Slashdot Mirror

Of course, at some point, we do have to trust someone.
Ken Thompson wrote an original speculative essay on this for CACM back in 1984 of all years.

It is really well worth the read. The short form is that there exists a way to subvert the compiler such that it is no longer trustable and it will build a back door into the OS forevermore. This paper is a must read.

You forgot UNIX. But we already know that came from outer space.

http://plan9.bell-labs.com/sys/doc/venti/venti.h tm l

Abstract

This paper describes a network storage system, called Venti, intended for archival data. In this system, a unique hash of a block's contents acts as the block identifier for read and write operations. This approach enforces a write-once policy, preventing accidental or malicious destruction of data. In addition, duplicate copies of a block can be coalesced, reducing the consumption of storage and simplifying the implementation of clients. Venti is a building block for constructing a variety of storage applications such as logical backup, physical backup, and snapshot file systems.

• Obvious references to Linux/Unix/X11 still seem pretty rare to me. Some movies have featured them prominently, but unless the computer is itself part of the plot, the interface is usually made to melt into the background. Here's some examples I can think of where the *nix interface seems obvious:
  • The movie "Hackers" is a standard one to cite here. The movie is really awful, but I'm willing to give it a pass not because of the silly computer displays, but because it has Penn Jillette in a small role, Hal (which automatically scores points for the 2001 reference). And the reason it's cool that Penn is in there is, well, because it's Penn, and he's really "in" on this silly little subculture. Witness his snarky comments on Richard Stallman, the comedic potential of the Turing Test and Markov chains ("Mark V. Shaney" -- get it?), the math behind public key encryption, and -- most of all -- is chummy with Unix co-designer Rob Pike, and has even pulled pranks on Nobel laureats with him. So, short of putting in someone like Pike, Dennis Ritchie, Ken Thompson, or Linus Torvalds, putting Penn Jillette in a geek role in a movie is pretty much close enough for me.
  • "Jurassic Park" had a famous scene where the girl sits down at a terminal, looks things over, then exclaims "This is Unix! I know Unix!". Silly, but then it was real, sort of: the screen shots were of an experimental 3D file manager from SGI. There was probably an xterm open somewhere offcamera or behind the file manager window so that a technician could enter commands in between the GUI clips that made it into the film.
  • There are other examples of Linux in movies, but unfortunately most of the movies are awful (Antitrust, Swordfish, <troll> The Matrix </troll>, etc).
• As has been noted all over, Macs show up a lot in movies & tv shows. This probably isn't a coincidence: the machines may look nicer than the typical beige box PC, but the product placement was probably paid for (also see here, at the bottom) in most cases, just as it would be for any other identifiable consumer product in a show. That said, random Mac sightings I can think of include:
  • Carrie's laptop in recent seasons of "Sex and the City" is an old black Powerbook G3 running OS9. Before that she had an older Powerbook. She was given a clamshell iBook as a gift when the G3 crashed, but returned it & fixed the Powerbook.
  • Harry Connick Jr's character had a G4 tower & cinema display on his desk in a recent "Will & Grace". The display wasn't up, so no idea what it was running.
  • In the movie "Zoolander", Apples show up all over the place. The funniest example was probably when Ben Stiller & Owen Wilson are told to break into an office & steal some files off someone's iMac: after staring blankly at it for a while, they call for help and are told that the files are "inside the computer". Like wisdom dawning on the apes in 2001, they get the idea -- and start beating on the case trying to break it open and cause the files to spill out.
  • In my favorite example, it has been observed that on the show "24", all the good guys use Macs and all the bad guys use Dells. An awareness of this pattern would have uncovered a turncoat who ended up betraying people at the end of the first season.
• A lot of shows have hard to identify OSes. Probably on purpose.
  • On "CS

It might have been interesting for ESR to have mentioned Plumbing as seem in Plan 9. Arguably (as with much in Plan 9), the idea is a natural evolution of the UNIX model (in this case pipes):

Plumbing is a new mechanism for inter-process communication in Plan 9, specifically the passing of messages between interactive programs as part of the user interface. Although plumbing shares some properties with familiar notions such as cut and paste, it offers a more general data exchange mechanism without imposing a particular user interface.

The core of the plumbing system is a program called the plumber, which handles all messages and dispatches and reformats them according to configuration rules written in a special-purpose language. This approach allows the contents and context of a piece of data to define how it is handled. Unlike with drag and drop or cut and paste, the user doesn't need to deliver the data; the contents of a plumbing message, as interpreted by the plumbing rules, determine its destination.

The plumber has an unusual architecture: it is a language-driven file server. This design has distinct advantages. It makes plumbing easy to add to an existing, Unix-like command environment; it guarantees uniform handling of inter-application messages; it off-loads from those applications most of the work of extracting and dispatching messages; and it works transparently across a network.

[Dennis] Ritchie and [Ken] Thompson joined Bell Labs within a year of each other, Thompson in 1966 and Ritchie in 1967. They worked closely together for several years on the design and development of UNIX. The C Language, in which the UNIX operating system is written, was invented by Ritchie. It grew out of an earlier language, B, written by Thompson. (link)

poor Plan 9 web server, you're giving it the run of its life...

well, the people at Bell Labs aren't stupid -- they've automagically turned off the web page in question, as soon as they saw the transfer to it peaking...

notice, however, that neither the network connection nor the web servers are saturated: stuff like http://plan9.bell-labs.com/plan9dist and http://cm.bell-labs.com/plan9dist works just fine :)

gotta lova plan 9 :P

poor Plan 9 web server, you're giving it the run of its life...

well, the people at Bell Labs aren't stupid -- they've automagically turned off the web page in question, as soon as they saw the transfer to it peaking...

notice, however, that neither the network connection nor the web servers are saturated: stuff like http://plan9.bell-labs.com/plan9dist and http://cm.bell-labs.com/plan9dist works just fine :)

gotta lova plan 9 :P

It's your users you should be worrying about, not the outside world.

Remote roots are the least of your headaches. Escalating privileges of logged in users is a very real threat. When OpenBSD talks of "no remote exploits" you have to rememeber to add the caveat "in the default install" which paints a different picture.

Root considered harmful is more than a cute saying.

Root is a design choice and it is an achilles heel of the Unix family.

Likewise administrator, though Windows has a fair few more bad design choices from a security perspective.

And guess what? When the creators of Unix decided to take what they had learned and start again root was one of the first things to hit the bit bucket.

However, noting that the state-space size is large isn't really a very useful observation, since chess programs these days don't try to map out the entire tree of possible outcomes. Instead, they operate on neurodynamic programming techniques, which basically try to extract which "features" of the game are important and weigh those features to decide which moves to make. This significantly reduces the complexity of the system, but requires that the person writing the program have some intuition about which "features" are important. In chess, for example, these include such things as material balance, piece mobility, king safety, and other positional factors. A period of training is usually required as well, where basically the computer goes over a lot of games that grandmasters have played and tries to "learn" how to weigh the different features in order to choose the optimal move.

For those who are interested in reading further about this (yeah, yeah, this is Slashdot, if people can't RTFA what are the odds they'll want to pick up a book? :) ) a good place to start would be Chapter 6 of Bertsekas' "Dynamic Programming and Optimal Control".

That being said you can run GNU/Linux and get rooted just as easily as you could with Windows if you don't patch your system.

if you don't have root you can't get rooted

Theo de Raadt & Linus Torvalds have both left their pet projects to start coding for Plan 9, another great operating system that never was.

Maybe the OS never was, but they had one hell of a mascot! :)

Anyone know why this is a good/bad thing ?

The stuff concerning dev_t on the LKML concerned itself with tty stuff.

Will it really mean the death of tty ? I do hope so.

isatty is a stupid anachronism (which is why plan9 got rid of such idiocy).

I vote for xfce as having the dang cutest logo yet.

My vote goes to Glenda, the Plan 9 Bunny. Astronaut bunnys, can't get much cuter.

Bell Labs develops UNIX as a development environment for next generation telecommunication systems. It grows to encompass a myriad of applications over the years. One day a bright young Finn hacks a derivative to run on commidity computing hardware. This derivative "Linux" is embraced first by hard core geeks for their own use (and as an expression of independence from various monolithic computing behemoths) and eventually, through the coding and evangelizing efforts of this user community, gains acceptance in enterprise level computing environments. Now NTT wants to use Linux as a platform for next generation data/telecomm applications...

They were. Read all about it.

Take a read. It's true, because the case was settled out of court, there isn't a formal ruling that it's public domain. But neither the law nor the facts have changed. Well of course the law has, copyright law has really gone to hell, but the changes don't apply - if AT&T had no copyright to begin with, as the judge in this case seems to have been convinced.

There are other routes to Public Domain, of course. But this is the one SCO is preparing to trod. Idiots.

I can't believe they're planning to go that far. I expect Darl will grab his performance bonus and dump his stocks and head for the hills long before any of this gets to trial.

1986 called, they want their screen shot back.

Like many of the new ideas in the BSDs (private namespaces as chroot for instance), the idea of separate networks stacks is taken from plan9.

It is even possible in plan9 to use the network stack of a remote machine as your own.

Using sshnet one can do "sshnet remote_host" then all subsequent network activity for the current process group and any children will use the remote hosts' network stack as though it were local.

In this way one can run tcp listeners on a remote machine (on IP N) that deny requests from anything but the IP N and leave SSH as the only external listener. No special tunnelling hoops to jump through.

Monolithic kernels are dying.

Like many of the new ideas in the BSDs (private namespaces as chroot for instance), the idea of separate networks stacks is taken from plan9.

It is even possible in plan9 to use the network stack of a remote machine as your own.

Using sshnet one can do "sshnet remote_host" then all subsequent network activity for the current process group and any children will use the remote hosts' network stack as though it were local.

In this way one can run tcp listeners on a remote machine (on IP N) that deny requests from anything but the IP N and leave SSH as the only external listener. No special tunnelling hoops to jump through.

Monolithic kernels are dying.

Dan Egnor says it best :

Somewhere deep inside the secret headquarters of the RedHat/GNOME/Ximian/Mozilla Cabal, there's a hidden document with a list of everything in Unix you know and love, marked with a date for its final expurgation. I think 'ls' is slated to be finally replaced with a symlink to 'nautilus' in 2007. Except that symlinks will have been replaced by ".shortcut" files, which are interpreted by the Mono implementation of GNOME-VFS.

Luckily the spirit on Unix lives on.

You should use a proper operating system. Not an amateur toy.

Replying to my own post... how crass...

There's a better picture of the mouse on this brochure for the successor, the 5630.

The clearly the best mouse ever was the hemispherical, red-clown-nose mouse that came with the AT&T/Teletype 5620 terminal. What a buzz using that thing was :)

so of course it has holes in it

The Unix family is insecure by design.

That is why it is better to start again than try and paper over the cracks.

Can someone explain to me *how* BPF is part of SysV UNIX, anyway? SysV UNIX network stack is based on the STREAMS architecture - originally developed by D.Ritchie. A BSD network stack (adopted by linux) is a totally different animal. In order to create a packet filter in STREAMS, you'll need to simply write a STREAMS module that plops in between the data link layer and network layer. The BPF doesn't even fit in with the STREAMS architecture, unless one were to somehow write wrapper code around it so that it conforms to the DLPI and NPI interfaces...and I think this would be highly inefficient. Admittedly, I have not looked at the BPF code, but I imagine that it would need to be highly modified / re-written to get it to work with STREAMS.

Original STREAMS Paper

Putting this sort of stuff in that device is a cool hack but totally the wrong thing.

It should run a little file server, serving something like 9p whihc would allow you to read/write settings and stream off the full data packets read for snorting.

fools.

Some may prefer Dennis Ritchie or even Richard Stallman...

[ shameless_plug ]
StarFish is a block-level storage system allowing on-the-fly geographic replication, that would work with any database. It was OpenSourced by Lucent a few months ago. It won the Best Paper award at Freenix '03.
[ /shameless_plug ]

You're half right there. The first implementation of Unix was in 1969 on a PDP-7, not a PDP-10. The two machines are quite different: the PDP-7 was a small 18-bit minicomputer designed for small groups of users, while the PDP-10 was a large 36-bit mainframe-type machine designed to serve big communities, like an entire university.

It was natural for Thompson and Ritchie to move from the PDP-7 to the 16-bit PDP-11, since the two machines were of similar scale in the beginning. Moving from the PDP-10 to the PDP-11 would have been like moving from your nice roomy house into your college friend's attic. :-)

There were a number of interesting operating systems on the PDP-10 (TENEX, TOPS-10, TOPS-20), but no one has ever implemented a Unix-style OS on it, as far as I know.

Dennis Ritchie's web site has a good account of the early development of Unix. It makes interesting reading if you're curious about such things.

Just like root, Secure OS's don't use it

In the article, the author digresses a bit into the history of the Teletype, and says, "The Unix system at Bell Labs was developed using the Model 35 Teletype, which was the next model produced after the Model 28..." However, the 35 was a rather large and fairly rare unit. I suspect most of the Unix work done at Bell Labs was on Teletype Model 33 variants. This is supported by the picture available at this link maintained by Dennis Richie himself.

See here.

The BSD code was in dispute, and is in the clear now.

Thanks; ihnp4!arpanet!pobox!bill.stewart

I'd rather not use the term "hobbyist". AT&T was bound by anti-trust regulations to supply their inventions to universities. Scientists took it the way they like the best - with emphasis on peer review and free circulation of information (free as in beer and as in speech). I think that "scientific" or "academic" is a better description. After all, they weren't amateurs.

The people involved certainly weren't amateurs, but the project itself started as an attempt to get the "Space Travel" game working on a PDP-7 and eventually crept into use through the "back door" of the company as Unix become more and more feature-rich. It took over minicomputer land, eventually, but started out as a small OS with small goals but with lots of portability behind it - quite similar to Linux. AT&T eventually supplied Unix to scientists (thanks to anti-trust requirements) and universities, but only after its worth had been established.

Although your comments seem authoritative and knowledgeable, your facts are fuzzy and wrong. Unix was derived from Multics, but Multics at the point was dead. AT&T had abandoned the project altogether. That's when Dennis Ritchie developed Unix. Since Ritchie helped to develop Multics for AT&T, how can anybody say that he ripped off Multics. Is like saying I'm stealing part of my car to use in my other car. If it's mine, it's not stealing.

The Berkeley Shareware Distribution (BSD) was sued by AT&T in the early 1990s, for openly distributing copyrighted code in its public-domain source releases. As if this wasn't enough, it turned out that AT&T had also broken the license on code they had taken from BSD, leaving both sides forced to essentially accept the other's illegal behavior in order to avoid stiffer penalties.

Again, a mispresentation of history. Actually, AT&T had sold the Unix copyrights to USL by then so it wasn't AT&T vs BSDI. It was USL vs BSDI. Get your history straight. In the preliminary ruling, the judge indicated he would rule in BSDI's favor since USL could not support most of their claims. Was code from both parties intertwined? Yes, but from all accounts BSD had very little code to worry about. USL had larger problems because not only did they borrow code from BSD, they also removed BSD's copyrights and sold the code to other parties.

Reputable software companies such as Microsoft, though initially interested in Unix, have learned to steer clear of the mess of standards, licenses, and conflicting intellectual property rights that Unix forms.

I get it, you're a Microserf! So how much is Billy paying you to troll? Microsoft is one of the few companies you should use with the term 'reputable' around here. To be honest, large companies sometimes do engage in unethical behavior. Microsoft is no exemption. They have been sued countless times for copyright and patent infringement. Stac, Goldtouch, Timeline, Softimage, etc. Stac, Timeline, and SoftImage have all won their suits by the way.

Microsoft Windows XP is the latest release of Microsoft's flagship version of Windows, built from the ground up in the early 1990s based on the most modern concepts in operating systems, without any legacy baggage from the 1970s.

Just because XP is newer doesn't mean it's better. Is XP the most stable version of Windows to date? Yes. Is it good enough for most PC users? Again yes. It is good to run enterprise systems? Maybe. Like any OS, you have to match the requirements with the capability.

From its inception, Unix systems have been designed from the ground up to provide stability, security, and power while handling multiple users and processes. Microsoft only started trying to incorporate those features with NT. So in other words, Windows is the new kid on the block but is trying to play catch up. Is like saying my Kia is so much better and safer than your Volvo because they started designing the Kia after the Volvo has been saving lives for decades.

And it is available essentially for free, preloaded on hardware from all major manufacturers.

I hate to tell you, but you are paying for Windows because its price is rolled into the PC price. If you don't believe, go shopping on Dell.com and try customizing a business server. One of the options is to change the OS. If you remove the Factory OS, it will subract $799 from the price. In my world $799 is not "essentially free". Also if you read /., a user posted how he got refunds on Windows

/proc is really a Plan 9 From Bell Labs thing, though Linux quite reasonably picked it up.

mouse 1 then drag selects the text
keeping mouse 1 down and pressing mouse 2 cuts
pressing 1 then 3 pastes

mouse 2 alone on select text runs it as a command

mouse 3 alone on selected text sends it to the plumber.

plumbing is like file types but instead uses regexes

If there is no plumbing action defined for that pattern and you are using the text editor (Acme) then the text editor tries to open that filename. If this fails the current document is searched for the text.

This system is much better than file associations because you have more control over what happens.

Maybe when you use a particular URL you want to always view the source rather than view the rendered html, no problem. The plumbing rules are per-process so you can have different rules in different windows.

It's great

Let's reveiw some of the lies in the parent post:

and the great thing is that SCO/Caldera can not move against the BSDs because the court already ruled against them (they bought what AT&T had).

There was no court ruling, it was settled out of court. There was a preliminary finding of fact which did indeed list numerous places where AT&T's code was in BSD. Read the truth for yourself.

So while the new owners of Unix, like many recent Linux converts, have not learned from history and are doomed to repeat it the BSD groups do not have to go through it all over again!

There was no final ruling, and just because AT&T backed down (because it was revealed AT&T had stolen BSD code without keeping the Regeant's copyright -- because the BSD license is now essential no license, that trick won't work for round two) nothing prevents SCO from taking BSD right back into court just as AT&T did. Unless you guys replaced that code. Nothing stops McBride from applying the same legal extortion, since his modus operandi doesn't require revealing the alledged violation. Apple would be the primary shakedown target there.

In summary, the GPL has always been too structured for the BSD people, who believe in a might-makes-right anarchy in which coders are relagated to hobbiest or peasant status and all their work is quickly stolen by the big companies they whorship. Needless to say, seeing a company like SCO take linux just as they have always wished to sacrifice BSD to a giant beast, fills their heart with joy.

The BSD coat tail riders will not be forgotten or forgiven when this is over.

Dunno why you GNU folks seem to prefer bc... it was originally just a preprocessor for dc, which is an RPN calculator.

It's sad, but I believe this will fail horribly. The past has shown that "committee" languages don't succeed (or at least only in some niches, like Ada).

In my experience systems (like PL, applications or even OS's) are best when designed by just a small group of people: not too few and too many (I guess four to six people is a good value). Too few (just one or two) tends to include just the exclusive view of thing from the designer, while too many just makes a great mess since it's hard for a group of people to settle on one point.

Another thing I learned while reading some of the stuff about C and its history on dmr's homepage is that languages which are defined first and implemented later often hold some promises in the form of "in theory this should be very elegant and nice, but it turns out to be annoying or very, very hard to implement".

dmr also said one of the reasons C succeeded was because it was created to fit a need, not to make a point or as demonstration. So if there is need for some features then this project might succeed, but I don't see any striking needs any more, there is already a PL for almost every problem out there (and interestingly nearly every PL out there is very strong/elegant in solving certain problems but fails horribly in others).

But nonetheless I wish the people participating in this project all the best, if they would come up with an interesting, useful and beautiful language that would make an improvement to the PL world this would be a very cool thing :-)

Here's an HTML version of Why Pascal is Not My Favorite Programming Language. There's a Postscript version on Kernighan's website

If you liked the book you might love the operating system

John W. Tukey ... is of course best known for his (re)invention, with IBM's Jim Cooley, of the fast Fourier transform, (FFT) which changed the topography of digital signal processing (never mind that Gauss had the FFT 150 years earlier). Tukey was also a great wordsmith: he coined the terms bit, byte, software and cepstrum, (the Fourier transform of the logarithm of the Fourier transform). But some of his cookier coinages, like quefrency (for cepstral frequency) and saphe (for cepstral phase) didn't catch on.

I first heard about the cepstrum from John, which he had invented to distinguish underground nuclear explosions from earth quakes in connection with the US-Soviet test ban negotiations. It became immediately clear to me that the cepstrum was ideally suited for extracting the fundamental frequency (the pitch) from speech signals -- a difficult task for distorted telephone signals. The cepstrum was an ingenious idea and today, 40 years later, it remains the best method for separating long delays (travel times of seismic waves in the earth's mantle or times intervals between the puffs of the human vocal cords) from short-delay and resonance effects (of the human vocal tract).

-- Memories of John Tukey

The seminal work coining the terms is:

B.P. Bogert, M.J. R. Healy and J.W. Tukey, "The Quefrency Alanysis of Time Series for Echoes: Cepstrum, Psuedo-Autocovariance, Cross-cepstrum and Saphe Cracking", in Proceedings of the Symposium on Time Series Analysis, edited by M. Rosenblat, 1963 (New York: Wiley), pp. 209-243.

this is a little OT, but what exactly does constitute a copyright violation? if you are an artist or musican, fair use allows for a certain amount of reusing someone elses work. also, in terms of coding structure, there maybe instances when something can only be written one way. read the proceeding from the USL vs BSDi for a more detailed clarification on this topic.

until this thing goes to court(sco vs ibm), paying sco a license fee of any sort is just feeding their fud. don't do it!

The entire system, including the default program that runs in the window the equivalent of xterm [Far89] with `cutting and pasting' between windows is well under 90 kilobytes of text on a Motorola 68020 processor, about half the size of the operating system kernel that supports it and a tenth the size of the X server [Sche86] without xterm.

The components of Plan 9 are connected by a common protocol based on the sharing of files. All resources in the network are implemented as file servers; programs that wish to access them connect to them over the network and communicate using ordinary file operations. An unusual aspect of Plan 9 is that the name space of a process, the set of files that can be accessed by name (for example by an open system call) is not global to all processes on a machine; distinct processes may have distinct name spaces. The system provides methods by which processes may change their name spaces, such as the ability to mount a service upon an existing directory, making the files of the service visible in the directory. (This is a different operation from its UNIX namesake.) Multiple services may be mounted upon the same directory, allowing the files from multiple services to be accessed in the same directory. Options to the mount system call control the order of searching for files in such a union directory.

8½ serves a set of files in the conventional directory /dev with names like cons, mouse, and screen. Clients of 8½ communicate with the window system by reading and writing these files. For example, a client program, such as a shell, can print text by writing its standard output, which is automatically connected to /dev/cons, or it may open and write that file explicitly. Unlike files served by a traditional file server, however, the instance of /dev/cons served in each window by 8½ is a distinct file; the per-process name spaces of Plan 9 allow 8½ to provide a unique /dev/cons to each client. This mechanism is best illustrated by the creation of a new 8½ client.

Indeed. Not only that, but since the line between OS and IE is so blurry, its so easy to wright these click and "oh shit" scripts. Some Operating Systems simply run the OS as a userspace program, the way it was meant to be. Of course, certain other operating systems limit what software users can install in the first place.

If patents on systems implemented in software were granted from day one, the software industry would be very different than it is today. Other industries, electronics for example, deal with patents and licensing on a day to day basis and the state of the art keeps advancing. One could argue whether it would be better or worse, but at least it would be consistent. What we have today is the worst of both worlds. We have a patent office accepting patents while they have no concept of the state of the art or prior art.

Personally, I feel what the patent office is missing is the idea that computers are designed to be infinitely configurable machines, and that software just sets the machine to a particular configuration. Being able to patent software is like being able to patent a particular Lego layout.

But just to let you know, if it would have been possible to patent Hello World, and if the patent to it had coincided with the publication of the first edition of the book The C Programming Language (long after string output had implemented in software), then the patent would have expired by now.

here

Don't underestimate the power of a programming system where all data is accessed in the same way

It's here in 32 bits already