There is a well-known and expensive requirements capture, documentation and traceability tool known as DOORS. Although the underlying database is secure and completely trustworthy (unlike Windows or Gates), it is expensive, its user interface is diabolically bad, and its interface to the generic terms "word" or "office" (to which I assume M$ have no rights) is even worse. (Sad, because it is the only credible tool so far in that area. No doubt an OSS equivalent will eventually appear, most of the infrastructure of such a program must already exist as OSS.)
In my last job, there were comments frequently made by the engineers, such as "Who needs DOORS, Windows or Gates?". We looked for, but never found, any more software packages with names relating to points of entry through barriers. I have for example not seen anything called "portcullis", it would perhaps be fun to write such an application.
Who is going to use the term "word" next, I wonder?
But, you may be right about non english countries, although if Windoze is called by some local translation, Lindows would not resemble that translation and so would not infringe. It is also likely that the local term for windows has been used generically. I don't know if M$ name their product in the local language or not, if they do it will only cost them more in legal fees. But, if they have any sense, they will capitulate when they have lost one or two major markets.
I predict that M$ will in fact lose the rights to the clearly generic name in the end, followed likely as not by word, office, works, access, excel, paint..... They can probably keep Wordpad, one of their better achievements in any case.
But, when they lose the branding, they will lose the corporate monopolist identity, followed soon after by the business, and good riddance. We don't need bug-ridden, time-wasting insecure trash.
Perhaps it is because everything that the Convicted Monopolist does has negative impact on the world as a whole. Like much of their business, even the very name of their vilest. most bug-ridden and insecure product, has not been obtained legally. Their current strategy, because they have now wakened up to the fact that they are incapable of developing good software or producing any form of innovation, is to try to bleed competitors dry with long and tedious lawsuits. This has diversified into funding others like Darl McFraud to do it for them. As an organisation they are morally and intellectually bankrupt, hopefully they will be economically bankrupt when Longhorn fails to gain acceptance due to breaking compatability with all that went before. The world would be a far better place without M$ or Gill Bates.
The fact is that windows were in use on the MAC and before that, the Lisa, and in the form of Xwindows on Unix. They were called windows. They may also have been used on other computers not so widely known, but the term was certainly generic. IIRC the dos extender/task switchers which existed before Windoze used the term. Desqview was probably one. So I think did GEM, which surely predated Windoze, and BTW was a lot more efficient and stable. It also had a far tidier and more coherent set of APIs, with teh result that developing software for it was far easier and cheaper than for the disorganised mess of disjointed dlls commonly called by the generic name windows. I may have an old GEM manual somewhere at home, it would be sufficient if that manual mentioned opening and closing windows, and was dated pre Windoze 1.0. End of story. End of corporate identity. End of the worst business ever to have existed. And good riddance!
I hardly think Fedora Core 1 can be held up as a shining example of how to do anything. It is the most badly broken distro I have seen in years, complete with an update system that is based on the vile RPM, with some HUGE binaries. Most of the world is still on dial-up at 56k or less, most ISPs in the world have a 2-hour timeout (or less), and there is no way of resuming an attempted update when it times out. In short, it does not work. I wasted several weeks on Fedora, half of the configuration utilities one expects in a modern distro were not there. It went to/dev/null where it belongs, in the end.
The only sensible way to work is with source patching, it is far easier to download smallish source patches and compile in situ than to solve the stupid issues of large binary downloads. Most end users would neither know nor care how it worked, as long as it worked. Most PCs are more than capable of supporting the C compiler, or whatever is needed in each case. A source patching system can be made to work transparently to the user, anything involving excessively large downloads can not be made to work. It is as simple as that.
This issue is one of the few serious problems that are hindering the uptake of Linux on the desktop. The other major one I have already referred to, badly broken configuration utilities. It may seem a boring issue, but it deserves the full attention of the best programmers out there, and for that matter the best user interface experts. Get it right, i.e. better than the efforts of the Convicted Monopolist, and we will see Linux on most desktops. Continue to get it wrong, and the Convicted Monopolist will remain exactly that, with the desktop monopoly remaining for ever.
"A lot of W3C standards seem overlooked by some pretty big sites."
Only because they are designed by morons who think that following the Convicted Monopolist is the One True Way, and are wilfully ignorant of the official standards. Some of them even use Frontpage, which must be about the worst web site development tool ever invented.
All web sites are supposed to be viewable by anyone, so they need to be validated to W3C standards, not for compliance with a nasty, bug-ridden browser like Incompetent Exploder. The only room for argument is to which W3C standards you should work, because people with older or even text-only browsers ought not to be excluded.
Ideally, no site would be allowed on the net unless it met the standards in force at the time.It would save a lot of trouble, and annoyance for users.
People ought to complain to the webmasters of sites which are non-compliant to a serious degree. If enough people bother them, they will have to fix it eventually.
Moves towards new standards, although not intrinsically a bad idea, need to be made carefully, so that alternative content is provided for those who can't access the new standards. That is very important. Unlike many things in life, the web has traditionally been inclusive, let us keep it that way, not make it exclusive.
Most importantly, the Convicted Monopolist must be prevented from asserting any proprietary standards, he has already done far too much damage with his vile perverted versions of Java, Javascript, non-standard HTML and other seemingly smart ideas which just cause misery for those who need to be standards-compliant.
You badly need a UK-style Data Protection Act in the US. Here, even an accidental leak of personal data is a serious criminal offence, and rightly so. It is not even legal to keep data for longer than strictly necessary.
Why do so many millions of people in the US put up with such a rotten system? Surely your politicians need votes to be elected (not excepting Dubya of course, as he was in fact not elected, and does not lawfully govern the country), so they must be susceptible to pressure from the millions who are fed up with the present system. Yet you seem to elect the same kind of vile scumbags, who give you bad laws, every time. Why?
For your own sakes, do something useful with your votes in the forthcoming elections. For a start, get rid of the unelected, mentally retarded, warmongering fascist scumbag that poses as president. An idiot like that would be unemployable anywhere else in the world, yet you almost made him president, a role which he then misappropriated as it was not his. You let him do that. It makes no sense. He should have been removed from office the day the votes were properly counted. Now he is playing with his dangerous toys worldwide, and can be utterly relied upon to do everything possible to create war and destruction. The man is sick, and should be in a secure mental hospital, yet some of you will vote for him again.
You will only get what you deserve if you don't vote, or vote for an imbecile.
It depends whether you really mean Javascript, Jscript or ECMAscript.... Are there any more variants?
Thanks of course to the malignant incompetence of the Convicted Monopolist who has to pervert every possible standard. I fail to see why, because no-one, not even the Illegal Monopoly, gains from this sort of thing, and it makes programming a lot of low-level things sheer misery, because you have to allow for the non-compliance of M$. The easy, and preferable, solution is to only use Mozilla, or other browsers which make good attempts to be standards-compliant.
Yet another reason for getting rid of Inept Exploder.
I think most people probably do have Java installed, as the majority of PCs in the world pre-date the court case between the Convicted Monopolist and Sun, and prior to that, most installations of Inept Exploder would have been accompanied by Java. I could be wrong of course, because my own experience is as far as possible with standards-compliant browsers and JVMs. I have missed almost all of the bugs and security problems, life is too short without any of these....
What I don't understand is why the world does not kick Bill in the teeth for his rotten, pernicious habit of wrecking everything which he did not invent, in some kind of apparently psychotic, infantile tantrum. The sensible thing to do is to dump his products, particularly the third-rate, insecure things that you don't need, like IE, and even more importantly, Lookout.
You mean Microsoft. Sun has nothing to gain or lose from this. Microsoft have already funded SCO. Darl is simply Bill's puppet, to be used against Linux, which he can't compete with in any other way. It also isolates Bill from a lot of the risk.
M$ have failed technically, and are spending their funds at a high rate. People are sick of their rotten, disfunctional, insecure products. The fact is that to survive, M$ needs a new business model, because they are not capable of competing on product quality, thanks to incompetence at the top. They are attempting to create a new model, which is based on suing others by proxy. Darl is the proxy, this time. There will be more.
But when users see that Longhorn breaks backward compatability, Outlook remains insecure, IE remains useless, bug-ridded and also insecure, etc, they will get fed up and adopt the several viable alternative solutions. In the end, M$ will implode upon itself, there will be a mass exodus of employees whose stock options are losing value so fast that they are working for less than nothing, and the end will come surprisingly quickly. But, in the meantine, they will have employed Darl.
The best reason to do a kernel recompile would be to allow source patching, instead of huge binary downloads for every upgrade. Unfortunately most Linux distros don't do it that way, they prefer the huge download of a new kernel and probably a new set of modules every time. They could learn a lot from FreeBSD, for example.
It is much quicker on a modern PC to download some source patches, apply them and recompile, it can be automated, and it might even encourage people to keeep their security patches up to date. If the process is properly managed, it can be much quicker than loading all those interminable patches, and patches to the patches, from the Convicted Monopolist. If the maintenance of Linux is made to be easier than Windoze (it cam be, IMHO), then it will eventually be used more than Windoze. This should be the area where developers focus their attention next.
I bought it because I was fed up with the bugs in 98 (with all the latest SPs). The computer press at the time said it was wonderful. The fact is that it was more unstable than everything that went before, or since.
I originally upgraded to 95 because 3.1 was unusable on decent screens (eg 1280*1024, 24 bit colour) because the icons became black rectangles due to the limitations of a 64k segment somewhere, not to mention the fonts...... (and 32 bit should in theory have been faster). 95 was useless, with every SP applied there were still constant BSODs and freezes, the TCP/IP memory leak was never fixed, and lots more. I upgraded to 98. Guess what, NONE of the serious bugs were fixed, and there were lots of new ones.
Meanwhile, I got a laptop with 2000, it also had lots of bugs, including strange freezes for about 10 seconds, but with every SP (totalling about twice the size of the OS IIRC) installed, it was just about tolerable. So, I upgraded the old desktop to XP, which promptly crashed and lost all the passwords so anyone could log in. It now runs Xandros Linux, (rock-solid BTW) with one manual tweak to make it the internet gateway (ICS is about the only thing that Xandros seem to have forgotten, or maybe I have just not found it yet), the laptop is now dual-booted with SuSE 9.0, the new desktop runs only SuSE 9.0, and some old machines run various Linux and xBSD versions.
I am well pleased with Star Office and OpenOffige.org, but a lot of old documents are in Word perfect, which I also like. Once I get Word Perfect 2000 running under Xandros and SuSE 9, all trace of Mr. Gates will be removed, except that I will keep Word 2000 for special occasions, running under Crossover Office, when I want to remember how bad it was.
The common thread running through all of the M$ problems is that they tried, again and again, to be too clever, and as always, failed miserably. The apparent removal of DOS was one of these attempts.I will never again purchase a M$ product. All my trackballs (much better than mice, give one a fair trial if you have not used one before) are the Logitech variety, which M$ made a very poor copy of, and claimed to have invented, so they will not even sell me any hardware. On second thoughts, I may change my mind, I may get some hacked X-boxes, which should make nice cheap, if low performance, web servers, because they are a loss-maker for M$....
Being mainly an analogue designer, I find this very amusing.
But there must be a solution. There are generally an almost infinite number of ways of doing something in hardware, as in software. Don't know much about the Amiga, but clearly the internal clocks would need to be synchronised to the incoming video, same as in an audio application. To get 16-bit audio with no corruption, at 44.1KHz sampling, needs about 200pS maximum jitter.But, I don't see why video needs to be so precise, well under one pixel would probably be about OK, say 10nS. The eye is much more easily fooled than the ear. You barely need 8-bit DACs either (24-bit colour), 7-bit is probably adequate.
I guess the Amiga uses a PLL to synchronise everything to the incoming video. It would be hard to do that on a modern PC, you would need to butcher the motherboard to inject your own clock signal, and the crude PLL clock multiplier in the CPU would probably mess it all up again. Sometimes older technology is best.
Proper video editing cards for PCs are expensive because they have to run synchronously to the video, not the PC, amongst other things.
Still, it is good that this software has been released. I wonder if the price of used Amigas is going to rise, as everyone wants a toaster?
McBride has not defined his terminology, simple as that.
Actually I can't see what you would necessarily need to do enterprise-scale testing, except lots of PCs, with lots of different configurations. Is that not how free or open-source software is tested anyway, by thousands or millions of testers, some of whom have state of the art equipment, others anything from Pentium 4 or Athlon down to old 386s?
The test coverage, although informally organised, must be orders of magnitude better than what is achieved by, for example, the Convicted Monopolist (and it shows, I have yet to experience a kernel crash in several distros, althout I have seen Xfree86 crash a couple of times).
I don't see what could be added by throwing expensive equipment at the problem, except maybe in areas like the F00F bug of a few years ago, and certain device driver work, where things like emulators and logic analysers can be very useful, but some developers have access to these things at work or university.
I would like to know what magical facilities SCO have which are not available to a proportion of Linux developers. I generally work in compamies which develop safety-critical hardware and software, even there we can accomplish all of the testing required by the certification authorities, and the extra bit for our own conscience and reputation, without anything mega-expensive or unusual. Low budget jobs, without all the fancy hardware, simply take a bit longer.
The kernel developers, if you include all who contribute code, or submit bug reports and suggest fixes, have an enormous number of man-hours at their disposal. Admittedly, most of the coding work is done by a smallish team, but I doubt that anyone has a full grasp of how many people are testing, they only become visible when they find and report a bug. Careful, conscientious work can accomplish a great deal without fancy facilities, and many of the people in the background, as well as the visible ones like Linus, are very talented indeed.
If they need fancy facilities, I am sure that the FSF or one of the other organisations will advertise for funding, I for one, as an end user who likes good code, would make a modest donation if it was needed.
Perhaps Darl could make a list of what he actually has at his disposal, that free software developers do not have access to? Or is he talking out of the wrong end of his anatomy again?
Funnily enough, I know some people who do that with their telephones, to block double glazing salesmen and other callers they don't want, by simply not picking up the phone on the first set of rings. It is not specific to computers, in fact it has probably been used since before computers were invented, but nor is it foolproof, although it does get rid of a useful number of unwanted calls.
Is it possible to run a 1 pixel monitor in X? I suspect creating the correct modeline would be an interesting challenge!
A tricolour LED, buffered as appropriate, wired to the Red, Green and Blue DAC outputs of any crummy old graphics card would do, provided you could set up a 1 pixel screen.
I know you can set up a 1 pixel window, trojans, spyware and nasty things do it all the time in a certain badly broken OS, but not so sure about the monitor size.
Of course I jest, it is much better to use the grotty old card to drive a grotty old 640*480 monitor with all sorts of possibilities.
But I don't actually see the point of a device like this anyway, just because it can be done, and technology could have done it many years ago, does not mean that it is useful.
Wrong. If you ignore the HF components you will find out the hard way that it is in fact an analogue world, becuase your digits will not work correctly, if at all. The more you attempt to make the analogue world approximate to a digital work of fiction by trying to generate real ones and zeroes, with no intermediate transitions, the more obvious it becomes that all digital signals are in fact special, limited cases of analogue signals. The reverse is simply not true.
The distinction is not always obvious. Your oil painting, being in an invariant state, can be considered to be digital, insofar as each molecule has its fixed position (OK, at absolute zero, it is a noisy fixed position otherwise). But, look at your stream on 1s and 0s on a good oscilloscope and you will see a succession of what are clearly analogue levels, with finite edges, ringing, noise.......
The issue is whether one or the other is a more accurate and therefore better expression of its designer's intentions. Both can be shown to have serious deficiencies. Analogue and digital quantities are merely controlled manipulations of the underlying statistically exact but deterministicaly inexact thermodynamics. A digital quantity, in the sense of for example binary data, always resolves into a finite number of analogue quantities, which in turn are mere statistical approximations to binary states.
For some reason, the salaries of Analogue designers are catching up with those of Digital designers, at least in the UK, so maybe it is finally becoming apparent that all real-world interfaces must be predominately analogue, because the gap between digital and thermodynamics is too great to bridge any other way. You can't have an individual digital device manipulating each air molecule in front of a loudspeaker, for example.
I would challenge anyone to design a digital system which does not rely on some analogue actions for its ability to function, e.g. the need for a reset pulse or power good signal, and a clock oscillator. Now these may be improvised with logic gates, that is not what I mean, I mean pure logic, no RC networks or any other analogue fiddles to get it initialised to the desired state, and then clocked. I don't know if it is mathematically provable or not, but experience suggests that it can't be done. One concession would be that the power supply can be assumed to be applied instantaneously and at the required voltage, even then it still can't be done. How do you get a flip-flop to set itself one way or the other, without designing in an analogue bias, or using a reset signal?
The wheel is turning full circle, and analogue will rule once more!
Incorrect, the world on that level is part analogue, part thermodynamic. A molecule can have any energy it wants, but the air pressure sensed by your ears is a statistical quantity, summed over many molecules which have random, not quantized, movement, and is therefore noisy, hence Brownian motion, which is what sets the threshold of hearing in a normal healthy person, or the sensitivity of an otherwise perfect microphone. But, as well as being noisy, air pressure and other physical quantities are also infinitely variable.
A speaker cone, and its effect on the air pressure, is not quantized, but rather analogue plus random, the random components coming from amplifier noise and Brownian motion of the air it is driving.
The movement of hairs in the ear is also mostly analogue. Older Slashdotters may remember analogue frequency meters, which had an array of tuned reeds covering a fairly narrow band around the power frerquency of 50, 60 or 400Hz, and by looking at the white painted ends you could see a crude spectrum display. The reeds might be tuned in steps of 0.25 Hz, but the eye would pick out the curve and interpolate its peak to less than that. The ear mechanism is similar, there are many overlapping tuned filters, but the amplitude detection is at least semi-analogue.
The real world only becomes digital when you get down to doing things like counting photons, for example in astronomy using an image intensifier in front of a CCD, but even there the distribution in both time and frequency/energy of these photons is thermodynamic or if you prefer, statistical.
Many digital designs fail or are troublesome because the whizz-kids who think they know VHDL and can drive a toolset, forget about ANALOGUE things like transmission line effects on the PCB, or statistical things like timing jitter, which is ultimately caused by amplitude noise, a statistical thing, being summed with a signal of finite slew rate.
It is DIGITAL systems which are unable to make a sufficiently exact representation of the real world, for example 16 bit encoding is woefully inadequate for high quality audio.
Energy also comes in any quantity we like, the physical velocity of hot gas, for example, contains a random element, but because of partial ionisation its summation is not limited to the sum of permissible quantum states of individual atoms, which of course would be quantised.
So, it is a noisy analogue world, and digits superimposed thereon are simply discrete groups of noisy analogue levels. If you don't believe that statement, look at a PCI bus on a good analogue oscilloscope!
If you want to see something else which in theory should be digital, and therefore precise, exact and consistent, look at the trash software products of the Convicted Monopolist. Here the statistical component is indirect, caused by numerous semi-random human contributions at the design and coding stages, and a large degree of something which is neither analogue nor digital, nor even statistical, neither is it a variable (because its purpose is truly deterministic/monoploistic), nor a constant (because it produces output which is truly random insofar as it is totally uncorrelated with anything else), but is in fact the product of pure greed (a quasi-digital quantity, usually approximate by 1), pure arrogance(a dimensionless variable approaching infinity) and total incompetence (a dimensionless variable approaching zero). I will call this new property of matter Billness.
Re:Maybe time to drop this "securitier than thou"
on
Remotely Crash OpenBSD
·
· Score: 2, Insightful
Tha analogy would be the way the press treat road and rail accidents. In the UK (BTW no passengers at all were killed in crashes last year) it is headline news for weeks, and then again all through the inevitable pubilc enquiry if 4 people are killed in a train crash, yet IIRC on the same day, or maybe the bnext day as 4 were killed in the crash I am thinking of, at least 10 died on the roads, 6 in one vehicle. That one got a small paragraph.... The average is 10 a day in the UK on the roads, about 2 or 3 per year in trains.
Now the specialist press, including web sites, who know of the existence of OpenBSD, are likely to treat this in much the same way. A BSD crash, any variant, is a rarity, 1000 times or more less likely to happen than a BSOD. Same sort of ratio fro security holes also. So, the same thing happens, the uncommon major event gets the attention, although it does far, far less harm overall than the very common everyday event.
Of course in this case the normal press remain in utter ignorance, some of them may know that Windoze is not the same as a MAC, a few will know of Linux, and very few indeed will know what BSD is, they probably think it is a shorter abbreviation for BSOD. So, the mainstream press will leave this well alone.
It is quite right and proper that crashes should be reported, and certainly it is only fair that a problem with a secure OS gets to be known, and fixed, but like the train crash, it needs to be kept in perspective.
I know that Theo allegedly has an attitude problem, however those who extrapolate from his remark that it is only a crash to suggest that he does not care are IMHO quite wrong. I think he was only putting the event in its true perspective, as being of slightly less importance than a security breach. I think he does care, very much, that "his" software works properly, that is what drives such people, who could earn much more financial reward elsewhere.
All of this is a matter of seeing the thing in its true perspective. If people did that, no-one at all would use the products of the Convicted Monopolist, and the world would be a very much safer place as regards computer security, and much more productive because there would probably be only one crash for 1000 or even 1000000 BSODs in inferior systems, which are riddled with fundamental design errors.
Put small non-metallic items in the microwave for a few seconds, not long enough to get hot. The high field strength will kill silicon semiconductors easily. Of, if you know where the thing is, use an ESD gun, blast it with a 200pF or so capacitor charged to about 25Kv and see how it likes it....
In any case, to activate one of these tags, it has to be able to receive sufficient power to activate itself, meaning that the reader must be very close, or be using seriously illegal and hazardous RF power. Remember that an RF-powered device must receive at least one diode forward voltage drop, to be able to rectify its power, say 0.3V for a schottky diode, plus at least one base-emitter drop (0.7V) or MOSFET gate threshold voltage (probably more) before it can do anything anyway. The transmitter in the reader likely as not generates only an H-field, which will probably fall off with an inverse 4th power law. All of this works very much in your favour. Distance is the key thing. You will not be tracked in the middle of an open space, for example (they will use your cellphone instead, that can be tracked quite accurately any time it is visible to 3 base stations.)
I read somewhere recently that Linux was being used fairly extensively, and successfully of course, within the BBC. It seems to follow from this convoluted reasoning that they, or some of their employees, must be prime suspects.....
Of course, what it really shows is the abysmal ignorance of the author of this disgraceful article, of what his employer is actually doing, and probably ignorance also of what Linux, and open source in general, actually is. He will probably be confusing Tony B. Liar with a Socialist, or Dubya with an elected president, next.....
It gets worse. Most graphics and picture editing programs, or the various books and articles relating to them, all talk a lot of rubbish about resizing to so many dpi. It is utterly irrelevant, yet I have seen serious articles purporting to tell you how to prepare photographs for publication, which really labour the point. The fact is that your camera takes a picture, in its best mode, of in my case 2560*1920 IIRC, which should be retained through the entire process until the final resize, to avoid losing detail. The DPI is irrelevant, yet the instructions were to resize to a certain number of DPI (then a bit later resize again.....)
It does matter at the output process, typically a printer, where it is best to resize the image to exactly match the pixel resolution of the printer, or an exact multiple or sub-multiple thereof.
I wonder where all this junk does originate? Yes, it is in X, but the screen does tend to be set up correctly, or can be by someone of modest ability. In most cases, certainly in Linux, the configurator program (which varies according to distro) asks the actual screen size, and anyone competent to use a PC can also use a ruler. As for using a positive value to get points and a negative value for pixels in the same call, that is about as disgracefully bad and iresponsible a programming practice as you will ever see, but rather good by Microtrash standards.
It is a careful move to ensure that, like their junk OS, you will need to replace it every few years, because in this case the flash memory will wear out, and the cost of removing and replacing the chips will exceed that of a new machine. A hard drive would be replaceable by the user, although it could be made slightly difficult.
It is sad that Sir Bill and his cronies can put so much effort into a toy, but not do anything constructive about their vile, bug-ridden, insecure products, particularly Lookout and Inept Exploder, which continue to damage the world. It suggests that Sir Bill has in fact not grown up yet, either that or is so arrogant that he does not care how much destruction occurs as a result of his delibarately incorporated automatic virus downloading facility.
I hope that the next court judgment (there will be one, because he is ignoring the last one) does order the breakup of the Convicted Monopolist. Bill should be assigned to the toys division, and never, ever let near real software.
Meanwhile, I would suggest that M$ be known as a toy company, because that is clearly the only area where they have the slightest amount of competence.
Much as I would like to, even I can't actually find any way to blame SCO for this font change. I can't blame Sir Bill either. But, I think it is stupid, because 14 point is a bit too big and so wastes paper, not that the US government ever cared about waste.....
In any case, there are more visually pleasing fonts, and I see no reason why official documents should not look good. Some organisations use their own custom font, I would have thought that the US government could afford to pay for a real expert to come up with a good one, which might also be more readable by the visually challenged.
What president? You have not got a legally elected one right now. In any case, how can a moron like him sign anything? May as well use Wingdings, so he can look at the pretty pictures.
Slashdot Mirror
In my last job, there were comments frequently made by the engineers, such as "Who needs DOORS, Windows or Gates?". We looked for, but never found, any more software packages with names relating to points of entry through barriers. I have for example not seen anything called "portcullis", it would perhaps be fun to write such an application.
But, you may be right about non english countries, although if Windoze is called by some local translation, Lindows would not resemble that translation and so would not infringe. It is also likely that the local term for windows has been used generically. I don't know if M$ name their product in the local language or not, if they do it will only cost them more in legal fees. But, if they have any sense, they will capitulate when they have lost one or two major markets.
I predict that M$ will in fact lose the rights to the clearly generic name in the end, followed likely as not by word, office, works, access, excel, paint..... They can probably keep Wordpad, one of their better achievements in any case.
But, when they lose the branding, they will lose the corporate monopolist identity, followed soon after by the business, and good riddance. We don't need bug-ridden, time-wasting insecure trash.
The fact is that windows were in use on the MAC and before that, the Lisa, and in the form of Xwindows on Unix. They were called windows. They may also have been used on other computers not so widely known, but the term was certainly generic. IIRC the dos extender/task switchers which existed before Windoze used the term. Desqview was probably one. So I think did GEM, which surely predated Windoze, and BTW was a lot more efficient and stable. It also had a far tidier and more coherent set of APIs, with teh result that developing software for it was far easier and cheaper than for the disorganised mess of disjointed dlls commonly called by the generic name windows. I may have an old GEM manual somewhere at home, it would be sufficient if that manual mentioned opening and closing windows, and was dated pre Windoze 1.0. End of story. End of corporate identity. End of the worst business ever to have existed. And good riddance!
The only sensible way to work is with source patching, it is far easier to download smallish source patches and compile in situ than to solve the stupid issues of large binary downloads. Most end users would neither know nor care how it worked, as long as it worked. Most PCs are more than capable of supporting the C compiler, or whatever is needed in each case. A source patching system can be made to work transparently to the user, anything involving excessively large downloads can not be made to work. It is as simple as that.
This issue is one of the few serious problems that are hindering the uptake of Linux on the desktop. The other major one I have already referred to, badly broken configuration utilities. It may seem a boring issue, but it deserves the full attention of the best programmers out there, and for that matter the best user interface experts. Get it right, i.e. better than the efforts of the Convicted Monopolist, and we will see Linux on most desktops. Continue to get it wrong, and the Convicted Monopolist will remain exactly that, with the desktop monopoly remaining for ever.
Only because they are designed by morons who think that following the Convicted Monopolist is the One True Way, and are wilfully ignorant of the official standards. Some of them even use Frontpage, which must be about the worst web site development tool ever invented.
All web sites are supposed to be viewable by anyone, so they need to be validated to W3C standards, not for compliance with a nasty, bug-ridden browser like Incompetent Exploder. The only room for argument is to which W3C standards you should work, because people with older or even text-only browsers ought not to be excluded.
Ideally, no site would be allowed on the net unless it met the standards in force at the time.It would save a lot of trouble, and annoyance for users.
People ought to complain to the webmasters of sites which are non-compliant to a serious degree. If enough people bother them, they will have to fix it eventually.
Moves towards new standards, although not intrinsically a bad idea, need to be made carefully, so that alternative content is provided for those who can't access the new standards. That is very important. Unlike many things in life, the web has traditionally been inclusive, let us keep it that way, not make it exclusive.
Most importantly, the Convicted Monopolist must be prevented from asserting any proprietary standards, he has already done far too much damage with his vile perverted versions of Java, Javascript, non-standard HTML and other seemingly smart ideas which just cause misery for those who need to be standards-compliant.
Why do so many millions of people in the US put up with such a rotten system? Surely your politicians need votes to be elected (not excepting Dubya of course, as he was in fact not elected, and does not lawfully govern the country), so they must be susceptible to pressure from the millions who are fed up with the present system. Yet you seem to elect the same kind of vile scumbags, who give you bad laws, every time. Why?
For your own sakes, do something useful with your votes in the forthcoming elections. For a start, get rid of the unelected, mentally retarded, warmongering fascist scumbag that poses as president. An idiot like that would be unemployable anywhere else in the world, yet you almost made him president, a role which he then misappropriated as it was not his. You let him do that. It makes no sense. He should have been removed from office the day the votes were properly counted. Now he is playing with his dangerous toys worldwide, and can be utterly relied upon to do everything possible to create war and destruction. The man is sick, and should be in a secure mental hospital, yet some of you will vote for him again.
You will only get what you deserve if you don't vote, or vote for an imbecile.
Thanks of course to the malignant incompetence of the Convicted Monopolist who has to pervert every possible standard. I fail to see why, because no-one, not even the Illegal Monopoly, gains from this sort of thing, and it makes programming a lot of low-level things sheer misery, because you have to allow for the non-compliance of M$. The easy, and preferable, solution is to only use Mozilla, or other browsers which make good attempts to be standards-compliant.
Yet another reason for getting rid of Inept Exploder.
I think most people probably do have Java installed, as the majority of PCs in the world pre-date the court case between the Convicted Monopolist and Sun, and prior to that, most installations of Inept Exploder would have been accompanied by Java. I could be wrong of course, because my own experience is as far as possible with standards-compliant browsers and JVMs. I have missed almost all of the bugs and security problems, life is too short without any of these....
What I don't understand is why the world does not kick Bill in the teeth for his rotten, pernicious habit of wrecking everything which he did not invent, in some kind of apparently psychotic, infantile tantrum. The sensible thing to do is to dump his products, particularly the third-rate, insecure things that you don't need, like IE, and even more importantly, Lookout.
M$ have failed technically, and are spending their funds at a high rate. People are sick of their rotten, disfunctional, insecure products. The fact is that to survive, M$ needs a new business model, because they are not capable of competing on product quality, thanks to incompetence at the top. They are attempting to create a new model, which is based on suing others by proxy. Darl is the proxy, this time. There will be more.
But when users see that Longhorn breaks backward compatability, Outlook remains insecure, IE remains useless, bug-ridded and also insecure, etc, they will get fed up and adopt the several viable alternative solutions. In the end, M$ will implode upon itself, there will be a mass exodus of employees whose stock options are losing value so fast that they are working for less than nothing, and the end will come surprisingly quickly. But, in the meantine, they will have employed Darl.
It is much quicker on a modern PC to download some source patches, apply them and recompile, it can be automated, and it might even encourage people to keeep their security patches up to date. If the process is properly managed, it can be much quicker than loading all those interminable patches, and patches to the patches, from the Convicted Monopolist. If the maintenance of Linux is made to be easier than Windoze (it cam be, IMHO), then it will eventually be used more than Windoze. This should be the area where developers focus their attention next.
I originally upgraded to 95 because 3.1 was unusable on decent screens (eg 1280*1024, 24 bit colour) because the icons became black rectangles due to the limitations of a 64k segment somewhere, not to mention the fonts...... (and 32 bit should in theory have been faster). 95 was useless, with every SP applied there were still constant BSODs and freezes, the TCP/IP memory leak was never fixed, and lots more. I upgraded to 98. Guess what, NONE of the serious bugs were fixed, and there were lots of new ones.
Meanwhile, I got a laptop with 2000, it also had lots of bugs, including strange freezes for about 10 seconds, but with every SP (totalling about twice the size of the OS IIRC) installed, it was just about tolerable. So, I upgraded the old desktop to XP, which promptly crashed and lost all the passwords so anyone could log in. It now runs Xandros Linux, (rock-solid BTW) with one manual tweak to make it the internet gateway (ICS is about the only thing that Xandros seem to have forgotten, or maybe I have just not found it yet), the laptop is now dual-booted with SuSE 9.0, the new desktop runs only SuSE 9.0, and some old machines run various Linux and xBSD versions.
I am well pleased with Star Office and OpenOffige.org, but a lot of old documents are in Word perfect, which I also like. Once I get Word Perfect 2000 running under Xandros and SuSE 9, all trace of Mr. Gates will be removed, except that I will keep Word 2000 for special occasions, running under Crossover Office, when I want to remember how bad it was.
The common thread running through all of the M$ problems is that they tried, again and again, to be too clever, and as always, failed miserably. The apparent removal of DOS was one of these attempts.I will never again purchase a M$ product. All my trackballs (much better than mice, give one a fair trial if you have not used one before) are the Logitech variety, which M$ made a very poor copy of, and claimed to have invented, so they will not even sell me any hardware. On second thoughts, I may change my mind, I may get some hacked X-boxes, which should make nice cheap, if low performance, web servers, because they are a loss-maker for M$....
But there must be a solution. There are generally an almost infinite number of ways of doing something in hardware, as in software. Don't know much about the Amiga, but clearly the internal clocks would need to be synchronised to the incoming video, same as in an audio application. To get 16-bit audio with no corruption, at 44.1KHz sampling, needs about 200pS maximum jitter.But, I don't see why video needs to be so precise, well under one pixel would probably be about OK, say 10nS. The eye is much more easily fooled than the ear. You barely need 8-bit DACs either (24-bit colour), 7-bit is probably adequate.
I guess the Amiga uses a PLL to synchronise everything to the incoming video. It would be hard to do that on a modern PC, you would need to butcher the motherboard to inject your own clock signal, and the crude PLL clock multiplier in the CPU would probably mess it all up again. Sometimes older technology is best.
Proper video editing cards for PCs are expensive because they have to run synchronously to the video, not the PC, amongst other things.
Still, it is good that this software has been released. I wonder if the price of used Amigas is going to rise, as everyone wants a toaster?
Actually I can't see what you would necessarily need to do enterprise-scale testing, except lots of PCs, with lots of different configurations. Is that not how free or open-source software is tested anyway, by thousands or millions of testers, some of whom have state of the art equipment, others anything from Pentium 4 or Athlon down to old 386s?
The test coverage, although informally organised, must be orders of magnitude better than what is achieved by, for example, the Convicted Monopolist (and it shows, I have yet to experience a kernel crash in several distros, althout I have seen Xfree86 crash a couple of times).
I don't see what could be added by throwing expensive equipment at the problem, except maybe in areas like the F00F bug of a few years ago, and certain device driver work, where things like emulators and logic analysers can be very useful, but some developers have access to these things at work or university.
I would like to know what magical facilities SCO have which are not available to a proportion of Linux developers. I generally work in compamies which develop safety-critical hardware and software, even there we can accomplish all of the testing required by the certification authorities, and the extra bit for our own conscience and reputation, without anything mega-expensive or unusual. Low budget jobs, without all the fancy hardware, simply take a bit longer.
The kernel developers, if you include all who contribute code, or submit bug reports and suggest fixes, have an enormous number of man-hours at their disposal. Admittedly, most of the coding work is done by a smallish team, but I doubt that anyone has a full grasp of how many people are testing, they only become visible when they find and report a bug. Careful, conscientious work can accomplish a great deal without fancy facilities, and many of the people in the background, as well as the visible ones like Linus, are very talented indeed.
If they need fancy facilities, I am sure that the FSF or one of the other organisations will advertise for funding, I for one, as an end user who likes good code, would make a modest donation if it was needed.
Perhaps Darl could make a list of what he actually has at his disposal, that free software developers do not have access to? Or is he talking out of the wrong end of his anatomy again?
Funnily enough, I know some people who do that with their telephones, to block double glazing salesmen and other callers they don't want, by simply not picking up the phone on the first set of rings. It is not specific to computers, in fact it has probably been used since before computers were invented, but nor is it foolproof, although it does get rid of a useful number of unwanted calls.
Why would anyone except politicians want imprecise information?
A tricolour LED, buffered as appropriate, wired to the Red, Green and Blue DAC outputs of any crummy old graphics card would do, provided you could set up a 1 pixel screen.
I know you can set up a 1 pixel window, trojans, spyware and nasty things do it all the time in a certain badly broken OS, but not so sure about the monitor size.
Of course I jest, it is much better to use the grotty old card to drive a grotty old 640*480 monitor with all sorts of possibilities.
But I don't actually see the point of a device like this anyway, just because it can be done, and technology could have done it many years ago, does not mean that it is useful.
Wrong. If you ignore the HF components you will find out the hard way that it is in fact an analogue world, becuase your digits will not work correctly, if at all. The more you attempt to make the analogue world approximate to a digital work of fiction by trying to generate real ones and zeroes, with no intermediate transitions, the more obvious it becomes that all digital signals are in fact special, limited cases of analogue signals. The reverse is simply not true.
The issue is whether one or the other is a more accurate and therefore better expression of its designer's intentions. Both can be shown to have serious deficiencies. Analogue and digital quantities are merely controlled manipulations of the underlying statistically exact but deterministicaly inexact thermodynamics. A digital quantity, in the sense of for example binary data, always resolves into a finite number of analogue quantities, which in turn are mere statistical approximations to binary states.
For some reason, the salaries of Analogue designers are catching up with those of Digital designers, at least in the UK, so maybe it is finally becoming apparent that all real-world interfaces must be predominately analogue, because the gap between digital and thermodynamics is too great to bridge any other way. You can't have an individual digital device manipulating each air molecule in front of a loudspeaker, for example.
I would challenge anyone to design a digital system which does not rely on some analogue actions for its ability to function, e.g. the need for a reset pulse or power good signal, and a clock oscillator. Now these may be improvised with logic gates, that is not what I mean, I mean pure logic, no RC networks or any other analogue fiddles to get it initialised to the desired state, and then clocked. I don't know if it is mathematically provable or not, but experience suggests that it can't be done. One concession would be that the power supply can be assumed to be applied instantaneously and at the required voltage, even then it still can't be done. How do you get a flip-flop to set itself one way or the other, without designing in an analogue bias, or using a reset signal?
The wheel is turning full circle, and analogue will rule once more!
A speaker cone, and its effect on the air pressure, is not quantized, but rather analogue plus random, the random components coming from amplifier noise and Brownian motion of the air it is driving.
The movement of hairs in the ear is also mostly analogue. Older Slashdotters may remember analogue frequency meters, which had an array of tuned reeds covering a fairly narrow band around the power frerquency of 50, 60 or 400Hz, and by looking at the white painted ends you could see a crude spectrum display. The reeds might be tuned in steps of 0.25 Hz, but the eye would pick out the curve and interpolate its peak to less than that. The ear mechanism is similar, there are many overlapping tuned filters, but the amplitude detection is at least semi-analogue.
The real world only becomes digital when you get down to doing things like counting photons, for example in astronomy using an image intensifier in front of a CCD, but even there the distribution in both time and frequency/energy of these photons is thermodynamic or if you prefer, statistical.
Many digital designs fail or are troublesome because the whizz-kids who think they know VHDL and can drive a toolset, forget about ANALOGUE things like transmission line effects on the PCB, or statistical things like timing jitter, which is ultimately caused by amplitude noise, a statistical thing, being summed with a signal of finite slew rate.
It is DIGITAL systems which are unable to make a sufficiently exact representation of the real world, for example 16 bit encoding is woefully inadequate for high quality audio.
Energy also comes in any quantity we like, the physical velocity of hot gas, for example, contains a random element, but because of partial ionisation its summation is not limited to the sum of permissible quantum states of individual atoms, which of course would be quantised.
So, it is a noisy analogue world, and digits superimposed thereon are simply discrete groups of noisy analogue levels. If you don't believe that statement, look at a PCI bus on a good analogue oscilloscope!
If you want to see something else which in theory should be digital, and therefore precise, exact and consistent, look at the trash software products of the Convicted Monopolist. Here the statistical component is indirect, caused by numerous semi-random human contributions at the design and coding stages, and a large degree of something which is neither analogue nor digital, nor even statistical, neither is it a variable (because its purpose is truly deterministic/monoploistic), nor a constant (because it produces output which is truly random insofar as it is totally uncorrelated with anything else), but is in fact the product of pure greed (a quasi-digital quantity, usually approximate by 1), pure arrogance(a dimensionless variable approaching infinity) and total incompetence (a dimensionless variable approaching zero). I will call this new property of matter Billness.
Now the specialist press, including web sites, who know of the existence of OpenBSD, are likely to treat this in much the same way. A BSD crash, any variant, is a rarity, 1000 times or more less likely to happen than a BSOD. Same sort of ratio fro security holes also. So, the same thing happens, the uncommon major event gets the attention, although it does far, far less harm overall than the very common everyday event.
Of course in this case the normal press remain in utter ignorance, some of them may know that Windoze is not the same as a MAC, a few will know of Linux, and very few indeed will know what BSD is, they probably think it is a shorter abbreviation for BSOD. So, the mainstream press will leave this well alone.
It is quite right and proper that crashes should be reported, and certainly it is only fair that a problem with a secure OS gets to be known, and fixed, but like the train crash, it needs to be kept in perspective.
I know that Theo allegedly has an attitude problem, however those who extrapolate from his remark that it is only a crash to suggest that he does not care are IMHO quite wrong. I think he was only putting the event in its true perspective, as being of slightly less importance than a security breach. I think he does care, very much, that "his" software works properly, that is what drives such people, who could earn much more financial reward elsewhere.
All of this is a matter of seeing the thing in its true perspective. If people did that, no-one at all would use the products of the Convicted Monopolist, and the world would be a very much safer place as regards computer security, and much more productive because there would probably be only one crash for 1000 or even 1000000 BSODs in inferior systems, which are riddled with fundamental design errors.
In any case, to activate one of these tags, it has to be able to receive sufficient power to activate itself, meaning that the reader must be very close, or be using seriously illegal and hazardous RF power. Remember that an RF-powered device must receive at least one diode forward voltage drop, to be able to rectify its power, say 0.3V for a schottky diode, plus at least one base-emitter drop (0.7V) or MOSFET gate threshold voltage (probably more) before it can do anything anyway. The transmitter in the reader likely as not generates only an H-field, which will probably fall off with an inverse 4th power law. All of this works very much in your favour. Distance is the key thing. You will not be tracked in the middle of an open space, for example (they will use your cellphone instead, that can be tracked quite accurately any time it is visible to 3 base stations.)
Of course, what it really shows is the abysmal ignorance of the author of this disgraceful article, of what his employer is actually doing, and probably ignorance also of what Linux, and open source in general, actually is. He will probably be confusing Tony B. Liar with a Socialist, or Dubya with an elected president, next.....
It does matter at the output process, typically a printer, where it is best to resize the image to exactly match the pixel resolution of the printer, or an exact multiple or sub-multiple thereof.
I wonder where all this junk does originate? Yes, it is in X, but the screen does tend to be set up correctly, or can be by someone of modest ability. In most cases, certainly in Linux, the configurator program (which varies according to distro) asks the actual screen size, and anyone competent to use a PC can also use a ruler. As for using a positive value to get points and a negative value for pixels in the same call, that is about as disgracefully bad and iresponsible a programming practice as you will ever see, but rather good by Microtrash standards.
It is sad that Sir Bill and his cronies can put so much effort into a toy, but not do anything constructive about their vile, bug-ridden, insecure products, particularly Lookout and Inept Exploder, which continue to damage the world. It suggests that Sir Bill has in fact not grown up yet, either that or is so arrogant that he does not care how much destruction occurs as a result of his delibarately incorporated automatic virus downloading facility.
I hope that the next court judgment (there will be one, because he is ignoring the last one) does order the breakup of the Convicted Monopolist. Bill should be assigned to the toys division, and never, ever let near real software.
Meanwhile, I would suggest that M$ be known as a toy company, because that is clearly the only area where they have the slightest amount of competence.
In any case, there are more visually pleasing fonts, and I see no reason why official documents should not look good. Some organisations use their own custom font, I would have thought that the US government could afford to pay for a real expert to come up with a good one, which might also be more readable by the visually challenged.
What president? You have not got a legally elected one right now. In any case, how can a moron like him sign anything? May as well use Wingdings, so he can look at the pretty pictures.