Hope you have Haskell soon, I've wanted to do some analysis of occurrence of certain patterns.

Love the idea! I tried it by searching for Node's require. It did point me to the right source [1], but the usage examples are all minified and unreadable. Might want to check for a maximum size on the usage examples that are presented.

[1] https://sourcegraph.com/github.com/joyent/node/.CommonJSPack...

One of the co-founders of Sourcegraph came to UC Berkeley and spoke at entrepreneurship event. He was very engaged and gave lots of real practical advice.

I've tried Sourcegraph and think it solves a real pain point but I have had problems integrating it into my workflow. I'm not sure when I should be using it. I've been meaning to use it to read more open source code code but it seems like they recently disabled support for Ruby.

I've just been able to add my repository but there's nothing happening even though it says it's indexed the code completely. No annotation happening anywhere. Does it support angular.js's development style?

https://sourcegraph.com/github.com/SchizoDuckie/DuckieTV

This is very cool, it is very close to a product I would spend money for. The product I would spend money for is an e-reader for code. I would copy a 'blob' to my tablet (ipad preferred by Android is ok too) which has all of the code in my example (could be small like an application, could be large like the entire 3.13 kernel source) There would be an ever present search bar or hot-gesture that would let me search for function, definition, callers, or callees. Think cscope on steroids for the language of my choice. Browsing around the source code would bring up source in syntax colored glory (either dark motif or light). It would use high resolution fonts on my retina display for crisp clear text. The typography would be optimized for reading code. It would remember my walks through the code, it would be able to recreate them on request, it would have a swipe-able interface to get to a notes pad "behind" the code where I could at notes about what does what. I would have the ability to fold entire functions, clauses, or loops, and unfold them as needed. It could bring up doxygen comments, when I gave a long press on a function name, it could jump between call graph and code mode. In short I could read code of arbitrary complexity while I rode the train home from work, or sat on the tarmac at JFK. I would pay $100 for an App that could do that for me.

This is cool, but I'm not sure what's going on here. Is this full static (i.e. AST) analysis of these languages or simple full-text search, or some middle-ground constraint on full-text search driven by common usage patterns?

For example, "d3.timer" doesn't show meaningful results?

What I love about sourcegraph is that between it and github search/searchcode you can find pretty much every bit of code you need without resorting to Google. Big improvement to my workflow personally as I always have preferred to find an example first and the documentation second.

I must be stupid, but can someone please tell me how I navigate to the list of examples? When I click on search result, I get to the relevant source code. I just can’t seem to find a way to get to the examples.

I'm 2 weeks into building the thing I just watched you demo. Cue sinking feeling.

It looks awesome. Bravo.

I've been using it to study good open source code.

Is it possible to filter search results by language?

Sourcegraph looks beautiful. Good job guys!

Congratulation Beyang and Quinn. Looks really slick!