Kuhn: Busybox GPL enforcement concerns resolved

[Posted February 22, 2012 by corbet]

From:		"Bradley M. Kuhn" <bkuhn-AT-ebb.org>
To:		busybox-AT-busybox.net
Subject:		Re: Amusing article about busybox
Date:		Tue, 21 Feb 2012 14:47:45 -0500
Message-ID:		<878vjwkly6.fsf@ebb.org>

I had a long meeting with Tim Bird of Sony at Embedded Linux Conference
last week.  While I think Tim will probably continue to give his
personal support the ToyBox project (note: Sony has *not* given its
support to ToyBox -- that was reported incorrectly), it was clear that
much of Tim's worries about Conservancy's enforcement efforts were based
on rumors spread by a few people and those rumors weren't accurate.  I
was able to give Tim an accurate story.

Tim indicated to me that most of the requests that Conservancy makes in
GPL enforcement efforts are reasonable, now that he has heard the "real
story" of how Conservancy does enforcement.  There was a minor thing
that concerned him, but I believe we were able to work out that issue in
our discussion; I'm happy to tell anyone off list about the details of
that if anyone is interested (just email me privately).

From my point of view, my discussion with Tim settles settles the
matter.  Tim got some incorrect information about BusyBox enforcement
efforts, and that's what led him to feel he needed to support a BusyBox
replacement initially.  Tim seems to be in completely reasonable about
the whole thing now that he's talked directly with me about the actual
GPL enforcement efforts by Conservancy for BusyBox.

I wish the ToyBox project the best: it's always a good thing when
someone writes new Free Software under any license, and BusyBox
shouldn't be afraid of healthy competition.  As Rob pointed out, his
decision to relicense ToyBox from GPL to BSD was his own, and he has
every right to license the project the way that he sees fit.


The LWN article called this whole situation a "Tempest in ToyBox",
referencing the saying, a "Tempest in a Teapot".  Having been through
the whole situation to the end, I think that was an accurate title.
-- 
Bradley M. Kuhn, Executive Director, Software Freedom Conservancy

to post comments

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 22, 2012 21:42 UTC (Wed) by jiu (guest, #57673) [Link] (3 responses)

Damn, what's so different about licenses that makes even the strongest proponents of free and open source software go all hush-hush? I have an inkling that this attitude is counterproductive, and will make the 'Tempest in a teapot' reoccur.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 22, 2012 22:59 UTC (Wed) by BrucePerens (guest, #2510) [Link] (2 responses)

I'm not sure I am clear what you're asking, but if there's anything you want to know about how SFC enforces the license on Busybox, I can tell you.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 22, 2012 23:45 UTC (Wed) by jiu (guest, #57673) [Link] (1 responses)

I think that what would be better than a response here would be a post summarizing and debunking the misconceptions that led to the initiation of the whole story, clarifying things about SFC's GPL enforcement requests. Maybe this could be collated in an FAQ section of the SFC's website. I understand that you may not want to quote anyone in person as to their misconceptions but the whole story points to a lack of communication on the SFC's part.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 0:40 UTC (Thu) by bkuhn (subscriber, #58642) [Link]

My feeling is that it's not worth it to build an FAQ merely to deal with temporary FUD. Tim Bird said some things that greatly incorrect in the thread. I talked to him, and I think he's unlikely to say them again, because now he knows they weren't true. If you look back at the thread, he's pretty much the only person who was saying such things. I forgave him for it, we sorted out the issue, and it's done. Why build a big FAQ about the whole thing because of some old confusions? I've put up a blog post that explains what Conservancy actually asks for and does in enforcement. That should be enough, IMO.

The saddest part for me now is that had I know my BusyBox mailing list post above was going to get made into an LWN story, I'd have checked it for typos better. There are two of them in there; including one quoted above the fold. :)

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 22, 2012 23:05 UTC (Wed) by pr1268 (guest, #24648) [Link] (3 responses)

Well I'm glad that's settled. Of course, now Kuhn may have to focus his energies towards quelling the perceived source of Bird's motivations for creating ToyBox ("...rumors spread by a few people...").

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 4:35 UTC (Thu) by vapier (guest, #15768) [Link] (2 responses)

not sure if a typo, but Tim Bird didn't create ToyBox

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 20:30 UTC (Thu) by pr1268 (guest, #24648) [Link] (1 responses)

I was inferring from the above article: "Tim got some incorrect information about BusyBox enforcement efforts, and that's what led him to feel he needed to support a BusyBox replacement initially."

If I was wrong to assume Tim created ToyBox then please accept my apologies to all those concerned.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 21:16 UTC (Thu) by khim (subscriber, #9252) [Link]

Tim absolutely have not created the toybox. He coopted it - and this was the plan from the start as you can see.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 2:14 UTC (Thu) by branden (guest, #7029) [Link] (27 responses)

Well, don't leave us in suspense.

Was Rob Landley one of the "few people" spreading false information about how the SFC does GPL enforcement?

Maybe it's just me, but that's what I see in screaming caps...between the lines.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 3:00 UTC (Thu) by raven667 (subscriber, #5198) [Link] (16 responses)

You know, this isn't a reality tv spectacle for our enjoyment, it's real peoples lives. If the problem has been resolved to all parties satisfaction then I don't feel the need to press for embarrassing details just for the entertainment factor.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 3:07 UTC (Thu) by bronson (guest, #4806) [Link] (4 responses)

It's not real peoples' lives, it's software licensing. And, since there are still unanswered questions, it seems like it's still worth discussing.

This summary is basically "Tim Bird was super wrong, we corrected him privately, please move along." I wonder if all parties involved would agree with this...?

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 3:36 UTC (Thu) by raven667 (subscriber, #5198) [Link] (1 responses)

Outing the name of the person who misinformed tbird serves no legitimate purpose in a discussion on software licensing, it is merely an attempt to embarrass and make spectacle. If you want a substantive discussion on software licensing then take a look the previous articles and comments on this subject.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 8:27 UTC (Thu) by bronson (guest, #4806) [Link]

I agree, there's certainly no need to out anybody or ridicule anything. It would just be nice to hear how the disagreement arose and was resolved. "Mistakes were made by certain parties" is not a satisfying ending.

That said, I don't personally have any horses in this race. I'm happy to drop it here.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 8:32 UTC (Thu) by nhippi (subscriber, #34640) [Link] (1 responses)

Worth discussing.. ..if we are interested in converting lwn from an technical publication into a daily mail style tabloid.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 13:29 UTC (Thu) by wookey (guest, #5501) [Link]

Indeed. This episode actually demonstrates the value of meetings-in-person. You can have a proper discussion about things with much higher bandwidth and much lower probability of getting the wrong end of any sticks. The fact that Tim and Bradley were at the same event shortly after this blew up (and that they are sensible people who talked about it rather than sulking in corners) enabled it to be quickly and efficiently dealt with.

And I didn't give Tim an ear-bending as originally promised (I was there too, and he offered me the opportunity :-) because it had become clear by then that the whole thing was just a concatenation of misinformation and incorrect assumptions by various people.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 18:40 UTC (Thu) by landley (guest, #6789) [Link] (10 responses)

Resolved is a strong word, I'd guess most of us just don't care. (I'm only reading this now because a co-worker emailed me a link to it.)

We care about code, and this isn't about code. I put out a toybox release over a week ago (and have enough material already for another), but you wouldn't know it from lwn.net because they didn't think it worth mentioning. Instead lwn.net and such keep posting statements from people who have no code in busybox _or_ toybox. This article is a lawyer writing about his meeting with a guy who organizes trade shows.

As the guy who maintains toybox, and who used to maintain busybox, and who initiated the whole chain of lawsuits, I find my complete irrelevance to this discussion amusing. (When I get mentioned at all, it's in passing about 2/3 of the way in.)

I find it particularly strange that people keep blaming Tim for being some kind of puppetmaster. For the record, his contributions to toybox so far:

A) ask me an unrelated question that made me reexamine an obsolete licensing decision,

B) contribute an incomplete id.c that wasn't standards compliant and two other people have since patched,

C) set up a wiki page for the purpose of organizing committee meetings at a recent conference I didn't bother to attend (which I'm told didn't even wind up actually voting on anything).

D) Mentioned toybox at an OLS 2008 BOF in his slides about upcoming new developments in the embedded world. (I remember because I was in the audience, and had to speak up that the project was mothballed at the time).

That's it. I haven't gotten a dime from him or anybody else to work on toybox. (As always, I'm open to free money, but that's not why I'm doing this. http://landley.net/notes-2011.html#26-06-2011 and http://landley.net/notes-2012.html#10-02-2012 and http://lists.busybox.net/pipermail/busybox/2010-January/0... are.)

Tim's been following toybox ever since I started it, but part of his _job_ is to keep tabs on the embedded world. I think he liked toybox's design. I like the design too, but I'm always trying to improve it. Me relicensing the thing was one such improvement: it opened up the possibility of deployment on Android and supplementing or replacing android's toolbox.

The toybox relaunch was _never_ about busybox. Back when it _was_ about busybox, I mothballed the project for years and tried to port the toybox code into busybox (they took a few bits, but mostly weren't interested).

The toybox relaunch is about _android_. Busybox predates android by years, the android guys made a conscious decision _not_ to ship it, and 5 years in they're not changing their minds. So I'm providing something better which they _can_ use, by writing new code, which freaked out the "I don't code but here's how you should do it" brigades.

This entire series of articles has been due to the fact that me writing new code to replace my own old code inconveniences people who didn't write either. Once I got over my initial annoyance, other than the occasional burst of http://xkcd.com/386 my lack of caring is epic.

Rob

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 18:54 UTC (Thu) by jake (editor, #205) [Link] (9 responses)

> I put out a toybox release over a week ago (and have enough material
> already for another), but you wouldn't know it from lwn.net because
> they didn't think it worth mentioning.

Umm, Rob, did you send a release announcement to lwn@lwn.net? Maybe you did and I (and others) missed it. But if you didn't, that's a pretty good way to make sure that a project release is noticed by us.

jake

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 19:24 UTC (Thu) by landley (guest, #6789) [Link] (8 responses)

Nope, I didn't know people did that. I just mentioned it on the project's webpage, and in my blog. I'll email you about the next release.

I'm sorry if my comment unduly singled out lwn.net, I'm just most familiar with it, and this is where this particular story seemed to originate. Places like http://www.h-online.com/open/news/item/Debate-over-non-GP... pick this stuff up from you guys. You're the one who deemed various blog and mailing list posts newsworthy. (My criticism wasn't "you're not doing your job" it was "this is what passes for news these days".)

Personally, I agree there's not much to cover on toybox yet. It's nice that people are noticing it, but there's a lot of obvious stuff to do yet before it's really useful. I'm aiming for a 1.0 release in the fall, but there's a hundred or more new commands to write between now and then.

I've mostly been focusing on writing code and letting everybody who _doesn't_ write code flame themselves out. I should go back to that now...

Rob

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 21:57 UTC (Thu) by bkuhn (subscriber, #58642) [Link] (7 responses)

I have to admit I agree with Rob on this point about my post to the BusyBox mailing list not being news. I posted the email to the BusyBox list because a few subscribers there had encouraged me to talk with Tim directly. I wanted to tell them that the discussion happened. That was it.

I don't think it was news, and I was surprised that LWN picked it up as such. It's just that, in the Free Software world, the easiest way to update a large group of developers is via the mailing list for the project in question, so that's what I did. Simply as that.

News

Posted Feb 23, 2012 22:05 UTC (Thu) by corbet (editor, #1) [Link] (6 responses)

The opening of this particular extended discussion was certainly deemed to be news. How could a note about an apparent peaceful resolution and closure not be news?

It was not us that kicked off this discussion - that was Matthew Garrett. But we did report on it; should we just stop and leave our readers under the impression that a disagreement still exists when there is a statement from one of the parties involved indicating that things have been worked out?

I must admit that I'm now even more confused than usual.

News

Posted Feb 23, 2012 22:30 UTC (Thu) by bkuhn (subscriber, #58642) [Link] (5 responses)

I don't mind that you quoted it; it was a post made publicly and you have every right to do so. But, I wasn't drafting it as a news statement on the subject. If you wanted to make an article, I'd have rather you'd asked me about it my post, and allowed me to say something more substantive. That email message was specifically just to answer questions that BusyBox developers had asked me, and was not designed to be a full-on press statement.

The thing that troubles me about the situation is this: I've now realized that I have to assume that every post I make to every mailing list has a chance of being above-the-fold quoted in LWN. I suppose that was always the case, but it's never actually happened to me: in the past, the emails of mine I've had quoted as news on LWN have been designed to be news. Had I know journalists were directly subscribed to BusyBox mailing list, waiting for my emails to immediately repost them as news articles, I certainly would have at least double-checked for typos! And, I probably would have sent a more complete public statement to LWN, and then retooled that to be a post to BusyBox.

Fact is: those of us who live in the public eye already waste a lot of time trying to say everything perfectly for fear we'll get quoted on it. That already wastes a lot of time. This experience has convinced me: wow, so even a simple mailing list post to a member project of Conservancy has a chance of being LWN's headline tomorrow. Indeed, Every byte I ship out of my computer to any mailing list is a potential news headline. I guess I should have known that, but I never really thought about it before.

That said: hey, you're the press: it's your job to dig and find news whereever it hides. OTOH, I now have to find the extra 15 minutes of redrafting of every mailing list post I make, because I'll have to assume every one might be an LWN story! For a guy that usually works 12 hours a day, often including weekends, that's a daunting thing to have to add to my already overflowing workflow!

News

Posted Feb 23, 2012 22:39 UTC (Thu) by dlang (guest, #313) [Link] (2 responses)

Even if LWN isn't directly subscribed to a list, it's not uncommon for people to notice things that they find 'interesting' or 'newsworthy' and forward them to lwn@lwn.net

I don't think it's quite a dire as you are seeing it, until an issue blows up that you are involved in, the chance of a message being quoted directly and quickly is fairly low. It's only after an issue has blown up when people are especially interested in details that the odds rise (especially if the post makes it sound like something significant has taken place)

News

Posted Feb 23, 2012 23:34 UTC (Thu) by bkuhn (subscriber, #58642) [Link] (1 responses)

Yeah, perhaps you're right. In my mind, this story was already over, and telling BusyBox what was up with my conversation with Tim was merely a loose end. I see why LWN saw it differently. No big deal.

News

Posted Feb 24, 2012 3:59 UTC (Fri) by rfontana (subscriber, #52677) [Link]

Perhaps we can say, to paraphase something famously said back in 1992, that LWN.net tonight has made Bradley Kuhn the comeback kid?

News

Posted Feb 24, 2012 1:09 UTC (Fri) by raven667 (subscriber, #5198) [Link] (1 responses)

That is so adorable! If I could make a suggestion though it would be that you can't live your life trying to be perfect for the public, sometimes there will be typos that get quoted in an article on LWN. No one is expecting perfection every waking moment.

News

Posted Feb 25, 2012 12:21 UTC (Sat) by jospoortvliet (guest, #33164) [Link]

maybe, but i get where Bradley is coming from. It can be an unpleasant surprise when a message written with little extra attention to detail, as its “just a reply on a ml”, gets a far wider audience. I guess we need to get used to this, and it is great john/lwn gets the news from the source. Still feels awkward i guess...

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 14:51 UTC (Thu) by branden (guest, #7029) [Link] (8 responses)

I have to say I disagree with the consensus view here.

Whoever was spreading these rumors was behaving in an irresponsible and destructive way to our community.

I want to know who it was so that:

1) Our social mechanisms of governing the actions of our members remain effective; and

2) So that I can advise my employer against adopting technologies (e.g., software packages) under the control of irresponsible loose cannons--it's why I advise people not to use software maintained by Joerg Schilling, for example. I'm not doing *my* job if I don't frankly apprise engineering teams of the risks associated with adopting certain technologies.

It's possible that the "few people" spreading these falsehoods where just "nobodies" within the community, upon whom both the of the factors above would have little impact and from whom the community derived little benefit in the first place.

But if that is the case, one wonders why folks like Tim Bird listened to them in the first place.

If someone wants to send me a private message, I'll accept that. I don't have the time or the inclination to put up a web page saying, e.g., "BOB GRANDLEY SUX".

But if you guys saying that a little embarrassment is too high a price for a person to pay for undertaking a deliberate effort combining wrongness, malice, and poor judgment with the calculated consequence of making GPL violations more widespread, than I exhort this community to think again.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 17:20 UTC (Thu) by raven667 (subscriber, #5198) [Link] (7 responses)

I don't think it matters who this was, only that their allegations of impropriety on the part of the SFC are believed to be false. If someone starts spouting misinformation we can call them on their BS. We don't need to ostracize individuals to do so. I'm just going to trust Bradley Kuhn and his judgement on this one, that naming names and playing blame games is not productive.

On a side note I also think its silly to avoid technologies just because you don't agree with the politics of the person who created them. If their software works then it doesn't matter if they are a loudmouthed a-hole and if their software doesn't work then their software doesn't work, regardless.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 19:15 UTC (Thu) by dlang (guest, #313) [Link] (1 responses)

I disagree with you on both points.

If the person making 'incorrect statements' is Rob Landley (as many seem to be assuming by reading between the lines), he was actually in a pretty good position to know what was happening, as a plaintiff in the busybox lawsuits, and as a contractor at a company that was hit by one of the lawsuits. Just dismissing his views as 'misinformation' because the people disagreeing with him say that it is seems one-sided.

As for the maintainer's personality and attitude (which I assume is what you are meaning by politics, as opposed to Democrat/Republican type politics). I see this as very important. There are maintainers who don't work well with other projects (and frequently these same maintainers don't work well with other contributors). The code produced by these people may work very well for the exact use case that they built it for, but inevitably there are problems when the code needs to work in slightly different situations (which can be running on the same OS after kernel changes provide new ways of doing things)

Joerg Schilling with his scary warnings about using /dev/cdrom under linux is a perfect example.

DJB and his many projects are another prime example. He writes very good code, as long as you only want to do exactly what he anticipated. If you need to do anything else you start running into problems, no matter how minor the change.

I had a Senior Manager at work that was the same way, he could knock out really fast code really quickly, but the results were not easily maintainable by someone else (which combined with his habit of working through the weekend and sending an e-mail "I re-wrote this critical part of the app and deployed it to production, since I worked the weekend I'm taking monday off" resulted in real problems for the ops staff)

All three of these individuals can be great programmers, but they aren't people I want to depend on as software maintainers

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 26, 2012 23:25 UTC (Sun) by Wol (subscriber, #4433) [Link]

Actually, Rob Landley would NOT "be in a position to know".

Do you know what's going on in a company where you left their employ ten years ago? Rob has been quite open that he has had nothing to do with Busybox for quite a while, and if you want to attribute current knowledge of it to him, you're clearly mistaken. After all, isn't that why he started Toybox?!

Might it not be mjg? Not saying it is, but he seems a more likely candidate to me.

And anyways, what does it matter. As others have said, there's been an almighty misunderstanding, and it's now sorted. Why does everyone have to assume malicious intent? Why can't people accept that someone might have put their foot in it? Heaven knows I'm guilty of foot-in-mouth often enough - I'm more inclined to accept that as an explanation than maliciousness.

Cheers,
Wol

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 20:40 UTC (Thu) by branden (guest, #7029) [Link] (4 responses)

"On a side note I also think its silly to avoid technologies just because you don't agree with the politics of the person who created them. If their software works then it doesn't matter if they are a loudmouthed a-hole and if their software doesn't work then their software doesn't work, regardless."

There are multiple problems with this statement.

Firstly, being a "loudmouthed a-hole" is constitutive of a person's politics (even if one tends notice an overwhelming number of such types when they hold a political ideology in conflict with one's own).

It is, instead, a personality issue.

And personality matters. If a copyright holder or software supplier is erratic, irrational, malicious, or even just prone to doing horribly embarrassing things with which one can end up associated, this *will* weigh on adoption decisions by people evaluating the code. Especially when the people doing the evaluation are not private individuals acting solely on their own behalf, but doing so for risk-conscious corporations.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 20:42 UTC (Thu) by branden (guest, #7029) [Link]

Sigh. s/constitutive/NOT &/

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 24, 2012 1:21 UTC (Fri) by raven667 (subscriber, #5198) [Link] (2 responses)

Maybe politics wasn't the best word to use. The concept I was trying to communicate was that you shouldn't pre-judge a piece of software by criteria that are not relevant to software development. Race, gender, religion, politics, etc. are not valid criteria and the authors status in these categories doesn't "taint" a software package. I would also count personal unpleasantness in that category as well. There are plenty of successful software projects that are run by strong personalities that many people find unpleasant. If the software is buggy or doesn't work well it doesn't matter if the author is nice or not.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 24, 2012 1:59 UTC (Fri) by dlang (guest, #313) [Link]

but even if the software works and has few bugs, and unpleasant enough maintainer can effectively kill a project (or force a fork, if the project got popular enough)

being nice isn't nearly enough, but working code isn't nearly enough either.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 24, 2012 10:50 UTC (Fri) by jschrod (subscriber, #1646) [Link]

I have to disagree. While it doesn't matter if the author is nice, it really does matter if he is cooperative, responsive, and not abrasive. One has to work with him in case of problems, after all.

Kuhn: Busybox GPL enforcement concerns resolved

Posted Feb 23, 2012 18:43 UTC (Thu) by landley (guest, #6789) [Link]

No, the screaming caps were probably http://landley.net/notes.html#10-02-2012 but nobody's offered factual corrections yet.

Rob