[vga15]

I've been at this exact same place. The best way you can prep for a 'web startup' is to leverage the skills you've got (javascript, java, html), and take action RIGHT NOW. [I've a feeling you know enough to get hacking on an idea, and learn along the way.]

We've no clue what the world would look like 2 years down the line. One thing we'd be certain of -- insane amounts of competition. The most important thing you'll learn right now is 'STARTING' with limited information. If you can couple that with an insane drive to 'COMPLETE', you'd be prepped & ready.

---------------

Since you mentioned 'web apps',

Get into action, the 'lean startup' way: [http://www.amazon.com/Lean-Startup-Entrepreneurs-Continuous-...]

- list a few target markets/niches you'd like to work with

- research them online/offline (forums, meetup groups etc.) & dig out their pain points

- build landing pages (unbounce.com) around your MAGIC SOLUTION to their problem

- get enough folks to sign up for your landing page (might cost you about 100$ worth of adspend)

- the inspiration you get from having actual folks that need your Magic Pain Solver would drive you to complete your product

- you'll relieve them of their pains by forcing yourself to build this ASAP, or you'd hire someone

- you'll have learned a ton about the essence of entrepreneurship through this experience

---------------

Basically, you're connecting a pain-point for an small-enough group of people, that communicates amongst themselves regularly (the market), to a transaction (business model).

And please, start Right Away.

[pge]

This is great advice, but if it's too daunting right now to think about a market, just build something you would use. The experience of building anything is the best start. Then, as you start to get ideas on what product might ultimately be a business, you'll have a better sense of what is possible and how one might go about executing on it. Most of all, as others have said - start building something.

[keiferski]

Thanks for the response. I actually have a ton of ideas and have done market validation. The problem I have is of implementation.

[vga15]

That's great. You're much more of an entrepreneur than I'd assumed.

Check out chapter 6 from 'Getting Real', by the folks at 37signals. [http://gettingreal.37signals.com/toc.php]

The problem I had starting out, was following an exact process for going from an idea to a Minimum Viable Product. Most importantly, I had trouble getting past my designs (PSD) being messy.

-----------

Since you have an idea already:

- figure out what features your minimum viable product should contain

- throw out a whole bunch of those features, till there's just a few main ones, and you feel a little uneasy

- turn those features into screens (paper mockups). as few as possible.

- go from paper mockups to PSD's (photoshop). use dribble, themeforest, forrst, etc. for inspiration. especially the 'minimalist' designs

- or you could just use one of the 'admin themes' from themeforest. or 'bootstrap' by twitter.

- chop the PSD to css/html. This is where it becomes (sorta) 'REAL'.

- since you know basic java, I'd recommend you use php (almost similar syntax, and tons of help online) to hack together your first app

//---EDIT---

I see you're interested in Rails. Personally, with history in c, c++, dotnet & java, rails just seemed a little 'off'. Since you're familiar with java, I'd highly recommend GRAILS: http://grails.org/ as a viable/quicker alternative to rails.

//---END EDIT---

- don't stress about languages and platforms just yet. I'd spent weeks trying to figure out which framework would make me sound more technically proficient to a third party(investor perhaps?). The right answer always is, whatever gets the job done ASAP.

- consider using facebook/twitter for user account management. makes a lot of stuff easier

----------

One thing I'd fix going back, would be rewiring my process to having the PSD chopped first(and played with), before writing any server-side code.