A friend pointed me to this blog posting on the latest developments in the SCO vs IBM-and-everyone-else-Linux case. My first response was "Is that STILL going on?"

Oh, yes, friends, the case is dragging into its third year and not only is it dragging, there's no likelihood of a trial on the facts any time soon. As near as I can make out, both sides have buried the judge in motion requests, each of which has to be considered on its merits and ruled on by the judge before anything else can happen.

One of the motions filed by IBM was a request for summary judgement - essentially saying "Judge, this is crap so please just tell them to go away." It is my sense from talking to various legal people that judges are usually extremely reluctant to issue summary judgements. If nothing else, a judge tends to feel that even a far-fetched case deserves a hearing and to be fair it may be that new things emerge at trial, or the threat of an actual public trial may force parties to settle or compromise. Not to mention that any summary judgement in a high-stakes case is itself going to be appealed and may put the judge him- or herself under scrutiny.

So if IBM wants to win this motion it really has to deploy extremely convincing evidence in support of the motion. Naively speaking they'd have to show that their case against SCO is a slam-dunk or that SCO's claims are utterly without merit.

Which brings us back around to the blog posting. Since the trial documents are not public (and may never be) we have at best a redacted version of IBM's supporting document (linked at http://www.zensden.net/misc/IBM-838-1.pdf and http://www.zensden.net/misc/IBM-838-2.pdf). There appear to be five reasons IBM is willing to argue why SCO's claims should be dismissed.

1. No Code, No Crime. SCO has yet to produce any evidence to support their claim that IBM copied protected code into Linux. In addition, IBM argues that SCO hasn't produced evidence to show what rights it has in the particular code it claims IBM copied.

2. Got License. IBM argues that its various licensing deals permit it to use the code. The judge may also be swayed by the argument that some of those licenses came from SCO's predecessor companies or organizations in which they were members. Caldera's actions in respect to that same code under the GPL may also be relevant.

3. Promissory Estoppal. IANAL and I don't pretend I can explain this concept in general, but it appears that IBM is claiming SCO's legal predecessor (Caldera) encouraged and endorsed use of the code. SCO can't go back on that.

4. No Copyrightable Code. IP Wars admits that there's a lot of relevant context in the material not in the public domain, but what appears to be happening here is that IBM is claiming the particular lines of code identified as contentious are themselves a functional entity and not expressive in the manner that would invoke copyright protection. This is not a position statement from IBM on the copyrightability of code in general, but rather an argument that SCO is picking-and-choosing particular statements (e.g. function prototypes) that don't _do_ anything. There's a technical point here about whether the code lines constitute part of an international standard and/or an API definition in which case IBM may be drawing on previous case law that has ruled such interfaces are not copyrightable.

5. Copyright Misuse. IBM appears to be claiming that SCO is trying to enforce copyright law in ways not permitted. IP-Wars claims that "the penalty enforced by federal courts for overreaching copyright claims is forfeiture in the instance." So if this is true and the court finds that SCO overreached itself, the court might dismiss the entire case on these grounds alone.

Now with all that said it's important to remember that SCO will be submitting material arguing the other side and it's possible they'll find persuasive counter-arguments to each of IBM's points. Or maybe the judge will chuck this entire mess in the trashbin where it belongs.