From Learning Rails to Deploying a SaaS App in 7 Days | Hacker News Reader

From Learning Rails to Deploying a SaaS App in 7 Days

vutran.me

57 pointsvu0tran20 comments

On a previous thread, [I mentioned](https://news.ycombinator.com/item?id=5942320) that at some point I'm thinking about writing Learn Enough Rails To Be Dangerous to go along with the Ruby on Rails Tutorial—but it looks like, by using the Rails Tutorial, you can already become plenty dangerous. ;-)

Seriously, though, this is cool, and every developer-entrepreneur should do something like this just to get the experience of actually shipping a web app.

Ah, so you're the guy I have to thank.

I'll come back to the site eventually and finally get around to learning how to write those unit tests. :)

I'd recommend purchasing the book and/or screencasts as your way of thanking :-)

This is the double edged sword that is Rails. It gives you enough to quickly deploy a basic SaaS app and lull you into the false sense of security (it's so easy to launch an app!).

Rushing through learning RoR and not a) learning Ruby and b) actually taking your time to write some tests goes against what is arguably the best practice for learning RoR properly.

Throwing together one app is a start down a path to learning RoR, it's not actually "learning Ruby on Rails". They are two different things.

I agree. There's also:

c) GemHell

Where you're so captivated in all the shiny Ruby Gems that you don't take the time to learn what they actually do.

Hey Vu,

I'm impressed. Would like some words of advice from you, as I am attempting something similar but less ambitious. Ok to contact you via email?

Yeah! Of course. Just tweet at me and I can direct message you my contact info.

@vu0tran

We've all gotten stuck there. It's too tempting to add easy functionality, until it breaks and you have to debug it.

Hopefully you are going to take a long journey with Rails!

It's just a screw up that RoR got security wrong at the start and had to change their paradigms to fix it. Following a few principals (no in-band signaling), newer frameworks are secure by default and just as simple to learn.

Agreed. That said, for people who aren't necessarily programmers at the start of the process, producing useful things makes it a lot easier to justify the continued learning required to get there.

I want to know why he spent no time listening to Kanye's new album? It's amazing!!

Oh, and seven days? Really? Even with such a thorough explanation I have to wonder how much time people have on their hands to buckle down and do this. This can't be something that's done sporadically. Not in 7 days.

Woot, 2.19s end user browser load time. Significantly better than crashing and burning this time around guys! Go dyno go!

As someone about to launch first big Rails app (I've done small ones before): given the timing of Rails 4, did you consider going Rails 4/Ruby 2 (I'm assuming Rails 3.2, as Hartl's tutorial for 4 was in beta until 4's launch 2 days ago)

The author's twitter account only has 160 some tweets but more than 19000 followers, a lot of which look like fake accounts...building the rails app is a good use of time but it seems it wasn't actually used (who would need to automates 100 tweets)?

So, Django or Rails for someone who wants what you're recommending this for (beginners with light coding background looking to build an app quickly)?

Is there something like that SaaSTemplate for Django?

Rails.

I don't know.

Bonus commentary: Rails is not for the faint of heart. It will suck, but just dive in and see if you can hack it. Then stay humble and keep on pushing yourself.

You can just as easily program something like this with Django. One criticism that you may encounter with Django is that it's not as extensible or have as robust of a community. In terms of plugins, you'll need distinguish between django specific plugins and just general python plugins.

For example, for Twitter - you'll encounter that there isn't necessarily a one-size-fits-all Twitter plugin that's specific to Django. Instead, you might need an authentication plugin which will store relevant Twitter information (id, token) and then use a Python Twitter plugin to do the kind of neat stuff the author has accomplished.

I prefer Django because it's more explicit and there is less magic. This is also good for beginners because it's easier for you to track down where things might be going wrong for you. Django might not have the best community-support (no railscast equivalent).

While this a great story, I'd focus more on what you shipped. I can see it being a valuable service for people.

I don't know if I should ask, but I am curious to know if you made any money with this app!

A single tear...

Reply on news.ycombinator.com