This weekend's witch-hunt

Rob Landley rob at
Tue Sep 19 03:34:44 UTC 2006

Sigh.  I composed a big long response to him on Friday, and then saved it in 
my drafts folder rather than sending it, because I just didn't want to 
continue the flamewar.  But apparently, he does...

On Monday 18 September 2006 9:34 pm, Bruce Perens wrote:
> Dear Busybox folks,
> Last week, I obejcted to Rob Landley changing the license statement on 
> portions of the busybox code that I wrote. After some (acrimonious, on 
> his part) discussion, I left him a final directive:

Ultimatum, actually.

>     Portions of this program are copyrighed by Bruce Perens. Those 
> portions may be used under the GNU General Public License, irrespective 
> of version.

Those would be the 1.2.1 and earlier versions.  I'd planned to have the actual 
announcement (which goes up on the website like the other things I want 
people who don't read the list to see) would have stated that 1.2.1 and 
earlier are still available under the license they've always been under.  
(Eventually they may move to the Legacy directory along with the other really 
old releases.)

I was already planning on doing this.  It takes serious effort to ask somebody 
to do something they're already doing in such a way that they don't say yes.  
You have a positive talent for this.

> That declaration would have respected my rights as a copyright holder, 
> and would have allowed Rob to make the license change to GPL2 only for 
> the overall program that he wished.
> Having added that declaration,

If you hadn't done the whole Schlamiel thing and had simply asked with a 
little less condescension and bluster, I was already planning to do something 
very similar to that.

However, I take objection to being treated like hired help.  I have now worked 
on this project twice as long as you ever did, and Erik's put in a half-dozen 
times as much work as you even if you'd written all the applets in BusyBox 
0.25 (which was maintained by two other people after you abandoned it before 
it got to Erik).  I _did_ ask Erik.  I warned Erik I was thinking about this 
months ago, and listened to what he had to say.  (Hint: walking in off the 
street and making demands after ten years was _stupid_.  Despite that, I 
responded politely the first few times.  You used up my strategic reserves of 

I decline the opportunity to be your Schlamazel.  Or at the very least, I'm 
washing the soup off.

> Rob could have gone on with his life as a  
> productive coder. Instead, Rob embarked on a days-long forensic 
> analysis, his log of which is visible at

Yup, it's only about half done though.

As a computer historian, it's an interesting learning experience.  And it was 
my weekend.

> and
> It appears that he's putting in a great deal of work to remove all 
> traces of Bruce Perens from the Busybox software.

Nope, just removing any leverage you may still have to fling soup on a list 
you never posted on in the entire seven years of Eric's tenure, over a 
project you abandoned ten years ago, over a license simplification that only 
applies to new releases and which someone who isn't even the maintainer could 
legally make to their own fork.

> He's missed pieces,

Remember when I called it a snapshot?

Sheesh, you're trying to make it sound like I'm hiding something.  How exactly 
does one hide something by posting it in their blog?  (And mentioning it on 
the #uclibc irc channel on sunday.)

I am _proud_ to prove that you have had no input into this project for almost 
a full decade, and yes I honestly believe that's _WHY_ the project has been 

> but I feel little motivation to assist him by naming them.

As I said, it's only about half done, but I have yet to find much that 
comparator (a legal grade forensic analysis tool) missed.  A few of the 
shorter help messages are the same.  Still about 70 files to look through.

But that's why I haven't announced it yet.  That's a snapshot of the current 
(unfinished) state.  I thought my blog made that clear, but making stuff 
clear to you is obviously a hit-or-miss proposition.

> I think it would be fair to classify his behavior as "over the top".

Pot, kettle.  Kettle, pot.

> I  
> don't believe my correspondence here has been sufficient cause for this 
> extremely nonlinear response.

You get that a lot, I take it?;708563949;fp;16;fpid;0

How'd the presidency of Linux Capital Group work out for you?  Is the Bruce 
Perens book series with Prentice Hall selling well?  How are you doing with 
Open Source Risk Management?

Did you accomplish a single concrete thing in your entire time at HP between and ?

News Flash: you have a knack for pissing people off, alienating your own base, 
and working against your own objectives.  "Messing your nest" is how I've 
heard it described.  You would appear to me to be a classic example of 
a "Schlameil".  (I'm not jewish, but my favorite wireless coffee shop's very 

Sure, plenty of people in the community are hard to get along with, from Theo 
De Raadt to Al Viro.  The thing is, they're constantly putting in a lot of 
hard work to offset this.  And they do it themselves, they don't order other 
people to do it for them.

I don't care what you _used_ to do, or what you've done elsewhere.  I 
requested that if you want a version of BusyBox under GPLv3, you either use 
any of the existing releases or you fork the project and maintain it 
yourself.  I was not joking, that was a serious suggestion.  If you don't 
like the way I'm maintaining the project, FORK IT.  (Go contact a guy named 
Vladimir Oleynik, the two of you would be perfect together.)

> However, it's free software, and Rob has  
> the right to do what he did.

Then why the public melodrama, on a list you have never previously been a 
member of?

> You, however, might want to think twice  
> before you go along with this strangeness.

I hope whoever you're referring to does.  I certainly never asked anybody to 
blindly follow me anywhere.

I put out the best versions of BusyBox I know how to do, and I do it because I 
love it.  That's the most I can manage.  I never claimed to be perfect, and 
I've never forced anybody else to use the versions I put out.  (Lots of 
people are still happy with the 0.6x releases.  I have no interest in 
supporting them, but they're up for download on the site, and will remain 

I have no problem with people disagreeing with me either.  I expect that, and 
respect it.  I disagreed with Erik about a few things when he was maintainer.  
(For example, "count" was a way better name than "pipe-progress". :)  But 
when Erik made a decision, acknowledging my concerns but continuing to 
disagree with them... he was the maintainer.

I never even tried to become maintainer, I was just the one doing the 
shoveling after Erik got busy with other things.  I wanted Erik to come back, 
and would _still_ give it back to him in a second if he had the 
time/inclination/energy to do it.  But he's still having trouble keeping up 
with his existing projects.  (The embedded world's getting big.  Keeping up 
with it's a lot of work.  I keep poking him about a 0.9.29 release of uClibc 
because I want to use it, but I try not to be annoying about it.  With mixed 
success, I suspect.  And the only help I can offer in that area is testing 
and bug reports...)

I want what is best for BusyBox.  And in my best judgement, after nine months 
of public debate on this list _before_ you showed up, that's GPLv2.

> I've been guilty of "over the top" behavior in my time, and can 
> sympathize with Rob while at the same time I don't want to leave him in 
> charge of the only development version of Busybox.

Please please please say you're forking the project.  Please.

> I have respect for  
> Rob as a coder and great respect for Erik who put in a tremendous lot of 
> work on this program.
> I do not feel a need to be the major coder of Busybox, but feel that a 
> fork is necessary and will put in the work to make one.


You'll want to talk to Vladimir Oleynik.  I don't think his email address has 
changed.  Last I heard, he was vowing to maintain his own tree.  He's a 
very... prolific coder.

> I once was very successful in distributing the Debian base system 
> development to many separate package maintainers, having got it as one 
> monolithic package. Perhaps this is possible. I see some new technical 
> directions that are possible.

Good luck.

>     Thanks
>     Bruce

Never bet against the cheap plastic solution.

More information about the busybox mailing list