yeah for those that are new, slashdot superbowl ad threads are an annual tradition that was codified in 2000 during the dot com bubble, featuring big-ticket flashes in the pan such as pets.com.
Can you explain why it's important to you that (A) you have to wait to open the start menu and (B) that it takes up only a portion of the screen?
This is incorrect. No mainstream models of dark matter are based directly in string theory. It's hard to demonstrate directly that this is the case but if you'll do a search for 'string' in the dark matter wiki article you'll see no mentions barring one at the end, a tangential mention under the 'alternate theories' section.
Also I think you overstate the case when calling dark matter a 'questionable' theory. It is widely accepted among cosmologists, at least as a tentative explanation that fits the available data.
tell me about it, i was potty trained via flashcards and now i can't write an exam without pooping EVERYWHERE.
what you are saying isn't strictly speaking meaningful. all physical constants are relational. which you pick as 'fundamental' versus 'derived' is at some point arbitrary. if this article was about the speed of light, you could just as well ask which constant isn't, the length of a unit of space or the duration of a unit of time?
Interestingly enough these considerations apply precisely to EVE as well.
fair enough, your annoyance is justified.
i still think you're being too harsh on computational analogies in general though. i'll admit that my knowledge of molecular biology is meager, but maybe you can point me in a good direction to dispel my misconceptions.
as i understand it, we have a strings of a formal language (DNA, RNA) which are operated on by state machines (proteins, protein/RNA complexes) that act (relatively) deterministically to either modify the original string or create/modify a new/existing one (proteins, DNA, RNA).
obviously this description is nowhere near complete; proteins are a much more functionally complex kind of 'string', for instance. but as far as i can tell the basic idea of molecular biology is first-and-foremost the basic idea of molecular chemistry: discrete, combinatorial entities composed of atomic constituents that interact with one-another in deterministic ways to produce other such entities. biology adds a new layer of abstraction by representing functionally distinct units (proteins) in a common, (relatively) functionally homogeneous formal language.
accepting certain glosses for the purposes of brevity, is this a fundamentally incorrect way of looking at it?
That may indeed be the case. The problem is, this usage is simply wrong.
yes, i agree.
considering genomes as programs
and any line of reasoning built on the assumption that they are fails on that basis.
i don't see how this follows. nothing is 'the same' as anything else but yet the same methods of reasoning can apply if one can draw a formal analogy, with clearly delineated constraints. this is the premise of mathematical modelling, the common feature shared by all hard sciences. this particular analogy is unsupported, and it is probably not terribly useful, but their clearly are similarities, and it might be at least didactively useful to address them.
i mean, admittedly the OP rings hollow, but the idea that the potential for 'total viral immunity' is suspect on a logical (mathematical) basis might have some grounding doesn't seem on-its-face ridiculous. but maybe you can enlighten me.
i think a lot of non-mathy people colloquially use 'godel's theorem/s' to refer to the pretty general notion that 'there exist simple formal problems which can be proven to admit no general solution.' like how there is no compression algorithm which can compress all strings of data. is there a good term for this situation?
i'm not totally convinced that even a direct godel reference is necessarily bullshit here either, or rather i could imagine that computability comes into the issue. like, trying to state an assertion in a slightly more formal manner: considering genomes as programs, there exists no 'host' program which can distinguish all viral programs from native programs without actually executing enough of some viral programs to ensure replication. something close to this statement should be true i think (?).