Forecasts are rather difficult, especially when they’re about the future. Or so the saying goes. While predictiong returns is a pointless exercise, there is some value in keeping an eye on correlations.

I use SSH for pretty much anything from VPN, server administration, database connections or iPython work on remote machines. When working from weird places and with weird internet connections SSH become painfully slow. I already use Mosh, but that also relies on ordinary SSH to initiate the connection. Pittsburgh University has this OpenSSH-patch to remove some bottlenecks and make it 1000% faster (they claim).

Last week we discussed Linux Debian’s apt-get update mechanism and how to fully automate essential updates. This week I’d like to demonstrate how to do the same thing for Python. I admit that keeping Python packages up-to-date is probably not half as essential as keeping internet-facing server infrastructure updated. Nonetheless I like to work with the latest versions of packages, as they might fix problems or add features.

This week’s massive SSL-security vulnerability showed how important regular security updates for all of our software is. Because – let’s face it – today’s world is largely powered by software. Software that is written by humans, who make mistakes when writing it. The rule should be: retire it or update it.

I’m back in Shanghai and faced with the pollution problem once again. Here a quick update on the last months. You can clearly see a reduction in pollution around Chinese New Year, when factories are shutting down. With warmer weather the readings also seem to be lower. I don’t know the reason, but anecdotal evidence gives the same effect for Beijing. An expert from Vienna University of Economics is currently analyzing the data and correlating it with weather observations. I’ll give an update, when some results emerge.

This morning I came across this post by Kelly Norton. He calculated the number of ‘pleasant’ days for each US zip-code area. California seems to win the race with more than 180 ‘pleasant’ days each year. A pleasant day is defined by the min- and max temperature not exceeding certain limits.

I’d like to officially name my current dev stack: A.. for AngularJS. Drives the user frontend. S.. for Supervisord. Takes care of processes. P.. for Python. Quick way to implement almost any business logic. E.. for Nginx. Fast web server for static files and to add SSL. L.. for Linux.

Few things are more annoying than hotels, who think they need to earn some extra cash by charging people for wireless internet. Given the low to nonexistent cost of providing the service, they should also charge for warm water or fresh air, when following the same logic. In the case of a local Howard-Johnson (HoJo) hotel, the wifi’s paywall was so badly implemented, it actually posed a threat to the rest of the hotel. Below, I will describe 4 options, anyone can use to get free internet at this particular hotel.

A word of warning: Updating from Mountain lion to Mavericks is by far the most buggy process I have ever seen from Apple. At times the installer quit in the middle, My time machine volume was not recognized, network settings are lost, … Make sure you have multiple backups and plan some downtime. On the plus-side Homebrew and my Python-packages all survived. Just make sure you use pip with a virtualenv or the –user option.

This is pretty crazy. Every conspiracy theorist was correct. The government is really watching everything. All internet data on the internet is compromised. Check out the presentation and read it nice and slow: http://www.theguardian.com/world/interactive/2013/jul/31/nsa-xkeyscore-program-full-presentation