Ask HN: What technologies to use for new startup ideas? | Hacker News Reader

Ask HN: What technologies to use for new startup ideas?

4 pointsspectrum12347 comments

I have many good and bad startup ideas. I took a web dev bootcamp but am a much better product manager (The Lean Startup blew my mind). I want to use oDesk to get some prototypes up. I want to think of each startup idea pretty meta, merely each as experiments to better learn the product market fit process of tech startups. Obviously I believe my best ideas are ahead of me, through hard pivots or otherwise. (Yet I also believe the technologies used for both each startup now and a future iteration of fresh ideas need to be the same as much as possible.) Lastly, I am aware marketing is a big factor to measure these experiments that I am leaving out of this right now.

My concern is over best technologies to use. Is it obviously just node + react? Other thoughts are meteor + maybe react.

Here are the factors I believe matter: -freelance dev knowledge of tools -future tech debt -resources (libraries and documentation, and stack overflow questions) -quickness in hacking out new features

Thanks

Just use the tool that you know and that you can deliver with. No need to jump on the bandwagon of the new shiny language/framework/methodology of the week. If you can ship software in php, use it. If you can ship with node + react, use it. Focus on building.

And stay away from oDesk, elance, freelancer.com, etc. If you want good technical help, go with gun.io. I'm not affiliated with them in any way, other than being a freelancer on their platform. You will pay more for sure, but you'll end up with something useful rather than a broken pile of hacks.

Exactly my thoughts. If your project gets to a point where the language you've picked for first release is slowing you down you should also be at a point where you're ready to bring in people smarter than yourself to keep it afloat long term.

When you're ready you can build v2 in the hot new language of the month or the old tried and tested, but for now that's not your concern. Walk before you run, crawl before you walk. There's no rush.

IMHO: If your plan is to outsource these projects, the most important factor is your level of comfort with the code. Outsourcing is tricky, and you might need to be very hands on (for code reviews, bug fixes, project planning, etc). You don't want to be building prototypes that you couldn't maintain/troubleshoot yourself.

I say that all to say: go with what you know.

Sure, the differences between certain platforms and technologies do matter (especially at scale)... But if your only present-concern is getting products out the door, then your comfort-level should trump all other concerns.

Out of all the technologies out there right now I would say Meteor is the fastest to learn to get a prototype out the door quickly. Rails is also always a good option since it's so popular.

If your priority is building a prototype (and not becoming a developer) I would focus on being practical, as opposed to spending months finding and learning the "best" possible technology.

Don't waste your money hiring contractors to build your product, especially off of oDesk.

Personally, what I'd recommend is find someone local (yes it'll cost more), and say that what you want to do is pair program with them with the intention of bringing up your skill and comfort level with programming.

Rust, Go, or any other new technology. Hell, maybe even some React, (Angular.js is old school)

Agree with Meteor, trying out RamNode now too for VPS

Reply on news.ycombinator.com