Thoughtbot best practices guide →
A guide for programming well.

❧ Programador, valenciano emigrado a Madrid. Trabajo en BeBanjo y parte del tiempo libre lo dedico a Toldo. ¿Quieres saber más? →
Back in January, Sandi Metz introduced her rules for developers in a Ruby Rogues podcast episode episode. Around the time Sandi’s rules were published, the team I am on was starting a new project. This post details the experience of that team applying Sandi’s rules to the new application.
The…
Hace un par de días me enviaron de Amazon un cupón de 10% de descuento para cualquier Kindle. Como yo ya tengo uno he hecho una pequeña landing para el primero que lo pille:
Si me quieres devolver el favor abajo hay unos enlaces de afiliación a los Kindle que puedes comprar. Los puedes usar para comprar el Kindle en Amazon.
A [work-in-progress] self-hosted, anti-social RSS reader.
Stringer has no external dependencies, no social recommendations/sharing, and no fancy machine learning algorithms.
But it does have keyboard shortcuts and was made with love!
Ocho principios nucleares:
In the last week we have been measuring the mean response time in the server of the public pages of Toldo shops, it is, how long takes Rails to render those public pages.
What we did is, by far, the simplest solution: each day we process the previous day logs using a simple bash script, which filters the requests from the shops and extracts the column with the response time.
values=`cat $1 | grep "Completed 200" | grep -v '\[tol\.do\]' | awk '{print $8}' | ruby mean.rb`
Also, a simple Ruby script helps calculating the mean:
# mean.rb
#!/usr/bin/env ruby
total = 0
sum = 0.0
STDIN.read.split("\n").each do |a|
total += 1
sum += a.to_f
end
puts "#{sum} #{total} #{"%.3f" % (sum/total)}"
Finally, we have been using the awesome dashboard service Ducksboard to generate some nice graphs. And this is the result (the time is in miliseconds):

In fact, we took profit of this measuring to introduce a nice improvement in the rendering time and check the impact (as you can see in the decrease of the rendering time of about the 40%).
The improvement was introducing fragment caching implemented with the gem cache_digests. We are still amazed by the fact that we hadn’t have to write any single line to expire the fragments. Take a look to this railscast if you want to learn how it works.
We are still considering new improvements, such as a better dealing of browser cache and finding a way to serve faster the fonts from Typekit.
But back to the post main topic, keep in mind that:
Linode anunció el pasado día nueva la última de sus tres actualizaciones en su infraestructura con las que pretenden competir con servicios como Digital Ocean y sus discos SSD a 5 dólares.
Primero, fue una actualización en la red, luego doblaron el número de cores, y por último han doblado la RAM de todos los planes. Si entras en tu panel y ves esta imagen, estás de enhorabuena:

Yo siempre he sido bastante fan de Linode, por la calidad de sus VPS, su servicio técnico y sus precios económicos. De hecho, en Toldo tenemos el servidor principal funcionando en Linode desde hace más de un año sin ningún problema. En breve reiniciaremos el servidor para volver con el doble de RAM por la cara.
People who struggle to work remotely often bemoan the lack of in-person collaboration jumping from this tool to that tech in an effort to recreate the magic that only happens when we’re all in the same room. There are definitely advantages to face time, but too often it seems like facial expressions and waving arms are substituted for clear thought and courtesy.
The next time you have a question for a coworker, try writing it out as if they were 1000 miles and 3 time zones away – even if they’re sitting right next to you. You might surprise yourself with the answer.