Entries RSS Comments RSS

Articles to 2015-07-02

First the link to this week’s complete list as HTML and as PDF.

***

I read Ahmed et al.’s numbers as probably one serious side effect per 1000 treatments. For a purely preventative, non-essential treatment of limited efficacy against a mostly harmless, transient illness, that seems a lot to me and I can’t quite agree with Wekerle calling it minute.

Correction: It is one per 10 000. So perhaps I can agree with Wekerle after all. For myself I’m only inoculated against life-threatening and permanently disabling diseases, not flue.

***

Lazar and Bakshy et al. go deep into technical detail but ignore the obvious question. By design and from its inception the Internet was decentralised, free, and libertarian, enabling everyone to seek and find all the information of interest and relevant to himself instead of what the editor of the local rag happened to consider appropriate for him. It also made it feasible and easy for everyone to publish his own stuff without any selection and filtering by the publisher. But instead of taking advantage the huge majority chose to make very few sites like Facebook or Youtube de facto monopolists and let algorithms and Big Brother choose their content for them. Not only that, they freely sold their souls in the process, as obviously the users are not Facebook’s customers but their merchandise which they manage to sell at large profit. As this tendency to hand everything over to Big Brother is one that can easily be exploited and misused, the obvious question is how it can be controlled.

That said there is one outstanding and significant result. It turns out that “conservatives” are far more liberal in their views while “liberals” tend to be totalitarian and favour (self-)censorship. Not that this comes as any surprise to those of an age group that has had the first generation of 68-ers as their teachers, but it is enlightening to find this subjective observation confirmed methodically and quantitatively.

***

While the Tim Hunt storm is blowing over leaving only debris in its wake, I consider Alessia Errico’s to be fitting final words: “Judge by actions, not words”.

***

Some time ago the colours of “the dress” caused a big stir all over the blogs and the yellow press. While the first considered and correct explanations quickly appeared in blogs, it is Current Biology that has now treated the issue experimentally and in depth in three short articles by Gegenfurtner, Lafer-Sousa, and Winkler.

***

As so often happens in the social sciences the explanations given by Voigtländer & Voth for their numbers are insufficient and meaningless. A confidence interval is defined for one sample from one community but not for hundreds of individual, different ones averaged. What figure 2 probably really shows is twice the standard error. As the error is meaningless here and the standard deviation about ten times as large, all values are statistically indistinguishable. How is this in any way relevant to the bombings and pogroms committed today and to the background of those actually committing and condoning them?

***

What I expect from a dating procedure is:

  • few preconditions easy to confirm
  • a simple, well understood process with few and precisely known parameters
  • few or no extra assumptions or auxiliary measurements

OSL conforms to none of these. Looking at the scatter plots of individual measurements in Fattahi, any attempt at actual dating seems wishful thinking at best. Yet despite all that and all the doubt it implies, his results do look consistent and meaningful and do not, in this case, confirm previous expectations

***

After having seen tracking in caves spread all over the yellow press and goggle box we are finally treated to a serious article worth reading by Pastoors et al.

***

As Leusch et al. point out the earliest large gold finds coinciding with the beginnings of copper metallurgy were already formed by casting. With the melting point of gold being far higher than that of cast iron this invalidates all the usual explanations for the lateness or iron metallurgy and the long reign of bloomery iron. (Granted, with gold you do not need a highly reducing atmosphere at the same time as the highest temperature, but I doubt that alone explains it.) So it would seem that before the advent of effective decarbonising cast iron was probably avoided as a useless waste material and iron seen as inferior to bronze and only used in a large scale in times of a scarcity of tin or of charcoal near the copper ores. See also Killick 2015 & 1991.

Leave a Reply

You must be logged in to post a comment.