing" / "fair use" provisions {which may vary from jurisdiction to jurisdiction}, then SCO are violating copyright law by distributing the Linux kernel. By imposing a "fee" of $699, they are also committing fraud, and quite likely demanding money with menaces.
SCO's only defence against the copyright violation charge is the GPL; but since they have broken the conditions attached to the permissions it grants, then those permissions are withdrawn. However, anyone who has received a copy of the Linux kernel from an unlicenced distributor is licenced to distribute it {unless and until they break the rules}, since it is the copyright holder, not the distributor, who grants the GPL permissions.
The GPL has already been shown to be valid in a Court of Law {albeit a minor one in Germany}; and if the GPL is not legally enforcible then no EULA will ever be legally enforcible either, since most EULAs ask you not to do things the Law of the Land gives you an explicit right to do. {Analogy: If a woman signs a piece of paper stating that she consents to have sex with a man at a certain time on a certain date, then changes her mind, and he has sex with her anyway, it's still rape; because you can't waive your right to say "no" to sex.}
Copyright is one of the greatest misnomers going. It's actually a temporary privilege granted on behalf of Society At Large, in exchange for the promise that a work will eventually enter the Public Domain.
Because sooner or later they would find a way around it. Haven't you looked at your e-mail lately? Why do you suppose spammers are sending strings of nonsense words, and referring to dog-worming tablets {sold under the pretence that they may contain some of the drug Sildenafil Citrate} as "\/i4gr/\" rather than the more-obvious "Viagra" ? Answer: On purpose: to get past filters that were designed to keep them out!
It'd take about 15 minutes to set up an anti-GNAA filter -- and, coincidentally, that's about how long it would work before someone found a way to bypass it. It then becomes a pointless game of cat and mouse, just like e-mail Spam filtering. And meanwhile, the content will become more and more offensive.
The best thing in the long run is to simply ignore it. GNAA and others are just sad, pathetic individuals with no life of their own, so the only way they can get their kicks is by spoiling things for other people. If you respond to their postings, then you will only encourage them. The trolls may claim to hate Slashdot, but we know this is false. Slashcode is Open Source, so they could always set up their own message board -- something they have consistently failed to do. There are also plenty of closed-source message board systems, for those who simply need to feel they are stealing something rather than accepting a gift offered with grace. They have consistently refused this offer because the truth is they need Slashdot, when they are down in their parents' basements with their tweezers and magnifying glasses.
{I once wrote a message board of my own, and I came up with a really good way to keep both trolls and normal people happy. I'm not going to describe it here, though, because that might reduce its effectiveness.}
Indeed.
Once upon a time, a group of cave-dwellers were sitting around a fire, cooking the day's catch. The men were fixing their weapons for the following day's hunt, sharpening their stone axes and spear heads in the flickering firelight; some of the women nursed infants. Already the group's numbers were growing. Thanks to two related discoveries -- that if you hit two of the right kind of rocks together, not only can you shape one of them, but you can also produce a lethal sharp edge; and that dried grass will catch on fire if you hit two of the right kinds of rocks together near enough to it -- they were all better fed than they had been just a few years ago. Of these children, playing, feeding and sleeping in the glow of the flames, more than twice as many would survive as compared to before these discoveries of simple stone weapons and firelighting techniques.
This was the moment when society had become absolutely dependent on technology.
I have also seen it spelt "riduculous" -- by a New Zealander, even, so it was kind of appropriate! This was in response to a complaint of racism made against another employee at my last place of work: he had mis-spelt "Australian" as "Austrilian" and some sad-act -- not even an Aussie, an Aussie would most probably have laughed -- interpreted it as a racial slur {it could be construed as making fun of the Australian accent}.
As for "teh", that seems to be a keyboard rollover thing. When you type fast, you actually strike the next key before releasing the previous one, and the keyboard controller has to work out what you meant. It sounds trivial; but remember, each key "bounces" for about 10ms. on release. Ten whole precious milliseconds, getting dangerously comparable to the duration of a fast typist's keystrokes, and you can't even use a simple R-C circuit for hardware debouncing because of the way the matrix scanning works. So the keyboard controller is having to compensate for key bounce in software, too, as well as rollover. Occasionally it messes up.
I did once write a piece of software that, if it saw the letters "t", "e", "h" typed within a certain timeout of each other, changed it to "the"; but if you typed it slowly and deliberately, one letter at a time, it assumed you really meant "teh" and left it alone. Unfortunately, I thought at the time that it was no more than an amusing novelty. Maybe I should have taken it further?!
I guess other languages must have similar rollover-related problems leading to common typos of their own. Anyone care to comment?
What crack is Ballmer smoking? He is saying computers are obviously too expensive, hence people are paring money off the total cost by "pirating" Windows. Hmm ..... That sounds like someone who owns a pay toilet complaining that beer is too expensive. Except Ballmer is keeping a £1-a-leak pay toilet. In the woods. With no roof.
IMHO if anyone is encouraging "piracy" of Windows, it is those manufacturers of cheap and nasty peripherals who provide only Windows drivers and say "f**k you" to Linux / BSD users. It's no skin off their nose if someone has to "pirate" a copy of Windows just to make their brand new, 10mm thick, 96-bit, 76800dpi scanner work. They haven't even got the excuse that "perhaps our customers are running a less expensive but nonetheless legal operating system" if there are no drivers supplied for such systems.
Maybe if purchasers of Windows software -- and that includes Windows drivers for hardware -- were obliged to show proof they are duly licenced to run Windows {and analogously for other OSes; except that very few people indeed are not licenced to run Linux, and nobody at all is not licenced to run BSD} then things would change. In practice, what would happen is that the OSS drivers would be on the same CD as the Windows drivers and thus not physically separable. Just the fact of having Linux and BSD drivers would get the hardware manufacturer off the hook -- it's a far more reasonable supposition that someone is going to use a cheap peripheral under a Free OS if a driver is already provided for that OS than otherwise.
The only thing you can do about stupid people is hope that sooner or later, they will make a mistake that either (a) teaches them not to be so stupid or (b) eliminates them from the respiration-photosynthesis cycle.
In the old days, when you had power tools with big blades and no safety guards, nobody ever made the same mistake twice. Computers have made things easier and safer for idiots, and idiots have made things harder and more dangerous for the rest of us.
To measure the power of a gas boiler, knowing the flow rate V litres per minute and the inlet and outlet temperatures, just multiply the temperature difference by the flow rate by 0.07.
..... (1)
Q = m * c * theta
In one minute (=60"), we have V litres of water flowing, or about V kg. So dm/dt = V/60 kg s-1. c is about 4168 J kg-1 K-1. But it'll be easier to handle power in kilowatts, so we shall divide everything by 1000.
Power = flow rate [LPM] / 60 * 4.168 * temp diff [C]
= flow rate [LPM] * temp diff [C] * 4.168 / 60
= (about) flow rate * temp diff * 0.07
That's as close as you're likely to get it anyway due to system inaccuracies.
Suppose you have something like this (apologies for loss of indentation)This is ripped off from a web application I once wrote. You should be able to modify the time by typing figures into the box or by using up and down arrows. What happens in practice is that adding 1 actually concatenates "1" on the end of the string, so you find that 1 + 1 = 11, and 11 + 1 = 111.
This is what you get when you borrow one idea from Perl about how the computer should be able to work out from context whether or not something is a number or a string; and one from BASIC about re-using operators obviously out-of-context {strings cannot be added} to mean something different {such as concatenation}. The result, as they say, is a mess. For all practical purposes, JavaScript lets you subtract, multiply and divide numbers; but if you want to add, you'd better subtract a negative number.
I mean, it's not freaking rocket science, is it? = is for telling, == is for asking. + is for adding, . is for joining strings. Sheesh!
The Linux driver API *is* stable -- provided that you have the source code available and can recompile anytime. IMHO that assumption is totally reasonable. No matter how many layers of abstraction you provide, something somewhere has to ride the metal.
The problem is that manufacturers are working on the {false} assumption that a compiled binary driver is secure against competitors working out what it does. Decompilation is mathematically similar to shape recognition, and it's a matter of time before a usable decompiler is released under an open source licence. What's going to be worse for some people is that they have released binaries under a BSD-like licence -- which effectively will give anyone who acquires the source code by force, permission to distribute it.
- Open the source code in your favourite text editor
- Locate the offending section of code
- Insert comment markers at the beginning of each line
- If necessary, add lines setting various variables to sensible values instead of what what you just commented out would have set them to
- (Optional but stupid not to) Add comments explaining what you just did
- make clean; make && make install (or whatever you do on Windows to compile and install a package)
- (Optional but good manners) Submit patch to appropriate Internet sites
Easy, isn't it.I'd most definitely buy a fully open-specced graphics card, even if it cost a little extra. Better that than a tainted kernel. Although to my mind the question should not be "should someone make an open spec card>", but "shouldn't everyone be obliged to make their hardware open spec?"
If I am the rightful owner of a graphics card {say}, then Common Law says that nothing concerning that graphics card is a secret from me -- and nobody can stop me using whatever techniques are at my disposal for discovering what I have a right to know {which is, basically, every true fact in the universe}. They can bind me to keep secret what I discover; but, as long as I own the card, it is not secret from me. Nor can it possibly be secret from anyone else who owns an identical card.
Proper enforcement of this law is what we need. {It never really mattered up until now, because nobody ever envisaged technology getting as complicated as it has become today; you could figure out how something like a mechanical clock works just by looking at it, hardly needing any instrument more sophisticated than a magnifying glass.} Anyone who owns an nVidia or ATI card already has the right to know how to program it -- they shouldn't have to fight for that.
And spare me the bitching about revealing things to competitors. Are nVidia really so naïve as to think ATI don't reverse-engineer their cards, and vice-versa? Come on, guys.
It still only invalidates the result from one polling station, and that might not even be enough to alter the eventual outcome for the seat {depending upon whether or not you were using transferrable votes, and how close the candidates were} -- but, then again, nobody is likely to try this sort of vote-spoiling tactic in a "safe" seat. It's not such a nightmare logistically. Again, this is a consequence of having many polling stations, each serving a few thousand voters.
Anyway, there are ways an electronic or mechanical voting machine could be tricked into accepting multiple votes from the same person. The problem is with the end, not the means.
In the UK, you cast your vote by making a single, unambiguous mark -- conventionally an X since the intersection of the two lines indicates a precise point on the page -- in a box next to one candidate's name from the pre-printed list. {The lists are randomised; two adjacent papers will have the names in different orders}. It's up to the person counting the votes to use some common sense, obviously; but dubious ones are placed in a separate pile and looked at by everyone at the very end.
In most of the rest of Europe {and in private elections in the UK, e.g. student unions, trade unions &c.}, Single Transferrable Votes are used. You cast your vote by writing a number in a box adjacent to each candidate's name on the pre-printed list. The lists are randomised, but "Re-Open Nominations" {or its local equivalent} is always last. Same procedure applies with "dodgy" papers. Numbers are pretty hard to confuse.
If you change your mind, you have to get another ballot paper from the presiding officer.
I work for an ISP. And knowing what I know, there is no way I would let the fact of who I voted for get out on the Internet.
In a paper ballot, you give every voter an identical ballot paper, and they give it back to you with their vote on it. To emulate this electronically, you would need to issue a token to each voter which is revoked as the vote is cast. But that isn't quite perfect. Paper ballots can't be copied {well, they can; but you aren't allowed to take the ballot paper out of the polling station, nor to bring the necessary wherewithal to make a copy of the paper into the polling station, and let's assume the presiding officer is running a tight ship here}, so you know just from whether or not the paper is in the ballot box, whether or not the token has been revoked; and you know from whether or not the voter's name has been crossed off the list, whether or not the token has been issued in the first place. To be able to know which electronic tokens are valid and which have been revoked, every one would need to be different - and thus identifiable. Furthermore, you have absolutely no idea what is going on in transit between the server and the client and back. If you aren't actually sitting there in a polling station, then you can't supervise voters to make sure they are not misbehaving. They could be "printing their own ballot papers" {generating valid tokens for themselves}, or "stealing other people's ballot papers" {misappropriating valid tokens issued to other people}. How do you know, given the proliferation of Windows PCs, that a client machine has not been infected with a Trojan horse that hijacks its internet connection? You could maybe give out bootable CD-ROMs with a secure OS and client software on them, but then how do you ensure that your special protocol doesn't get reverse-engineered and misapplied between issuing the CD and the vote being cast?
The problem is simply that every time you fix one of the existing failure modes, you introduce some more new ones that you hadn't thought of before. Some of the limitations are not limitations of technology, which can be overcome by invention, but limitations of the universe, which can only be overcome by breaching one or more fundamental laws. At least the failure modes of paper ballots are well understood, and mitigated to the fullest extent possible.
So it's just like an "old-fashioned" modem that you used to plug into a serial port, and with an entry in /dev/ that acts like a ttyS* port? Does that mean I can dial my spare phone line at normal mobile-to-landline rate {which on my tariff is nil in the evening and at weekends for the first 300' a month}, have mgetty auto-answer and gateway myself through my broadband connection? Or am I stuck with using my own telco's dial-up access, or is the call charged extra anyway just because it's a modem call?
But the serial number on the ballot paper is not recorded against your name anywhere. You can always exchange your ballot paper for a different one, which would spoil the records. And nothing is stopping you "swapping votes" with someone you trust -- so you vote for their party, they vote for your party, and everybody's happy.
In the UK, we have hand-counted paper ballots. We have had them since we started having elections. It is a system that everybody can understand, and it's evolved over the years to be surprisingly robust.
Each ballot slip is placed whole into the box. So it's verifiable if necessary, by re-counting. The fact of your voting is recorded, but in such a way as not to be able to link your name to a particular ballot paper. In case the ballot slips are secretly marked or anything like that, you can pick your own if you feel sufficiently paranoid {you aren't forced to accept the one the presiding officer gives you}; so it's secret.
Each polling station takes votes from an area no bigger than the volunteers working there could comfortably count by hand all the votes from. So it's scalable -- if you have more voters, you just add more polling stations. It's also quick -- in each polling station, there are only a few thousand votes to count. All this is going on in parallel, results are initially telephoned through and then the ballot papers are sealed back up in case they need to be re-counted.
The numbers involved mean that to "buy" an election, you would have to pay off a lot of people. So it's actually quite tamper-proof. And if any shenanigans are suspected, a recount can be ordered -- or, in the worst case the ballot repeated -- in just the known affected polling stations.
It is not clear to me how this system could be improved on without introducing new failure modes. Any kind of vote-counting machine is susceptible to tampering. Even if it is absolutely open to public scrutiny for the days when it is not being used for an election, there are stunts that could be pulled on the day. And even if the machine is verified by a hand-count, it will still takes the same number of people to hand-count the ballots after the machine is done, so what have you saved?
If you're going to rely on human honesty, it's best to distribute that reliance as widely as possible, i.e. to trust several thousand people to be just a little bit honest rather than trust a few people or just one person to be very honest indeed. After all, the majority of human beings are generally honest, and more so when the stakes are low. What benefit is there to dishonesty in counting a few thousand votes among tens of millions? On the other hand, if you are the managing director of the company that makes the only officially-approved voting machines, you effectively have every election in your hands -- and that is where the benefits of being dishonest do start to show.
I am seriously thinking of writing to my MP to try to get it made law that manufacturers must provide full disclosure of driver specs if they want to sell their goods in this country. Placing the driver source code under BSD licence or GPL would of course satisfy the requirements. {I know BSD can be poison; but it isn't actually too bad for situations like drivers, where there is no danger of proprietary extensions taking over from the main code base. The manufacturer can't add proprietary extensions themself, since that would break the requirement for full disclosure.}
Meanwhile, has anyone got any success stories re. getting a Linux laptop on the internet with a mobile phone? Any mobile at all? If I use my own dial-up server, will it just be counted as any old mobile-to-landline call {and therefore covered in the monthly allowance on my current tariff}? I don't need a feature-packed phone with a colour display, camera and integrated self-rolling ashtray, just something that lets me send and receive text messages and answer voice calls. Oh, and go on the internet, obviously
It's all very well that the phone itself runs Linux underneath. What's more important from where I'm sitting is, can I use it with my Linux {single-booting and proud of it} notebook? For instance, does it use standard {or at least, well-documented and free for the asking} APIs so I can write my own perl scripts to do cool things with it? Can I use it to get on the Internet? Do Motorola provide the drivers as source .tar.gz files which I'll be able to compile on any system, not just the "commercially viable" ones? Does the phone have a scripting language {I mean, more sophisticated than ash} on-board?
My wishlist would be for something that looked to the host PC as though it was a USB network adapter plugged into some sort of network. You would assign it an IP address in a subnet of your choosing, from the phone's "console". There would be standard servers on standard ports {21 for FTP, 22 for SSH, 80 for web, 3306 for database, &c.; maybe even an Asterisk proxy for hardcore VoIP users} so as to keep the interface clean and simple {I'm implying that they would just be using regular files and databases for storing ringing tones and contact info; there's no reason to assume otherwise}. While you were on the Internet, the phone would do NAT, just like one of those little ADSL gismos.