Feb

6

We’re looking to add to our team again. If you are passionate about Ruby and Rails, come join me and the team:

http://jobs.barefootproximity.com/2010/01/27/senior-rails-developer/

Dec

17

Rails Edge introduces Rails Metal, based on Rack. Think of it as a way to bypass the Rails framework on certain requests. DHH gives the example of the Campfire chat application. I’m thinking about Flash Remoting (AMF). No reason to really call a controller, walk through the Rails stack, just to have separate logic query the database and return a serialized object. This could make Remoting through Ruby really really fast.

Riding Rails: Introducing Rails Metal.

Nov

25

_why explains the recent negative comments on Ruby and the community.


Katz Got Your Tongue? » Stop Watching Sophie’s Choice And Get Some Work Done.

Nov

12

I haven’t used this yet, but it has been on my radar. Relic provides a Rails plugin that simply monitors the performance of your application.

New Relic Raises $6 Million For Rails Performance Monitoring .

Nov

11

GIANT ROBOTS SMASHING INTO OTHER GIANT ROBOTS.

Sep

26

We are using solr (and acts_as_solr) as the search engine on the backend of a Rails site we recently built. While writing the functional tests I discovered TestUnit needed extended for solr functions to be included and executed. I found an example of the extension, but it had some errors (the syntax wasn’t valid), so I fixed it. Here it is:

gist

Remember to include test case info in your solr.yml and start solr in the test Rails environment (rake solr:start RAILS_ENV=test).

Jul

25

I’ve been using Git and Capistrano to manage and deploy sites as of recent. One of the main reason’s for using a repo is to allow new developers to grab all the source needed for a project. That of course can mean Flash source files (FLAs). Git can’t do binary changes, but can still keep track of FLA versions. This is kind of nice. But Flash files can be rather large in file size. Which makes the repo bigger and hence cap deployments can take a long time. Tarring the whole repo up is inefficient. Capistrano offers three ways of dealing with this. One, don’t put your binary sources (FLA, PSD, etc…) in the repo and use the remote_cache deploy option to make deployments only use diffs and be really really speedy. Awesome. Option two, put the binaries in the repo and deploy using a local cache option that excludes certain directories. So something like:

set :deploy_via, :copy

set :copy_cache, "/tmp/caches/#{application}"

set :copy_exclude, ["flash_work", "art_work"]

This works, but still bogs down deployments. I’m not sure if cap is really excluding the directories before it creates the tarball, I need to dig threw the source and see. The third option, is to deploy the entire repo, binaries and all, and then add a after_update hook and remove the directories. Again, you are still tarring everything, which is slow, and then SFTPing everything, which is slow.

In the end, remote_cache would be absolutely awesome if it could exclude directories, which is a whole ‘nother Git issue. Perhaps Jamis has something planned for remote_cache and excludes. I’d be happy to test it out.

Jan

28

we’re hiring a Rails developer, come join our team:

Location: Cincinnati, OH
URL: http://www.thinkbarefoot.com

Description

Barefoot has an immediate opening for a Senior Web Application Developer. We’d like you to have 5 years experience developing database-driven, dynamic web sites. We’re strong in a number of technologies, the primary ones being Ruby on Rails, PHP and .NET. You need to be very strong in Rails, and if you’ve got substantial experience in a second language, bonus points. You’ll need to be able to point us to multiple Rails sites in production. Strong SQL application development experience is also a must. You’ve also got to be at least comfortable with client-side Web development (JavaScript, CSS, XHTML), although that won’t be the main part of your job. You’ll receive excellent benefits and a salary commensurate with your experience.

To apply
Send your resume, and URLs to recruiting@thinkbarefoot.com

Posted Here: http://jobs.37signals.com/jobs/2744

Aug

19

Under the guidance of my superior, Sean ‘bash script’ Brown, I set out this afternoon to get Rails and PHP 5 running under Apache 2 on my Fedora 4 development server. I already had Apache with PHP5 running, so I jumped in during the fcgi download step. One of the only changes I had to make was changing the top path to reflect where I already had Apache installed. Only other change was, I had to add a LoadModule call for the FastCGI module in my Apache config file. As so:

LoadModule fcgid_module modules/mod_fcgid.so

I also had to change some file permissions that were not owned by Apache. Thanks to this post on Rails Forum.

Lastly, I wanted my Rails apps running on port 81, so I had to edit Apache config to listen to port 81, and set up the Rails app virtual host to use 81 as well.

All worked easily and I now feel my development server is complete. Let the building begin!

P.S. Listening to Thom Yorke’s album ‘The Eraser’ while doing so, totally set the mood.

Jul

25

I played with Tracks some more last night, and discovered there are actually instructions included for using Rake to do the setup schema. I was unable to use the migrate command, however db_schema_import worked for me. Looks like my hacking of the SQL script to setup the db schema was 99% accurate. However, I would suggest everyone use the instructions included and try and use Rake.

I’ve found a couple of things, that maybe I’ll jump into the code and change, like the login controller should check to see if there are any existing users and if not, forward to the signup controller for first time set up. Also, when adding an action while on page ‘home’ should refresh the page (eventually use Ajax) to refresh the list of actions.

Also, my feed is invalid. I need to look into that. But overall I am going to stick with this solution for GTD. Looks promising, this is a great start.

keep looking »

Linkroll

Recent Projects