Burt blog

We’ve been using Cassandra more and more lately. It’s pretty good write-wise, but it isn’t a very verbose Database and it’s bothering me. Here’s a transcript of a little conversation we had over SSH, nodetool and DataStax OpsCenter recently. No data was harmed in the recording of this incident. 

We have done our Master’s Thesis project about Browser Fingerprinting as an alternative to cookie-based identification methods. Now, I know what you’re thinking; tracking Internet users without them knowing is evil. Well it might be, at least if you’re not transparent about what you collect and why you want to track your users. Another crucial feature which needs be provided not to be evil, is some way for the users to opt-out (e.g. DNT). Anyhow, the ethical issues have not been our main concern for the thesis as we have been looking at the technical challenges.

Here’s the slides and a link to the video from Theo’s presentation at DeNormalised London.

The awesome people at SkillsMatter also recorded the talk and have put the video online.

“G*d damn it, you’re a programmer. Start using an editor you can damn well program.”- Magnar Sveen

If you’re a true programmer you gotta love that quote - I sure do!

To start off with, we have to get it clear what Emacs is. According to the official website:

“Emacs is an extensible, customizable text editor - and more. At its core is an interpreter for Emacs Lisp.”

But what does that mean? Emacs is extensible. Well… most editors are, you can just download a plugin and install it. Emacs is customizable. Again, most editors are. You can just grab your mouse, browse through the menu system and hopefully you’ll find an option for it.

At Burt we’re heavy into RabbitMQ. We use it as a series .. of tubes .. that we dump .. all our .. data .. on.. -ish. It’s been an integral part of our system for a long time and it’s served us well. Previous load tests had us worried though. We knew that if we overloaded the inputs we would sink the system. That was old RabbitMQ though.. 2.8 has flow control. Well. Lets see what 2.8 can do.

The combined JRubyConf EU and Eurucamp conference is over and we’re home again in rainy Sweden. We had a blast, the location at lake Müggelsee was awesome, the weather amazing, and the conference was great. Read on to find the slides from our presentations.

This weekend, for the first time, JRubyConf will be held in Europe, and we’re excited that not only one, but two, of our Burtlings will be there to speak!

JRubyConf EU will be held as part of the annual Eurucamp, where there’s also one more Burtling speaking. This year, Eurucamp will be taking place at a lake just outside of Berlin, and we’re all looking forward both to the conference and to all the activities on and around lake Müggelsee.

Fellow Burtling Jakob was representing with in style at the Spain.js summer javascript conference in Spain last week. We’re all proud he is now sitting next to us at the office and doing some “serious inception”. Unfortunately the talk was not video documented but he has written down some of his thoughts just after the presentation on his blog.

jakobmattsson:

Yesterday, at Spain.js, a library called Scoped HTTP Client got mentioned during one of the talks. While I agree that the native HTTP client in node.js is too low level for the most common use-cases, I find that there are fare better alternatives than the mentioned library.

Read more…

Relational databases. How’s that for a puke inducing concept? Foreign keys. On Delete Cascade. Brings back memories? This was all the rage a couple of decades ago. All the data was to be normalized and linked together, keeping the individual tables as small and general as possible. Database operations and schemas were thoroughly thought out beforehand and set in stone by a league of blue-tied men in short sleeved shirts and slack pants. Join:s were the shit, fields were typed and preallocated and universities were graduating, God help us, database engineers. Database-fucking-engineers.

When we started building version 2 of Rich (our site and campaign reporting app), one of our main goals was to make developing new features for it very agile.

The first version of Rich, which has been in production since late summer 2009 started off as a classic Rails app running MySQL and rendering views using ERB. It was fine for the first year, but as Burt grew immensely during 2010 so did the requirements of Rich and we started feeling the limits of our current architecture.

Starting last fall, we decided it was time to stop developing new features for our baby version 1 of Rich and instead focus our energy on version 2, which is currently being beta tested actively by several customers.