I understand Larry's desire to set the record straight, as he feels
Eric has claimed undeserved credit on a public list.
However, I do request that this be the end of the topic, at least on
the Subversion development list. As far as finishing Subversion goes,
it is useless noise, and it's important for everyone that this list
remain Subversion-specific.
Now that the entire list is aware of the disagreement, neither side
can claim that lies are being allowed to go unchallenged. Actually
sorting out the truth in this matter should not happen here (and I
daresay most of us don't feel a pressing need to get to the bottom of
this dispute anyway).
Thank you in advance for your restraint,
-Karl
Larry McVoy <lm@bitmover.com> writes:
> > I've signed up because I want to help head BitKeeper off at the pass.
>
> We're flattered you think it is worth your time. With such a luminary
> against us, how can we possibly succeed? Maybe we ought to drop our
> "insidious" source license and make it a binary only product. I'm sure
> that slashdot would love to hear that ESR has stamped out another bad
> license and the only cost was that you can't get source anymore.
>
> > So my goal is to find out where you guys are, and if I judge that you
> > look like the best bet to funnel some people and resources your way.
> > I may contribute a few ideas (I gave McVoy the idea for the changeset
> > abstraction in BitKeeper back in 1998) but probably not heavy coding;
>
> Eric, let's set the record straight on this one. First of all, you
> can't possibly be claiming you invented the changeset abstraction, right?
> See below before you consider doing so. Second of all, as to you giving
> me ideas, have you ever considered that perhaps the policy of giving
> others credit even when they don't deserve it is a better thing than
> taking credit when you don't deserve it? That's what was happening,
> I was being nice and you turn it into you invented the idea. Sheesh.
>
> Perhaps you are unaware of this, but 7 years before we talked, I had
> *finished* writing my first source management system, something which
> Sun still ships today, generates 10s of millions of dollars a year, has
> been used for Solaris development for 10+ years, etc. It may come as
> a shock to you, but I'm well versed in the area, in fact, better versed
> in theory and in practice than you are. I'll happily stack my code
> against yours, my knowledge against yours, I love the idea of a panel
> question and answer session with you and me on the panel. I think it
> would be a hoot and quite the crowd pleaser. Let's do it.
>
> I'm well aware of the conversations we've had on the topic, and I can
> assure you that you added little, if anything at all, to the effort.
> I let you believe that you did more than you did simply to stroke your
> ego in hopes of you fixing up that mess you call VC mode so that it would
> work with BK. That's all I wanted and it still isn't done, and we're
> writing code to work around how badly VC mode is written. I was wrong
> on two counts: stroking your ego and believing you'd ever fix that code.
> I guess I could have predicted the latter, but it was stupid of me to
> do the former. Who would have thought you'd take a few nice comments
> and try and turn them into you doing things you didn't do? Still,
> I should have known better. I won't do it again, at least not with you.
>
> For those who wish to live a little closer to reality , it may be
> interesting to note that the changeset abstraction Eric claims he
> invented or passed on to me, has been around for at least 12 years,
> Aide de Camp is a cset based system which first shipped around '88.
> There were conferences 5 years ago where all papers were comparing
> "change sets" vs. "change packages", perhaps Eric should contact those
> people and let them know that he invented the abstraction, they will
> surely print retractions. I've read some of those papers long before
> Eric and I ever met, BTW. Here's a good one:
>
> http://www.continuus.com/developers/developersACEG.html
>
> And the basic concepts on which a changeset is built, what is commonly
> called a changeset engine, are present in SCCS, in CDC Update, etc.
> These go back around 30 years. One of the guys who works here patented
> some stuff based on changeset ideas in '96. I was writing Sun's Teamware
> system in '91, are you going to claim credit for that as well? Come on,
> Eric, we all know you are smart guy and all, there is no need to claim
> credit for things you didn't do. heck, I can't open a traderag without
> hearing about the "substantial contributions" you have made to the Open
> Source world. Isn't that enough for you?
>
> As much as you seem like you would like to help us with your great
> ideas, maybe you could consider sticking with the areas where you excel,
> such as fetchmail, rather than trying to help us. We're doing fine.
>
> We'd be happy to see you working with Subversion, you are a sharp
> guy and I'm sure the Subversion people will welcome your assistance.
> --
> ---
> Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
Received on Sat Oct 21 14:36:23 2006