TravisCI – a continuous integration service – shares some of the insights from over 2,000,000 builds they’ve run, in an blog post called “What We Learned about Continuous Integration from Analyzing 2+ Million Travis Builds“. For me, the most valuable bit is about the reasons for failing builds, which clearly indicates the need for and the importance of unit, integration, and UI tests:
Around 20% of all builds fail. There is a variation based on the language – for some programming languages, testing is part of the process and culture – for others it’s an acquired tool. Once you do implement testing, most of your builds will run. You’ll cancel very few. But about 20% will fail due to failed unit tests, configurations, or environment setups. Catching these 20% before it hits production is super important.
GitHub blog says that from now on your profile can include the private repository contributions on your profile.
When enabled, these can make quite a difference in the number of the green boxes, showing your GitHub activity. Here’s an example from mine. Before enabling those, showing only Open Source contributions:
And here’s one after, including private repository contributions:
Indeed, it is a more accurate representation of my GitHub activity. Given that these days most of my private repository activity happens on BitBucket and not on GitHub, this is quite surprising.
Jordi Boggiano looks at some common files in PHP packages, using Packagist as a data source. There are some interesting metrics in there. For example:
- 58% of packages include a src/ directory and 5% a lib/ one. That’s surprisingly low to me, that means a lot have the code simply in the root folder.
- 4% have a bin/ directory, including some sort of CLI executables.
- 55% have a LICENSE file, that’s.. pretty disastrous but hopefully a lot of those that don’t at least indicate in the README and composer.json
- 49% have some file or directory indicating the presence of tests (phpunit.xml & co). I am not sure if this is good or bad news to be honest, that depends on your expectations.
The flow towards Europe project provides a vivid visualization of the refugee migration. It is an interactive map with breakdowns by country, and with a timeline covering the years 2012-2015.
Europe is experiencing the biggest refugee crisis since World War II. Based on data from the United Nations, we clarify the scale of the crisis.
Stack Overflow published the results of their developer survey for 2016. Over 50,000 participants from more than 170 countries answered the questions this time around. Some of the results are quite predictable, while others less so.
Here’s some not so light coffee time reading on IPv6 – IPv6 non-alternatives: DJB’s article, 13 years later – an article that links, among other things to this Ars Technica article, which features some IPv6 statistics. Summary? Sure. IPv6 RFC celebrates 20 year birthday this month with 10% global penetration.
Exponential growth year-on-year is good. But the absolute numbers aren’t so bright yet. Especially considering some of the areas where it wasn’t so successful.
This year’s Jetpack annual report for this blog is ready – have a look. Here’s a teaser:
It’s been a busy year, so I haven’t been blogging as much as I wanted to, but overall, I think I did good (have a look at 2014 and 2013). Just to give you a quick comparison:
I blog mostly for myself, but it’s nice to see a slight grow in traffic. Although the fact that the most popular post in this blog throughout the years – how to check Squid proxy version – is a little concerning, yet funny. Well, at least people still find my “Vim for Perl developers” useful, even though it’s been more than 10 years since I wrote that (and probably five years since I promised to update it soon).
But as I said, I’m quite satisfied with my blogging this year. Hopefully I can continue to do the same in 2016.
Here’s an interesting set of experiments trying to answer the question of how far can you go with HAProxy setup on the smallest of the Amazon EC2 instances – t2.micro (1 virtual CPU, 1 GB of RAM). Here’s the summary.
At 460 req/second response times are mostly a flat ~300 ms, except for two spikes. I attribute this to TCP congestion avoidance as the traffic approaches the limit and packets start to get dropped. After dropped packets are detected the clients reduce their transmission rate, but eventually the transmission rate stabilizes again just under the limit. Only 1739 requests timeout and 134918 succeed.
It seems that the limit of the t2.micro is around 500 req/second even for small responses.
Back in 2013 I linked to some (not so) surprising facts about social media. Two years is a lot of time and a lot of things has changed since. So here comes 100 social media facts and statistics for 2015. These spread from general statistics to service-specific ones, including Facebook, Twitter, LinkedIn, Instagram, Pinterest, Google+, and others. Unlike many other similar collections, this one actually links to sources for every single fact, and provides an easy one-click share to Twitter button. Here are a few to get you started:
- 189 million Facebook users are ‘mobile only’.
- There are 4 billion daily video views on Facebook.
- 50% of unique LinkedIn visitors access it via mobile.
- There is a 50% average increase in comments when a LinkedIn page post contains a question.
- Over 40 billion photos have been shared on Instagram.
- Google+ has 300 million monthly active users around the world.
- Google+ grows at a rate of 33% each year.
- Average time spent on YouTube per mobile session is 40 minutes.
- There are 300 hours of video uploaded to YouTube every minute.
I get my stats anywhere I can find them. Here’s a source that I haven’t seen before – Pornhub Insights, showing all kinds of differences and similarities between Apple iOS and Google Android users.