Hacker News new | comments | show | ask | jobs | submit login
Alcatel-Lucent releases source for 8th, 9th and 10th editions of Unix (tuhs.org)
360 points by adamnemecek 11 months ago | hide | past | web | favorite | 57 comments

The people complaining that this isn't free-as-in-freedom should remember that there's a lot of code in here that Nokia/Alcatel-Lucent does not and has never owned. 10th edition, specifically, was never 'distributed' and probably could not be because it contained gcc. You'll note these archives are not even hosted by the corporation. They STILL aren't 'distributing' any of this. There's no way to know a priori whether there's someone else's IP in here... the packaging method for these versions of unix was "Dennis makes a copy of a running system, including whatever happened to be on that disk."

So, this is a kind gesture made for the benefit of software archaeologists. Retroactively applying some kind of modern-hippie license would cost a tremendous amount of time and money.

Folks aren't necessarily complaining about the release, rather, the issue was with the title which previously incorrectly called the release "open source". HN admins have since changed the title, see comments further down.

Maybe there needs to be a "title history" feature, which lists the revisions to the title below the actual title. Would end many confusions.

V10 hardly contained gcc. People imported whatever external software they wanted. Norman Wilson's tape doesn't include gcc, Dan Cross' tape does. However, gcc wasn't used by anything in v10, the system compiler was based on pcc.

While the engineer in me abhors the lack of process in their releases, I like the idea of their almost genetic distribution on a, like, LSD tripping level.

Almost genetic distribution?

Like a single-celled organism splitting off children.

So you can't use the code to make money. Who cares? You can still learn from the design of this stuff.

"Software archaeologists"

As a matter of fact, whenever I think about the original UNIX and how far we've come from that to modern day operating systems, I get a little nostalgic, and feel like back then software development was done for much more than just money.

Also, it's pity that this work of art (and many other pieces of software alike) have always been under the ownership of some "corporate" guys.

Also it is for a reason that the term FOSS had emerged because unlike "Open-Source" it's not subject to interpretation depending on your agenda.

"Open Source" isn't supposed to be subject to interpretation. For something to be Open Source it has to follow the OSI Open Source definition.

Go tell that to companies that deliver crippled unusable community edition only so they can put an Open-Source sticker on their paying premium product...

Technically Acatel-Lucent didn't release source. They simply agreed not to sue over the source releases in question. The folks made the source available online have been holding onto those sources for years, and have been collecting copyright non-assertion letters from various companies who might have an IP interest source in the sources. Acatel-Lucent is just the most recent company who agreed that they aren't going to sue.

This is roughly the same as signing a quit-claim deed. How much significance it has depends on how strong your previous ownership interest was in whatever you are saying you won't sue over. (For example, if I sign a quit-claim assertion over the Brooklyn Bridge, it doesn't mean much. :-)

But given this was sufficient so that the people who had been keeping private copies of Unix source, confident enough that they wouldn't be sued into oblivion, it's certainly significant in that sense.

This isn't open source, as the "no commercial use" violates a central tenet (#6) of the Open Source Defintion[1].

I believe this would be closer to "Shared Source"[2] than anything else.

[1]: https://opensource.org/osd-annotated

[2]: https://en.wikipedia.org/wiki/Shared_source

Where is Richard Stallman when you need him? This thread seems indefensibly void of "free" vs. "open-source" software ranting. (Seriously though, the restrictive license on this is really a bummer.)

> Seriously though, the restrictive license on this is really a bummer.

I question how much it really matters. If you wanted to create something like xv6 (the x86 remix of sixth edition Unix), you wouldn't want to keep too much of the original code anyway, would you?

Most of the C code I saw there are full of ancient practices, for one, I haven't seen a single #include in about ten or more files I looked at (IDK if CPP existed at all).

Link to the original Alcatel-Lucent statement: https://media-bell-labs-com.s3.amazonaws.com/pages/20170327_...

This is a little off topic but I want to take a moment and say thank you to Warren Toomey. He is responsible for TUHS and it is a wonder resource for people who enjoy UNIX.

Thank you sir!!

Should be a boon to this project to create a git history of Unix:


(previous discussion:) https://news.ycombinator.com/item?id=10995483

The README indicates that there are files for 1st and 2nd Plan 9 editions but those are not made available. [1] I guess Lucent's lawyers still want to keep their rights over those...

I have a shrink-wrapped 2nd edition distribution with manuals, but no source :(


1: http://www.tuhs.org/Archive/Distributions/Research/Dan_Cross...


So.. Anyone have any insight on what these actually provide, feature wise over v7?

Have often wondered about these 'mystery unices'..

Am sure I will trawl the source archives.. but pointers would be useful.

/proc, user-level filesystems (used for netb/netfs, and i think upas too), dmr's streams (not STREAMS), and a whole TCP/IP and DataKit stack using streams different from BSD's networking stack, the rc shell used later in Plan 9 (somehome missing from these archives though), mk (also used in Plan 9), upas (also used in Plan 9), GUI on the blit/jerk terminal, the sam text editor, and a lot of other stuff that I'm forgetting right now.


Just also found this:



Note that many of the details on wikipedia are wrong, especially time-wise. Now with the source available it's a good time to correct them, however.

Back when I was in University, we had an Amdahl mainframe with Unix running under VM. The directory structure included an awful lot of source code. I remember porting source for lex and yacc to my PC-XT running Borland's Turbo C. I assume it was licensed to Universities and source was included under an educational clause, though I'm not exactly sure.

I wonder which version of unix I was using. This would have been around December of '87.

> I assume it was licensed to Universities and source was included under an educational clause, though I'm not exactly sure.

I think that is exactly how BSDs got started, and later led to the famous AT&T vs BSDi lawsuit.

'87 - probably a System VR1 variant (though it might have been from System III)

It was either CMS (a single-user UNIX-like commonly distributed with VM) or VM/IX aka AIX/370 (an IBM-flavored SysV, but almost totally unrelated to the other AIX products.)

Good news, but it's not open source. The statement at the root of the projects says only:

"[...] that it will not assert its copyright rights with respect to any non-commercial copying, distribution, performance, display or creation of derivative works of Research Unix®".

ahh, the gems one finds in old source code: /games/trek/trek.h:

    #define ever (;;)

#define never (; 1 < 0;)

for never { foo(); }

Why not just `#define never (;0;)`? It even looks like an alien emoticon.

True, that's better, should have thought of it :)

Huh, TIL :)

These are the operating systems the Blit (https://www.youtube.com/watch?v=Pr1XXvSaVUQ) was used with.

We had BLITs at Wisconsin talking to a VAX running some 4BSD release, might have been Uwisc-BSD.

Are you sure it wasn't a DMD-5620 (also called the "jerk") and not a blit?

Note that this is not available under a conventional open source license, but one of the "non-commercial use" variety. Don't rush to incorporate it into your products ;-).

Thanks, we've updated the title from “Alcatel-Lucent open sources 8th, 9th and 10th Editions of Unix”.

Is it too soon to ask the question if it is possible to compile these and run them in some emulator?

We got v8 (for VAX) running in SIMH. We could probably bootstrap v10 from that on the same emulated hardware.

We also have a blit emulator, but it's for Plan 9 only at the moment.

It's unclear whether there exists a sun m68k emulator that could run v9.

Yes there is, it's called TME, and it runs fine!


V8 runs in simh. I don't have installation instructions though :/

Title should be: Nokia releases source for 8th, 9th and 10th editions of Unix.

Nokia bought Alcatel-Lucent over year ago. See for yourself: http://www.alcatel-lucent.com

> Nokia Bell Laboratories

The paths of mergers and acquisitions are indeed meandering.

i'd love to see code standards comparison dones for similar code, how does open source stack up.

Awesome... But only at least 25 years too late...

A better title would be:

Alcatel-Lucent makes the source code of 8th, 9th and 10th Editions of Unix public

Since the general usage of the word open source has implications about the a "free" license to use too.

Well, specifically, it implies an open source license under the OSI definition, which happens to be almost identical in a licenses covered to the Free Software definition from the FSF.

It might to you, but it shouldn't. The definition of open source is not controlled by the OSI; they only organize licenses and certify licenses as OSI-approved. However, you don't need an OSI-approved license to be open source. There are many open source packages that are not under an OSI license--most famously, SQLite.

*edited for clarity and a typo.

You don't need to use an OSI-approved license to be Open Source, but you need to use a license that complies with OSI's Open Source definition. Otherwise, the term would be meaningless.

Well, since they invented the term, to a certain degree they do.

They didn't invent the term. The term was invented first. The OSI was founded later. See the OSI's own post on this [1].

A much more detailed discussion of the origin of the term and its initial use appears here [2]. The latter link in particular is interesting reading, because it includes the political dimensions (especiall w.r.t O'Reilly's difficulties with the FSF).

[1] https://opensource.org/history [2] http://www.oreilly.com/openbook/freedom/ch11.html

Applications are open for YC Summer 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact