Chris Brown, Crocodile Dundee, our Aussie visiting SLC, and cxd all attended the hearing and their reports are now coming in. Chris says first they talked about scheduling. Everyone seemed to agree that there needs to be a postponement of the trial date, to make room for hearings on all the summary judgment motions. Kimball asked whether the SCO v Novell case should go first. Of course SCO's attorney Stuart Singer said SCO v IBM should go first, and IBM's David Marriott said Novell should, because of two issues: whether SCO even has any copyrights and second Novell's waiver of any alleged breach. The judge took that under advisement. They also discussed how long each side would be able to take for argument. On SCO's objections to the Wells Order, that too was taken under advisement. More details to follow.
Crocodile Dundee tells us this:
The hearing ran from 10 to 12 and the main points SCO raised are:
1) the things Judge Wells dismissed are (in effect) claims, thus the motion of IBM's was dispositive and requires de novo review;
2) Judge Wells' decision effectively excluded the information SCO *had* presented, rather than excluding information it has allegedly withheld. SCO contends that IBM could simply object to that at trial.
There were also some scheduling discussions at the end where SCO noted that they wanted to get to trial as fast as possible. When IBM and SCO were asked for their opinion as to whether Novell should go first, SCO said No, and IBM said that there were some significant reasons why they should.
SCO asked for a 1 week delay (and then another). Judge Kimball asked if he should schedule for the 23/24 December? (He was just kidding).
SCO reckons that the trial could still go forward on Feb. 7, but IBM says realistically the court won't hear the 10 outstanding motions (this was the 2 days 23/24 Dec -- probably will be the week earlier) and then rule on this and then allow SCO/IBM to get ready for trial. The Judge agreed with IBM.
It look like the trial will be delayed, and if this occurs, there may be no openings left for a 5 week trial.
My feeling is that SCO desperately wants this to go ahead before the Novell hearing, even though they say the Novell hearing has almost nothing to do with this. SCO vs. Novell is all about copyrights, they say, and SCO vs. IBM is all about contracts you see.
Chris adds this information: that regarding the summary judgment motions, what was decided was that SCO and IBM will present a
proposed schedule to allow hearings to be completed before Christmas. And did SCO bring up spoiliation again? Yes, of course.
Chris also says that his personal impression is that Kimball is inclined to be more lenient than Judge Wells regarding the evidence that she tossed out. He thinks he's inclined to go for a de novo review, but of course, this is just his personal impression. IBM called the 180+ items evidence that wasn't properly shown by the deadline; SCO called them claims. Singer argued that if you throw out all the evidence of a claim, you have in effect thrown out the claim. And Chris felt that Kimball seemed to be influenced by that argument. However, he says it's very hard to know what Kimball is thinking. He doesn't broadcast it.
SCO raised another issue: their position is that Wells was required to present particularized findings, explaining why each item was excluded. Also that she was required to hold an evidentiary hearing. Marriott said she did review them individually, and she found that SCO hadn't presented any lines of code on any of them, no lines of AIX code on any of those 180+ and only 3 that referred to Linux, without saying what version or line. They just referred to files. Because they didn't provide details, she wasn't required to particularize. What could she particularize about?
cxd tells us this:
Some of the best stuff happened at the end of the hearing.
Kimball asked both if the Novell litigation should go first.
Singer said no way -- these issues are more related to IBM and IBM should
go to trial first.
Marriott said that letting Novell go first would eliminate much of the
case. He thought that would be a great idea. So Kimball is thinking
about how the cases relate. That is good, I would think.
Kimball also asked about the time required for all of the summary judgment hearings and how
that would affect the trial date. Mr. Singer said that SCO wanted as
little delay as possible because they want to have their day in court
as soon as possible. They actually said it: "We want the trial as
soon as possible."
Kimball then proposed they hear the motions on the 23rd and 24th of
December, takes a few moments to let that sink in to both parties, and
then says he was just joking. I wonder if that was in relation to the
Christmas holidays past?
Mr. Marriott did not see how the current schedule could hold, given the
amount of work before the court. He said that each motion would
require a hearing needing about 45 minutes of initial argument, let alone
rebuttal. This all took place before Judge Kimball asked about the
Novell trial going first.
Much of this is going to depend on if Kimball agrees that Wells was
not throwing out causes of action, that all of the causes of action
are unrelated to the individual unsupported with line and file
information items that IBM are wanting eliminated from the case.
Update2: And now, after reviewing his notes, cxd adds more:
SCO was very upset that there was not an evidentiary hearing, that
each item should have been heard on an up-or-down basis.
They wanted an up or down on each, based on the merits of each item.
They said that often the evidence is included in an email or web site and that
only IBM can know what its developers knew. "It is in the mind of the
engineers -- how can SCO be expected to know what is in their minds?"
SCO did say that Judge Wells did not follow 10th Circuit law when
throwing out the 194 items. And she denied them an evidence
hearing. This came up over and over again.
Mr. Marriott later made it clear that SCO could have deposed these
people if it wanted to.
SCO complained that Judge Wells failed to consider lesser sanctions
and said this was enough to overthrow her ruling.
SCO also brought up willful noncompliance again. They said that IBM
has no evidence that SCO had evidence and that they did not produce
it. We are going around and around again.
Mr. Marriott said the ruling of Judge Wells was not dispositive, and it
fit a clear standard, and that even the cases cited by SCO
support IBM's view. Marriott later explains how all of the claims
remain in the case, just the items of evidence that were not backed up
were removed. I love it when they use case law against SCO when it is
More details to follow. It sounds like our guys had fun.
And now Chris is home and here's the more detailed version:
Since the arguments on SCO's Objections to Magistrate Judge Wells'
Order of June 28, 2006 mostly followed the line of reasoning in their
respective motions, I will first tell you about the other issues raised
by Judge Kimball for discussion.
After the hearing on the scheduled matter, Judge Kimball said that the
current trial date of Feb 26th, 2007 was set mid-2005 before the present
Summary Judgement motions were raised. He pointed out that these
Summary Judgement motions are scheduled to be fully briefed by December
8th, then hearings held, then Judge Kimball has to give them due
consideration before ruling. He pointed out that when the schedule was
written, there would have been 5 months from the summary judgment motions
before trial, and now there are two and a half. He was concerned that
there would not be sufficient time to prepare for trial. He asked both
David Mariott and Stuart Singer to comment. Mr. Singer said that SCO
very much wants to get to trial as soon as possible, Mr. Marriott said
that IBM does as well, but being realistic, that he believes the trial
may need to be delayed.
Judge Kimball asked about the Novell case and what their thoughts were
on whether to have the Novell case proceed first or IBM. Mr. Singer
said that the IBM case should proceed first. Mr. Marriot said that the
Novell case should proceed first for two reasons: first, that the issue
of whether SCO received the copyrights to UNIX bear on the copyright
issues in the IBM case; and second, whether Novell retained the right to
act on SCO's behalf would bear on the contract issues in the IBM case.
Judge Kimball asked how much time each party would need at a hearing
to argue the summary judgment motions. Mr. Marriott opined an hour and
a half between the two of them for each motion. Mr. Singer estimated an
hour. (Note: Having watched these two I don't know how they could have
said that with a straight face). Judge Kimball asked if they're
indicating that two full days could cover it, to which Mr. Singer opined
that one day would do. Judge Kimball said that with 10 motions and an
hour and a half apiece, it amounted to 15 hours and he believes people
likely couldn't handle that, what with needing to go to the restroom and
They agreed that Mr. Marriott and Mr. Singer, since they seem to still
be on speaking terms, will work together to come up with a new proposed
schedule for the hearings on the summary judgment motions and trial date.
After this was agreed to, Mr. Singer said that the current schedule
for SCO to respond to the summary judgment motions was agreed to in
light of the schedule for a February trial date, and that if that's to be
delayed, SCO would like an additional week to respond. Mr. Marriott said
that if it was only an additional week, IBM could agree to that.
As you may have surmised from my earlier email, David Marriott spoke
for IBM and Stuart Singer for for SCO. Also present for IBM was Todd
Shaughnessy, Michael Burke, and Wing Chiu. SCO also had Sashi
Boruchow, Mark James with Ryan Tibbetts in the audience. [Please excuse any misspellings. Feel free to contact us for corrections.]
Mr. Singer handed out a book for items for reference while he spoke.
He said that the order of Judge Wells infers that SCO was willfully
withholding information from the court. He said that is not the case
and that the present opposition relates to the significant number of
tech claims dismissed without their day in court. He said that SCO has
been as complete and forthcoming as possible. While IBM insists on a
nine-point coordinate to identify the claims, the source code
coordinates are simply not known to SCO. He argued that Judge Wells did
not consider other 10th circuit alternatives to dismissal and it was not
based on an evidentiary hearing. Mr. Singer said they took the
obligations of this court very seriously, that they had experts working
thoughout 2005 on this extensive list of 294 items. He said that Marc
Rochkind was the principle expert working on that list.
He said SCO is not holding anything back and that Mark Rochkind
testified so. He said for instance that only IBM's technologist knows
what lines of Dynix code were in his mind when he wrote Up to the
Linux community saying, "Here's how we did it in Dynix," that there is no
evidence that IBM asked their technologist what code he was relying on.
Only IBM knows this. He refered to a chart in the book he handed out
saying that for each of the 198 items, SCO provided what they know, including
Linux files. He pointed to Item 146, Differential Profiling. In that
item are the scripts from Dynix and where they came from. That those
appear in Linux, and they'd identified the location on IBM's server where
the code exists but that SCO could not access it because it was behind
He pointed out Item 53, Semaphores, where IBM's Tim Wright disclosed
that these are not currently used in Linux and they are from Dynix. SCO
identified the Linux files it appears in. SCO said that this claim was
struck (by Judge Wells). He pointed out Item 38, Checking Updates in
Memory, same thing.
He said that 10th Circuit law requires particularized findings for
sanctions of this type and that they don't have that here, that all
there is is a Magistrate's statement that said all 198 were defective. He
said that SCO expressly requested an evidentiary hearing on this. He
said that it's required with disputes between respective experts over
Mr. Singer said that if there's a concern of sandbagging, the
appropriate remedy is to prevent the party from supporting it with
evidence and should be addressed at the time of trial.
At this point Mr. Singer brought up the details of their allegation
of IBM spoiliation of evidence. I will not repeat it here as I believe
it is under seal. Mr. Singer said that IBM's [alleged] spoiliation of
evidence pertains to why SCO doesn't have more details on their claims.
He said that Judge Wells' order is dispositive and therefore a de novo
review is required, that IBM has backtracked on this but that IBM filed a
motion to limit SCO's claims. He said that each and every one of the
188 claims were a breach of contract, that issues of law, of legal
principles, require de novo review, and that there should also be an
Mr. Singer stated that SCO has not violated any court orders, that
IBM provided proposed language of version, file, line of code but that
the court's orders do not include this language. He said that IBM's
principal interrogatory of version, file, line of source code uses the
words "where appropriate."
He said that there is no evidence of willfull non-disclosure. That
the court must have intentional non-compliance showing that SCO has the
evidence within their ability to provide, that they withheld.
Judge Kimball asked Mr. Singer about SCO's public statements [of
having more evidence]. He asked Mr. Singer "How would you interpret
those statements?" Mr. Singer replied that those statements were made
several years ago and have not been made since, as it's a matter for the
courts now, that most of those statements related to methods and concepts
that were identified in 2005, that none of those statements say SCO has
the source code for these claims and were not going to provide it.
Mr. Singer brought up Judge Wells' analogy to a Nieman Marcus
shoplifter in her order. He said that a better analogy would be the
shoplifter was told what it was he stole but was let go because he was
not told the page in the catalog which it appears.
Next: Mr. Marriott replies.
Mr. Marriott responded. He said that SCO accuses Magistrate Judge
Wells of no fewer than 20 errors. Mr. Marriott said that Judge Wells
relied on rules 37(b)2 and 37(b)3, which state that such evidence is
"automatically excluded" unless SCO could provide convincing reasons
otherwise. IBM's Mr. Wing Chiu held up an approximately 3 ft by 5 ft chart
(he held it up on the table for the entire 45 minutes of Mr. Marriott's
talk! Arms of steel!) to which Mr. Marriott referred often. I could
not see the contents of the chart. Mr. Marriott said the undisputed
facts are that Unix, AIX, Dynix, and Linux together constitute hundreds
of versions, millions of files, and billions of lines of code. I forgot
the second undisupted fact he mentioned. The third is that SCO did not
provide a single line of System V code in any of the disputed items, nor
did they provide a single line of AIX or Dynix code.
He said that Judge Wells' order is not dispositive, that her order
does not strike, remove, or dismiss any of SCO's claims. He said that
it is not the relief requested (by IBM). He said that if SCO believed that
IBM's motion was dispositive, or that Judge Wells' could not properly
hear it, they should have said something in court, but they didn't.
Mr. Marriott quoted page 2 of Judge Wells' order, where she said,
"...what I don't want is either side to use information that has been
withheld... all evidence needs to be on the table..."
He made reference to three cases SCO cited, Ocelot, Smith, and Lister
and he tells what happened in each case and says that none of those
cases is dispositive or supportive of SCO's position.
He said that the rules are that it is automatic. "Violate order,
willfull, prejudice to IBM," that if she says that, it's automatic. Mr.
Marriott reviewed each order of Magistrate Judge Wells since 2003
quoting terms "with specificity", "respond to IBM's interrogatories",
etc. He said that SCO was required to identify with specificity with
file, line, and version of code. He read from her order of 3/3/04
"fully comply within 45 days," "SCO is to provide and identify ...
specific lines... from Unix", other orders specifying "lines of code."
Mr. Mariott said that three orders of the court and Rule 26(e) required it.
Mr. Marriott pointed out that Magistrate Judge Wells issued these
orders in 2003 and 2004 and that SCO had 10 days to object but did not
do so, and objections are therefore waived. He said SCO cannot make an end-run
around these orders now.
He said that methods and concepts are implemented in source code, and
Randall Davis, in his declaration, said it cannot be otherwise. He
pointed out that SCO's own witness uses code examples in his books to
discuss methods and concepts and that SCO's employee, Mr. Gupta,
indicated he'd need the code for methods and concepts.
Mr. Marriott contended that SCO has withheld its allegations.
He pointed out SCO's Item 98 [92?] of its Final Disclosures, where SCO
claims IBM took code from Dynix and put it in Linux by "Michael" who
purportedly is a former employee (of Sequent or IBM). That's all it
gives, an email from someone named "Michael" on the Internet who
says he worked for (Sequent or IBM) and that IBM put code in Linux.
Contrast that claim with the size of the codebase. He says that it's
worse than trying to find a needle in a haystack; the needle is nebulous.
Next: Mr. Singer responds.
Mr. Singer responded to Mr. Marriott by first quoting, as Mr. Marriott
had, Judge Wells' quote shown on page 2 of her order: "... What I don't want
is either side to use information that has been withheld in support of a
summary judgment motion or in support of their case at trial, all
evidence needs to be on the table..." He said that this order takes the
evidence SCO has put on the table and takes it off.
He said that the sanctions for sandbagging is when the party brings up
the withheld evidence, an objection is made, and it is excluded.
[Referring to IBM's example] Mr. Singer says that this is the first time
Item 92 has been brought up.
He said that IBM claims that that de novo review is not required but SCO's position is that when
you've thrown out all the evidence of a claim, you've thrown out the claim.
He said it must be a de novo review, that the order requires it.
He said the 10th Circuit requires particularized findings, that an
evidentiary hearing is required. Pertaining to IBM's assertion of SCO
waiving their rights in 2003 and 2004, he said SCO is not objecting to
the orders, but to their interpretation [cb: I'm not sure if he meant
IBM's interpretation of the orders, SCO's interpretation, or the Court's
interpretation of it's own orders]. He said there is nothing on the
record that they have information that they have not provided. SCO has
Rochkind saying they've provided everything. SCO is not withholding
anything. He said SCO's claims should be heard on their merits.
Judge Kimball asks Mr. Marriott about SCO's request for an evidentiary
hearing. Mr. Marriott replied that SCO made no request until their
reply papers and therefore it's waived. He said IBM has answered this
more fully on page 55 [of IBM-748-1]. In this Mr. Marriott says that SCO's
"Mopix" case supports IBM, not SCO.
Mr. Marriott points out that SCO's claims are intact however many
times SCO refers to their evidence as "claims", all their claims are intact.
Referring to SCO's example where they'd identified an item that
referenced IBM code that can be found at a specific address but that SCO
could not get to it because it was behind IBM's "firewall" -- Mr. Marriott
said that what's "behind the firewall" IBM provided to SCO to the extent
they could identify what SCO's talking about. They provided it. He
quoted Bates ranges for the production.
Mr. Marriott reminded the court of the expense and effort they went
through, as a result of an order of the court, to provide CMVC access to
SCO. He said IBM provided all the code ordered so SCO could provide the
code in its claims but SCO didn't do so.
When Mr. Singer responded, he said that if the code is in CMVC, then
IBM has it. [There was scattered laughter from the spectators.]
Then Judge Kimball proceeded to discuss the other issues of schedules.
And that's it. I think
it'll wind up being a de novo review and that Judge Kimball will not
hold SCO to the fire over not providing version, file, and line
information on every one of the items. I suspect he'll provide
particularized findings for each of the disputed items. Of course, like
most of us, I'm not a lawyer.
There was one point in the hearing where Mr. Singer brought up IBM's
reference that the 7th Circuit and saying that the Magistrate Judge's ruling must
smell like a week old, unrefrigerated, dead fish before overturning it.
Mr. Singer had said that that's the *7th* Circuit, but here in the 10th
Circuit the court must follow the law (particularized findings,
evidentiary hearing, etc). Refering to the "Dead Fish Rule", Judge
Kimball said something to the effect: "But you've got agree, it's a
good rule." I got the impression, as I have at the other hearing I
attended opposing an order of Judge Wells, that he's not keen to
overturn what she's already ordered. We'll see. I expect he'll rule
quickly on this one with the schedule as it now is.
And now Crocodile Dundee has arrived "home" and he shares his more detailed notes with us, as is, no warranty implied or otherwise, so to speak:
All quotes indicate what I believe was said; in most cases it will be abbreviated, and in some cases slightly restated. Quotes should be read as my observation without personal editorial input. That which is not shown as quotes will probably reflect my personal take on what was said. (if that makes any sense). I've had to insert some clarification inside some quoted comments. [I do that like this.]
09:50:00 SCO's attorneys appear. There's quite a few people in SCO's side of the gallery, just the 2 of us on IBM's side. I'm told that some of these attorneys look new.
09:54:30 IBM's arrival must be imminent. I'm told they always turn up *exactly* 5 minutes before a hearing.
09:55:30 Sure to form, IBM turns up in force 5 minutes before the hearing. I am pointed to a dapper gentleman who is identified as David Marriott.
10:03:00 Judge Kimball rocks on in and we're off and racing!
Marriott and Singer announce that they'll be arguing. Singer says he will require quite a bit of time (Judge Kimball expresses surprise), saying there's a lot of things to cover. He gets 40 minutes then 15 minutes. Marriott says he will need 40 minutes.
Singer starts off by telling the court that SCO have been "looking forward to this argument for some time".
He says a "significant number of tecnical claims were dismissed without opportunity for their day in court". He continues by saying that this is wrong.
They see the ruling as being "predicated on SCO flaunting court orders."
"This could not be further from the truth."
"We have sought to be as forthcoming ... as possible."
"Our objections are based on the following points
[These went pretty fast and my notes look like chicken scratchings]:
"* not p..."
"* did not consider alternative sanctions" (I think)
"* not for evidentiary issues"
"* did not " (something -- I was getting behind here)
"* did not follow from each other"
"* did not rest on factual basis of flaunting orders"
Singer then went on to describe how the (information given to IBM) was prepared.
They "took things *very* seriously"
"We got substantial discovery"
something about the work done by experts
in "2005 we prepared a comprehensive list of 294 items" ... and ... "thousands of pages of exhibits" including email.
"Mark Rochkind's declaration states that"... in almost all cases" there is email (presumably disclosing something -- I didn't catch exactly what he was talking about).
"for each of the 294 items [we] did everything we could to disclose all matters"
"Version, line and file were provided when available"
"We have provided everything we have -- we're not holding back"
"IBM's 9 coordinate model is useful for argument" (although they don't agree with it -- 9 coords are V, L & F for Unix, Dynix/AIX and Linux).
"where there has been no source code disclosure the IBM Person" (in emails from IBM staff to Linux developers) "doesn't give code in the email, therefore no code to offer"
At this point SCO basically claimed that anything in Dynix or AIX was theirs
SCO contends that if an IBM person had some Dynix/AIX code in mind when writing an email to a Linux developer detailing some Method/Concept, it was not possible for SCO to tell IBM what exactly that was.
At this point I wondered what depositions were for if not for this, but SCO continues...
"IBM hasn't asked its own people!"
"so, how can SCO know?"
SCO can "say where this information impacted Linux, but they're not certain of which lines" (!!)
SCO claim to have produced:
"A Code provided"
"B Some positions"
"C express admission from IBM of disclosure from DYNIX/AIX"
"D file locations in Linux"
Point C was footnoted by the comment "presuming our contract theory holds"
something about "146 differential profile methods and concepts" moved(?) "Dynix --> linux"
"IBM has access to Linux (I presume from the context) that SCO doesn't"
I got a whispered comment from beside me that SCO was looking for more discovery.
"hundreds of lines of code"
"experience(?) from Sequent"
"specificity and admissions, but struck out"
(following numbers are references to something)
"not originally [?] used in linux"
"38 Updates in memory"
"46 bug fixes"
"and about a dozen more"
SCO contend that Judge Wells erred by her "failure to make particularized findings" before "throwing things out"
second major point: "Failure to hold an evidentiary hearing"
"specifically requirested by SCO"
"IBM presents a 10 page motion"
"IBM's reply has a declaration by an expert [Davis]"
"SCO gets leave to submit Rochkind declaration"
"Since there was an evidentiary conflict between the two experts, therefore an evidentiary hearing should have taken place".
"The evidentiary hearing was required for:
"i Davis vs. Rochkind conflicting expert declarations"
"ii failure to disclose (sandbagging claim)"
"this should stop further evidence, *not* throw out the claims"
[Note here that SCO is categorising each of the items thrown out as claims]
"iii deciding if prejudicial to IBM on a point by point basis"
iv matters arising from sealed evidence that I shall not report
third point started somewhere here I think...
"If our case is without merit, why do IBM have 11 expert witnesses?" I'll admit that this sounds a bit funky to me. Singer may have said something with nuances I didn't catch.
"This ruling is dispositive"
(Judge Kimball) "Subject to de novo review?"
(Singer) "Yes, de novo review is required"
"188 claims of disclosure"
"we say breach of contract"
(I think it was here, Singer read the first part of the title of IBM's motion and stated that the name indicated it was clearly dispositive. This assertion was repeated several times.)
"Even less serious cases have had de novo review"
"even for just 1 claim"
"ref NY patent case" (sorry didn't catch the magic numbers)
"ref IBM's 'dead fish' standard"
(Judge Kimball) "An interesting standard"
(Singer) "Yes, but not the law."
"Even mixed questions of fact and law require de novo review (although limited)"
"IBM has disclosed misused material with specificity"
"Court did not include [perhaps consider??] this in the order"
"We're being sanctioned as if this were in the order"
"No express order on the basis of '9 coordinates of code'"
"No argument about the level of disclosure for methods and concepts"
"Nothing says you need to produce X, Y and Z or your claims will be dismissed"
"IBM asks for file, line & version 'where appropriate'"
"SCO is alledged to have asked for the same"
"This 'request' came in an expert statement containing our discovery request" (that seems garbled)
"The Magistrate Judge assumed that IBM provided this and that SCO would object. They didn't and we didn't"
"No evidence of willful non-compliance"
"must be willful and intentional"
"IBM must show we have evidence that we didn't provide"
"The Magistrate Judge [something about stating or drawing conclusions] no evidence of providing everything [and comments about] public statements"
(Judge Kimball) "What would be the best way to describe those public statements?" (referring to statements in 2003)
(Singer) "That we have a lot of source and tech items -- just not on these specific items." They "stand by them [their public statements]". Comments differentiating the source code issues (public statements) and the methods and concepts issues (an issue today). "The methods and concepts were not even identified until 2005". "Those statements do not support the dismissal of 188 items".
Singer then addressed the shoplifter analogy. He said it was more a case of the shoplifter being told exactly what he's stolen, but the police [SCO - ha!] not being able to identify the exact catalogue number of the item.
(I think by way of suggesting that re-including these would not delay things)
"IBM should be able to quickly respond as they would have (at least partially) prepared for this in the event that the order had not been granted."
At this point he seemed to be arguing that SCO didn't want delay, but that IBM might be granted more time. But he wanted "No delay, or at most, a very short delay".
10:40 Marriot responds
He first states that "SCO has accused the Magistrate Judge of 27 errors"
"3 simple points"
"Magistrate Judge formed opinion based on 37b2 and 37b3" (although he later seems to refer to 37c a lot)
"37c: ...fails to disclose... is automatically precluded unless [mitigating factors]'
"1) SCO failed to provide file, line and version"
"2) SCO failed to show justification for failure"
"3) SCO failed to show that failure was harmless to SCO"
"The crux is that IBM is said to have misused 3 different sets of OS's, Unix SysV, Dynix, Linux"
"because of the size of these [mention of billions of lines] IBM's discovery requested specificity via version, line and file"
"In response, SCO's final disclosure failed in 187 out of 2?? cases to provide the most basic identification"
"not one of the 187 had:
"* line info for SRV"
"* line info for AIX/Dynix"
"3 out of 187 had line information for Linux"
"All failed to provide version information for Linux, and there are 500 plus versions of linux"
"This is of great consequence to this motion"
Somewhere around here (perhaps earlier) Marriott had read the full title of the motion they put to Judge Wells, asserting that it was clear it was not to remove actions, but the evidence (or non evidence) supporting them
"The Magistrate Judge's decision was non-dispositive"
"a firm and definite error is needed"
"...dead fish standard..."
"*Even if* a de novo [review is needed???] the Magistrate Judge's [decision] is proper under *any* standard of review"
"A motion is dispositive if it disposes of a cause of action"
"Magistrate Judge's decision did not dismiss, remove, or strike any of SCO's 9 clauses or 14 defenses"
"SCO is wrong in their argument because this was not done under rule 12 but under rule 37" (very loose quoting)
"If it had been dispositive, wouldn't SCO have argued it was dispositive at the hearing?" (Since Judge Wells could not address dispositive motions).
The assertion that this "affects claims is an interesting effort at relabelling. 294 items which are collections of evidence are now called actions."
Marriott quoting Magistrate Judge: "All evidence needs to be on the table"
"they are collections of evidence *NOT* causes of action"
"If they were causes of action, they would needed to have been the subject of an amended claim, but SCO's request to do so was earlier denied."
"The three cases quoted by SCO actually went the other way to that which SCO claims."
"The Magistrate Judge need not have mentioned the sanctions because they were automatic"
Marriott then goes on to show why IBM believe Version, Line, File was required. (Actually I think this was a little weak. It relies on assumptions. Those assumptions would be made by anyone with two (programing) brain cells to rub together, but maybe not by lawyers)
The orders require identification with specificity. They require SCO to respond fully and in detail as stated in IBM's interrogatories. The interrogatories request "detailed and complete".
"It can't be detailed and complete without line, version and file"
"They were specifically required to produce line and file"
"ref interrogatory #12 (and it's the same for all)."
referring to a large diagram, Marriot begins to explain the 9 coordinates.
"File and line are specifically required. For Linux verion was implied and not argued against by SCO. Derived from SRV, file and line required, also from Dynix *or* AIX"
He then argues that version is implied since without it the other information is meaningless (example of a street address that doesn't specify which street it's on, only the house number and city)
"SCO arguments with the 03 and 04 [2003 and 2004] orders should have been raised within 10 days. They're trying an end run."
"SCO argue that these requirements 'simply do not apply to methods and concepts'"
"* however some of those [187 items] are specifically about source"
"* whether code, method, or concept, specificity is still required"
"SCO failed to justify its failure to provide File, Line & version"
"SCO faced the burden, not IBM"
"SCO says ambiguous[?] -- no, and waived"
"SCO says file, line, version cannot apply to methods and concepts but many are about code"
"17 port of [???] code"
"26 port of NUMA code"
"So, SCO says code not methods and concepts. They could still provide file, line and version"
(Davis on methods and concepts quoted by Marriott) "exist as they are embodied in source code" and "in the source code, it couldn't be otherwise"
Also quotes Gupta and Rochkind who seem to be in some agreement.
(quoting SCO) "SCO must read IBM's mind"
(Marriott's comment) "A frolic and descent into [lost it :-(]"
"What has IBM done?"
"could not provide it prior to IBM telling it. Magistrate Judge threw that out, SCO didn't appeal"
"SCO failed to show it's failure was harmless"
(Marriott paraphrasing SCO) "IBM can figure it out"
Notes that one email "the Michael admission" was from "an anonymous Internet posting from a person claiming to be an ex-employee [of Sequent/IBM -- can't remember] stating [IBM doing bad things]"
11:00 Singer's reply (which will have to wait. need food...)
Update 4: And now our fourth witness, Justin Findlay, sent me his notes too:
I came in about 5 minutes after 10 AM. Singer was already arguing for
SCO at the time. Both Singer and Marriott argued with some measure of
passion. Judge Kimball was calm and refreshingly colloquial. Where
possible I have tried to remain true to the words and phrasing of the
speakers so as to give a more accurate impression of the dialog only
supplying my own words where my memory is not sufficient to
reconstruct. Still, I'm afraid that the attempted reconstruction of
these furious and arcane shorthand digraphs has made me mad. (-:
Singer: There has been lots of jockeying in this case.
Kimball: That occasionally happens.
Singer: There is substantial discovery material dating from early
2005. SCO's counsel has been working diligently with their experts
and have produced 294 items--thousands of pages of supplementary
SCO's expert, Mr. Rochind, helped to produce very specific
documentation for each item of infringement including public email
exchanges, URLs, dates, versions were cited where available.
Tab 3 refers to the 9 coordinates of the source code: versions, files,
etc. Chapter and verse -> version, file, and line.
SCO couldn't provide more specific citations for 198 exhibits from
Unix System 5, Dynix, and AIX because they didn't have access to all
the resources within IBM they would have needed.
It is unreasonable to expect SCO to tell what lines of System
5/Dynix/AIX source code an IBM engineer had in mind when they wrote
code which was contributed to Linux. IBM needs to consult its
developers to collect this evidence.
I will comment on a handful of examples which disclose full
specificity of infringement claims.
Tab 6: The methods and concepts claims are different from literal
source code claims in that it is harder to provide specific evidence
of code copied to Linux. Mr. Rockind could not get access to the IBM
intranet to pursue evidence or discover origins of a source.
IBM has access to this information that SCO needs to more specifically
establish these claims, not SCO. SCO can't read the minds of IBM's
Tab 7: [file] locking. SCO has source code patches with coordinates.
Mr. Linsley admits he wrote this code at Sequent, yet the claim was
stricken from the case.
Tab 8: semaphores, Tim Wright. This code is not currently used in
Linux. The source code was revealed to originate from Dynix. SCO has
identified the specific files that are derived from this code
contribution, yet, as item 53, it also was struck from the case.
Tab 9: memory
Tab 10: bug fixes
SCO has over a dozen more examples of claims dismissed in its
appendix. According to 10th circuit case law, these claims were
wrongfully thrown out.
The claims have been dismissed without sufficient consideration.
Tab 12: Failure to hold an evidentiary hearing. The initial 10-page
motion talks about the July 2005 order. SCO disputes that they have
been sandbagging IBM. Even were SCO sandbagging, the proper
resolution is to stop the claims at trial rather than throw out the
claims. The claim should be considered on an item by item basis.
[Some discussion of spoliation.} The Linux Technology is at the heart of the case. This
affects our ability to conduct discovery.
Tab 13: Aaronhouse [sp?] case from the 10th circuit. A lesser sanction must
be considered if possible before throwing out claims. The exclusion
of these claims from the case is improper without sufficient evidence.
I would like to discuss the scope of review. IBM claims these claims
are without merit. These issues are highly technical and must be
considered with expert discovery.
This hearing is dispositional in order to limit SCO's claims.
Kimball: Subject of review.
Tab 15: 10th circuit case law. The dismissal of the claims is invalid.
188 claims from technical discovery will never have their day in
court--never see a jury.
Tab 16: No de novo review.
A New York patent case whose one claim thrown out of many. The court
found that dismissed claims of potential liability must be reviewed.
A deferred review is better than no review. Stinking fish standard.
Kimball: interesting standard.
SCO's case should merit de novo review.
Tab 21: mixed questions. The law requires de novo review. SCO hasn't
violated the court's orders.
Tab 23: There is no explicit order to produce the 9 coordinates of
source code. It is not really possible to provide so much specificity
for methods and concepts.
Tab 25: IBM asks for version, file, line but qualifies with "where available".
Tab 29 Our discovery has been limited by IBM. IBM's Interrogatory
#1. There is no clear order for SCO to provide the 9 coordinates of
source code. Westinghouse case.
Tab 33: June 28 order. SCO had made numerous public statements about
the evidence it had found. The Rockind statements contains all the
evidence we have.
SCO's public statements cannot be read as [I'm really racking my mind as to his exact statement, but
Kimball: What is the best way to read the statements?
SCO has lots of evidence to offer. We reserve them for future
subjects of the case. There has been much public inquiry over SCO's
statements. These statements are true and may [not?] be related to
just source code.
The claims mentioned here have only been discovered as late as 2005
after SCO has had time to review the discovery provided by IBM, not
the subject matter from the 2003 public declarations. SCO has never
said "we have items of infringing source code but we won't tell".
IBM has shown prejudice in the case. SCO's provisions of items is
not, as IBM suggests, like a shoplifter being accused of shoplifting
outside the storefront without being told what he has shoplifted but
rather as if the shoplifter wasn't told the exact inventory number of
what he has stolen in the store's catalog.
The ruling dismissing SCO's 188 claims has been effected without due
deliberation. SCO expects to be able to address these items. SCO is
willing to be limited to the Rockind declaration for its claims.
IBM should be able to respond to these claims in the space of three weeks.
SCO should get a pretrial conference or evidentary hearing. This
would cause no or little delay.
Thank you, your honor.
Kimball: Thank you.
Marriott: SCO has been liberal in their accusations of the court.
They have accused Judge Wells of 27 errors.
Rules 37(b)2, (b)3
If the party fails to follow the guidelines the evidence is
automatically precluded from court unless there is substantial
justification or harmlessness.
1. SCO failed to provide version, file, and line as required by rule
26(b) and as ordered by Judge Wells
2. SCO provided no substantial demonstration of justification
3. SCO failed to show harmlessness
The crux of SCO's case involves IBM's alleged misuse of three sets of
operating systems. [Here Marriott with assistance from another lawyer
of IBM's counsel sets up a prodigious chart which Marriott, and later
Singer, refers to frequently.] The three operating systems in
question are Unix System V, Dynix, and Linux. There are hundreds of
versions, millions of files, and billions of lines of code. The size
of the universe of the evidence is indisputably enormous.
Tab 5. IBM asked in specific terms for version, file, and line
coordinates for *each* item. In SCO's final discovery they failed to
provide version, file, and line information on 187 of the 297 items.
Tab 6 illustrates SCO's failings. Not one of the 187 items provided
line information. For AIX/Dynix there is not a single line indicated
from the 187 items. For Linux only 3 out of 187 items provides line
information--none provide version information. There are over 500
version of Linux.
As regarding the standard of review, SCO argues that Judge Wells'
order is dispositive. It is clearly not. The test from the 7th
circuit is the evidence must have the stench of 5 week old dead fish.
SCO wants their evidence interpreted as claims. SCO has 9 causes of
action and 14 affirmative defences. Rules 27 and 36.
SCO accuses on page 13 that the order is plainly intended for dispositive
purposes. Rules 26, 30, and 37. Judge Wells' order is not over a
SCO interprets the 294 "items" of evidence as "claims". All evidence
was to be on the table according to the court's prior order. The time
for dispute over discovery has passed. SCO has failed to provide
version, file, and line information.
Tab 16: foldout exhibit, rule 26(c). The order required specificity.
IBM didn't make up the requirements. The court required expressly and
unequivocally that each item should have precise citations of source
code. You can't provide the file and line without providing the
version as well. That is like saying your address is at 26 without
telling the street name. SCO failed to provide version, file, and
line information for their method and concepts items.
[Marriott mentions a "second" order from March 3, 2004 and later
another July 1, 2005 order, then a 12-12-03 order.] Judge Wells
issued 03 and 04 orders asking for SCO to produce its evidence.
SCO's complaints should have been raised within 10 days of the issuance
of the orders.
SCO failed to make substantial justification for their version, file,
and line arguments. SCO owns the burden of substantiation of rule
SCO claims the order is ambiguous about version, file, and line
information. Version, file, and line information is not available for
method and concepts items, but in SCO's own words, "code" is used to
describe methods and concepts.
Tab 18: R. Davis. Item 17 memory code from PTX and item 26, NUMA.
Professor Davis says that methods and concepts don't exist in the
ether but in the source code.
Tab 19 Davis says methods and concepts must be implied by source
code. SCO's own technical witness, Mr. Gupta, says that methods and
concepts can be found in the source. Mr. Rockind also says that
source code describes methods and concepts. SCO says they must read
IBM's mind to get the evidence they want. Marriott doesn't want to
hazard a frolic detour into irrelevance.
IBM has repeatedly requested version, file, and line information over
the years including 2 motions to compel and 3 of [another type of
motion] yet SCO had not complied.
Judge Wells asked for all evidence to be on the table. Evidence
identification implies version, file, and line information. IBM must
figure out itself what it did.
Item 92, tab 26: SCO presents as evidence an anonymous post to the
Internet from someone named Michael who claimed to have worked on
Tab 24: scope of code implied. Substantial: millions of files,
billions of lines of code. It is not possible for IBM to find the
needles in the haystack when the needles are undefined. There are
many additional reasons the claims lack merit. Rule 37(c) is
Kimball: Thank you.
Singer: Strange that Mr. Marriott mentions the coversion of evidence
to claims. SCO wants evidence the other party has in order to look at
The order dismissing many of SCO's items excludes much of SCO's
evidence. There is nothing automatic about rule 37(c).
Second point. The move for summary dismissal is improper action.
Sequent and Dynix were licensed to IBM stipulating that methods and
concepts can't be disclosed. IBM's developers are the ones who know
the version, file, and line information for methods and concepts
items. [They are the ones who know what they were thinking when they
wrote their Linux code.] We've done all that we could. We can't tell
what an IBM developer had in mind. Only Wright [and another IBM
developer cited by SCO] knew what they did. IBM has not addressed the
fact they haven't asked their developers. This is not unworkable for
Tab 16: de novo review. There is no legal support in this circuit for
the dead fish rule. SCO can't penetrate IBM's firewall to trace down
the origins of claims. The items thrown out must be considered item
by item according to 10th circuit precedent.
Rules 37 and 26. Judge Wells dismissed basic stuff from off the table.
The 9 coordinate system was IBM derived [and not specifically required
by the court] in the July order.
Tab 27. No product, file, and line information [possible?]. The
contribution is admitted to come from Dynix.
Tab 16: AIX/Dynix. JFS, an entire filesystem.
Methods and concepts can be described without code. Many items have
come up since discovery.
Mr. Marriott's example is too abstract, not item by item
consideration. These items can only be considered after the expert
reports and discovery.
Identification of methods and concepts can happen without specific source code.
According to June 2003 Wells order the most important fact about
specificity is "where appropriate". [SCO should get an evidentiary
hearing or one or two other courses of action.] There is no precedent
for the Draconian strike of claims.
Kimball: [question about evidentiary hearing versus SCO's stricken claims]
Marriott: [a long string of numbers ending with the year 1971] Wells
made it abundantly clear she wanted version, file, and line
Exhibit 6: no single System V lines from the 187 items--no AIX/Dynix
lines--only 3 Linux items with line information, yet they have no
Kimball: [speaking about a request for an evidentiary hearing]
Marriott: No request was made until after the reply papers. The
request was waived
[says New York patent case is not relevant somehow]
One, SCO claims IBM developers know the coordinates. CMVC. IBM gave
this information to SCO. The minds of the developers are to be
questioned during deposition.
[a string of numbers again]
Kimball: Mr. Singer, this is your motion, you get the last word.
Singer: File, line, and version information wasn't stipulated exactly.
[stuff about an evidentiary hearing, the case is about limiting SCO's
claims] The [Ampex?] patent case findings ultimately fall within the
same cause of action as this copyright case.
Mr. Rockind used CMVC a great deal although CMVC doesn't say what the
IBM developers were thinking when they wrote Linux code. IBM isn't
charged with finding undefined needles in the haystack.
Last point. Item 146, Davis' declaration.
Kimball: [will take under advisement] The current trial date is
February 26. There's not a lot of time for a full scale trial.
Singer: 6 million pages, 40,000 pages of exhibits. Additional review
could be done in 1 week. Very short 2-3 page briefs relevant to
remaining issues. Hate to lose trial date.
Marriott: The last thing we need is more briefs. I don't know
realistically what we're getting into by taking on more briefs.
Kimball: How long would it take to hear arguments for the 10
remaining? How long do you want?
Marriott: That will vary by motion. There are very few facts in
dispute in the motions. Any of these motions will take 1 1/2 hours.
Singer: A reasonable amount of time.
Kimball: Two full days.
Singer: One day.
Kimball: 15 hours.
Kimball: The most brilliant jurist from December 9, 10 will be
difficult to be properly prepared.
Marriott: Can't expect [it], not feasible. Trial date should slip.
Counsel should collaborate, coordinate.
Kimball: Do that. I can see that you're still talking to each other.
Talk to each other before leaving town to arrange motions before
Singer: [inquire from Judge Kimball's office about his schedule]
Kimball: [determine how long]
Singer: [three months out, five week jury trial]
Kimball: There's always a bunch of trials. Let's plunk it down and do it.
Marriott: [not extend dates]
Kimball: December 23, 24? Should Novell be tried first? What's your
Singer: IBM case should go first.
Marriott: Novell case should come first to clean up/dispose of a lot of claims.
Kimball: [clean up is what Marriott wants, not Singer]
Marriott: [Novell has right to waive]
Singer: The issues are relevant to IBM, but not litigated in the IBM case.
Marriott: The issues belong to both cases.
Kimball: Thank you. [recess]
Singer: [40,000 pages, over 100 depositions, ask for week to respond]
Marriott: Additional week will be fine.
Kimball: 15th for replies?
Marriott: Get done a week early and be briefed by holidays.
Kimball: Thank you. [recess]
Update 5: I wrote to Crocodile to ask him some questions about what the lawyers and the judge look like. And he also gave me some information on what the evidence mainly is that SCO wants salvaged from Judge Wells' Order. In order of importance, then, first the information about the evidence:
I'm not sure much has been written about the "evidence" supporting them -- it may well be sealed (I'm not sure). It's typically emails from inside IBM to Linux developers (this is paraphrasing the words of SCO) saying words to the effect of, "This is how we did it in AIX (or Dynix)". The emails (apparently) don't reference any code. These are the points where SCO says IBM should ask it's own people. But why didn't they depose them? Is it a case of the lawyers realising that the actual answers might be damaging to their case, but the allegations look pretty good?
I don't quite see how you can put that back in without deposing the people that wrote the emails, and in one case at least, it's an email from an anonymous person who claims he at one time worked on Dynix, but there's no last name and no proof even that it's true. On the Internet, after all, no one knows you're a dog. And here's Crocodile's descriptions of the cast of characters:
Kimball has what appears to be shortly cropped grey hair and a round face. He seems to be a person with a good sense of humour.
David Marriott is tall and has dark hair with grey through it. He has a voice that's very easy to listen to.
Singer is shorter and a little heavier. He presents well and quite powerfully; however he is sometimes limited by his material.
Oh, and the person holding up the board during his marathon talk was Choo (I believe).
The legal teams seemed on good terms with each other. They greeted each other and shook each other's hands. All very civilised.
SLC is quite a nice place.