Hackathons Taking Center Stage, The Transformation Of The Computer Scientist | Hacker News Reader

Hackathons Taking Center Stage, The Transformation Of The Computer Scientist

techcrunch.com

4 pointsgailees3 comments

So, part of the annoying thing about this is that "computer science" as billed at any university is a product so diluted in quality and standards that it's a crapshoot to see if somebody with a CS degree is good at anything you need.

Assuming a competent student and good curriculum, CS at a school could mean either theoretical computer science (computability, graphics), computer engineering (OS design, networking, HPC), security, software engineering (how to structure large projects, design and documentation in depth), or even applied math (crypto, numerical methods).

It usually means some blend of the above.

And for all of that, the things that are most financially lucrative these days don't require anything other than the ability to read the fucking manual and bolt together your favorite ruby gems or node packages and then apply a pretty design to it.

Hackathons are a really good way of getting students exposed to the state-of-the-art in our industry and getting them out of their comfort zones in a controlled environment where they can fail freely.

All that said, it's still a bummer to see a full panel of toy mobile apps. Then again, we all had to start somewhere.

  And for all of that, the things that are most financially 
  lucrative these days don't require anything other than the 
  ability to read the fucking manual and bolt together your 
  favorite ruby gems or node packages and then apply a pretty 
  design to it.

This is simply untrue... financially lucrative things almost always have to be highly scalable, very robust, and implement functionality that is not "in the manual". Making a trivial webapp is easy; making one that implements useful functionality and scales to millions of users is not so easy.

Scaling is a problem that happens once you get users, and by that point you can pay real engineers to solve your scaling problems. Scaling is a phantom bugaboo used by engineers to help massage their egos that some punk-nosed kid is kicking their ass at capitalism using Ruby or PHP or Node or Perl with a shitty, inferior product...that happens to have more users and mindshare and potential then that engineer's pet project, oh-so-carefully engineered and constructed.

If you set out to build Twitter or Faceobook or Snapchat or whatever at scale, from the start, you're doomed.

But, cranking out a stupid little Rails app with enough hook and flair to get your initial userbase and attract investors is pretty much just bolting together gems and bootstrap.

Worse is better, the crowds have spoken.

Reply on news.ycombinator.com