Re: Language Oriented Programming

Laurent Bossavit wrote:
> Isaac,
>
> > Is there any chance of informed discussion when we are simply guessing
> > that the statements we make are correct?
>
> Sure, as long as they are informed guesses.
>
> > My rather bad-tempered point was that if "you don't know the ratio of
> > robust to fragile" then why oh why are you claiming that "most
> > ecosystems can afford to lose one species".
>
> There's a background rate of species extinction which I've seen quoted
> at one to ten per year in prehistoric times,

I don't really understand what that means:
- does "background rate" suggest that there's some other "species
extinction" going on which isn't included in the "background rate"?
- "one to ten per year" looks like a ball park estimate, which (if the
number of species has increased over time) seems to imply that
extinctions have decreased over time?

I can see that someone might have come up with something like that by
making both a low estimate and high estimate of extinctions over the
fossil record, and then working out an average per year.

If "one to ten per year in prehistoric times" is that kind of ballpark
average, then it doesn't seem to tell us anything about how those
extinctions were distributed in space or time. It doesn't seem to tell
us anything about how one species extinction may or may not have been
related to another.


> and I've not come across a study of an extinction event or ecosystem collapse
> which attributes the cause to loss of a single species.

Neither have I, but in my case it's simply because I haven't looked.

It would be more interesting to hear that you've looked at X out of N
studies of documented species extinctions and none of them
attributes...


> Logically, it's also part of what we mean by "ecosystem"; an ecosystem
> is a stable community of species, plus environment. Typically, the
> identity and continued existence of a community does not crucially
> depend on any single member of that community.

That just seems to be a different phrasing of the initial statement,
not an argument - it's begging the question.


> > Is there some specific situation where you can say that you know "when
> > one module fails (even a small one) typically the whole thing fails"?
>
> I'll retract "typically". Read "too bloody often" in its place.

So, somewhere between 1 and N where N>1

(And the specific situation would be - Ariane?)


> You said: "When one module fails (even a small one) typically the whole
> thing continues on as though nothing happened (aka log the error)."
>
> If that was true, Ariane would certainly be *atypical*, in that the
> computation which caused the exception was no longer relevant when the
> program blew up. If it had logged the error and continued, rather than
> shut the whole subsystem down, nothing would have happened.

I said "I just turned your statement around".

What is a "typical" software system?

.

Relevant Pages

  • Re: The Ratz Ten Favorites of 2004
    ... scientists to understand an unprecedented series of 'blasts', energy surges, ... detected in the upper troposphere large positive ions with mass numbers up ... Mass Extinction Underway, Majority of Biologists Say ... The rapid disappearance of species was ranked as one of the planet's gravest ...
    (sci.astro)
  • Re: An idle question (Paging Mark Isaak)
    ... Frankly, if you really want to know what I think; I think there *was* a relatively big loss of species, but it happened a few million years before *you* think it happened and/or I'm quite convinced that it happened over a much larger period of time than you claim. ... they are virtually extinct due to man "hunting" them into virtual extinction and virtually no other reason. ... The assertion that "almost exterminating" is somehow different than "for all intents and purposes it WAS exterminated" is simply evidence of your desire to deny what is accepted as fact by virtually everyone, in a vein effort to make your point-of-the-moment. ... For example, if you look at the quote I was responding to, there is no mention of megafauna or flint or "a time when the world was underpopulated by man". ...
    (talk.origins)
  • Evolutionary consequences of sixth mass extinction
    ... How Will the Sixth Extinction ... Affect the Evolution of Species? ... Human activities have brought the Earth to the brink of biotic crisis. ...
    (sci.bio.evolution)
  • Article: Biologists Map Mammal Extinction Hot Spots
    ... Biologists Map Mammal Extinction Hot Spots ... Modern conservation efforts focus on saving as much biodiversity as ... possible--and preventing living species from going the way of the dodo. ... first estimated the overall risk of extinction a given mammal faced based on ...
    (sci.bio.evolution)
  • Re: Sean Pitman: definitions wanted
    ... So maybe this explains most of the background extinction events? ... say there are 1000 different species all occupying the same ecological ... Then the environment changes slightly, ... neutral gap problem, affecting a vast majority of species, but not all ...
    (talk.origins)