Holiday reading, book 1: The Square and the Tower (Niall Ferguson)

I really enjoyed this book. Nevertheless, compared to the author's other compelling works the pace is quite uneven. Parts of it can easily be skipped, in fact. Where it reviews the theory of networks, for example, I was quite bored and disappointed but possibly because I know the topic somewhat well. It is really very good in the main however and most of the conclusion is excellent.

Ferguson makes the point that our era of technological advance in communication is not perhaps that novel at all.I would differ from the author on three points though, in his conclusion and general approach.

1. I am not nearly as bullish as he is about the progress of AI. It could also easily be that it all ends up, if it even happens, as some kind of intelligence living happily and separately in machines and knowing and caring about us as much as we care about the cockroaches or beings on other planets. Moreover, I do not see a great leap forward between now and 2030 although I would not completely discount the possibility that by that time (or even now) it will only be the computers reading this.

2. Tall buildings as a symbol of power do not strictly belong to the past but are still a symbol of (now) consumption power for the rich of Manhattan, Dubai, Bangkok and (lately) even Johannesburg.

3. The current American president does not strike me as of immense importance in the long run but neither does his predecessor.

The author is an admirer of the political philosophy of Henry Kissinger and here I agree with him almost entirely. He very nicely describes the dangers in the near future that Kissinger had identified and I am afraid that these are very real. Further, I think that the book pays insufficient attention to modern Germany as an economic and cultural power, instead concentrating overly much on the UN Security Council countries. I would not have placed much emphasis on France but one of the nice things about the book is that South Africa features surprisingly frequently. I really enjoyed this and think that the country is an interesting case in the history of the world over the past few hundred years. French readers might have inverse sentiments.

Medical information confidentiality – little of it at Bonitas

Bonitas medical fund has for years now been sending me loads of confidential medical information regarding one of their members by e-mail. Presumably, an address had either been given incorrectly by the patient or captured incorrectly by Bonitas. Despite the fact that I have written to inform them of this, the e-mails have not stopped coming.

A recent one deals with a claim for OTHER SPECIFIED MISADVENTURES DURING SURGICAL AND MEDICAL CARE (codeY65.8) which leaves me wondering about the code for unspecified misadventures might be and why this is not something that the medical services provider should fix for their own account. Nevertheless, the privacy of the patient is being egregiously violated and (of course) she has no way of knowing this. Both the e-mail and the letter contain the following text.

This information is confidential and only intended for the beneficiary specified above, or in the case of a minor, the legal guardian or parent. The sender hereof cannot be held responsible for what happens if the recipient shares this information with any other person.

In other words, Bonitas cavalierly indemnifies themselves and me – surely a legal impossibility – from whatever happens to this information. In any event, service providers can avoid a lot of this mess by using the Yimi or Yimi1234 systems for obtaining patient information. Bonitas might want to get their house in order.

Banks should let customers input amounts as displayed

An older lady of my acquaintance recently had to drive 20km to visit the branch of a well-known South African bank in order to make a transfer to another account at the same bank even though she uses Internet banking. It was a somewhat larger amount than usual and her mistake was using the comma as thousands' separator. Why? Since that was how the bank also displayed numbers on the same page. Very recently, I had the same problem at another bank. If you include a space in the amount, it is rejected even though the bank displays the space itself, right at the next step. Now, since we officially use the comma as decimal symbol one can understand a reluctance to allow it as a thousands' separator when using the decimal point as most banking applications do. Why the space is forbidden, I do not know as it clearly makes the number more reasonably and a few lines of JavaScript can remove it when the transaction is submitted. The same is true for telephone numbers. User Experience: fail.

Skype for Business is nie Skype nie

Mense wonder soms waarom gewone Skype en Skype for Business nie met mekaar praat nie. Die rede is dat die twee eintlik min met mekaar te doen het – buiten dat beide tans deur Microsoft besit word. Skype is 15 jaar gelede begin in Estland en eers heelwat later deur Microsoft gekoop.

Gewone Skype het baie mense verlos van duur internasionale oproepe en vriende en familie nader aan mekaar gebring. Die sagteware en veral Skype se algoritmes vir die verspreiding van die netwerkverkeer het eenvoudig net goed gewerk.

Skype for Business is eintlik Lync, die opvolger van Office Communicator 2007. Dit is 'n Microsoft-produk wat gebaseer is op uibreidings tot die SIP-protokol. Dit is dus nie 'n diensuitbreiding van die gewone Skype-produk,  soos die naam dalk aandui, nie.

Bibliography tools for easy referencing of an academic paper

Nowadays, when I quickly need to write a report or paper with academic references, my methodology is to first complete a list of references and try to download all the papers and other sources that I think I might use. Since it is very convenient to write using the Markdown language and use BibTeX source files for citations (since this allows for later easy formatting and reformatting) it helps to first get the BibTeX references for each paper.

This is easiest using doi2bib which provides the BibTeX from a DOI number or address. The DOI is a generally used universal identifier for much technical literature. Incidentally, the easiest tool for obtaining the papers themselves is the cooperative or pirate (depending on your point of view) site Sci-Hub which uses the same DOI to get the paper itself from behind the paywall. Naturally, Sci-Hub is probably in serious violation of ordinary copyright but I think the users, especially if they are already entitled to access the material they seek but in a less convenient way, might consider it to be fair use and moderately unobjectionable.