Xinu Is Not Unix | Hacker News Reader

Xinu Is Not Unix

xinu.cs.purdue.edu

45 pointscolinprince21 comments

I took the course at Purdue that built on Xinu (CS 354, Operating Systems) — Xinu is really just used for teaching purposes. Doug Comer (the creator of Xinu) was my favorite professor — it was inspiring to hear from someone who was pretty close to the development of the Internet and TCP/IP in the 70s (https://en.wikipedia.org/wiki/Douglas_Comer).

Took this epic course. Getting virtual memory to work on x86 is still one of the most satisfying projects ever done! Thanks Dr Comer and everyone.

How does it compare to os161 as a teaching OS ? What I really like about it is how fast it builds and runs and how comprehensive the course is with a decent amount of assignments.

I never did CS161. They look similar from the course page.

Huh - my old RTOS professor from undergrad contributed a lot to Xinu and used to shill it to us in class. Didn't think I'd ever see it again.

We focused on it in undergrad too. If I remember right the entire kernel was like 8 pages in the textbook, including commentary.

I took another class from a professor who had never written an OS. It was decent but Xinu was much better.

Agreed, I was an EE at Purdue but still familiar with Xinu from CS clubs. Didn't expect to see it on HN, but a fun throwback!

Not to be confused with the Berkeley software company "Mt Xinu". (The operating system's name is a recursive acronym, while the company's name is a backwards spelling.)

"We know UNIX TM backwards and forwards." -Mt Xinu

https://en.wikipedia.org/wiki/MtXinu

Famous for the great posters they handed out at Usenix:

"4.2 > V" BSD -vs- System V, X-Wing / Death Star Poster

https://raw.githubusercontent.com/ashaferian/Drive/master/Mt...

https://www.ericconrad.com/2008/12/

I love all the old telephone equipment in the explosion!

Very little info on the website and Wikipedia about this OS.

You could find more in the book. The OS was done mostly for teaching purposes. You could write the building blocks of the OS: virtual memory, context switch etc in an easier environment than Linux/UNIX.

I'm sorry, but I don't see why anyone should waste a moment with a proprietary operating system with such limited use.

If you need a simple educational operating system with a book, Minux is available and open source software.

If you need a fully capable operating system kernel, Linux is available and is open source.

There are also many other open source software systems, including the *BSDs, Zephyr, and so on.

It takes time to learn the details of an operating system. I read parts of those books years ago, but I wouldn't waste any of my time with them today. Better to learn details of something you can use, or at least something you are legally allowed to change and share.

If it is open source software, and my comments don't apply, but then they should make that much clearer on their front page.

$ tar xzvf Xinu-code-Galileo.tar.gz

$ cd Xinu-code-Galileo/

$ cat COPYRIGHT

<<<begin COPYRIGHT>>> Portions of this software are covered by the following copyright:

        Copyright (c) 2012, 2015 Douglas E. Comer and CRC Press, Inc.
                        All rights reserved.

Redistribution and use in source and binary forms are permitted provided that this notice is preserved and that due credit is given to the copyright holders. The names of the copyright holders may not be used to endorse or promote products derived from this software without specific prior written permission. This software is provided `as is'' without express or implied warranty. The author assumes no liability for damages incidental or consequential, nor is the software warranted for correctness or suitability for any purpose. Portions of this software are documented in the book:

        Operating System Design -- The Xinu Approach, CRC Press,
        Boca Raton, FL, Linksys version 2012, Second Edition 2015.

This software may not be sold or published in printed form without written permission from the copyright holders. <<<<end COPYRIGHT>>>

It never ceases to astound me how many minor variants on 3-clause BSD retain clause 4 (the endorsement clause), despite not carrying the historical baggage of clause 3 (the advertising clause).

It made sense for 4-clause BSD (and its unbulleted predecessor) as a qualifier on clause 3: although advertising materials must contain the acknowledgement, it should not be presented such that could be misconstrued as an endorsement. Even when striking clause 3, clause 4 remained inoffensive enough to not warrant making more than the minimal change.

But for new work without that history, it's a total non sequitur.

You have an odd definition of "new work". My copy of the first Xinu book is dated 1984, pre-dating the invention of the GNU General Public Licence by half a decade.

Whatever the provenance of the content, the license is almost certainly newer than 4-clause BSD, evidenced by +endorse/-advertising clauses.

This has nothing whatsoever to do with GPL.

You were the one that brought up the provenance, by erroneously classififying Xinu as "new work without that history". Now you are trying to have your cake and eat it by trying to hand wave away the very thing that you yourself brought up.

Quite clearly, something from 1984 (actually earlier) is not "new work", and has a history: a history that pre-dates by half a decade even the very invention of one of the major free software copyright licences.

If you do not understand why the non-existence of the GPL at the time of first publication is relevant to the history of something being copyright licenced the way that it is, then you have a lot of computing history to learn, including all of the things that the world went through after the first publication of Xinu when it came to ideas in software copyright licencing. Xinu dates from a time when in order to run BSD one had to have an AT&T source licence.

The first part is a standard BSD license, however it has a weird extra restriction. It cannot be sold or published in printed form. That is a restriction on use that almost certainly makes it not open source software. It certainly isn't a common license. Game over.

But even if it was open source software, it's unreasonable for people to have to untar that file on a phone. As I posted earlier, if it's OSS, that info needs to be a lot clearer. It certainly needs to be on the front pagd. Requiring a untar is absurd for such basic information.

As a piece of history, it's cool, though I think the Lyons book is more important.

In years gone by, it was common practice to look inside archive files to inspect README files and suchlike. Certainly, one is quite reasonably expected to look inside an archive of a teaching operating system where reading and understanding the source tree is part of the process. I think that you are unwittingly making a strong case for your 'phone being the wrong tool for the job. (-:

> In years gone by, it was common practice to look inside archive files to inspect README files and suchlike.

That's true, I did it many times. But those years are long ago. Good riddance.

Most of the time today, for most people, a smartphone is the only tool you have available at the moment. When I made that comment I was nowhere near my laptops or desktops or tablet, and I think that's true for many people today. According to Statcounter the majority of pages viewed today are viewed using a smartphone: https://gs.statcounter.com/platform-market-share/desktop-mob...

> Certainly, one is quite reasonably expected to look inside an archive of a teaching operating system where reading and understanding the source tree is part of the process.

Not to find out the license. A license is important information that you need to know before you look at the archive.

I'm don't think it's even reasonable today to ask that people unarchive just to look at some code. GitHub, GitLab, SourceForge, and many other forges/repo sites make it quite easy to look at code without having to download and look inside an archive. By all means provide a simple link to download its archive (GitHub does this automatically), but it's also important to give people an easy way to look at its individual source files using a URL. I'll note that the Linux kernel does this just fine.

Things are hard enough on potential developers and potential users. Developers should work to make it easy for potential users and potential users. If nothing else, their projects are more likely to do well that well.

“This software may not be sold or published in printed form without written permission from the copyright holders.“ This is almost certainly because parts of the source code appear in the textbook, and publishing laws apply.

Reply on news.ycombinator.com