h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

If you are planning further reaction to the
unkind comments of others I have to suggest
you save your breath. It is the nature of the
interweb and even this wonderful h2g2 site
that there is no permanent record, no permanent
solutions. It is in a state of flux but keeps
on moving forward inexorably. So don't get stuck
in past events or hung up on past comments that
have flowed like water down the dirty duck's back.

Onward and upward.

~jwf~

Hi jwf,

It seems appropriate that on H2G2 we truly have parallel universes.
If you are having a conversation with someone in the same universe (e.g. alabaster to alabaster) then there are no issues. If you are
having a conversation with someone on another universe (i.e.
alabaster to barlesque) then there is an issue. Anyway I have now
managed to break through into the barlesque universe and have
managed to posts my replies (three links into the alabaster world to
my replies).

What I have learnt in this exercise:
a) better understanding of the various skins.
b) debugged a bug on barlesque.

I was confused why MM hadn't picked up on one of my replies in his
comments (this turned out to be unrendered on barlesque) and I
couldn't be certain that SWL and Orcus were not using barlesque.

Barlesque does look fine on ipad but not so good on a PC/Laptop. I
generally use a laptop.

>> on H2G2 we truly have parallel universes. <<

Indeed. And it is not only appropriate but essential!
There are as many versions as there are readers.

That way, every anonymous researcher can ventilate into
the void and be ignored or hated or loved of respected
or admired or whatever it is they hope to achieve.

The important thing is to realise that the interface
ends at the end of our noses. No one nose what really
lies beyond the cold sweaty noses we press against
the glowing screen.

It is as Shakespeare said nothing more than 'a foul and
pestilent congregation of vapours'. A lot of sound and
fury signifying nadda!

And yet how vital and invigorating it is!


~jwf~

>>This is the type of work I would like to steer my interests towards
in the long run (i.e. developing a predictive capability to group
behaviour) http://www.bbc.co.uk/news/technology-14841018 <<


Considering the BBC report, the title is oversimplified as well as
wrong: "Supercomputer predicts revolution".

Firstly, a "supercomputer" is just a tool like a hammer or a
screwdriver although, of course, it is a lot more versatile. It is
the researcher applying a particular methodology to a set of data
(extracted from the internet in this case) that hopes to be making
such predictions.

Secondly the study did not "predict revolution". The study was
retrospective and looked for "patterns" in past data, matching these
patterns to specific past world events (including revolution). The
researcher hopes to make use of it for prediction purposes.


Something possibly to be concerned about is the following comment
in the BBC report: "Mr Leetaru said he also hoped to improve the
resolution of analysis, especially in relation to geographic
location. The next iteration is going to city level and beyond and
looking at individual groups and how they interact"

... which is sounding a little like "hacking". Google can track
individual usage of the internet (whenever you use Google Search)
but it focuses on targeted advertising - that's how Google makes its
money.


In terms of working in a predictive capacity it could work a little
like the following: Imagine there is an analysis of the words "Nick
Clegg" and it is correlated with the words "idiot" and "never vote
for them again". Before the general election the number of
incidences (correlations of "Clegg" with "idiot" or "Clegg"
with "never vote for them again") was likely quite low around the
world. A few months after the British general election the number
of incidences would have begun to rise in the geographical region of
the UK.

When the Liberal Democrats, as part of the British coalition
government, announced that they would renege on their pre-election
promise by tripling student tuition fees, then there would have been
an immense surge in the number of these incidences in the UK,
especially in England. It is likely that the number of incidences
will drop to a steady but high level, to be followed by a further
increase towards the next general election. From this we could
confidently make the prediction that the Liberal Democrats will
be "wiped out" at the next the general election.

HS>>"Did you watch the recent docos, 'All watched over by machines of
loving grace'? It seemed to suggest there is always a flaw in it."<<

No but I just had a look at a three and a half minute "trailer" of
it. The trailer would perhaps suggest the documentary was more
"style" over "substance", with perhaps a focus on the ideology
of a few "offbeat mavericks" and engendering a sense of anxiety and
paranoia. But that's just based on the trailer (and the title):
http://www.bbc.co.uk/blogs/adamcurtis/2011/05/all_watched_over_by_machines_o.html

In terms of "flaw" I would guess that would be associated with the
particular computer programme (generally a bit of mathematics plus a
set of "rules") and what that programme is supposed to "represent"
in the real world. As mentioned in my previous comment, ultimately
a computer programme is just a "tool" and shouldn't be used as a
replacement for not thinking:
http://www.youtube.com/watch?v=9nS0VVTyPnY

No it wasn't style over substance. I detest that. They were quite playful with a lot of different ideas.

There were 3. If I recall, each an hour long..about completely different aspects to do with quite fragmented ideas.

I talked about the docos quite a bit on the Dawkins thread. I thought I'd kept a link..but I can't find it.

And I'm not searching. And the discussion wasn't on google search.

Never mind.

Hi Okay, I will have to wait until the series is repeated on the BBC