So on Saturday I took part in matchnhack's hackathon - they also have this startup speed dating session, which I'm not sure works all that well, but a hackathon is always fun time.
Anyway, the big thing I discovered on the hackathon was Firebase - a pure-Javascript in-cloud schema-less NoSQL database for agile AJAX-based Web 2.0 development - if you can say this stream of buzzwords with a straight face that is.
It is actually really good for throwing something together real fast. It cuts just about all the bullshit - there's Javascript database in Cloud, a library which establishes connection with this database and streams update events to your event handlers in your app for both initial setup and any changes, and you use jQuery (or whatever you want) to do all the user interface. There is no server-side involved. Synchronization between clients on different browsers happens automagically.
They have a build a sample app in 5 minutes interactive tutorial, which is just as impressive as Rails screencasts were six years ago, and if you do any kind of web development, for work or for fun, you should definitely spend that 5 minutes on it.
Now contrary to what they say this is currently entirely unsuitable for any production use, they provide zero security whatsoever (open Firebug console and you can delete the database), and their claims to scalability should be treated very skeptically, but for throwing something together fast it's truly amazing.
We made an chat app for complaining about London 2012 Olympics, which currently lives here and for some mysterious reason people still use it even after hackathon ended. It's not a complicated app, but just look how trivial the code is compared with any server-based solution even with a decent framework like Rails.
This app I made with @woodrat84, venkatesh, and GiftApp team even won hackathon's top prize, probably more for hilarity of the chats than for any technical achievement.
Other than Firebase, the second thing I'd like you to see is Style Nibble - it's a startup targeted for women which tries to do fashion recommendation by some combination of machine learning and in-house experts. I'm a huge fan of recommendation engines of all kinds and forms, and I'm really delighted someone is taking them in this direction.
They even have clear path towards monetization from day one by actually selling stuff, and it's high margin market, and one which so far Amazon and the rest of market leaders have been unsuccessful at. I fully expect something like this (buying clothes based on recommendation engines) to get really huge in a few years. Maybe it will be one of the startups, maybe it will be Amazon after all after who knows which attempt (possibly by buying one of these startups), but there's no way in hell something like this can stay unexploited for long.
No comments:
Post a Comment