KOffice is basically dead commits-wise. What you might be interested in is the Calligra Suite, a fork which is under heavy development. More info: http://calligra-suite.org/

If you are legally allowed to. Some journals require copyright trasnfer upon acceptance of a manuscript, which makes such things illegal.

System Settings > Window Behavior > Window specific overrides.

That's exactly what you're looking for, and oh, it's been there for several releases.

In fact, I think it's just hype. To do proper analysis runs you *need* parallel computing, as these tasks are primarily CPU bound. But a whole load of RAM doesn't hurt, either, especially if you analyze more than one genome at once.

There's another problem to bear in mind: coverage. What is coverage exactly? In terms of "next generation sequencing" (to which this machine belongs to) is how well a part of the genome is covered (sequenced), and that in turns means that the number of fragments (100-150 in the case of Ion Torrent) read for a specific region must be high as possible (those are called "reads").

A good coverage allows you ensure that what you're seeing is real and not some sequencing errors (all technologies suffer from certain types of errors). In the case of Ion Torrent, you usually sequence fragments up to 1 Gbp of sequenced material (G base pairs of DNA), which isn't enough to cover the inter-genic regions or other structural DNA. So usually you do a "reduction step" by either only selecting a handful of genes to be sequenced (mutation screening for example) or by sequencing only the "exome" (exons are the parts of DNA that are actually transcribed to mRNA, the exome is a collective term for all the exons in the genome). That may be useful per se but it's nowhere near a complete genetic map for one individual.

Equipment to do that is much more expensive (500K USD, not counting reagents and other machines).

The name for the technology is "Ion Torrent", and albeit the wording there is terrible, it detects, using a semiconductor, the flows of protons in solution to determine which DNA base is being read.

not be capable of doing worse than FFXI again.

At least from a commercial perspective, XI is not bad at all, being still in operation and (partially) developed. Having played it for about 7 years, I would add that it was not that bad despite clearly a very unrewarding set up. In particular, the main scenario (and even more in particular the Chains of Promathia expansion) are well worth the FF name.

I beg to differ: I've run R600 and R700 (read: 3000 and 4000 series) cards with the FOSS driver and I find it miles better than fglrx: at least it does play nicely with all other components in the system.

I recently set up a wiki on a server of mine to be used to take notes to replace Google Notebook because I was afraid it might shut down, and now I read this.

Looks my hunch wasn't that wrong, after all.

at least the hysterical can plead ignorance. what's your excuse asshole?

That physics does not follow hysteria. The situation is serious, but the media are blowing it way out of proportion.

And as someone said, detectable means nothing. The detection systems are so precise they can identify the natural background.