Star Citizen's feature creep... and other signs of a software project doomed for disaster?

It seems that Star Citizen is suffering from severe "feature creep" (where features keep getting added long after work has started on a project), as they are doing things like:

• Modifying the game engine to be double-precision (64-bit) for much larger game 'maps'.
• Trying to add orbital mechanics (i.e. actually moving planets & moons).
• Trying to add procedural generation of planets.

I first found about this here (skip the first big post, but read the rest of the page).

Apart from this sounding to me like a severe case of "Elite Dangerous envy", feature creep is a classic way for a software project to go way over-budget in time & money, with the end result being a horrible mess, or even simply running out of money before it is ever completed.  Feature creep (and changing project goals) happens to many government IT projects that have tens or hundreds of millions of pounds to spend, and is why so many government IT projects end-up being a disaster.  So having masses of money is not (on it's own) a good predictor of success (or failure).

And as someone pointed out, this also sounds suspiciously like what happened to a certain well-funded game about chewing bubble gum & kicking ass (i.e. Duke Nukem Forever).

Early signs do not seem good, as someone reports that they "have [yet] to produce ... properly working core functionality - even the hangar is a buggy mess".

Also, my impression of Star Citizen is that they're a big brand new team who have mostly never worked together before.  This is not a good way to start a massive project, as demonstrated by Daikatana:
http://en.wikipedia.org/wiki/Daikatana#Development
http://en.wikipedia.org/wiki/Daikatana#Development_hell

In fact Daikatana may have more similarities to Star Citizen than Duke Nukem Forever:

• Massive hype before the project had even started, mostly coming from one famous name.
• Massive project for a newly formed team (which also kept adding more people).
• Even Daikatana's changing of game engine mid-project isn't so far off, as Star Citizen's engine has been heavily modified to be double-precision (not an easy undertaking).
• Missed deadline(s).
• The release of an (IMHO) underwhelming early 'demo'.
• Daikatana was even funded to Star Citizen levels, to the tune of $44 million.

Having said all that, I do hope Star Citizen is (at least moderately) successful, otherwise the whole space game genre might be tarnished & disappear for another decade or two (why would publishers take the risk when another First Person Military Shooter is bound to sell well?).  But if S.C. is successful, then we are bound to see many more space games (even if most of them will be uninspired crap).

(Up front admission: Although I am a space game fan, Star Citizen has never excited me, but I did back Elite Dangerous, and I am crossing my fingers for Infinity Battlescape.)

UPDATE:
At least one Star Citizen backer is (IMHO rightly) getting nervous, as H1itman_Actual says:

The 6 month delay of the dog fighting module along with ambiguous release dates have test a lot of us backers.

If this date doesn't hold I'm really loosing faith.

I'm an original backer. But CR release dates the past 6 months have been broken. He should have never of told backers "2 weeks from now" "1 month from now" " 1 or 2 weeks after PAX"

that kind of leading backer on creates scepticism.