Hacker News Reader

Comment by olegp | Hacker News Reader

yogsototh thank you for submitting this. We are seeing 150 or so concurrent users on the site due to this at the moment. I'm retweeting some of the other online reactions we're getting at https://twitter.com/akshell_com

View on HN

150? It's only the beginning :)

You were right ;) I'll write a blog post about it once things calm down a bit.

Interesting tweet: "... someone made the old @heroku for node.js" -- did Heroku start out as a browser-based app builder?

Yes they did, they were in the same batch as AppJet as well.

Actually AppJet was summer 2007, Heroku was winter 2008.

Oh, then I've had it wrong for a while now. Thanks for pointing it out.

Yes, Heroku was the first. I tried it out. It was fun, but I think they realised that though it was a novel idea, it was not the _Big Idea_, and they replaced it with what they now have. That's pretty inspiring when you consider how much work they threw away and how that decision has been rewarded.

Maybe the timing was just wrong? The browsers are fast enough now and users, including developers, are much more comfortable with web apps.

I think it was less about speed and more about trying to replace peoples' editor / workflow. People are very particular about the tools they use and putting the editor and deployment in the browser doesn't do much in terms of productivity (and probably more counterproductive than anything else)

Screenshot: http://weblog.rubyonrails.org/2008/2/7/write-rails-apps-in-y...

Yes, I remember speed being an issue, and I wasn't trying to subtract from the usefulness of this creation.

I would say what they threw away was essentially a gimmick. Assuming the functionality of Heroku was the same before and after, publishing from github rather than publishing based on editing files online doesn't change their offering.

Reply on news.ycombinator.com