Think about it this way: how does the QNX kernel load any other module (fs included) in the first place?
My guess: a microkernel has the most basic ability to mount the root fs itself.
Oh, and you just damn well make sure that the vm server doesn't crash:-P
Seriously though, how often does the linux mm code fail? Core kernel code like this is fun. It gets a lot of attention, and many eyes make bugs shallow.
OO, for example, will NEVER surpass, or even equal traditional functional programming, until the hardware is itself OO. (The Crusoe is one step in this direction, where the end result is placing one method in one instruction, one class on one processor, and one instance in a selectable set of registers.)
Interesting, but I've lost what you mean by this. As I understand it, the Crusoe is just a standard VLIW processor - smart, nothing unusual in the way it executes code. The only really neat bit that I am aware of is the mmu, which is capable of mapping certain memory regions as being memory mapped IO regions, helping support the code morphing. How are you suggesting that the Crusoe supports OO better? placing one method in one instruction - uh? what are you saying here? memory-indirect function calls? you can get that functionallity out of the x86. one class on one processor urgh? a class is just a description of how a region of memory is being used?
Not wanting to offend - just not quite understanding your point.
Then, too, there is an assumption in computing that you have to move from beginning to end. Yet when you draw up a specification, you don't give a flow, you give mathematical rules which are valid. Nor do you say -how- something is to be done, you simply define the consequences of performing a given operation. Extend the concepts a bit, and you totally seperate the "whats" from the "hows", and the content from the presentation. Allowing the user to control THEIR end, and the machine ITS end will be the next major step forward in computing.
Yeah - have you ever coded in a functional language like ML? Just like this. For those who haven't, the difference between a procedural and a functional language goes something like this:
The source code itself doesn't do anything but communicate, so it's protected speech.
I just have one question about this:
Imagine I write a book telling people how to make a nuclear bomb. I am allowed to sell this, this is free speech.
Next, someone buys my book, uses it to build bombs, and sells the bombs. This is not legal - a bomb is not speech, it is a tool.
Now imagime that when someone bought my book, it could automatically, magically turn into the the bomb that it described. I am only distributing speech, information, a description - but people receive a product, a tool, a bomb.
That is what happens with software. A compiler is a magical device, with no comparison outside the computer. You hand it a description of what you want, and 'poof!' what ever you wished for pops out of the air.
I'm a programmer, and I consider a piece of code to be speech, but I can see why the law may have a problem with this.
I know that I'm taking a quote out of context here, but really, I am not being flamebait.
I understand the reasoning in allowing people to use pointers, but can't say that I agree with it. There may be times and places that you need the power of pointers, but are they really necessary in the kind of programming situations that C#/Java are addressing?
I remember hearing that 50% of programmer errors in c/c++ were memory allocation problems. From experience, sounds about right to me.
Give people a hammer, and they will go around knocking screws into the wall with it.
Caveat emptor.
</disalaimer>
I must say I agree with one point made in the Register article you point to - Micros~1 should have looked at using python.
A lot of people will say, 'release often, release early', etc (particuarly with reference to the linux model). I personally always feel that people ignore the amount of work done on GNU prior the the Linux kernel being released.
Linux was, to a fair extent, born full grown, in that there was a full UNIX just sitting there, waiting for a kernel. This let Linux explode off the blocks, and gain a lot of momentum. (Okay, 10yrs or so is quite a slow burn for me to use the term 'explode', but hey)
I am interested in OS development. So many great projects that seem to burn out as people seem to lose interest, as the project grinds on slowly.
The more people drop out --> the less work getting done --> the less results seen --> the less enthusiasm --> more people drop out...
I know that slashdot repeats itself every so often, but the javascript RSA one was in a bunch on quickies posted by emmett on June 30, barely more than a month ago.
Yeah, this reminds me of something said, I think by Rob Malda, on geeks in space, with regard to DVD-Audio. The big music companies were holding off due to fears of piracy.
The quote was along the lines of:
"In order to listen to it we have to be able to decrypt it. If we can decrypt it, we can rip it."
(This may not be verbatim.)
This is so true.
The only thing that is different about this is that the watermark is mixed in to the audio, not just the file - so if you decompress to a 16bit sample, and recompress back to a mp3, you will probably not lose the watermark.
The clever thing about this is that it still uses the mp3 format, instead of trying to replace it, like people ([cough] Micros~1) have in the past.
Interestingly, this is not a new idea. The Daily Telegraph, in the UK used to have a computing section, which used to run a 'Techno Turkey' column. I remember reading about a similar type of copy protect tapes/vinal recordings. Sadly, I cannot remember why it failed. Probably lack of uptake.
I've been meaning to ask Slashdot the same question as Cliff for a while, but slightly broader. I would like to be able to hack code where ever I am, but I am not fussy about the language. I will cheerfully code in dhtml, python, perl, C, C++, java, asm, or whatever is available.
I was guessing that the psion would a good choice, due to the quality of both the keyboard and the OS [not meant as a dig at winCE: epoc32 is meant to be *really* good.]
Having a Java runtime is good, but you'll also need a compiler. Is there one? I've done some Java bytecode hacking, so I could probaby write one if there is a demand.
Do you (/anyone else) have any experience of writing code on a Psion, good or bad?
I'm sure that the telecoms giants will give this the same warm reception that the music industry has extened to Napster.
But it is the future.
In the same way that digital CDs replaced analogue vinal, and compressed computer formats (eg MP3) are replacing CDs, the same will happen in the telecoms industry.
Analogue phone lines are being (/have been, depending on where in the world you live) replaced by IDSN/xDSL, and these will be replaced by dedicated internet connections, with no capability to carry uncompressed voice data.
I wonder if we will reach a point where people will find a way to start squatting.
Cybersqutting is usually meant to mean buying a domain that you do not intend to use, to sting money out of people who will later want that name. This is more like cyber-real-estate-speculation.
Squatting is when you make use of a property that you don't own, but that the owner isn't using. If people are going to register names, then leave them unused so they expire like this, then is it possible to start squatting in the domain?
I guess squatting involves:
Breaking in. (brute force password in some way?)
Moving in. (set it to point at your IP)
Swearing blind to the cops that you didn't actually break in, you just found it that way:P
I'm a 2nd year CS student at Manchester University, and one of the final year projects on offer this year is to port Linux to the Amulet.
I didn't look too deeply into it, but I assume from the fact that this project is on offer that some work must need doing to get the present arm ports running on the Amulet.
I believe EDSAC came one year later. Check the links on the above posts and you'll find Manchester Baby dates back to 1948 & EDSAC to 1949.
I think Cambridge claim it was the first 'proper' computer as it was more like modern computers. Possibly had a teletype style keyboard/printer? Can't remember exactly.
I'm not trolling, I'm deeply interested in OS & OSS stuff, and I've been going through the kernel a lot lately. I'll admit that all of the kernel code I've seen seems pretty damn solid , but as I understand it the network stuff is total spaghetti.
Don't get me wrong - linux networking support is brilliant, but the code wasn't designed with multiprocessor in mind (this is how NT beat Linux on those high publicity MS funded tests a while back). Apparently the reason no one has fixed the TCP/IP stack to multiproc better is that the code is such a mess.
Linux may be good, but that doesn't mean the code is well written from a maintainance perspective.
disclaimer: I haven't look at the net stuff yet myself, and don't know if this is true or not - but you can see the point I'm making:-)
An analogy: If a student wants to have a half decent OS (ie. non-win9x) on their computer the choice boils down to this: either install Linux or get a pirate of Windows NT. (I'm simplifying - be/bsd/whatever people please don't bother flaming)
Technically the piracy of NT is stealing - but it doesn't cost Microsoft anything. The student learns the microsoft product and not linux. When he becomes an IT proffesional he uses WinNT systems not Unix.
The piracy of Microsoft software by people who could not afford to buy it in the first place, who are learning and experimenting with OSes, is GOOD for MS.
People who cannot afford the music would not be buying it anyway. The band lose nothing, and gain a bigger fan base.
Sorry for mentioning MS on/. - I know it offends a lot of people, but it had to be done:-)
When I was a kid I had hundreds of copied tapes, and the MP3s are just a modern extension of this. I'm sure most people copy music for the same two reasons:
I used to try buying CDs, then find out they were junk.
I remember hearing a story about the band U2. BBC Radio 1 were broadcasting live a concert they were playing, so they got a uk music magazine (Q/nme?) to print an inlay card for a tape, with a full track listing of the set they intended to play. They wanted to encourage people to tape and copy their music, bypassing the established music industry, and associated means of distribution, to get their music heard by as many people as possible.
The more a bands music is distributed over the net -> the more people listening to their music -> the more fans they have. Simple:-)
I'm poor.
Yup. Why deny someone the music if they can't afford it anyway?
everywhere I look, record sales are booming. They're having no problems pushing CDs, even though they're generally $3 - $5 more than 5 - 10 years ago when I was in my teen popular artist CD buying phase.
The thing is taped recordings were always poor quality. If you liked the music and could afford to you would go out and buy the CD. So long as record companies can keep producing a commercial product which is better than the 'pirated' alternative (eg. music videos/interactive content/dvd audio), then they have nothing to fear, and people copying music will help promote artists more than it hinders sales.
It's not a patch on the older IBM "clicking" keyboards for overall feel, but it's not terrible. A matter of personal choice more than anything else
I was thinking about getting a mac. I know it's not a problem to get a decent USB mouse these days - but does anyone know if there are any decent USB click keyboards available?
When I'm coding I need that reassuring little clicky sound to tell me that i've pressed a key as much as i need caffine:-)
Slashdot Mirror
My guess: a microkernel has the most basic ability to mount the root fs itself.
Oh, and you just damn well make sure that the vm server doesn't crash :-P
Seriously though, how often does the linux mm code fail? Core kernel code like this is fun. It gets a lot of attention, and many eyes make bugs shallow.
Interesting, but I've lost what you mean by this. As I understand it, the Crusoe is just a standard VLIW processor - smart, nothing unusual in the way it executes code. The only really neat bit that I am aware of is the mmu, which is capable of mapping certain memory regions as being memory mapped IO regions, helping support the code morphing. How are you suggesting that the Crusoe supports OO better? placing one method in one instruction - uh? what are you saying here? memory-indirect function calls? you can get that functionallity out of the x86. one class on one processor urgh? a class is just a description of how a region of memory is being used?
Not wanting to offend - just not quite understanding your point.
Then, too, there is an assumption in computing that you have to move from beginning to end. Yet when you draw up a specification, you don't give a flow, you give mathematical rules which are valid. Nor do you say -how- something is to be done, you simply define the consequences of performing a given operation. Extend the concepts a bit, and you totally seperate the "whats" from the "hows", and the content from the presentation. Allowing the user to control THEIR end, and the machine ITS end will be the next major step forward in computing.
Yeah - have you ever coded in a functional language like ML? Just like this. For those who haven't, the difference between a procedural and a functional language goes something like this:
procedural:i l();
get_pan();
fill_with_water();
bo
place_egg_ in_water();
wait();
remove_egg();
shell();
fun ctional:) ))))));
shell(remove_egg(wait (place_egg_in_water(boil(fill_with_water(get_pan(
The source code itself doesn't do anything but communicate, so it's protected speech.
I just have one question about this:
- Imagine I write a book telling people how to make a nuclear bomb. I am allowed to sell this, this is free speech.
- Next, someone buys my book, uses it to build bombs, and sells the bombs. This is not legal - a bomb is not speech, it is a tool.
- Now imagime that when someone bought my book, it could automatically, magically turn into the the bomb that it described. I am only distributing speech, information, a description - but people receive a product, a tool, a bomb.
That is what happens with software. A compiler is a magical device, with no comparison outside the computer. You hand it a description of what you want, and 'poof!' what ever you wished for pops out of the air.I'm a programmer, and I consider a piece of code to be speech, but I can see why the law may have a problem with this.
cheers,
G
Uh - seriously now, I made the mistake of both:
- Trying out Win98
- Having LILO in the partition, not the mbr
Windows 98 had a habit of changing the boot partition from the linux one, to the it's own one, most of the times I booted into Windows.This feature has already been added!
Hejlsberg: "caveat emptor, it's unsafe"
I know that I'm taking a quote out of context here, but really, I am not being flamebait.
I understand the reasoning in allowing people to use pointers, but can't say that I agree with it. There may be times and places that you need the power of pointers, but are they really necessary in the kind of programming situations that C#/Java are addressing?
I remember hearing that 50% of programmer errors in c/c++ were memory allocation problems. From experience, sounds about right to me.
Give people a hammer, and they will go around knocking screws into the wall with it.
Caveat emptor.
</disalaimer>
I must say I agree with one point made in the Register article you point to - Micros~1 should have looked at using python.
A lot of people will say, 'release often, release early', etc (particuarly with reference to the linux model). I personally always feel that people ignore the amount of work done on GNU prior the the Linux kernel being released.
Linux was, to a fair extent, born full grown, in that there was a full UNIX just sitting there, waiting for a kernel. This let Linux explode off the blocks, and gain a lot of momentum. (Okay, 10yrs or so is quite a slow burn for me to use the term 'explode', but hey)
I am interested in OS development. So many great projects that seem to burn out as people seem to lose interest, as the project grinds on slowly. The more people drop out --> the less work getting done --> the less results seen --> the less enthusiasm --> more people drop out...
Very saddening to watch projects die this way.
G
I know that slashdot repeats itself every so often, but the javascript RSA one was in a bunch on quickies posted by emmett on June 30, barely more than a month ago.
Should I expect to see it again in 30 days time?
CmdrTaco should have used the preview button.
> Fact #1: To hear it, you have to stream it.
> Fact #2: If you can stream it, you can copy it.
Yeah, this reminds me of something said, I think by Rob Malda, on geeks in space, with regard to DVD-Audio. The big music companies were holding off due to fears of piracy.
The quote was along the lines of:
"In order to listen to it we have to be able to decrypt it.
If we can decrypt it, we can rip it."
(This may not be verbatim.)
This is so true.
The only thing that is different about this is that the watermark is mixed in to the audio, not just the file - so if you decompress to a 16bit sample, and recompress back to a mp3, you will probably not lose the watermark.
The clever thing about this is that it still uses the mp3 format, instead of trying to replace it, like people ([cough] Micros~1) have in the past.
Interestingly, this is not a new idea. The Daily Telegraph, in the UK used to have a computing section, which used to run a 'Techno Turkey' column. I remember reading about a similar type of copy protect tapes/vinal recordings. Sadly, I cannot remember why it failed. Probably lack of uptake.
Anyone else read about this?
Damn, where is that taken from?
cheers,
G
What OS / hardware combinations have you tried out?
How easy was it to replace winCE?
is it *really* a viable platform for coding on the road?
I mean - I don't need a pretty IDE, I don't even need X, but I at least need multiple virtual terminals, so I can code on one and compile on another.
Any stories of experience doing this will be appreciated.
cheers, G
I've been meaning to ask Slashdot the same question as Cliff for a while, but slightly broader. I would like to be able to hack code where ever I am, but I am not fussy about the language. I will cheerfully code in dhtml, python, perl, C, C++, java, asm, or whatever is available.
I was guessing that the psion would a good choice, due to the quality of both the keyboard and the OS [not meant as a dig at winCE: epoc32 is meant to be *really* good.]
Having a Java runtime is good, but you'll also need a compiler. Is there one? I've done some Java bytecode hacking, so I could probaby write one if there is a demand.
Do you (/anyone else) have any experience of writing code on a Psion, good or bad?
cheers, G
I'm sure that the telecoms giants will give this the same warm reception that the music industry has extened to Napster.
But it is the future.
In the same way that digital CDs replaced analogue vinal, and compressed computer formats (eg MP3) are replacing CDs, the same will happen in the telecoms industry.
Analogue phone lines are being (/have been, depending on where in the world you live) replaced by IDSN/xDSL, and these will be replaced by dedicated internet connections, with no capability to carry uncompressed voice data.
just my $0.02
Errr... am I the only person seeing this post about OS X in the story about Pizza Hut?
:-P
Either slash is going slightly mad, or this is the most surreal troll I have seen yet
Something weird is going on, or you are just _so_ offtopic.
The nearest that QDOS gets to power commands.....
As a fan of "rm -rf *", "fdisk -mbr" is the best that QDOS gets to providing me with dangerous power.
:-)
Also... weren't you /not/ supposed to have the acronym within it's own definition?
It's what is called a 'recursive acronym'. To quote Stallman:
The name GNU was chosen following a hacker tradition, as a recursive acronym for "GNU's Not Unix."
oh, and your link doesn't work for me. Try this one.
Apple Computer do not have the domain www.apple.co.uk
It is owned by a UK graphics design company called Apple. They have a perfectly legitimate right to the name (but I bet it pisses Apple off a bit :P ).
Cybersqutting is usually meant to mean buying a domain that you do not intend to use, to sting money out of people who will later want that name. This is more like cyber-real-estate-speculation.
Squatting is when you make use of a property that you don't own, but that the owner isn't using. If people are going to register names, then leave them unused so they expire like this, then is it possible to start squatting in the domain?
I guess squatting involves:
they [both] would say that though wouldn't they.
Cheers for the info.
I'm a 2nd year CS student at Manchester University, and one of the final year projects on offer this year is to port Linux to the Amulet.
I didn't look too deeply into it, but I assume from the fact that this project is on offer that some work must need doing to get the present arm ports running on the Amulet.
just a little insider information
I believe EDSAC came one year later. Check the links on the above posts and you'll find Manchester Baby dates back to 1948 & EDSAC to 1949.
I think Cambridge claim it was the first 'proper' computer as it was more like modern computers. Possibly had a teletype style keyboard/printer? Can't remember exactly.
Are you sure?
I'm not trolling, I'm deeply interested in OS & OSS stuff, and I've been going through the kernel a lot lately. I'll admit that all of the kernel code I've seen seems pretty damn solid , but as I understand it the network stuff is total spaghetti.
Don't get me wrong - linux networking support is brilliant, but the code wasn't designed with multiprocessor in mind (this is how NT beat Linux on those high publicity MS funded tests a while back). Apparently the reason no one has fixed the TCP/IP stack to multiproc better is that the code is such a mess.
Linux may be good, but that doesn't mean the code is well written from a maintainance perspective.
disclaimer: I haven't look at the net stuff yet myself, and don't know if this is true or not - but you can see the point I'm making :-)
Hmmm.
/. - I know it offends a lot of people, but it had to be done :-)
An analogy:
If a student wants to have a half decent OS (ie. non-win9x) on their computer the choice boils down to this: either install Linux or get a pirate of Windows NT. (I'm simplifying - be/bsd/whatever people please don't bother flaming)
Technically the piracy of NT is stealing - but it doesn't cost Microsoft anything. The student learns the microsoft product and not linux. When he becomes an IT proffesional he uses WinNT systems not Unix.
The piracy of Microsoft software by people who could not afford to buy it in the first place, who are learning and experimenting with OSes, is GOOD for MS.
People who cannot afford the music would not be buying it anyway. The band lose nothing, and gain a bigger fan base.
Sorry for mentioning MS on
When I was a kid I had hundreds of copied tapes, and the MP3s are just a modern extension of this. I'm sure most people copy music for the same two reasons:
I used to try buying CDs, then find out they were junk.
I remember hearing a story about the band U2. BBC Radio 1 were broadcasting live a concert they were playing, so they got a uk music magazine (Q/nme?) to print an inlay card for a tape, with a full track listing of the set they intended to play. They wanted to encourage people to tape and copy their music, bypassing the established music industry, and associated means of distribution, to get their music heard by as many people as possible.
The more a bands music is distributed over the net -> the more people listening to their music -> the more fans they have. Simple :-)
I'm poor.
Yup. Why deny someone the music if they can't afford it anyway?
everywhere I look, record sales are booming. They're having no problems pushing CDs, even though they're generally $3 - $5 more than 5 - 10 years ago when I was in my teen popular artist CD buying phase.
The thing is taped recordings were always poor quality. If you liked the music and could afford to you would go out and buy the CD. So long as record companies can keep producing a commercial product which is better than the 'pirated' alternative (eg. music videos/interactive content/dvd audio), then they have nothing to fear, and people copying music will help promote artists more than it hinders sales.
Wasn't meaning to troll on the 'ripped off' remark, apologies to the people it seems to have offended :-)
I was thinking about getting a mac. I know it's not a problem to get a decent USB mouse these days - but does anyone know if there are any decent USB click keyboards available?
When I'm coding I need that reassuring little clicky sound to tell me that i've pressed a key as much as i need caffine