The thing about 10 having only two factors has been a red herring ever since we started writing fractions in decimal notation. Express any fraction decimally, round to the greatest accuracy that you can actually measure, and -- provided your subunits are on a scale of 100s, 1000s, or some other power of ten -- there's your answer.
The canonical carpentry problem used to discredit the metric system is to divide a space one metre into three equal portions. Grab any old idiot-calculator; evaluate 1/3; and you get (assuming an 8 digit display) 0.3333333. Now, you can read a tape measure to 0.001m., and you might just be able to estimate 0.0001m. if you really try; but in any case, the width of the pencil mark will compromise your accuracy. So I will go with 0.333m.
For fairness' sake, let us divide a space 4 feet wide into five equal portions. Then each one should be 0.8 feet, and this is more problematic since most rulers are marked in feet and inches rather than fractions of a foot. (Actually, most rulers are marked in metric units only, so we should qualify this and say most rulers that are marked in feet at all.) So we have to convert 0.8ft. into inches, by multiplying by 12 to get 9.6 inches. Then we find that this measure is marked in 32nds of an inch, so we must find the nearest 32nds to 0.6. 0.6 * 32 = 19.2, so the answer is 9 and 19/32 inches. With only an idiot-calculator at our disposal, we needed three calculation steps plus the use of a pen and paper.
To show there is no cheating, let us take a really contrived example {or maybe it is not that contrived.....} and divide a space 1.2192m. wide into five equal portions. The idiot-calculator gives us 0.24384m., or 0.244m. after rounding.
Metric system => easy to manage using simple 4-function calculator.
American system => needs programmable scientific calculator.
Loose fruit and veg can only legally be sold by the kilo since 1 Jan 2000. The funny thing is, before then, it was actually illegal to sell it by the kilo -- it had to be in pounds. Some supermarkets used to weigh in fractions of a pound, which is neither here nor there but at least made it easy to verify the figures with a pocket calculator. I've often wondered how easy it would be to run a little scam wherein you added a couple of pence to the total anytime the reading wasn't bang on a whole pound..... which is most of the time..... after all, who's going to check suchg a difficult sum? And in a store that sells a lot of groceries, it would soon mount up.
Beer and cider are sold in 568ml (= 1 pint) measures, and you can be busted for selling it in litres no matter how accurately you measure it:( Which is a bit of a shame, because a 568ml to-the-brim glass would make a great 500ml to-the-line glass, thus keeping both the metrication lobby and the full-measure lobby happy (but not the landlords who profit by selling short measures).
I stand corrected on the ports issue..... looking at the plastic moulding, there could well be other variants with extra ports, and possibly even a headset option. Our distributor must just have been supplying the really basic one.
But one 10Mb segment shouldn't slow down your whole LAN. If it does, you need to get yourself a more modern switch with proper store-and-forward ability. Any 10Mb packets coming in will be artificially speeded up to 100Mb, and 100Mb packets will be slowed down to 10Mb..... the switch will fake a busy status while this is going on, but only on the port which is actually trying to send data faster than can be dealt with. Two other ports can still be exchanging data at 100Mb/s while all this is going on.
Even if you don't own the content, you own the medium upon which it is recorded. You have paid for the right to appreciate the content, regardless of the means used to achieve that end. Whether you watch the content on a store-bought DVD player or a home-made one, makes no difference to the content licensor. (It makes a difference to the DVD player vendors..... if I bought a Philips DVD player, Sony would suffer, and if I bought an Alba DVD player, Philips would suffer..... but that's just the free market economy in action).
Part of the problem is that Windows has traditionally been so lax on security that programmers have got away with bodges that would be considered unforgivable on a system that had been designed with security in mind from the word go. At some stage, though, something has to give. If all this legacy software is depending for its very operation on the same things as the viruses, worms, adware and spyware -- and it is -- then that is the choice you have to make: whether to allow sloppily-written programmes to take advantage of the security holes but unavoidably also permit malware to use them; or to prevent malware taking a hold, but in the process, unavoidably break sloppily-written legacy software. The two are indistinguible.
Now, if SP2 breaks compatibility with so much legacy software, then surely this spoils one of the arguments against switching to an alternative operating system that also would break compatibility with legacy software?
On a slightly different topic, why is anti-virus and spyware removal software closed source? If I cannot view the source code of an anti-virus programme then how do I know it is not simply going to infect my system with a virus every so often just so it looks like it has done some good? How do I know it is not going to infect other people's systems with viruses just so they will buy their own copies of anti-virus software? How do I know it is not installing its own spyware? If the software is not a Trojan horse then why will the makers not just show me the source code?
As a non-US citizen, I am puzzled by this. Surely the very fact that you own the disc on which the material is recorded, precludes anybody from telling you what you may or may not do with it so long as that use does not adversely affect others?
I mean, fair enough: if I own a knife, that does not give me the right to stab other people with it, nor does it give me the right to use it to cut up other people's property without their say-so. If I own a DVD, I can't legally throw it through somebody's window: I would be disrupting their common law property rights by damaging a window that they own. But I can legally watch the film that is recorded on it: that is my common law property right. And regardless of whether I watch that film using a player I bought in a store; or a player I made out of common household materials; or by looking at the pits and lands, translating the zeros and ones in my head, painting pictures on sheets of card and flicking the edges with my thumb; I am acting within my right to view the picture. It is the end that counts, not the means.
I can (almost) understand a prohibition against attempting to defeat encryption techniques, but the fact is that as the rightful owner of the DVD, I am the intended recipient of the encrypted message and I may use any reasonable means at my disposal to do so. Ownership of the DVD gives me the right to defeat the encryption, just as I cannot be arrested for picking the lock of my own front door.
And this is coming from a land without a written constitution! Surely the US constitution guarantees common law property rights?
Theoretically you can go even faster because you have more flexability in changing the parameters and algorithms used in the modem
Thou smokest crack. The telephone line is plugged into a..... well, let's call it a sound card for now..... at the exchange, which takes 8000 samples to 7 bits accuracy every second. That's 56000 bits a second. Adding parity bits takes it up to 64000 samples a second, which -- spookily enough -- is the same rate as an ISDN B-channel. When 30 of these B-channels (each changing 64 000 times a second) and their associated D-channels (much lower speed) are switched in turn, one at a time, onto a pair of wires for 1/2 048 000 of a second at a time, this is called an E1 line.
If you changed the signal amplitude more often than that 8000 times a second mentioned above, some of the changes would be missed by the phone exchange. Likewise, if you had more than 128 possibilities for signal levels, some of them would look the same when sampled. So you simply cannot get more than 56000 bits a second down a POTS line..... that is a limitation of the POTS itself. (You can maybe cheat and appear to get more using on-the-fly compression, but this is subject to the entropy of the data. Most of what gets uploaded/downloaded already is compressed anyway, and thus has naturally high entropy.)
You could have a phone with the digital-to-analogue conversion stages inside the phone, connected straight to a B-channel and a D-channel of an ISDN line.
If you wanted higher speeds out of an analogue phone line, you could take advantage of the fact that much of the bit bottleneck is in the exchange, and that the line from your home to the exchange is just two copper wires which will usually handle much higher frequencies than your voice can generate or your ear can hear. You would have to modulate the data onto a much higher frequency carrier, combine it with the voice signal from your telephone, blast that down the line and use filters to split the signals apart again. Then the high (data) frequencies could be sent to a special demodulator, which could recover the zeros and ones and chuck them at a router; and the low (voice) frequencies would just go to the ADC as before. You don't even have to use phone lines, you could use any wiring that will handle a few megahertz..... such as a cable TV cable, or even the mains wiring.
Oh, wait a minute...... That's called ADSL, isn't it?
Ever since my boss got friendly with the Asterisk developers, my company's internal telephone network is now almost entirely VOIP. We have a server running Asterisk, with a Zaptel line card (needed a 3V3, 6MHz, 32-bit expansion slot; something you apparently only find on high-end mobo's, as most of the low-cost ones are 33MHz and/or 5V) plugged into an E1 line giving 30 ISDN lines. But you only need this to connect to POTS phones -- connecting to other VOIP phones is just done over the internet. The Asterisk machine also currently runs our intranet, though I'm ordering a new server for all the non-telephonical functions as something keeps crashing (not often enough to be serious, but we need to narrow it down).
As for phone clients..... we use dedicated hardware telephones. The Grandstream BudgeTone 101 was the first we evaluated, be aware that this comes with a Continental-style mains adaptor so you may need to get a new power pack (regulated 5 volts 400mA DC + --o)-- - polarity). This works lovely as a SIP telephone but doesn't as standard allow for a headset, which we kind of need in a call centre. The handset does use a standard RJ01 connector, but there seemed no easy way to deal with the receiver switch. We also evaluated every softphone we could get our hands on. In general they seem to be a bitch to get to compile; I had the best result with Linphone, it wasn't as polished as KPhone but it seemed to crash less often; and got absolutely gnowhere with Gnophone. Bear in mind also that a telephone headset will reveal the limitations of the sound chipsets on modern mobo's: you will require a real SoundBlaster-compatiable if you want to be able to understand what anybody is saying. I am running Debian Sid, my boss is running some perversion of Mandrake with a load of stuff from Cooker, and all our workstations run Mandrake 9.2 (hackerish systems are fine for us hackers, but it's more important to have Stuff That Just Works for the masses). We also got a softphone client from Zultys, called LIPZ; which looked stunning but was problematic in practice. It seems to bogart memory and CPU cycles. And when I came to do some hacking on it, I found the real kicker: it doesn't include the source code, so who knows what the hell it's really doing? In the end, we wound up using Zultys ZIP4X4 hardware SIP telephones. These are very expensive for "just" a telephone, but they are stuffed with features, all known codecs, 4 virtual lines, even an integral 5 port (one for the phone, four brought out on RJ45 jacks) 100Mb/s switch, and they are hardware -- my favourite programming language is still 63% tin, 37% lead.
My honest recommendation would be go for something like the Grandstream, which does everything an "ordinary" phone should do and, being hardware is truly cross-platform. But note, it doesn't have any integral switch so you will take up an extra jack on your ADSL router.
If you did that, then it would not be free anymore. So why would you want to?
The whole point is that everyone should have access to the source code. The compiled version is just there for convenience's sake.
What you are saying is a bit like "What good is a marriage certificiate if my wife can still refuse to have sex with me if she doesn't want to?" It doesn't work like that!
it isn't the simplicity of an API that leads to worms, but rather the size of the install base.
Yeah, right. Apache webserver installations outnumber IIS by two-and-a-bit to one (the actual number varies with time), and how many exploits are there for Apache vs. how many for IIS?
Just because A happens and B also happens, doesn't mean A is the cause of B. If you're still not convinced, I've some elephant repellent you might be intrested to buy.....
He also could have said that "Open source software is so difficult to interoperate with that worm and trojan developers don't both."
You actually were much closer than you think with that remark. Open source software is difficult to interoperate with and not get noticed. Nobody ever thought to conceal anything from anybody, so everything is nice and transparent, and there are few places to hide.
[C]leaning up after a virus / worm / trojan / spyware / adware infection isn't regarded by blinkered bosses as "maintenance", it's "disaster recovery".
Which is rather at odds with the usual meaning of disaster. In most other industries "disasters" which happen frequently are considered "maintenance issues".
Bzzzzt! You are thinking like an inhabitant of planet Earth, not planet Pointy-Hair. Although Windows fails on an almost-daily basis, and does so by design, it must be better than Linux -- because everybody else uses Windows.
I don't understand it, either. But then again, my company has a written policy explicitly favouring Open Source -- which I wrote:) The only exception is for the beancounters, who need Sage for compatibility with Group Head Office -- and the very next time a beancounter's Windows box goes Tango Uniform, Group Head Office are going to be requested to rethink that policy. Possibly with a cricket bat:)
Wow..... I did not know that! So I take it that this means it's OK to shoplift food if you're hungry, and there is no such offence as conspiracy? Cool. Perhaps I will visit the USA after all!
How about everyone just cuts their losses. Live with the fact that artists are not going to get compensated for their work.
It's tough, but them's the breaks. The only reason anybody ever made money out of distributing recordings of artistic performances was that the ability to manufacture recordings was a scarce commodity. With the ready availability of CD and DVD recorders, that is simply no longer the case.
So let's let the whole music industry just pack their bags and go home. It's been a good ride while it lasted, but now it's over. So long and thanks for all the tunes. The world will not end just because there are no more Britney Spears clones, and people will not stop playing good music just because they are not getting paid for it.
Linux maintenance is an ongoing cost
Maintanance is an ongoing cost, regardless of OS.
Agreed.
and [Linux maintenance] does tend to be more expensive, when it is required, than Windows maintenance, but that is to be expected.
I doubt the cost of cleaning up after Windows malware is exactly "cheap".
Agreed. But cleaning up after a virus / worm / trojan / spyware / adware infection isn't regarded by blinkered bosses as "maintenance", it's "disaster recovery". Contracting a computer virus should IMHO be a sackable offence, or at the very least something that affected employees would be expected to pay for out of their own pockets. It isn't, because PHBs are as susceptible as the next person, and such phenomena are -- wrongly -- regarded as part and parcel of using computers, when they really are part and parcel of using Windows.
The product of cost of recovering from virus attacks on a Linux network times probability of occurrence is negligible, but good luck convincing a Gates-worshipper that it's so.
Linux doesn't go wrong very often; and when it does go wrong, it usually goes wrong for a reason. Windows is inherently unreliable. Sorry to upset the MS fanboys, but it is. The reason is that it is closed source.
Part of the problem is the closed source nature of Windows, which means there is a temptation for bad code and hacks to wind up in there. Because "no one is going to see this anyway". But there have also been cases of Microsoft deliberatly choosing to write bad code in the name of "integration".
Agreed. I was holding my tongue to avoid accusations from the closed-source fanboys; but now you have spoken, I feel free to say: GEDDINTHEREMYSAN! Another reason why bad code pravails in closed-source projects is that, since developers aren't allowed to borrow code that is recognised to have been Done Properly {either because the good code exists in a closed-source programme which they are not allowed to look at; or -- slightly, but deliciously, ironically -- because the good code exists in a GPL programme but the developer insists to release a closed-source programme}, everyone has their own go at inventing the wheel from first principles; and, barring supreme good luck, introduces their own special little errors.
Anyone noticed that almost no notebook has a line-in port these days? (Mic inputs are mono, preamplified and band-limited to fuck.)
I don't know about pre-built desktops {never bought one, never likely to}. Mobo-integrated sound chipsets do have a line in, but it picks up so much static and power hum that it's not worth using. In my experience, they're more for looking at and saying "look, it has these all these ports built in already" than for actually using. Even separate sound cards can be noisy, as none of them have stuff like tin cans and power chokes anymore. And the old ones that did, were all 16-bit cards and new mobos are all 32-bit. Unless someone makes a converter to plug old 8/16 bit cards into a 32-bit expansion slot? No, thought they didn't. Pity really, 'cause the old 16-bit bus made it extraordinarily easy to interface to homebrew hardware. Not that it would have made any sense not to; because up until then, they basically were still trying to encourage people to invent new stuff so as to sell more PCs, whereas now it's not in any established manufacturer's interest to have people inventing new stuff in case it takes away sales.
I saw a copy of [Nievy Ynivtar]'s new album "[Haqre Zl Fxva]" with a big copy-protection warning splattered all over it in H.M.V. a couple of weeks ago, and I just had to buy it for the "hacking challenge" factor.
Cdparanoia read the audio without a hitch -- all spaces, not even so much as a single minus sign, and cdrdao made a.toc and.dat pair that burned OK.
I still felt an overwhelming sense of disappointment..... this was not so much shooting fish in a barrel, as standing by a barrel of fish and waiting for them to jump out.
Of course I made sure to check I could get a refund "if it wouldn't play on my DVD recorder". The warning said it might not work in anything other than a home audio CD player..... though actually, my DVD recorder is built only to recognise the first session on a multisession CD {to paraphrase the manual: if you want to make an MP3 CD to play on this machine, you must burn all the songs to the disc in one go, otherwise it will only see whatever you recorded up to the point where you first ejected the disc} which sounds like they were anticipating some sort of copy-prevention attempts involving a bogus second session. And the machine even has a digital audio-out which I haven't investigated; I think it's electrical rather than optical. Anyone know of a good sound card with digital-in and full Linux support including open source drivers?
Disclaimer: by reading the ROT-13ed text above, you are agreeing not to laugh.
When I was 14 years old, that would have been 1985, I invented a self-destructing audio cassette. It was a regular audio cassette with a piece of ceramic magnet glued inside the shell just upstream of the takeup spool, where the tape would pass over it during playback. This had the effect that the recording could be listened to once only (at least without my "specially-modified" recorder).
I only used it a couple of times for a bit of fun, but does it count as prior art?
I originally thought of using it for computer games that would transfer themselves to disk (so you could buy a tape, which was cheaper, then have the game save itself on a disk you had purchased elsewhere more cheaply than the software companies, who used to have to charge an extra GBP4.00 to cover the cost of the horrendously expensive floppies they used to use). But you would only be able to play it once, so you would only get to make one copy. Of course I realised as soon as I had thought of it, that it was flawed, because the cassette doesn't know if the audio signal is going into one computer, fifty computers with their tape-ins wired in parallel, or another tape deck with a non-self-destructing cassette inside it.
The article says there is a small ampoule containing a chemical, which is released when the package is opened (the DVD is screwed in place to begin with) and attacks the reflective coating on the top surface of the disc. In one of the two they received, the process had been started already, probably by improper tightening of the nut. The other one was still playable after the 8 hours were up, just by skipping the introduction and going straight to chapter 2.
Build air-tight cabinet around DVD player with special rubber gloves for manipulation. Put DVD inside cabinet. Close seal. Flush cabinet with nitrogen. Set pressure regulator just above 1 atmosphere (in case of any leaks). Put hands in gloves. Unwrap DVD. Play. After watching, put DVD in air-tight container. Charge with extra nitrogen.
All that nitrogen may be getting a bit expensive, so:
Buy the disc. You have 8 hours to make a longer-lasting copy of it. Return the disc to the manufacturer at their own expense, claiming it is faulty. Watch copy at your own leisure.
Or, if mindless destruction is more up your alley:
Find a stack of them in a store. Puncture the wrapping with a pin (or, for extra shock value, a syringe).
Christians like to pretend they have a monopoly on good deeds. This is not so. In fact, most christians really are not nice people, and this is exactly the behaviour I would think to associate with christians.
At the end of the day, christians are some of the most self-centred people you could hope to avoid having to meet. If an atheist does you a favour, it is because they want to do you a favour. If a christian does you a favour, it is because they think it will improve their own personal chance of getting into heaven. No more, no less. If their priests told them they could get into heaven by stealing hub caps off people's cars, that is what they would do.
I found a couple of problems with Konqueror. The company where I work used to use an old, obsolete (but it did most of what they wanted) Windows application, together with some Perl on a web server -- till I replaced it with a roll-me-own solution, using mostly PHP with a bit of Perl and MySQL. Eventually it got to the point where we no longer needed Windows at all.
Unfortunately, my Web based front end used a fair amount of JavaScript, and did things like copying data from a form on one page into a different form on another page and triggering a submit when a button was clicked on yet another page. It was fine in Mozilla (my testing platform), and what was then called Firebird, but Konqueror barfed. I'm presuming this is due to overly-paranoid security settings. Cross-site scripting vulnerabilities are one thing; but on an Intranet with just one web server, it doesn't matter. Mozilla correctly deduced that the servers were one and the same and permitted the operation. {BTW, I fully expect someone to explain in one paragraph how to fix this.}
Also, Konqueror and Mozilla behave a bit differently with some JavaScript / CSS fancy menu code I wrote {this wasn't related to my "main" work}. But I can't see what bits aren't valid, so maybe it's KHTML's fault?
Installing Linux is only the beginning. There is going to be a period of transition where people have to learn to use the new systems. It's not as bad as some people seem to think -- I have first hand experience, having to support ex-Windows users in a nearly-all-Linux company {Windows is still used by about half a dozen people, for a few legacy applications; but you can bet your arse, we fully intend to make use of our statutory right to perform reverse-engineering for the sake of interoperability -- which is not affected by anything the EULA may say.....} and once you have explained the obvious differences, people will get used to it.
Legacy documents will have to be converted to the new standard, quite probably by retyping them from scratch. This will represent a substantial one-time cost; but once documents are held in an open format, the hardest bit of the work is over. There is a big decision to be made: for each document, should we (a) go to the effort of converting it accurately, (b) put up with some idiosyncrasies arising from an imperfect conversion or (c) scrap it. Such triage itself costs money, but it might lead to a considerable saving if there are many B or C answers.
Linux maintenance is an ongoing cost -- and it does tend to be more expensive, when it is required, than Windows maintenance, but that is to be expected. Linux doesn't go wrong very often; and when it does go wrong, it usually goes wrong for a reason. Windows is inherently unreliable. Sorry to upset the MS fanboys, but it is. The reason is that it is closed source. Application and driver developers can't see the OS code, and have to trust that it matches the documentation exactly. OS developers likewise have to trust that application and driver developers have specified their interfaces correctly. What happens in practice is that you get "logic traps" -- sets of ordinarily innocuous events which, if and not unless they all happen together, will cause something different to happen instead of what should have happened. It may be as obscure as pressing the W key on a Belgian keyboard as the CRT raster comes to the end of the 87th line while the sound card is playing F# above middle C out of the right hand speaker and a PCL5 printer driver has been swapped out from memory -- and the only way anyone could ever spot it, short of actually experiencing it, would be to look at the source code for the OS and the keyboard, graphics card, sound card and printer drivers. It's not hard to see how the severity increases geometrically as more devices and programmes are added to a system. Even motherboard drivers are not immune.
So Windows has these designed-in random crashes that happen for no reason other than source code bogarting, and they account for by far the vast majority of Windows crashes. Without the source code, the only thing you can do is tut, scratch your chin, and whip out the power lead. And a trained monkey could do that. Replacing the hardware, or installing a newer version of the software, will cure most of the worst ones by introducing a different set of trigger criteria. Linux doesn't have the same potential for hidden logic traps, because the developers can see one another's code and so prevent them from ever arising in the first place. Of course they may be present in early versions, but someone is certain either to fix the problem themselves, or alert someone who can fix it. If something is amiss with Linux, there is likely to be a reason which will not go away if you just cut the power. And fixing it takes someone who really groks the system, bellyfeels it, is one with it. That costs money -- but since Linux doesn't have built-in unreliability, most of the time it will just work. This means you will need fewer, more expensive support personnel, as opposed to more, cheaper personnel with a comparable Windows installation.
Slashdot Mirror
The thing about 10 having only two factors has been a red herring ever since we started writing fractions in decimal notation. Express any fraction decimally, round to the greatest accuracy that you can actually measure, and -- provided your subunits are on a scale of 100s, 1000s, or some other power of ten -- there's your answer.
.....} and divide a space 1.2192m. wide into five equal portions. The idiot-calculator gives us 0.24384m., or 0.244m. after rounding.
The canonical carpentry problem used to discredit the metric system is to divide a space one metre into three equal portions. Grab any old idiot-calculator; evaluate 1/3; and you get (assuming an 8 digit display) 0.3333333. Now, you can read a tape measure to 0.001m., and you might just be able to estimate 0.0001m. if you really try; but in any case, the width of the pencil mark will compromise your accuracy. So I will go with 0.333m.
For fairness' sake, let us divide a space 4 feet wide into five equal portions. Then each one should be 0.8 feet, and this is more problematic since most rulers are marked in feet and inches rather than fractions of a foot. (Actually, most rulers are marked in metric units only, so we should qualify this and say most rulers that are marked in feet at all.) So we have to convert 0.8ft. into inches, by multiplying by 12 to get 9.6 inches. Then we find that this measure is marked in 32nds of an inch, so we must find the nearest 32nds to 0.6. 0.6 * 32 = 19.2, so the answer is 9 and 19/32 inches. With only an idiot-calculator at our disposal, we needed three calculation steps plus the use of a pen and paper.
To show there is no cheating, let us take a really contrived example {or maybe it is not that contrived
Metric system => easy to manage using simple 4-function calculator.
American system => needs programmable scientific calculator.
Wouldn't it be more sensible to spell it "megagramme" ? I have always wondered about this. I bet the Germans call 1000kg 1 Mg.
Loose fruit and veg can only legally be sold by the kilo since 1 Jan 2000. The funny thing is, before then, it was actually illegal to sell it by the kilo -- it had to be in pounds. Some supermarkets used to weigh in fractions of a pound, which is neither here nor there but at least made it easy to verify the figures with a pocket calculator. I've often wondered how easy it would be to run a little scam wherein you added a couple of pence to the total anytime the reading wasn't bang on a whole pound ..... which is most of the time ..... after all, who's going to check suchg a difficult sum? And in a store that sells a lot of groceries, it would soon mount up.
:( Which is a bit of a shame, because a 568ml to-the-brim glass would make a great 500ml to-the-line glass, thus keeping both the metrication lobby and the full-measure lobby happy (but not the landlords who profit by selling short measures).
Beer and cider are sold in 568ml (= 1 pint) measures, and you can be busted for selling it in litres no matter how accurately you measure it
I stand corrected on the ports issue ..... looking at the plastic moulding, there could well be other variants with extra ports, and possibly even a headset option. Our distributor must just have been supplying the really basic one.
..... the switch will fake a busy status while this is going on, but only on the port which is actually trying to send data faster than can be dealt with. Two other ports can still be exchanging data at 100Mb/s while all this is going on.
But one 10Mb segment shouldn't slow down your whole LAN. If it does, you need to get yourself a more modern switch with proper store-and-forward ability. Any 10Mb packets coming in will be artificially speeded up to 100Mb, and 100Mb packets will be slowed down to 10Mb
Even if you don't own the content, you own the medium upon which it is recorded. You have paid for the right to appreciate the content, regardless of the means used to achieve that end. Whether you watch the content on a store-bought DVD player or a home-made one, makes no difference to the content licensor. (It makes a difference to the DVD player vendors ..... if I bought a Philips DVD player, Sony would suffer, and if I bought an Alba DVD player, Philips would suffer ..... but that's just the free market economy in action).
This is a real opportunity for the Open Source movement to make a killing, if we handle it right.
Part of the problem is that Windows has traditionally been so lax on security that programmers have got away with bodges that would be considered unforgivable on a system that had been designed with security in mind from the word go. At some stage, though, something has to give. If all this legacy software is depending for its very operation on the same things as the viruses, worms, adware and spyware -- and it is -- then that is the choice you have to make: whether to allow sloppily-written programmes to take advantage of the security holes but unavoidably also permit malware to use them; or to prevent malware taking a hold, but in the process, unavoidably break sloppily-written legacy software. The two are indistinguible.
Now, if SP2 breaks compatibility with so much legacy software, then surely this spoils one of the arguments against switching to an alternative operating system that also would break compatibility with legacy software?
On a slightly different topic, why is anti-virus and spyware removal software closed source? If I cannot view the source code of an anti-virus programme then how do I know it is not simply going to infect my system with a virus every so often just so it looks like it has done some good? How do I know it is not going to infect other people's systems with viruses just so they will buy their own copies of anti-virus software? How do I know it is not installing its own spyware? If the software is not a Trojan horse then why will the makers not just show me the source code?
As a non-US citizen, I am puzzled by this. Surely the very fact that you own the disc on which the material is recorded, precludes anybody from telling you what you may or may not do with it so long as that use does not adversely affect others?
I mean, fair enough: if I own a knife, that does not give me the right to stab other people with it, nor does it give me the right to use it to cut up other people's property without their say-so. If I own a DVD, I can't legally throw it through somebody's window: I would be disrupting their common law property rights by damaging a window that they own. But I can legally watch the film that is recorded on it: that is my common law property right. And regardless of whether I watch that film using a player I bought in a store; or a player I made out of common household materials; or by looking at the pits and lands, translating the zeros and ones in my head, painting pictures on sheets of card and flicking the edges with my thumb; I am acting within my right to view the picture. It is the end that counts, not the means.
I can (almost) understand a prohibition against attempting to defeat encryption techniques, but the fact is that as the rightful owner of the DVD, I am the intended recipient of the encrypted message and I may use any reasonable means at my disposal to do so. Ownership of the DVD gives me the right to defeat the encryption, just as I cannot be arrested for picking the lock of my own front door.
And this is coming from a land without a written constitution! Surely the US constitution guarantees common law property rights?
Relay my arse. Modern ones use an opto-isolator and four rectifier diodes. I know this from experience.
If you changed the signal amplitude more often than that 8000 times a second mentioned above, some of the changes would be missed by the phone exchange. Likewise, if you had more than 128 possibilities for signal levels, some of them would look the same when sampled. So you simply cannot get more than 56000 bits a second down a POTS line
You could have a phone with the digital-to-analogue conversion stages inside the phone, connected straight to a B-channel and a D-channel of an ISDN line.
If you wanted higher speeds out of an analogue phone line, you could take advantage of the fact that much of the bit bottleneck is in the exchange, and that the line from your home to the exchange is just two copper wires which will usually handle much higher frequencies than your voice can generate or your ear can hear. You would have to modulate the data onto a much higher frequency carrier, combine it with the voice signal from your telephone, blast that down the line and use filters to split the signals apart again. Then the high (data) frequencies could be sent to a special demodulator, which could recover the zeros and ones and chuck them at a router; and the low (voice) frequencies would just go to the ADC as before. You don't even have to use phone lines, you could use any wiring that will handle a few megahertz
Oh, wait a minute
Ever since my boss got friendly with the Asterisk developers, my company's internal telephone network is now almost entirely VOIP. We have a server running Asterisk, with a Zaptel line card (needed a 3V3, 6MHz, 32-bit expansion slot; something you apparently only find on high-end mobo's, as most of the low-cost ones are 33MHz and/or 5V) plugged into an E1 line giving 30 ISDN lines. But you only need this to connect to POTS phones -- connecting to other VOIP phones is just done over the internet. The Asterisk machine also currently runs our intranet, though I'm ordering a new server for all the non-telephonical functions as something keeps crashing (not often enough to be serious, but we need to narrow it down).
..... we use dedicated hardware telephones. The Grandstream BudgeTone 101 was the first we evaluated, be aware that this comes with a Continental-style mains adaptor so you may need to get a new power pack (regulated 5 volts 400mA DC + --o)-- - polarity). This works lovely as a SIP telephone but doesn't as standard allow for a headset, which we kind of need in a call centre. The handset does use a standard RJ01 connector, but there seemed no easy way to deal with the receiver switch. We also evaluated every softphone we could get our hands on. In general they seem to be a bitch to get to compile; I had the best result with Linphone, it wasn't as polished as KPhone but it seemed to crash less often; and got absolutely gnowhere with Gnophone. Bear in mind also that a telephone headset will reveal the limitations of the sound chipsets on modern mobo's: you will require a real SoundBlaster-compatiable if you want to be able to understand what anybody is saying. I am running Debian Sid, my boss is running some perversion of Mandrake with a load of stuff from Cooker, and all our workstations run Mandrake 9.2 (hackerish systems are fine for us hackers, but it's more important to have Stuff That Just Works for the masses). We also got a softphone client from Zultys, called LIPZ; which looked stunning but was problematic in practice. It seems to bogart memory and CPU cycles. And when I came to do some hacking on it, I found the real kicker: it doesn't include the source code, so who knows what the hell it's really doing? In the end, we wound up using Zultys ZIP4X4 hardware SIP telephones. These are very expensive for "just" a telephone, but they are stuffed with features, all known codecs, 4 virtual lines, even an integral 5 port (one for the phone, four brought out on RJ45 jacks) 100Mb/s switch, and they are hardware -- my favourite programming language is still 63% tin, 37% lead.
As for phone clients
My honest recommendation would be go for something like the Grandstream, which does everything an "ordinary" phone should do and, being hardware is truly cross-platform. But note, it doesn't have any integral switch so you will take up an extra jack on your ADSL router.
If you did that, then it would not be free anymore. So why would you want to?
The whole point is that everyone should have access to the source code. The compiled version is just there for convenience's sake.
What you are saying is a bit like "What good is a marriage certificiate if my wife can still refuse to have sex with me if she doesn't want to?" It doesn't work like that!
Just because A happens and B also happens, doesn't mean A is the cause of B. If you're still not convinced, I've some elephant repellent you might be intrested to buy
I don't understand it, either. But then again, my company has a written policy explicitly favouring Open Source -- which I wrote
How about everyone just cuts their losses. Live with the fact that artists are not going to get compensated for their work.
It's tough, but them's the breaks. The only reason anybody ever made money out of distributing recordings of artistic performances was that the ability to manufacture recordings was a scarce commodity. With the ready availability of CD and DVD recorders, that is simply no longer the case.
So let's let the whole music industry just pack their bags and go home. It's been a good ride while it lasted, but now it's over. So long and thanks for all the tunes. The world will not end just because there are no more Britney Spears clones, and people will not stop playing good music just because they are not getting paid for it.
The product of cost of recovering from virus attacks on a Linux network times probability of occurrence is negligible, but good luck convincing a Gates-worshipper that it's so. Agreed. I was holding my tongue to avoid accusations from the closed-source fanboys; but now you have spoken, I feel free to say: GEDDINTHEREMYSAN! Another reason why bad code pravails in closed-source projects is that, since developers aren't allowed to borrow code that is recognised to have been Done Properly {either because the good code exists in a closed-source programme which they are not allowed to look at; or -- slightly, but deliciously, ironically -- because the good code exists in a GPL programme but the developer insists to release a closed-source programme}, everyone has their own go at inventing the wheel from first principles; and, barring supreme good luck, introduces their own special little errors.
Anyone noticed that almost no notebook has a line-in port these days? (Mic inputs are mono, preamplified and band-limited to fuck.)
I don't know about pre-built desktops {never bought one, never likely to}. Mobo-integrated sound chipsets do have a line in, but it picks up so much static and power hum that it's not worth using. In my experience, they're more for looking at and saying "look, it has these all these ports built in already" than for actually using. Even separate sound cards can be noisy, as none of them have stuff like tin cans and power chokes anymore. And the old ones that did, were all 16-bit cards and new mobos are all 32-bit. Unless someone makes a converter to plug old 8/16 bit cards into a 32-bit expansion slot? No, thought they didn't. Pity really, 'cause the old 16-bit bus made it extraordinarily easy to interface to homebrew hardware. Not that it would have made any sense not to; because up until then, they basically were still trying to encourage people to invent new stuff so as to sell more PCs, whereas now it's not in any established manufacturer's interest to have people inventing new stuff in case it takes away sales.
I saw a copy of [Nievy Ynivtar]'s new album "[Haqre Zl Fxva]" with a big copy-protection warning splattered all over it in H.M.V. a couple of weeks ago, and I just had to buy it for the "hacking challenge" factor.
.toc and .dat pair that burned OK.
..... this was not so much shooting fish in a barrel, as standing by a barrel of fish and waiting for them to jump out.
..... though actually, my DVD recorder is built only to recognise the first session on a multisession CD {to paraphrase the manual: if you want to make an MP3 CD to play on this machine, you must burn all the songs to the disc in one go, otherwise it will only see whatever you recorded up to the point where you first ejected the disc} which sounds like they were anticipating some sort of copy-prevention attempts involving a bogus second session. And the machine even has a digital audio-out which I haven't investigated; I think it's electrical rather than optical. Anyone know of a good sound card with digital-in and full Linux support including open source drivers?
Cdparanoia read the audio without a hitch -- all spaces, not even so much as a single minus sign, and cdrdao made a
I still felt an overwhelming sense of disappointment
Of course I made sure to check I could get a refund "if it wouldn't play on my DVD recorder". The warning said it might not work in anything other than a home audio CD player
Disclaimer: by reading the ROT-13ed text above, you are agreeing not to laugh.
When I was 14 years old, that would have been 1985, I invented a self-destructing audio cassette. It was a regular audio cassette with a piece of ceramic magnet glued inside the shell just upstream of the takeup spool, where the tape would pass over it during playback. This had the effect that the recording could be listened to once only (at least without my "specially-modified" recorder).
I only used it a couple of times for a bit of fun, but does it count as prior art?
I originally thought of using it for computer games that would transfer themselves to disk (so you could buy a tape, which was cheaper, then have the game save itself on a disk you had purchased elsewhere more cheaply than the software companies, who used to have to charge an extra GBP4.00 to cover the cost of the horrendously expensive floppies they used to use). But you would only be able to play it once, so you would only get to make one copy. Of course I realised as soon as I had thought of it, that it was flawed, because the cassette doesn't know if the audio signal is going into one computer, fifty computers with their tape-ins wired in parallel, or another tape deck with a non-self-destructing cassette inside it.
The article says there is a small ampoule containing a chemical, which is released when the package is opened (the DVD is screwed in place to begin with) and attacks the reflective coating on the top surface of the disc. In one of the two they received, the process had been started already, probably by improper tightening of the nut. The other one was still playable after the 8 hours were up, just by skipping the introduction and going straight to chapter 2.
Verdict, in any language:
:(
Build air-tight cabinet around DVD player with special rubber gloves for manipulation. Put DVD inside cabinet. Close seal. Flush cabinet with nitrogen. Set pressure regulator just above 1 atmosphere (in case of any leaks). Put hands in gloves. Unwrap DVD. Play. After watching, put DVD in air-tight container. Charge with extra nitrogen.
All that nitrogen may be getting a bit expensive, so:
Buy the disc. You have 8 hours to make a longer-lasting copy of it. Return the disc to the manufacturer at their own expense, claiming it is faulty. Watch copy at your own leisure.
Or, if mindless destruction is more up your alley:
Find a stack of them in a store. Puncture the wrapping with a pin (or, for extra shock value, a syringe).
Christians like to pretend they have a monopoly on good deeds. This is not so. In fact, most christians really are not nice people, and this is exactly the behaviour I would think to associate with christians.
At the end of the day, christians are some of the most self-centred people you could hope to avoid having to meet. If an atheist does you a favour, it is because they want to do you a favour. If a christian does you a favour, it is because they think it will improve their own personal chance of getting into heaven. No more, no less. If their priests told them they could get into heaven by stealing hub caps off people's cars, that is what they would do.
I found a couple of problems with Konqueror. The company where I work used to use an old, obsolete (but it did most of what they wanted) Windows application, together with some Perl on a web server -- till I replaced it with a roll-me-own solution, using mostly PHP with a bit of Perl and MySQL. Eventually it got to the point where we no longer needed Windows at all.
Unfortunately, my Web based front end used a fair amount of JavaScript, and did things like copying data from a form on one page into a different form on another page and triggering a submit when a button was clicked on yet another page. It was fine in Mozilla (my testing platform), and what was then called Firebird, but Konqueror barfed. I'm presuming this is due to overly-paranoid security settings. Cross-site scripting vulnerabilities are one thing; but on an Intranet with just one web server, it doesn't matter. Mozilla correctly deduced that the servers were one and the same and permitted the operation. {BTW, I fully expect someone to explain in one paragraph how to fix this.}
Also, Konqueror and Mozilla behave a bit differently with some JavaScript / CSS fancy menu code I wrote {this wasn't related to my "main" work}. But I can't see what bits aren't valid, so maybe it's KHTML's fault?
Installing Linux is only the beginning. There is going to be a period of transition where people have to learn to use the new systems. It's not as bad as some people seem to think -- I have first hand experience, having to support ex-Windows users in a nearly-all-Linux company {Windows is still used by about half a dozen people, for a few legacy applications; but you can bet your arse, we fully intend to make use of our statutory right to perform reverse-engineering for the sake of interoperability -- which is not affected by anything the EULA may say .....} and once you have explained the obvious differences, people will get used to it.
Legacy documents will have to be converted to the new standard, quite probably by retyping them from scratch. This will represent a substantial one-time cost; but once documents are held in an open format, the hardest bit of the work is over. There is a big decision to be made: for each document, should we (a) go to the effort of converting it accurately, (b) put up with some idiosyncrasies arising from an imperfect conversion or (c) scrap it. Such triage itself costs money, but it might lead to a considerable saving if there are many B or C answers.
Linux maintenance is an ongoing cost -- and it does tend to be more expensive, when it is required, than Windows maintenance, but that is to be expected. Linux doesn't go wrong very often; and when it does go wrong, it usually goes wrong for a reason. Windows is inherently unreliable. Sorry to upset the MS fanboys, but it is. The reason is that it is closed source. Application and driver developers can't see the OS code, and have to trust that it matches the documentation exactly. OS developers likewise have to trust that application and driver developers have specified their interfaces correctly. What happens in practice is that you get "logic traps" -- sets of ordinarily innocuous events which, if and not unless they all happen together, will cause something different to happen instead of what should have happened. It may be as obscure as pressing the W key on a Belgian keyboard as the CRT raster comes to the end of the 87th line while the sound card is playing F# above middle C out of the right hand speaker and a PCL5 printer driver has been swapped out from memory -- and the only way anyone could ever spot it, short of actually experiencing it, would be to look at the source code for the OS and the keyboard, graphics card, sound card and printer drivers. It's not hard to see how the severity increases geometrically as more devices and programmes are added to a system. Even motherboard drivers are not immune.
So Windows has these designed-in random crashes that happen for no reason other than source code bogarting, and they account for by far the vast majority of Windows crashes. Without the source code, the only thing you can do is tut, scratch your chin, and whip out the power lead. And a trained monkey could do that. Replacing the hardware, or installing a newer version of the software, will cure most of the worst ones by introducing a different set of trigger criteria. Linux doesn't have the same potential for hidden logic traps, because the developers can see one another's code and so prevent them from ever arising in the first place. Of course they may be present in early versions, but someone is certain either to fix the problem themselves, or alert someone who can fix it. If something is amiss with Linux, there is likely to be a reason which will not go away if you just cut the power. And fixing it takes someone who really groks the system, bellyfeels it, is one with it. That costs money -- but since Linux doesn't have built-in unreliability, most of the time it will just work. This means you will need fewer, more expensive support personnel, as opposed to more, cheaper personnel with a comparable Windows installation.