Wednesday, November 26, 2008

If EMail is EFail, What's EFTW?

Twitter-GoogleImage by Nils Geylen via Flickr"The underlying problem is that individual human beings don't scale." - Jeff Atwood

Not only is email inefficient, it leads to lost information. If I send you something by email, but Bob needs that info, how is he supposed to get it? In this day in age, if you can't get it through a Google search bar, it doesn't exist.

Now, I don't want to dump all of my company's information into the big G's index, and their appliances are a little pricey. So what do we do about internal data? Here are a couple of ideas I've come up with recently.

  • Set up Wordpress and post time sensitive notices there.

  • Set up a Mediawiki and use it to document EVERYTHING.

  • When someone asks you a question, reply with the link, not the info.

  • When you get information by email, put it in the wiki and reply with the new link

  • When the volume of information starts to grow, set up Nutch, the open source search engine, or invest in a Google Mini.



This doesn't solve every problem. For instance, you still don't get a private Twitter or FriendFeed, but with a little creativity, I'm sure you can hack something together.

Final thought: Email is an expensive way to transfer information, and your time is to valuable.



Reblog this post [with Zemanta]

Tuesday, November 25, 2008

Follow my Tumblelog

I've started posting some of the random stuff I find online to a Tumlelog. You can find it at http://pfeff.tumblr.com.

The stuff I post there comes mostly from my collection of RSS feeds. Using the bookmarklet, its really to just post an interesting article with a quick thought. I have trouble keeping up with blogging, but RSS + Tumblr makes it really easy to post throughout the day.



Reblog this post [with Zemanta]

Sunday, November 16, 2008

Code Formatting for Blogger

The code in my last post was formatted using Format My Source Code for Blogging.

Accessing the FriendFeed API from Rails

I was hacking around with the FriendFeed API last night, and after quite a bit of frustration I got ActiveResource to download the public and user feeds. The main problem I ran into was that ActiveResource assumes a very specific URL format, and I've never actually encountered an API that actually conforms to it.

This tutorial about ActiveResource and the YouTube API tells you how to manipulate the URL format. I didn't go nearly as far as they did though. To get ActiveResource to do what I needed it to do, I only needed to override one method:

class FriendFeed::User < FriendFeed::FriendFeedApi

class << self
def element_path(id, prefix_options = {}, query_options = nil)
prefix_option, query_options = split_options(prefix_options) if query_options.nil?
"/api/feed/user/#{id}"
end
end

end


This class wraps the user feed api. It makes a request to http://friendfeed.com/api/feed/user/{id} and converts the response to a ruby object when you make a call to FriendFeed::User.find(id).

The parent class, FriendFeed::FriendFeedApi, inherits from ActiveResource::Base and takes care of setting some defaults used across each of the API categories. For example, it sets the site variable on ActiveResource::Base to "http://friendfeed.com" and the format variable to :json.