The EULA restriction was in fact illegal. It is against some act that says that you can not dictate a supplier for another product in the process. That's why it was withdrawn.
The event occured around the early 1990's (late 80's MS DOS programs feature it. QHelp actually has an entry in its help system saying it has detected a non-ms opsys, and your licence may be void.
Microsoft products generally do not work as intended.
Case in point:
Write a script for a interpreted language (eg perl, rexx), and then try to pass parameters to it (eg REXX MYCALC 3+3). The thing works, and displays the result, which you can just see as the window closes. This is documented in Technet for Winnt 3.1, it is still there in NT4, 2K, and presumably, XP.
I find the spam somewhat worryingly high, rather than low. Since the filter is the subject line must contain "Microsoft Settlement" or something, this means that over 1000 spam messages had been modified to include this in the title. This is disturbing.
The subject "Hose your grandmother's account" would be filtered out because it does not contain the required subject header.
Of those who said "I hate Microsoft" or "Linux Rulz", these give no constructive comment on either the settlement's comments or what has been excluded. Simply saying "I hate (some company)" may be an ethical statement that you hate them because they sell a product you hate, and is no indication that they are break the law. eg, "I hate Ford", because they sell cars, and I hate cars. This is not a reflection on Ford's business practices.
My comment largely centered on possible antitrust comments in upgrades. For example, there is nothing stopping MS from doing things in "required" upgrades, such as shutting down competitive dual boots [Win2k], applications, &c. Upgrades and retail versions should be subject to the same technical restrictions as OEM versions viz Abiltity to not install assorted middleware, honouring multi-boots, etc.
The units you refer to belong to a single system: the CGS-Practical system. Volts and Ohms are older than SI or MKSA, the former date to 1860s, the latter to 1904 at the earliest.
Formerly, the practical electric units were additionals like miles and hours. That selected emu have names (Oersted, Gauss, Maxwell), and the absence of names for more obvious measures (eg charge), is the give-away here. Basically it was cgs + Practical units + some named cgs units.
You may be right: it's been 20 years since I dabbled in the field. But not withstanding, it is better to posit and test a quantum of a given size, or order of magnitude. And for this, this experiment is as much a quantum gravity result as we get today.
You're wrong about the Heaviside Lorentz units. The thing with speed of light as 1 is an Electrodynamic system. What they do is actually use a rationalised form of the CGS Gaussian system, ie
F=QQ/4pi r^2.
The unit of charge is 1/(2sqrt(pi)) esu.
It's not hard to give CGS units names: Kennelly did this in 1904 {ab-, stat-}, but in practice, scientists do not use units in calculations, and this is why the prefixes (or any naming convention unit), ever caught on.
People who actually do the experimental work will invent units suitable for the task at hand. Refering to any decent units dictionary will dispel any contary belief.
But the external gravitional field is a source of gravitons. The basis of the experiments were that if you worked with small enough matter, you should see individual gravitons at work.
It is quite possible that it may be a spectral effect similar to ionisation, but if the spectra proves to be regularly placed spikes, then it is better explained as a neutron hit by 1, 2, 3, 4, of the gravitons that make up the earth's gravitional field.
The same experiment done at midday could randomly reveal neutrons hit by gravitons from the Sun, and therefore falling upward!!!
The evidence for photons lies in the photoelectric effect. If you shine light at different wave-lengths onto a material, than it will not issue current until the wave-length becomes shorter than a certian length: ie it has enough energy to knock the electron out of its orbit. This is indirect proof of the photon.
The experiment is basically the same one that discovered the electron (with a few details changed).
In essence, if you select a mass small enough, you may be able to observe its interaction with individual quanta. What these people did was to slow the neutron down so much that they could see it fall under gravity. Their idea was that in stead of falling in a parabola, you should be able to see the polygon sides as the individual quanta hit, or downwards speeds quantised at different multiples of a base. It is this second element that they observed.
Since in the past, this yielded the experimental evidence for the electron, here it yields what could be experimental data for the graviton.
I recall someone actually studied "toon-town" physics, and found that there was a consistant basis for it. Not that it applies much to the "real" world, but it's still pretty important to the way minds think.
Your proposition about standing still is correct. I'm not sure about the speed it falls at, but a collision between a graviton and a neutron might cause that sort of speed in the direction the graviton arrives from. The graviton might also be scattered out in another direction as well.
The article says they used the downwards drift of slowly moving neutrons. The idea was that they slowed neutrons down enough that they could see it fall under gravity. Being relatively small, the neutrons should interact with individual gravitons. If this were the case, the quantum nature would be visiable and measurable. They did not measure the energy directly [as you can't: Energy is a product of two measurables].
Quantum falling was first used to measure the charge of the electron, where charged balls fell in gravity against a field. No-one knew at the time that it was the electron that was doing it.
The other amusing thing is the diversity of units, none of which are "SI": cm/s, electron volts, rather than m/s, J.
Because a byte is not necessarily eight bits. In older computers, it varies from six to twelve bytes.
That the byte might adjust the class of prefix is not unknown, and not without historical precedent.
For example, "minute", "second", "third" were applied to represent x^-n divisions of the degree, for both the grade (x=100), and the degree (x=60). No confusion arose, because the grade was never divided 60-wise, and the degree never divided into 100 minutes.
If one regards the 'kilo' sequence as representing x^n, where x is normally 1000, but can be 1024 on speficic units, then one does not have to remember the different families of prefixes for different units.
The confusion arises because "kilo" can mean 1000 or 1024 when applied to the computer units,
It's not only Linux.... I think it's more to do with "Non-microsoft" than "Open Source".
In the days of Windows 3.1, Microsoft released a series of packs of small games, called Windows Entertainment Packs. These typically had seven or eight games about the size of solitaire, and some screen saver modules. Four of these WEP packs were released.
By the time that OS/2 v 3.0 came out, some wag had dubbed it "WEP 5", basically on the assumption that you could spend hours configuring it.
The funny thing was, I never had any problems installing or configuring OS/2 v 3, but the wep packs and Windows itself at times caused endless grief.
On another note, the latest thing about Wireless Encryption Protocol brought back memories of the older WEP. Over use of TLAs.
Free software is better coded than commercial code, and is getting better all the time. Linux is out of the hobbyist area and is now a serious threat to the system.
But this does not reflect onto the user interface, and it is here that the lingering impression of "Bad Software" is created.
You see, while you may be able to say this is good or bad, you may not know exactly what makes it good or bad.
The reality is that a good UI should present the user with a series of controls, and each control should do what it appears to do. If it is part of a larger environment, then it should conform to this.
This holds as much for command line utilities as it does for GUI apps. For example, you would get annoyed with a terminal program whose output you could not redirect or work with, or was unable to take redirected input.
Writing documentation and UI is about second-guessing the great variety of user tastes and needs. Things like the CUA simplifies some of this.
Apart from having buttons, the program must communicate what it does. For example, most cd burnig programs allow you to make an image of the file as a separate item in the menu. NERO does not do that. Instead, you select burn to a virtual burner, and do it like that. I spent three days fiddling around with this, until someone enlightened me.
The point is, that when there is more than one step involved, the order can be confusing and yield incorrect results. [Think of incorrect order in a pipe].
What we need is some sort of effort done similar to what Knuth did with TeX. Some sort of arbitary language of UI that one can put this interface or that, a la LaTeX style. There is already enough different interfaces around to be able to do it. What one needs to do is be able to is to be able to construct enough of the UI functionality to define it for programmers to implement.
The problem with free software is not that the code is poor, but there seems no way of making a good communication style, and this is, what UI means.
Whether or not bugs were placed by terrorists does not mean that there are grave problems that we should be concerned about.
First, reducing biodiversity, whether of wheat or windows, is a bad thing, because viruses spread quickly and do lots of damage. The simple fact is that we need redundancy in the forms of at least two different technologies.
Secondly, when we do go one way, there is an encumbered trust to ensure that all is safe. This appears to be lacking from the MS world, since their model is to force use of their product.
Let's face it, there is no reason why I should not be able to use any POP3 client to collect mail from Hotmail, rather than specific clients.
Even if the current threat is a joke, there is still underlying issues that need to be address.
I would suggest that we need to get some biodiversity, or face the rath of bugs, whether deliberate (eg malicious), exploited (eg viruses) or accidental.
Remember: Never attribute to malice what can be attributed to stupidity.
3.10 Windows NT: Win16 hacked onto OS/2 1.3. Some 32-bit support.
3.50 Windows NT: Support for Win95 style 32-bit code
3.51 Windows NT: Win95 style shell.
4.00 Windows NT: Rewrite of lower levels. Mostly supports NT 3.51 stuff. HPFS still works, but not supplied. Some utilities from this work under Win95 and 98.
5.00 Windows 2000: Win98 style shell, with IE integration. OS/2 subsystem still works, HPFS does not work. Bitmaps are the same as v4
5.01 Windows XP: New shell, better DOS support, but no OS/2 subsystem. This might work if copied from Win2k. Bitmaps are same as v4.
Version Nr according to JPSoftware.
Upgrade from 5.00 to 5.01 does not sound like a big upgrade.
Rexx is very powerful when it comes to reading and creating strings. It has a huge calculator in it [ie bignum], and you can do things like edit files in it. On the other hand, it is slow.
What makes it useful is that you can write a filter or batch processor quickly.
As far as batch processors go, you can write one that has just one command, eg "$#$". Anything that follows $#$ is processed as a sub command, and anything else just passes through unprocessed. You can then use this to produce RTF output (since most lines are unprocessed, and only the values to be substituted get processed).
You can do the same in PERL, but in the DOS/Win/OS/2 world, REXX is the most widely known.
Slashdot Mirror
I ran 98 quite successfully in 16MB without the bloated shell.
The event occured around the early 1990's (late 80's MS DOS programs feature it. QHelp actually has an entry in its help system saying it has detected a non-ms opsys, and your licence may be void.
Microsoft products generally do not work as intended.
Case in point:
Write a script for a interpreted language (eg perl, rexx), and then try to pass parameters to it (eg REXX MYCALC 3+3). The thing works, and displays the result, which you can just see as the window closes. This is documented in Technet for Winnt 3.1, it is still there in NT4, 2K, and presumably, XP.
The subject "Hose your grandmother's account" would be filtered out because it does not contain the required subject header.
Of those who said "I hate Microsoft" or "Linux Rulz", these give no constructive comment on either the settlement's comments or what has been excluded. Simply saying "I hate (some company)" may be an ethical statement that you hate them because they sell a product you hate, and is no indication that they are break the law. eg, "I hate Ford", because they sell cars, and I hate cars. This is not a reflection on Ford's business practices.
My comment largely centered on possible antitrust comments in upgrades. For example, there is nothing stopping MS from doing things in "required" upgrades, such as shutting down competitive dual boots [Win2k], applications, &c. Upgrades and retail versions should be subject to the same technical restrictions as OEM versions viz Abiltity to not install assorted middleware, honouring multi-boots, etc.
Prehaps we could farm out the intellegence space to interested parties.
Formerly, the practical electric units were additionals like miles and hours. That selected emu have names (Oersted, Gauss, Maxwell), and the absence of names for more obvious measures (eg charge), is the give-away here. Basically it was cgs + Practical units + some named cgs units.
F=QQ/4pi r^2.
The unit of charge is 1/(2sqrt(pi)) esu.
It's not hard to give CGS units names: Kennelly did this in 1904 {ab-, stat-}, but in practice, scientists do not use units in calculations, and this is why the prefixes (or any naming convention unit), ever caught on.
People who actually do the experimental work will invent units suitable for the task at hand. Refering to any decent units dictionary will dispel any contary belief.
It is quite possible that it may be a spectral effect similar to ionisation, but if the spectra proves to be regularly placed spikes, then it is better explained as a neutron hit by 1, 2, 3, 4, of the gravitons that make up the earth's gravitional field.
The same experiment done at midday could randomly reveal neutrons hit by gravitons from the Sun, and therefore falling upward!!!
The experiment is basically the same one that discovered the electron (with a few details changed).
In essence, if you select a mass small enough, you may be able to observe its interaction with individual quanta. What these people did was to slow the neutron down so much that they could see it fall under gravity. Their idea was that in stead of falling in a parabola, you should be able to see the polygon sides as the individual quanta hit, or downwards speeds quantised at different multiples of a base. It is this second element that they observed.
Since in the past, this yielded the experimental evidence for the electron, here it yields what could be experimental data for the graviton.
Your proposition about standing still is correct. I'm not sure about the speed it falls at, but a collision between a graviton and a neutron might cause that sort of speed in the direction the graviton arrives from. The graviton might also be scattered out in another direction as well.
Quantum falling was first used to measure the charge of the electron, where charged balls fell in gravity against a field. No-one knew at the time that it was the electron that was doing it.
The other amusing thing is the diversity of units, none of which are "SI": cm/s, electron volts, rather than m/s, J.
So it's not vapourware, but it probably provides a bit of publicity......
At least ThinkGeek helps to expose the Linux name in a tasteful way. Also, /. is not selling it to Windoze companies.
That the byte might adjust the class of prefix is not unknown, and not without historical precedent. For example, "minute", "second", "third" were applied to represent x^-n divisions of the degree, for both the grade (x=100), and the degree (x=60). No confusion arose, because the grade was never divided 60-wise, and the degree never divided into 100 minutes.
If one regards the 'kilo' sequence as representing x^n, where x is normally 1000, but can be 1024 on speficic units, then one does not have to remember the different families of prefixes for different units.
The confusion arises because "kilo" can mean 1000 or 1024 when applied to the computer units,
Cylinders * heads * Sector * allocations
= 80 * 2 * 18 * 512
= 1474560 = 1.44 * 1024
Seems the actual style is to use k indifferently for 1000 and 1024, as the end suits the need.
In the days of Windows 3.1, Microsoft released a series of packs of small games, called Windows Entertainment Packs. These typically had seven or eight games about the size of solitaire, and some screen saver modules. Four of these WEP packs were released.
By the time that OS/2 v 3.0 came out, some wag had dubbed it "WEP 5", basically on the assumption that you could spend hours configuring it.
The funny thing was, I never had any problems installing or configuring OS/2 v 3, but the wep packs and Windows itself at times caused endless grief.
On another note, the latest thing about Wireless Encryption Protocol brought back memories of the older WEP. Over use of TLAs.
But this does not reflect onto the user interface, and it is here that the lingering impression of "Bad Software" is created.
You see, while you may be able to say this is good or bad, you may not know exactly what makes it good or bad.
The reality is that a good UI should present the user with a series of controls, and each control should do what it appears to do. If it is part of a larger environment, then it should conform to this.
This holds as much for command line utilities as it does for GUI apps. For example, you would get annoyed with a terminal program whose output you could not redirect or work with, or was unable to take redirected input.
Writing documentation and UI is about second-guessing the great variety of user tastes and needs. Things like the CUA simplifies some of this.
Apart from having buttons, the program must communicate what it does. For example, most cd burnig programs allow you to make an image of the file as a separate item in the menu. NERO does not do that. Instead, you select burn to a virtual burner, and do it like that. I spent three days fiddling around with this, until someone enlightened me.
The point is, that when there is more than one step involved, the order can be confusing and yield incorrect results. [Think of incorrect order in a pipe].
What we need is some sort of effort done similar to what Knuth did with TeX. Some sort of arbitary language of UI that one can put this interface or that, a la LaTeX style. There is already enough different interfaces around to be able to do it. What one needs to do is be able to is to be able to construct enough of the UI functionality to define it for programmers to implement.
The problem with free software is not that the code is poor, but there seems no way of making a good communication style, and this is, what UI means.
First, reducing biodiversity, whether of wheat or windows, is a bad thing, because viruses spread quickly and do lots of damage. The simple fact is that we need redundancy in the forms of at least two different technologies.
Secondly, when we do go one way, there is an encumbered trust to ensure that all is safe. This appears to be lacking from the MS world, since their model is to force use of their product.
Let's face it, there is no reason why I should not be able to use any POP3 client to collect mail from Hotmail, rather than specific clients.
Even if the current threat is a joke, there is still underlying issues that need to be address.
I would suggest that we need to get some biodiversity, or face the rath of bugs, whether deliberate (eg malicious), exploited (eg viruses) or accidental.
Remember: Never attribute to malice what can be attributed to stupidity.
- 3.10 Windows NT: Win16 hacked onto OS/2 1.3. Some 32-bit support.
- 3.50 Windows NT: Support for Win95 style 32-bit code
- 3.51 Windows NT: Win95 style shell.
- 4.00 Windows NT: Rewrite of lower levels. Mostly supports NT 3.51 stuff. HPFS still works, but not supplied. Some utilities from this work under Win95 and 98.
- 5.00 Windows 2000: Win98 style shell, with IE integration. OS/2 subsystem still works, HPFS does not work. Bitmaps are the same as v4
- 5.01 Windows XP: New shell, better DOS support, but no OS/2 subsystem. This might work if copied from Win2k. Bitmaps are same as v4.
Version Nr according to JPSoftware.
Upgrade from 5.00 to 5.01 does not sound like a big upgrade.You could {turning logic}, but it would be nearly as slow as Windows XP {MSCE bait}
- The line can still be hacked, because it is possible to put a TEE into the circuit, just as long as STDOUT looks like STDIN.
- It may be possible to hack elsewhere, al la Magic Lanten.
- It would be very succeptable to stray photons, including those made by quantum effects...
Or am I wrong....What makes it useful is that you can write a filter or batch processor quickly.
As far as batch processors go, you can write one that has just one command, eg "$#$". Anything that follows $#$ is processed as a sub command, and anything else just passes through unprocessed. You can then use this to produce RTF output (since most lines are unprocessed, and only the values to be substituted get processed).
You can do the same in PERL, but in the DOS/Win/OS/2 world, REXX is the most widely known.
My mistake :(
This was a laptop that supported a fat of 32 files, the interface was to move the highlight around the screen and select it.
I never suggested that there was a disk involved. The files were saved and edited in RAM, as the tandy 100 did not have a hard disk.