Planet IITR Update

So, I was just going through my old blog posts, and saw the Planet IITR Update, which I created out of a need for people to be able to find blogs from other people in IITR.

Since no one has ever submitted a single link to the planet , I just thought, why shouldn’t I just crawl all my Facebook friends from IIT-R, and check their website URLs. The Facebook part took ~20 minutes (getting list of users from my 2 friendlists, followed by getting the website url for each of those friends). After that came the link checking part, which tool ~15-20 minutes as well. A list of all the websites (very few people fill up that that field on fb) I found during the search is here (just 39).

After updating the planet, I had to update the spreadsheet as well. And here’s a list of the blogs, just in case.

In case someone is interested in taking over maintenance (meaning link curation) for Planet, please contact me at, and I’ll be glad to share the authentication details with you.

Link to Planet : Planet IIT-R

In case you were wondering what this planet stuff is all about, it allows one to add a single curated feed to one’s feed reader and get all updates via that. So planet iitr is a curated collection of blogs pertaining to IIT-R.

Why you should learn HTTP?

I see people learning RoR, PHP, Django, with a single intent: getting their own website. Of course, it is the million dollar idea that will blow everyone away, as always. But what I find fascinating is that too many upcoming web developers are testing the waters with opinionated coding frameworks. The entry barrier for software development has been lowered exponentially in the last decade, leading to a slew of web frameworks, tutorials, and screencasts. Today is arguably the best time to be involved in software development. Lots of people are learning to code their first web-site with rails, or django. There are lots of benifits with this approach: as a beginner, you are kept isolate from all the complexities, and can focus more easily on your application.

But, it also leads to shallow learning. You could have written a dozen sinatra apps, and still not understand how it all works. And as it stands, it is not essential to learn it. You can easily develop entire websites thinking only in terms of urls, hyperlinks, routes and controllers. This is all good for starting up, when you don’t wanna deal with the complexity of it all, but I’d expect any competent web-developer to understand HTTP.

You see, HTTP is the foundation for all of web. It is how the internet tubes work. Learning HTTP is uncovering the hidden layer behind your browser. It is understanding how cookies, and sessions work in PHP; how xsrf attacks happen and mitigating against them; the magic that rails does when it creates objects from the submitted form parameters transparently for you. And the best part is that its not all that difficult to learn at all.

There was a lot of debate concerning REST recently. I don’t claim to understand REST fully. I’m yet to meet someone who does. But I can comfortably build RESTish APIs, and consume them with ease without breaking a sweat. And smile at the fact that its all just HTTP. You cannot move to REST, HATEOAS unless you are comfortable with HTTP.

So, if you are a beginner in web-development, here’s my advice to you: Understand HTTP. A few pointers:

  • Read a good book on HTTP.
  • Read the HTTP RFC.
  • Wikibooks and Wikipedia entries on HTTP are quite good.
  • Use the network tab in Webkit Inspector/Firebug. And understand each of the damn headers.
  • Start using curl -i, if you don’t already
  • Above all, be curious

Question the web.

Shift to bundler 1.1 (Ruby)

In case someone out there is still stuck with bundler 1.0, and hates seeing the Fetching source index for screen, please update to bundler 1.1.

The following command should do the trick:

gem install bundler --pre

Bundler 1.1 is faster by a huge margin in comparision to 1.0.