Tee hee

Blast, I was going to repeat what you said on Victor's blog, but slightly rephrased, but I see you've beaten me to it!

If Curry does not get the message, she never will.

What is the story with Climate Lab Book, it does not appear to belong in the yellow subnet.

Harry,
I think it's because you can end up on the diagram both by taking part in the survey and mentioning 3 other blogs that you read, or by being mentioned by other blogs, even if you don't take part. I'm guessing that Ed Hawkins didn't take part in the survey but was mentioned by some of those in the yellow as being a blog that they read (because he's nice and had dinner with some "skeptics" :-) ) , but wasn't mentioned by any in the purple. So his blog ends up in the yellow, even if it really doesn't fit there.

Ed Hawkins's Climate Lab Book is actually an excellent blog by a working scientist who explains climate science (in particular modelling) for a lay audience.

Yes, Ed Hawkins got into the wrong cluster. He clearly presents good science. He provides the only link between all of the sciences and the WUWT & Co. cluster. I would say looking at the network, you can just see that that puts him on the boundary and the automatic coloring scheme of the cluster analysis cannot tell which cluster he belongs to and accidentally chose the wrong one.

Yes, Ed Hawkins's blog is excellent. Just to be clear, I wasn't suggesting otherwise by my earlier comment, simply trying to understand how it ended up where it did. Having looked at the graph again, my suggestion may be wrong. There seems to be 3 links to Ed's blog which might be because he took part, mentioned 3 other blogs which happened to be RealClimate, Mallemaroking (Mark Brandon), and WUWT and therefore ended up where it is because of the mention of WUWT. That does suggest, though, that noone mentioned his blog as one that they read.

ATTP, that would make sense, especially given the colors of the links. The 3 yellow links are likely the outgoing links of Ed Hawkins and the one purple link is the incoming link. Which would mean that the yellow WUWT & Co. cluster has not mentioned any blogs from the science cluster. Maybe they read them sometimes, but science seems not to be high on the priority list.

Mark Lynas is just mentioned once (by WUWT). He cannot help that.

Like I already wrote, you should not interpret the details. Even the main clusters are likely not very robust, but at least more so.

At Judith Curry they think like WMC above :) and mainly complain about the size of their cluster and I am the boogieman responsible for this size. While I did not mention the size on purpose, the sizes are easy biased by who the survey got too. If more mitigation sceptics' blogs were sampled more links may have appeared, but it seems likely that it would remain a highly self-referential cluster.

Also fitting to the compactness of the WUWT & Co cluster is that this Stoat post gives me more visitors than the post at Climate Etc.

[They don't get out a lot -W]

While in case of Climate Etc. it would make more sense for a reader to expect that my post is misrepresented and thus to click on the link to check what was really written. This gives the impression that the members of WUWT & Co. fear going out and being confronted with information that could produce [[cognitive dissonance]].

No chemistry blogs to be seen easily. Fing physicists and biologists

Heh. I was wondering if I had the wrong Mark Lynas in mind (since 'only one link' seemed impossibly unlikely) so I ran a quick google search. Not relevant at all - but I'm sure some here will enjoy this sub-headline to one article about him:

"Is Lynas just a slick self-promoter willing to say anything for attention, or did he sell his soul to the biotech industry?"

(That little gem courtesy of Alternet, unsurprisingly enough.)

Hola from the "ghetto" @jonova.
teehee!

[It appears that JN is avoiding discussing this little survey. Presumably because she doesn't even make the ghetto -W]

Did "Musings on Quantitative Palaeoecology" make the list? I like that one - it's often interesting though not so prolific in number of posts. I don't see it but the typeface on many of the entries is too small for us presbyopic folk.

[I think not. Its on my to-read list, but not on my top-3, though that's not a very meaningful concept as I read everything on my to-read list -W]

Well my name is longer than yours though :-)

But the whole thing seems a bit shonky, as it was just based on the random pals of someone I've never heard of (and who has presumably not heard of me either). And how anyone can say they "regularly" read a blog with a whole 4 posts in 2014...

[The selection methodology does seem a bit... as you describe it, I agree. I think I made it onto the invitees list due to a tweet I twitted at just the right time. Or maybe not -W]

The original post has been updated. It now contains a link an an interactive Gephi graphic http://bit.ly/MySciBlogREAD You can search individual nodes and check all their info like in and/or outbound links.

My guess would be that the sampling is quite good. Perfect is hard to reach for such a topic. The blog of Paige Brown herself is not dominant. Thus apparently she reached out well beyond her normal readers. There was an estimate that there are on the order or 5 to 6 thousand science blogs. If that is right, about 10% is sampled, that is a lot, even if also some non-science blogs got into the sample.

Here's one for Eli:
http://www.eurekalert.org/pub_releases/2014-12/dnl-bip122214.php

(oops [fixed - W]) Social Network Analysis is quite useful, if not over-interpreted, and if good, points the way to future research that will do better.
It's worth reading classic texts like Wasserman & Faust or de Nooy, Mrvar and Batagelj (the ones mis-used in Wegman Report).
See Relational Ties (W&F p.18, , Granovetter's strength of weak ties hypothesis (W&F p.449) or discussion in deN, M, and B, (pp.138-142 on strong ties.)
The challenge is that "read X" is an indication of a tie, but strength is unclear.

The same problem comes up in citation analyses, which generate directed (mostly) acyclic graphs ... but there is a big difference between a citation that acknowledges a source as a key reference, and one that mentions it to say that it is junk (politely, perhaps).

Gephi looks useful, as the challenge of these graphs has often been to create useful and readable groupings.

I hear the echo of Wegman's grad student laughing. She was not the only one to fail SNAP, apparently. But it's beautiful as both art and propaganda, so what matter if truth is nowhere evident in such a lovely compendium of circles?

[You don't like the results, obviously; neither does anyone in the yellow ghetto. But tarring it with Wegman? Shirley you wouldn't stoop so low. Or is it low, from your viewpoint? The correctness of Wegman is an article of faith amongst the Watties and the Mann-is-fat crowd. Do you disagree with Wegman? -W]

The WUWT cluster looks large, it has 8 blogs and many links between them, but this feature is less robust than you would thus think. It is only based on the responses of 3 bloggers.

Wow! A devastating analysis. Lots of squiggly lines and REAL CLIMATE in BIG CAPITAL LETTERS. And then STOAT in smaller (but still big!) CAPITAL LETTERS. You must be very proud. And I am going to have to re-think this whole global warming thing.

But in the real world, we use web traffic/Alexa Ratings:
Alexa World Rank:
WUWT: 19,427
RealClimate: 25,218,512

http://en.wikipedia.org/wiki/Web_traffic
(Careful, it's a Wiki page and you never know who has been editing it)

[Err, no. You use Alexa, because it provides a number that you like, and you avoid the caveats. Or perhaps don't even know about them. But even casting those aside, it obviously provides no measure of quality, only quantity; let alone the "connectedness" which is what we're talking about here -W]

Yes, connectedness. It takes a special someone to brag about being connected to the likes of HotWhopper and Sks.

[Sou is good; it would be even better if she had a better target; WUWT is too much like fish in a barrel. For more connectedness, see “Dr” Roy Spencer is sad and lonely and wrong -W]

Tom C:

Wow! A devastating analysis. Lots of squiggly lines and REAL CLIMATE in BIG CAPITAL LETTERS.

Heh. So much for the field of Social Network Analysis. What do you say to that, John Mashey 8^D?

I guess we can count Tom C among the dismissives.

Mal -

As Annan points out in the comment above, the methodology here leaves a bit to be desired, to put it mildly. But, for a "community" that accepts Lewandowsky, Cook, Mann et al. this is not surprising. WC is even happy to trumpet the results, especially because of the BIG CAPITAL LETTERS!. I guess we all cling to our comforts, eh?

Mal:
1) Well, I've been using SNA informally for 40 years, I've read key texts and traded emails with some top SNA researchers who've been very helpful. This study is typical of early probes in some direction: interesting enough to help encourage further work.

2) Having been on bulletin boards ~30 years, I long ago had to learn to recognize commenters never worth reading, from SNR ~ 0. Sometimes it only takes a few sentences.
I do miss usable KILLFILEs. :-)

Tom C:

I guess we all cling to our comforts, eh?

Only nihilists take comfort in the prospect of AGW. It is AGW-deniers (at least the ones who aren't fossil-fuel billionaires or their paid propagandists) who refuse to abandon their fool's paradise. I'm with Sagan, a genuine skeptic:

"For me, it is far better to grasp the Universe as it really is than to persist in delusion, however satisfying and reassuring.”

YMMV.

G*ddammit! Sagan!

But John, every science blog needs its pet troll, especially one who demonstrates the DK effect so well 8^).

Kidding aside, I haven't looked into SNA much, but when I see people like yourself taking it seriously, I know better than to dismiss it as nothing but lots of squiggly lines and BIG CAPITAL LETTERS. I didn't miss your your caveat, either:

Social Network Analysis is quite useful, if not over-interpreted, and if good, points the way to future research that will do better.

The arrogance of ignorance, OTOH, has no use for nuance.

Mal: thanks.

You might try Strange Scholarship ... pp.143-145, .

Later there was:
Strange Tales and Emails, p.7, where I wrote:
"Wegman recognizes Kathleen Carley as an expert, the sort of person who should have been asked to review (or write!) WR or Said (2008). Her opinion of the paper was similar to that or Robins and the second expert I asked last year, who recently sent an aptly concise email quote:
'Too bad you can only retract papers when it turns out they were plagiarized, when they should be retracted for not having any coherent or sensible argument!'
This is sad, but has much truth. I wish I‘d written it myself."

D-K afflictees know their knowledge is superior,, but some of us find good textbooks and especially figure out who real experts are and ask them a few reasonable questions, which many are happy to answer.

just looked this up again, saw the new link and ithe "analysis" is hopeless - it excludes all of my outgoing links, of which there are plenty (eg at least you, rabett, hawkins, RC, out of those included in the analysis).

> I do miss usable KILLFILEs. :-)

Oh, yes.

I've never understood why they weren't considered invaluable retained in the tools that superseded newsgroups.