If you need to ask that question, you're not prepared to carry the burden of executing such a project to a successful completion.
I've no exact numbers, but you can expect it to take several months to years, a team of maybe dozens of people, and a budget of millions of Euros.
And that's just to build it. The marketing, initial concept idea and artwork, etc. etc. add more time and a lot more money.
And of course running it costs money and resources as well.
Also, you'll need to learn a few languages here:
1) HTML (Most definitely)
4) MySQL (or another good database)
Also, you'll need to read up on servers and apache, find what to use, etc etc.
You will also need to know some jquery and or ajax to create a more fluid functionality in the game. If you are new to programming you should try a smaller easyer game with less functions just to get your feet wet
A lot of realistic advice has been given already. Though, I'd like to encourage you to try and develop something. If developing a cool online game is what fires you up, go for it even if you won't be able to complete it. Just making a serious attempt, will help you with learning a lot of valuable skills like programming, web design, user interaction. All of it is very marketable, e.g. can be translated to good money down the road.
One good way to start is to research if there is an open source version of an online game. Get involved with that, look at the code already written, if at first you can't contribute the new code, make yourself helpful by volunteering as a beta tester. You'll be appreciated by the community and learn along the way.
I'd suggest, as others have, in not attempting to jump into a huge project as your first. We all tend to have general illusions of grandeur when deciding what to attempt to create and indeed, many people would be capable of creating such a task in its individual components, if only time wasn't such a burden. The problem is that it'd take a single person years to create, by which point, the technology will have moved on and better versions of the same idea will be in practice, thus rendering the idea a mere lesson learnt.
I however, would not say to scrap the idea at all; all you need is some careful planning and you'll be able to create something with great personal wealth to yourself.
If you plan iteratively so that individual components can be created in release worthy quality, then molded together as and when each component is done (not simultaneously making a lot of parts, complete one, move on...etc), you'll avoid the pitfalls of a part completed project destined to live its life in an idle folder somewhere on your desktop. Instead, you'll have faster, if smaller, portions of your project complete and more importantly, many opportunities to stop when you see fit, without having the heavy feeling of guilt when abandoning a project which you've poured a lot of love into.
This is quite a common issue with programming; we all realise that we COULD create a massive game which would surpass all previous attempts at breaking a genres limits but as a sole developer, you have to be smart about it. By all means, put all your eggs into one basket; just be sure to extensively plan exactly where each one is going to go, how each will reinforce the other, that the basket will be amply 'filled' should you stop moving your eggs at any one time and be extremely precise as you move each individual one in turn... crap analogy but I'm sure you get the point. ;)
Any way you look at it though, years will be an accurate timeframe for a new technology...but then, that's half the fun. :)