SCO, Linux and Rob Enderle: A Conclusion (TG Daily)

• Because we are in an interesting settling of the accounts phase with regard to SCO and I was curious to see how this guy would explain himself.

• Because it is interesting to see how groups like SCO manage to rally "analysts" and others to its side. We will see it again.

• Because I don't like how loudmouths "representing" our community give this sort of person the excuse to turn against us while claiming to fear for his life.

• Because it is, in its own distorted way, a piece of the historical record.

Sorry if it bothered you.

The lowly entry level women who apparently sniffed out SCO lies instantly would not go and see the "evidence" because SCO required the signing of what they were calling an NDA but was really a gag order.

His description of "winning" the argument against Linux in the Enterprise is just bizzar. This was 2003. Guess he considers being wrong "winning".

I find it odd also that the first three people he either questions the judgment of or "defeats" in argument are women.

The "I had done nothing wrong" statement sort of rhymes with "I am not a crook". He claims he saw copied code and reported on it. The problem for him is that the code he saw was not "copied" in the sense that the people who he called thieves "did nothing wrong" in copying it.

That he accuses other people of not "fact checking" is just priceless. As are his mechanations around Groklaw. He accuses Groklaw of being misleading, unlike apparently his own insightful analysis which has survived the test of time.

Enough of this guy. It is really a shame such arrogance is rewarded.

Kind Regards

Sadly, I'm sure people will still listen to him since he's an "analyst".

Meanwhile... his article didn't seem quite as bad to me as many seem to be
portraying it. There's certainly some faults in it; I don't particularly
like the way he talks about that junior analyst for one thing. He
says he recognizes that he let others manipulate him into taking a
position he otherwise wouldn't have taken, and I see no harm in taking
that at face value. He now recognizes that manipulation, and that it lead
him to take the wrong position, a position a more mature person or one
with different weaknesses wouldn't have taken. That's coming some way and
it looks like he recognizes he's still got a way to go, some maturing to
do.

While I'm not going to be trusting any recommendations he makes right
away, I'm not going to be casting stones, either. Quoting the (now
disbanded) "Up with People" theme song, "If more people people were for
people, all people everywhere, there'd be a lot less people to worry
about, and a lot more people who care." He's "people" and certainly not
faultless (the junior analyst thing remains one here), but he's "people",
and as such, he's allowed mistakes, even ones as bad as this, and
as "people", he deserves a chance to mature and to change his ways.
Others do what they will, I choose to give it to him.

Duncan

Oops, I meant WWF (it's been a long time).

But yes, this NDA is almost certainly why the junior analyst Enderle mentions said she wasn't able to talk to SCO -- she wasn't able to talk to them on reasonable, sane grounds. Enderle didn't care about that, drank the kool-aid, and then apparently has a personality such that he can't admit mistakes and so had to keep chugging even long after it was clear the stuff was acid-laced.

There was an NDA to see the "violating code". Reading between the lines Enderle talks about that here:

... I had to make sure we could stand behind what our Open Source analyst was saying: She took the initial position that SCO’s copied code claim was a lie. ...

Our analyst indicated that SCO wouldn’t speak to her, so I contacted them myself to see if this claim was true. They said 'nope' and told me that they would show us the code. She refused to meet with them and then did everything in her power to keep me from meeting with them. I actually never had this happen before in the years I had been an analyst: I carried the title of vice president and she was near entry level. Preventing the meeting should not have been possible, not even remotely. She almost was successful.

I suspect that if LWN were to track down this "near entry level" analyst (and it took me five minutes of Googling to figure whom that is likely to be) we would find that she would not sign the NDA since it would restrict the analyst firm's ability to give advice to its customers.

It has long been suspected that the code shown to the analysts that did sign the NDA was same as presented in obscured form at SCO Forum 2003. An analysis of that code showed is was (1) the Berkeley Packet Filter code with the UCB copyright and BSD license removed and (2) a small amount of System V code in ate_utils.c, contributed by SGI. An analysis of the kernel code by SGI for other infringing code found one other small fragment, which was promptly removed. It was later discovered that the removed code was part of the Ancient UNIX release and thus no copyright infringement occurred.

Rob's article goes some way to confirming that suspicion:

... I met around the same time with SCO. They showed me what they claimed was copied code. As it turned out later, the code they showed me was mostly in the public domain and, as it eventually turned out, anything that was left was not theirs in the first place. (SCO actually had a relating internal memo I did not know about until years later that detailed the first part of this, the second part had to wait for the recent Novell ruling.)

There was a lot of discussion about the examples at the time, but from memory they broke into three pieces:

1) standard macro defintions like errno.h and signal.h
SCO doesn't own this (nor does Novell), and it was also obvious that it wasn't an exact copy. Of course Linux uses the same names -- they are in multiple published standards. And Linux uses the same (mostlly) numbers as other x86 Unix-like systems -- it's a defacto standard.

2) the BPF code
SCO doesn't own this (nor does Novell), and in fact the two versions they were comparing weren't the same. Again, it's an exported interface which is a de-facto standard. The original was published under a BSD license. Notably SCO's "original" copy had that license and copyright information removed.

3) a short kernel-space implementation of malloc for SGI's Altix platform
SGI contributed this code, and it did come from Unix, though it had been there since _very_ early days, and those versions are not in copyright. In any case, the kernel developers were not comfortable with it, and the code should have been using the standard allocators anyway so it was ripped out.

All in all is was a tiny amount of code, and it demonstrated SCO's ineptitude (and possibly intentional deception) more than anything else.

The fact that the analysts kept clinging to SCO's claims (which continued to expand) after this debunking is what really raised red flags, at least in my mind. I now instantly notice when an analyst is quoted, which is far too often, and how they never say anything very insightful. In short, I learned to not assume that people reporting the news have done their homework, and to not rule out the possibility that they have an undisclosed relationship with what they are reporting on.

"Pass the source, please"

There is little doubt the open-source software movement has anti-capitalist elements. It even has a manifesto, the well-known tract The Cathedral and the Bazaar, written by the movement's very own Karl Marx, Eric Raymond (find it at catb.org/~esr/writings/cathedral-bazaar).

Enderle seems to be trying to say he's not responsible for anything he did because he was just doing what he though was his job.

"Especially Didio"? Why does she get a pass?

These people are not just incompetents and liars. Anybody can be that. They made their living at being incompetents and liars. If anything, they experienced much less meanness than they earned: all are still employed, right?

Enderle has not a single word in this long piece for his readers. Lots of words about market players, editors, analysts, Linux meanies, his own importance, but about the sheep who believed him? Nothing.

He's not just any journalist. He's supposed to give valuable advice to people. He's not writing for entertainment, or hunting scoops, but advising people. And here we see those people count for nothing in his mind.

He reminds me of the late middle-age knights that cared only about the artificial rules of their tourneys, and had completely forgotten they were supposed to protect people and win actual battles.

Right.

> I think his piece is not entirely well written, but this sounds like a
> human being who is torn in between ethical principles and personal
> involvement. He did not manage to free himself back then and he is
> obviously still no able to look on the story in an unbiased manner. But at
> least he is able to think about this fact and explains his personal
> problems with situations as this.

You're making the rather large assumption that what he writes and what he thinks / feels are at least somewhat related.

> I came to see that it is extremely difficult to stand up and state that the
> people who treat you nice are assholes and that the people who treat you
> like an asshole are actually nice. It takes a very clear sight and a big
> deal of detachment. Most people do not even get to see the gap. Let alone
> let go of their friends.

It's not difficult if you're lying. (and, ostensibly, lying comes as easily to you as it seems to come to Rob)

> So if you are able to state, that you have always upheld truth and morale
> against your friends and in favor of people who hated you. Then you are
> free to cast the first stone. Otherwise, I think you should not judge Rob
> Enderle.

Judge him? No.

Exact redress for the measurable harm Mr. Enderle has done each and every person invested in anything linux-related? Sure.

Ensure that evaluations of Mr. Enderle's accuracy as an industry analyst accurately reflect his years of gross misjudgement? You bet.

Ridicule him and call him silly names? It's called socialization. :)

These people were conspiring to steal the work of other people. Plain and
simple. Remove all the legal verbiage, Novell drove them into bankruptcy
because they were spending Novell's money on lawyers to cause them harm.
They were conspiring to steal the copyrights of kernel developers for
their own gain. They found a willing stooge with a reputation and ego to
match.

Enderle deserves the reputation he gained. He was a fool, a pompous fool,
and has taken a very long and painful fall.

Derek

As a person? Yes. Though the lack of regret seriously tempers any sympathy.

As a journalist who is paid for (theoretical) objectivity? No. Not to excuse the threats he may have received, but if he reacts to flames with bias and taunts, he's in the wrong industry. Those that trolled him provided as much unbiased and factually based information as he was (and by this essay, is) providing others.

That he uses this non-mea-culpa to lash out at others that don't deserve it (the "analyst" that "didn't have the writing skills" and apparently "did all she could" to avoid meeting with SCO, NDA not mentioned) without taking personal responsibility for not performing fact checking only cements the impression.

In a better world, such an unrepentant lack of objectivity would be a serious impediment to the career of a journalist or analyst.

Better worlds don't happen on their own.

For all we know, it could have been Darl McBride himself. Or it could have been Steve Ballmer that convinced McBride to do the trick. Or (in a strange twist) maybe Stallman tricked Ballmer into convincing McBride to pose as a Free software fanatic, knowing that in the end Linux (the kernel) and Free software in general would come out fortified. It is hard to be sure of anything these days, with the internet and all.