Introduction

Licensing conditions

As the name suggests, FreeBSD is free. You don't have to pay for the code, you can use it on as many computers as you want, and you can give away copies to your friends. There are some restrictions, however. Here's the BSD license as used for all new FreeBSD code:

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

This software is provided by the FreeBSD project "as is’ and any express or implied warranties, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose are disclaimed. In no event shall the FreeBSD project or contributors be liable for any direct, indirect, incidental, special, exemplary, or consequential damages (including, but not limited to, procurement of substitute goods or services; loss of use, data, or profits; or business interruption) however caused and on any theory of liability, whether in contract, strict liability, or tort (including negligence or otherwise) arising in anyway out of the use of this software, even if advised of the possibility of such damage.

The last paragraph is traditionally written in ALL CAPS, for reasons which don’t seem to have anything to do with the meaning. Older versions of the license also contained additional clauses relating to advertising.

A little history

FreeBSD is a labour of love: big commercial companies produce operating systems and charge lots of money for them; the FreeBSD project produces a professional-quality operating system and gives it away. That's not the only difference.

In 1981, when IBM introduced their Personal Computer, the microprocessor industry was still in its infancy. They entrusted Microsoft to supply the operating system. Microsoft already had their own version of UNIX, called XENIX, but the PC had a minimum of 16 kB and no disk. UNIX was not an appropriate match for this hardware. Microsoft went looking for something simpler. The "operating system "theychose was correspondingly primitive: 86/DOS, a clone of Digital Research’s successful CP/M operating system, written by Tim Paterson of Seattle Computer Products and originally called QDOS (Quick and Dirty Operating System). At the time, it seemed just the thing: it ran fine without a hard disk (in fact, the original PC didn't have a hard disk, not even as an option), and it didn't use up too much memory. The only thing that they really had to do was to change the name. IBM called its version PC-DOS, while Microsoft marketed its version under the name MS-DOS.

By this time, a little further down the US West Coast, the Computer Systems Research Group (CSRG) of the University of California at Berkeley had just modified ATT's UNIX operating system to run on the new DEC VAX 11/780 machine, which sported virtual memory, and had turned their attention to implementing some new protocols for the ARPANET: the so-called Internet Protocols.The version of UNIX that they had developed was now sufficiently different from ATT's system that it had been dubbed Berkeley UNIX.

As time went on, both MS-DOS and UNIX evolved. Before long, MS-DOS was modified to handle hard disks - not well, but it handled them, and for the PC users, it was so much better than what they had before that they ignored the inefficiencies. After all, the PC gave you your own hard disk on your desk, and you didn’t have to share it with all the other people in the department. Microsoft even tried to emulate the UNIX directory structure, but succeeded only in implementing the concept of nested directories. At Berkeley, they were developing a higher performance disk subsystem, the Fast File System, now known as the UNIX File System.

By the late 80s, it was evident that Microsoft no longer intended to substantially enhance MS-DOS. New processors with support for multitasking and virtual memory had replaced the old Intel 8088 processor of the IBM PC, but they still ran MS-DOS by emulating the 8088 processor, which was now completely obsolete. The 640 kB memory limit of the original PC, which once appeared bigger than anybody would ever need, became a serious problem. In addition, people wanted to do more than one thing at a time with their computers.

A solution to both problems was obvious: move to the 32 bit address mode of the new Intel 80386 processor and introduce real multitasking, which operating systems on larger machines had had for decades. Of course, these larger machines were only physically larger. The average PC of 1990 had more memory, more disk and more processing power than just about any of the large computers of the 70s. Nevertheless, Microsoft didn't solve these problems for its "Windows" platform until much later, and the solutions still leave a lot to be desired.

UNIX, on the other hand, was a relatively mature operating system at the time when the PC was introduced. As a result, Microsoft-based environments have had little influence on the development of UNIX. UNIX development was determined by other factors: changes in legal regulations in the USA between 1977 and 1984 enabled ATT first to license UNIX to other vendors, noticeably Microsoft, who announced XENIX in 1981, and then to market its own version of UNIX. ATT developed System III in 1982, and System V in 1983. The differences between XENIX and System V were initially small, but they grew: by the mid-80s, there were four different versions of UNIX: the Research Version, used almost only inside ATT, which from the eighth edition on derived from 4.1cBSD, the Berkeley Software Distribution (BSD) from Berkeley, the commercial System V from ATT, and XENIX, which no longer interested Microsoft, and was marketed by the company that had developed it, the Santa Cruz Operation, or SCO.

One casualty of UNIX’s maturity was the CSRG in Berkeley. UNIX was too mature to be considered an object of research, and the writing was on the wall: the CSRG would close down. Some people decided to port Berkeley UNIX to the PC—after all, SCO had ported its version of UNIX to the PC years earlier. In the Berkeley tradition, however, they wanted to give it away. The industry’s reaction was not friendly. In 1992, ATT's subsidiary USL (UNIX Systems Laboratories)!! led a lawsuit against Berkeley Software Design, Inc. (BSDI), the manufacturer of the BSD/386 operating system, later called BSD/OS, a system very similar to FreeBSD. They alleged distribution of ATT source code in violation of licence agreements. They subsequently extended the case to the University of California at Berkeley. The suit was settled out of court, and the exact conditions were not all disclosed. The only one that became public was that BSDI would migrate their source base to the newer 4.4BSD-Lite sources, a thing that they were preparing to do in any case. Although not involved in the litigation, it was suggested to FreeBSD that they should also move to 4.4 BSD-Lite, which was done with the release of FreeBSD release 2.0 in late 1994.

Now, in the early 21st century, FreeBSD is the best known of the BSD operating systems, one that many consider to follow in the tradition of the CSRG. I can think of no greater honour for the development team. It was developed on a shoestring budget, yet it manages to outperform commercial operating systems by an order of magnitude.

The end of the UNIX wars

In the course of the FreeBSD project, a number of things have changed about UNIX. Sun Microsystems moved from a BSD base to a System V base in the late 80s, a move that convinced many people that BSD was dead and that System V was the future. Things turned out differently: in 1992, ATT sold USL to Novell, Inc., who had introduced a product based on System V.4 called UnixWare. Although UnixWare has much better specifications than SCO's old System V3 UNIX, it was never success, and Novell finally sold their UNIX operation to SCO. SCO itself was then bought out by Caldera (which recently changed its name back to SCO), while the ownership of the UNIX trade mark has passed to the Open Group. System V UNIX is essentially dead: current commercial versions of UNIX have evolved so far since System V that they can't be considered the same system. By contrast, BSD is alive and healthy, and lives on in FreeBSD, NetBSD, OpenBSD and Apple's Mac OS X.

The importance of the ATT code in the earlier versions of FreeBSD was certainly overemphasized in the lawsuit. All of the disputed code was over 10 years old at the time, and none of it was of great importance. In January 2002, Caldera released all "ancient" versions of UNIX under a BSD license. These specifically included all versions of UNIX from which BSD was derived: the first to seventh editions of Research UNIX and 32V, the predecessor to 3BSD. As a result, all versions of BSD, including those over which the lawsuit was conducted, are now freely available.

Other free UNIX-like operating systems

FreeBSD isn't the only free UNIX-like operating system available—it's not even the best-known one. The best-known free UNIX-like operating system is undoubtedly Linux, but there are also a number of other BSD-derived operating systems. We'll look at them first:

You could get the impression that there are lots of different, incompatible BSD versions. In fact, from a user viewpoint they're all very similar to each other, much more than the individual distributions of Linux, which we'll look at next.

FreeBSD and Linux

In 1991, Linux Torvalds, then a student in Helsinki, Finland, decided he wanted to run UNIX on his home computer. Atthat time the BSD sources were not freely available, and so Linus wrote his own version of UNIX, which he called Linux.

Linux is a superb example of how a few dedicated; clever people can produce an operating system that is better than well-known commercial systems developed by a large number of trained software engineers. It is better even than a number of commercial UNIX systems.

Obviously, I prefer FreeBSD over Linux, or I wouldn't be writing this course , but the differences between FreeBSD and Linux are more a matter of philosophy rather than of concept. Here are a few contrasts:

FreeBSD system documentation

FreeBSD comes with a considerable quantity of documentation which we'll look at in the following few pages:

Reading online documentation

You'll find a number of HTML documents in the directory /usr/share/doc/en/books:

In addition to the directory /usr/share/doc/en/books, there's also a directory /usr/share/doc/en/articles with a number of shorter items of documentation.

Note the component en in the pathnames above. That stands for English.A number of these books are also installed in other languages: change en to de for a German version, to es for Spanish, to fr for French, to ja for Japanese, to ru for Russian, or to zh for Chinese. Translation efforts are continuing, so you may find documentation in other languages as well.

If you're running X, you can use a browser like mozilla to read the documents. If you don’t have Running yet, use lynx. Both of these programs are included in the CD-ROM distribution. To install them, use sysinstall, which is described on page 92.

lynx is not a complete substitute for complete web browsers such as mozilla:since it is text-only, it is not capable of displaying the large majority of web pages correctly. It's good enough for reading most of the FreeBSD online documentation, however.

In each case, you start the browser with the name of the document, for example:

$ lynx /usr/share/doc/en/books/handbook/index.html
$ mozilla /usr/share/doc/en/books/handbook/index.html 

Enter the after the invocation of mozilla to free up the window in which you invoke it: mozilla opens its own window.

If you haven’t installed the documentation, you can still access it from the Live File system CD-ROM. Assuming the CD-ROM is mounted on /cdrom, choose the file /cdrom/usr/share/doc/en/books/handbook/index.html.

Alternatively, you can print out the handbook. This is a little more difficult, and of course you'll lose the hypertext references, but you may prefer it in this form. To format the handbook for printing, you'll need a PostScript printer or ghostscript. See page 271 for more details of how to print PostScript.

The printable version of the documentation doesn't usually come with the CD-ROM distribution. You can pick it up with ftp (see page 433) from ftp://ftp.FreeBSD.ORG/pub/FreeBSD/doc/,which has the same directory structure as described above. For example, you would download the handbook in PostScript form from ftp://ftp.FreeBSD.ORG/pub/FreeBSD/doc/en/books/handbook/book.ps.bz2.

The online manual

The most comprehensive documentation on FreeBSD is the online manual, usually referred to as the man pages. Nearly every program, file, library function, device or interface on the system comes with a short reference manual explaining the basic operation and various arguments. If you were to print it out, it would run to well over 8,000 pages.

When online, you view the man pages with the command man. For example, to learn more about the command ls, type:

$ man ls
LS(1)   FreeBSD Reference Manual   LS(1)
NAME
   ls - list directory contents
SYNOPSIS
   ls [-ACFLRTacdfiloqrstu1][file ... ]
DESCRIPTION
   For each operand that names a file of a type other than directory, ls
   displays its name as well as any requested, associated information. For
   each operand that names a file of type directory, ls displays the names.
(etc)

In this particular example, with the exception of the first line, the text in constant width bold is not input, it's the way it appears on the screen.

The online manual is divided up into sections numbered:

In some cases, the same topic may appear in more than one section of the online manual. For example, there is a user command chmod and a system call chmod(). In this case, you can tell the man command which you want by specifying the section number:

$ man 1 chmod 

This command displays the manual page for the user command chmod. References to a particular section of the online manual are traditionally placed in parentheses in written documentation. For example, chmod(l) refers to the user command chmod, and chmod(2) means the systcode call.

This is fine if you know the name of the command and forgot how to use it, but what if you can't recall the command name? You can use man to search for keywords in the command descriptions by using the -k option, or by starting the program apropos:

$ man -k mail $
apropos mail

Both of these commands do the same thing: they show the names of the man pages that have the keyword mail in their descriptions.

Alternatively, you may browse through the /usr/bin directory, which contains most of the system executables. You'll see lots of file names, but you don't have any idea what they do. To find out, enter one of the lines:

$ cd /usr/bin; man -f * 
$ cd /usr/bin; whatis *

Both of these commands do the same thing: they print out a one-line summary of the purpose of the program:

$ cd /usr/bin; man -f *
a2p(1)        - Awk to Perl translator
addftinfo(l)  - add information to troff font files for use with groff
apply(l)      - apply a command to a set of arguments
apropos(1)    - search the whatis database
...etc

Printing man pages

If you prefer to have man pages in print, rather than on the screen, you can do this in two different ways:

$ man ls | lpr

$ man -t ls | lpr

GNU info

The Free Software Foundation has its own online hypertext browser called info. Many FSF programs come with either no man page at all, or with an excuse for a man page (gcc, for example). To read the online documentation, you need to browse the info files with the info program, or from Emacs with the info mode. To start info,simply type:

$ info

In Emacs, enter CTRL-h i or ALT-x info. Whichever way you start info, you can get brief introduction by typing h, and a quick command reference by typing ?.

Other documentation on FreeBSD

FreeBSD users have access to probably more top-quality documentation than just about any other operating system. Remember that word UNIX is trademarked. Sure, the lawyers tell us that we can't refer to FreeBSD as UNIX, because UNIX belongs to the Open Group. That doesn't make the slightest difference to the fact that nearly every book on UNIX applies more directly to FreeBSD than any other flavour of UNIX. Why?

Commercial UNIX vendors have a problem, and FreeBSD doesn't help them: why should people buy their products when you can get it free from the FreeBSD Project (or, for that matter, from other free UNIX-like operating systems such as NetBSD, OpenBSD and Linux)? One obvious reason would be "value-added features." So they add features or fix weak points in the system, put a copyright on the changes, and help lock their customers in to their particular implementation. As long as the changes are really useful, this is legitimate, but it does make the operating system less compatible with "standard UNIX," and the books about standard UNIX are less applicable.

In addition, many books are written by people with an academic background. In the UNIX world, this means that they are more likely than the average user to have been exposed to BSD. Many general UNIX books handle primarily BSD, possibly with an additional chapter on the commercial System V version.

In Appendix A, Bibliography, you’ll find a list of books that I find worthwhile. I'd like to single out some that I find particularly good, and that I frequently use myself:

There are also many active Internet groups that deal with FreeBSD. Read about them in the online handbook.

The FreeBSD community

FreeBSD was developed by a world-wide group of developers. It could not have happened without the Internet. Many of the key players have never even met each other in person; the main means of communication is via the Internet. If you have any kind of Internet connection, you can participate as well. If you don't have an Internet connection, it's about time you got one. The connection doesn't have to be complete: if you can receive email, you can participate. On the other hand, FreeBSD includes all the software you need for a complete Internet connection, not the very limited subset that most PC-based "Internet" packages offer you.

Mailing lists

As it says in the copyright, FreeBSD is supplied as-is, without any support liability. If you're on the Internet, you're not alone, however. Liability is one thing, but there are plenty of people prepared to help you, most for free, some for fee. A good place to start is with the mailing lists. There are a number of mailing lists that you can join. Some of the more interesting ones are:

You can find a complete list of FreeBSD mailing lists on the web site, currently at http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/handbook/eresources.html. This address is part of the online handbook and may change when the handbook is modified; follow the link if it is no longer valid, or if you can't be bothered typing in the URI.

The mailing lists are run by mailman (in the Ports Collection). Join them via the web interface mentioned above. You will receive a mail message from mailman asking you to confirm your subscription by replying to the message. You don't need to put anything in the reply: the reply address is used once only, and you're the only person who will ever see it, so the system knows that it's you by the fact that you replied at all. You also have the option of confirming via a web interface with a specially generated URI. Similar considerations apply in this case.

FreeBSD mailing lists can have a very high volume of traffic. The FreeBSD-questions mailing list, for example, has thousands of subscribers, and many of them are themselves mailing lists. It receives over a hundred messages every day. That's about a million messages a day in total for just one mailing list, so when you sign up for a mailing list, be sure to read the charter. You can find the URI from the mailman confirmation message. It's also a good idea to "lurk" (listen, but not say anything) on the mailing list a while before posting anything: each list has its own traditions.

When submitting a question to FreeBSD-questions, consider the following points:

How to follow up to a question

Often you will want to send in additional information to a question you have already sent. The best way to do this is to reply to your original message. This has three advantages:

There are more suggestions, in particular for answering questions, at http://www.lemis.com/questions.html. See also Chapter 26, and the support page at http://www. FreeBSD.org/support.html.

In addition, a number of companies offer support for FreeBSD. See the web page http://www.FreeBSD.org/commercial/consulting_bycat.html for some possibilities.

Unsubscribing from the mailing lists

There's a lot of traffic on the mailing lists, particularly on FreeBSD-questions. You may find you can't take it and want to get out again. Again, you unsubscribe from the list either via the web or via a special mail address, not by sending mail to the the list. Each message you get from the mailing lists finishes with the following text:

freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscribe@freebsd.org"

Don't be one of those people who send the unsubscribe request to the mailing list instead.

User groups

But how about meeting FreeBSD users face to face? There are a number of user groups around the world. If you live in a big city, chances are that there's one near you. Check http://www.FreeBSD.org/support.html#user for a list. If you don't find one, consider taking the initiative and starting one.

In addition, USENIX holds an annual conference, the for more details.

Reporting bugs

If you find something wrong with FreeBSD, we want to know about it, so that we can fix it. To report a bug, use the send-pr program to send it as a mail message.

There used to be a web form at http://www.FreeBSD.org/send-pr.html, but it has been closed down due to abuse.

The Berkeley daemon

The little daemon at the right symbolizes BSD. It is included with kind permission of Marshall Kirk McKusick, one of the leading members of the former Computer Sciences Research Group at the University of California at Berkeley, and owner of the daemon's copyright. Kirk also wrote the foreword to this course .

The daemon has occasionally given rise to a certain amount of confusion. In fact, it's a joking reference to processes that run in the background—see Chapter 8, Taking control, page 150, for a description. The outside world occasionally sees things differently, as the following story indicates:

Newsgroups:   alt.humor.best-of-usenet
Subject:   [comp.org.usenix] A Great Daemon Story
From:   Rob Kolstad <kolstad@bsdi.com>
Newsgroups:   comp.org.usenix
Subject:   A Great Daemon Story

Linda Branagan is an expert on daemons. She has a T-shirt that sports the daemon in tennis shoes that appears on the cover of the 4.3BSD manuals and The Design and Implementation of the 4.3BSD UNIX Operating System by S. Leffler, M. McKusick, M. Karels, J. Quarterman, Addison Wesley Publishing Company, Reading, MA 1989.

She tells the following story about wearing the 4.3BSD daemon T-shirt:

Last week I walked into a local "home style cookin' restaurant/watering hole" in Texas to pick up a take-out order. I spoke briefly to the waitress behind the counter, who told me my order would be done in a few minutes.

So, while I was busy gazing at the farm implements hanging on the walls, I was approached by two "natives." These guys might just be the original Texas rednecks.

"Pardon us, ma'am. Mind if we ask you a question?"

Well, people keep telling me that Texans are real friendly, so I nodded.

"Are you a Satanist?"

Well, at least they didn't ask me if I liked to party.

"Uh, no, I can't say that I am."

"Gee, ma'am. Are you sure about that?" they asked.

I put on my biggest, brightest Dallas Cowboys cheerleader smile and said, "No, I'm positive. The closest I've ever come to Satanism is watching Geraldo."

"Hmmm. Interesting. See, we was just wondering why It is you have the lord of darkness on your chest there."

I was this close to slapping one of them and causing a scene—then I stopped and noticed the shirt I happened to be wearing that day. Sure enough, it had a picture of a small, devilish-looking creature that has for some time now been associated with a certain operating system. In this particular representation, the creature was wearing sneakers.

They continued: "See, ma'am, we don'texactly appreciate it when people showoff pictures of the devil. Especially when he'slookin' so friendly."

These idiots sounded terrifyingly serious.

Me: "Oh, well, see, this isn't really the devil, it's just, well, it's sort of a mascot.

Native: "And what kind of football team has the devil as a mascot?"

Me: "Oh, it's not a team. It's an operating—uh, a kind of computer."

I figured that an ATM machine was about as much technology as these guys could handle, and I knew that if I so much as uttered the word "UNIX" I would only make things worse.

Native: "Where does this satanical computer come from?"

Me: "California. And there's nothing satanical about it really."

Somewhere along the line here, the waitress noticed my predicament—but these guys probably outweighed her by 600 pounds, so all she did was look at me sympathetically and run off into the kitchen.

Native: "Ma'am, I think you're lying. And we'd appreciate it if you'd leave the premises now."

Fortunately, the waitress returned that very instant with my order, and they agreed that it would be okay for me to actually pay for my food before I left. While I was at the cash register, they am used themselves by talking to each other.

Native #1: "Do you think the police know about these devil computers?"

Native #2: "If they come from California, then the FBI oughta know about 'em."

Theye scorted me to the door. I tried one last time: "You're really blowing this all out of proportion. Alot of people use this 'kind of computers.' Universities, researchers, businesses. They're actually very useful."

Big, big, big mistake. I should have guessed at what came next.

Native:"Does the government use these devil computers?"

Me: "Yes."

Another big boo-boo.

Native:"And does the government pay for 'em? With our tax dollars?"

Idecided that it was time to jump ship.

Me: "No. Nope. Not at all. Your tax dollars never entered the picture at all. I promise. No sir, not a penny. Our good Christian congressmen would never let something like that happen. Nope. Never. Bye."

Texas. What acountry.

The daemon tradition goes back quite a way. Asrecently as 1996, after the publication of the first edition of this book, the following message went through the FreeBSD-chat mailing list:

To:   "Jonathan M. Bresler" <jmb@freefall.freebsd.org>
Cc:   obrien@antares.aero.org (Mike O'Brien),
    joerg_wunsch@uriah.heep.sax.de,
    chat@FreeBSD.org, juphoff@tarsier.cv.nrao.edu 
Date:   Tue, 07 May 1996 16:27:20 -0700
Sender:   owner-chat@FreeBSD.org
    >details and gifs PLEASE!
   If you insist. :-)
   Sherman, set the Wayback Machine for around 1976 or so (see Peter Salus' A 
Quarter Century of UNIX for details), when the first really national UNIX 
meeting was held in Urbana, Illinois. This would be after the "forty people 
in a Brooklyn classroom" meeting held by Mel Ferentz (yeah I was at that too) 
and the more-or-less simultaneous West Coast meeting(s) hosted by SRI, but before 
the UNIX Users Group was really incorporated as a going concern.
   I knew Ken Thompson and Dennis Ritchie would be there. I was living in Chicago 
at the time, and so was comic artist Phil Foglio, whose star was just beginning 
to rise. At that time I was a bonded locksmith. Phil's roommate had unexpectedly 
split town, and he was the only one who knew the combination to the wall safe in 
their apartment. This is the only apartment I've ever seen that had a wall safe, 
but it sure did have one, and Phil had some stuff locked in there. I didn't hold 
out much hope, since safes are far beyond where I was (and am) in my locks mithing 
sphere of competence, but I figured "no guts no glory" and told him I'd 
give it a whack. In return, I told him, he could do some T-shirt art for me. He readily agreed.
   Wonder of wonders, this safe was vulnerable to the same algorithm that Master 
locks used to be susceptible to. I opened it in about 15 minutes of manipulation. 
It was my greatest moment as a locksmith and Phil was overjoyed. I went down to my 
lab and shot some Polaroid snaps of the PDP-11 system I was running UNIX on at the 
time, and gave it to Phil with some descriptions of the visual puns I wanted: pipes, 
demons with forks running along the pipes, a "bit bucket" named /dev/null, all that.
   What Phil came up with is the artwork that graced the first decade's worth of 
"UNIX T-shirts," which were made by a Ma and Paoperation in a Chicago 
suburb. They turned out transfer art using a 3M color copier in their basement. 
Hence, the PDP-11 is reversed (the tape drives are backwards) but since Phil left off 
the front panel, this was hard to tell. His trademark signature was photo-reversed, 
but was recopied by the T-shirt people and "re-forwardized," which is why 
it looks a little funny compared to his real signature.
   Dozens and dozens of these shirts were produced. Bell Labs alone accounted for 
an order of something like 200 for a big picnic. However, only four (4) REAL 
originals were produced: these have a distinctive red collar and sleeve cuff. 
One went to Ken, one to Dennis, one to me, and one to my then-wife. I now possess 
the latter two shirts. Ken and Dennis were presented with their shirts at the Urbana conference.
People ordered these shirts direct from the Chicago couple. Many years later, 
when I was living in LA, I got a call from Armando Stettner, then at DEC, 
asking about that now-famous artwork. I told him I hadn't talked to the Illinois 
T-shirt makers in years. At his request I called them up. They'd folded the operation 
years ago and were within days of discarding all the old artwork. I requested its return, 
and duly received It back in the mail. It looked strange, seeing it again in its 
original form, a mirror image of the shirts with which I and everyone else were now familiar.
I sent the artwork to Armando, who wanted to give it to the Ultrix marketing people. 
   They came out with the Ultrix poster that showed a nice shiny Ultrix machine 
contrasted with the chewing-gum-and-string PDP-11 UNIX people were familiar with.
They still have the artwork, so far as I know.
   I no longer recall the exact contents of the letter I sent along with the artwork. 
I did say that as far as I knew, Phil had no residual rights to the art, 
since it was a 'work made for hire', though nothing was in writing 
(and note this was decades before the new copyright law). I do not now recall 
if I explicitly assigned all rights to DEC. What is certain is that John Lassiter's daemon, 
whether knowingly borrowed from the original, or created by parallel evolution, 
postdates the first horde of UNIX daemons by at least a decade and probably more. 
And if Lassiter's daemon looks a lot like a Phil Foglio creation, there's are a son.
   I have never scanned in Phil's artwork; I've hardly ever scanned in anything, 
so I have no GIFs to show. But I have some very very old UNIX T-shirts in 
startlingly good condition. Better condition than I am at any rate: 
I no longer fit into either of them.

Mike O'Brien
Creaky antique

Note the date of this message: it appeared since the first edition of this book. Since then, the daemon image has been scanned in, and you can find a version at http://www. mckusick.com/beastie/shirts/usenix.html