One renown scientist told me that simulations are just models that you tweak to get results you want.

In a very narrow interpretation, this is correct. If you're playing the good scientist for the day, you tweak a model to get results which agree with known values, which are the results you want. Then, using the refined model, you simulate the situation you want to predict. When those numbers come out, you tweak your model to find out how sensitive the simulated results are to each of the simulation parameters to estimate error bounds. Lather, rinse, repeat, publish.

Many scientists (I am among them) toss around glib quotes that serve more as cautionary tales, but the non-science public gets to hear these out of context and thinks of them as evidence of fraud. As an example which applies to simulation, in physics: "Theorists don't believe each other, experimentalists don't believe their results and computational physicists don't believe themselves." Which is really to say that computational physicists are always on the hunt for artifacts (read: wrong answers) due to the simulation process; these are distinct from incorrect theories.

Before you conclude the innocence or guilt of the CRU folk from something you heard from some other person who wasn't even commenting on their situation, let the investigation look at all the data.

Yes, as opposed to 28% in proprietary software (http://www.openwebvancouver.ca/sites/default/files/byron-women_in_open_source.pdf, p. 14-17). If it were as simple as "this trait [of willingness to do significant unpaid work] is far less common in women than men," then you'd expect that no volunteer organization to have a significant population of female members. But peace corps volunteers are 60% female (http://www.peacecorps.gov/wws/enewsletter/archives/may09.html).

I'm inclined to agree with you that women are not attracted to programming in general because of social factors external to the programming world, but there are definitely internal factors as well. (As evidence, I cite the fact of active research in the field of physics education toward understanding why women initially do well in introductory courses, but progressively drop off in higher-level courses. Mind you, this research is done by real physicists.) Most technical fields suffer from this. Non-commercial programming doesn't require a different level of competence than commercial programming; free software approaches problems of similar difficulty and scale as non-free software, therefore a difference between men and women (if one exists!) in interest or competence doesn't account for the large difference in female participation. I'd bet money that there are numerous causes which the FOSS community will need lots of time to sort out and repair, but they're not all invisible.

Bruce Byfield touches on part of the problem (http://itmanagement.earthweb.com/osrc/article.php/12068_3838186_2/Sexism-Open-Source-Softwares-Dirty-Little-Secret.htm): When an incident occurs where a joke unintentionally offends, the person who made the joke inevitably says something to the effect of "it was just a joke, chill out." This only serves to alienate women (and other offended groups), rather than re-gain their confidence. This applies in the cases of all mistakes, particularly on the part of leadership, which affect others. A failure to acknowledge the error will only persuade people to leave the community.

Defending the mistake is the wrong thing to do, and will only make the offended people more angry because most human beings do not like having their feelings and opinions ignored (I offer a lifetime warranty on this technique for alienating people). The excuse, "it was just a joke," essentially invalidates the opinion of the offended people, whether or not the defendant intends (you might notice that people who make excuses excessively have few to no friends). The correct thing to do is actually apologize. To understand this process the apologetic party must make an attempt to stand in the other party's shoes, then coherently address their concerns. To effectively show concern for the complainant, the person apologizing must be brave about it and never mount a defense. The process follows:

1) Do NOT defend actions. Do NOT put the burden on the other party; specifically, use active first person verbs ("I made...", "I will...", etc.), and do not use the passive voice ("I was misunderstood...") as it contains an implicit "by [someone]," effectively deflecting responsibility.

2) Acknowledge the other person's feelings. (Examples: Formal; "I understand that you are uncomfortable with something I [said/did]." Less Formal; "I get that you're pissed at me.")

3) Acknowledge responsibility. (Formal; "This arose from [an action I took/a statement I made]." Less Formal; "I messed up.")

4) Cite specific failure. (Formal; "I did not foresee the [negative outcome(s)/implicit message(s)] which I [caused/expounded]." Less formal; "I didn't know [it] was meaningful to you.")

5a) If a lesson was learned state the lesson. (Formal; "In the future I will be more aware of the impact my [words/actions] have [in this regard]." Less formal; "I won't do [it] again.")

5b) If a lesson was not learned, ask questions. (Formal; "I don't fully understand how this mistake came to be. What can I think about to avoid this in the future?" Less formal; "I don't get it, but I care. What shouldn't I do?")

6) Apologize. (Formal; "I regret my actions and hope that they have not caused a permanent rift. How can I work toward our renewed collaboration?" Less formal; "Sorry, dawg, can I buy you a beer?")

7) Do NOT expect the other person to accept the apology. Chill out, be prepared for either outcome.

I hope that Stallman will man up and sincerely apologize to the offended FOSS development women, and learn from this mistake. It's not only true for software development, but in all environments, that valuable contributors actively learn from their mistakes. It's our one advantage over the other critters on the planet. In the words of an old research advisor of mine: "Don't be a god-damned animal."