Comment ils ont eu le code d'Etherum? C'est open source?

Yes. The correct way to implement this kind of distribution model is to pick a block number (eg. 290000), and say that bitcoins at that specific block are what gives you aethereum. Otherwise, you can claim some AETH with your BTC, deposit and withdraw from an exchange, claim more AETH, and repeat ad infinitum. For the system to be trust-free and secure, it would have to be based on private key signing, which means that you would have to have them in a personal wallet.
If you don't want to set up a new wallet and deal with different addresses another somewhat technical alternative is to use my pybtctool toolkit to generate a brainwallet and put your money all in the one address just before the nucleus happens. Pybitcointools/pybtctool includes Electrum message signing functionality with the "ecdsa_sign" command; here's an example session:
You can independently verify that 04334... is the public key corresponding to sha256(cow). Then you can easily sign a transaction to move funds back out. It's literally as easy as a command line utility possibly can be. But I personally would recommend for this group to do something similar to what we're doing and release an application which people can send their BTC to (it's client side so no security risk involved) such that the application signs at the appropriate time and then sends the BTC back out. Having easy-to-use tools for doing this kind of signing would really do a lot to help Bitcoin-based issuance models gain wider adoption.

Agreed. The only truly "fair" coin is one that gives X per person per year forever. And, before you ask, yes it is one of our major interests to get at least a few of those launched as sub-currencies on ethereum, and you're free to create forks of each one on aethereum which are distributed via proof of ownership of the corresponding ethereum sub-currency at block N. With Merkle Patricia state trees, we even make "proof of subcurrency ownership at block N" trivial for you.

C'est quoi un boobstrap?

Spin-offs: bootstrap your alt-coin with a bitcoin-blockchain-based initial coin distribution

There are alt-coin lovers, alt-coin haters, alt-coin pumpers and alt-coin players.  At this early stage in the development of cryptocurrency, our community hasn’t come to a clear consensus on these coins.  Many people claim that alt-coins are important for conducting experiments that are too risky to do with bitcoin, while others say the primary purpose of alt-coins is to transfer bitcoins into the pockets of alt-coin developers.  

Recently Daniel K published a compelling article describing the coming demise of alt-coins based on bitcoin’s superior liquidity, market capitalization, and user base [1].  Although I agree with all points Daniel made, I believe he failed to recognize the importance of the bitcoin blockchain itself.  Over 5 years, 3 crashes, and $600,000,000 of unrecoverable mining costs, bitcoins have diffused across our user base and the resulting distribution has been logged to the blockchain.  By tempting you with unimaginable wealth during a rally and then threatening to take it all away during a crash, the free market has become a highly-efficient computer continuously refining the answer to the question “what is the most efficient distribution of coins in a cryptocurrency?”  

Through discussion in Cypherdoc’s “Gold down. Bitcoin UP” thread, we came up with an interesting method of solving the problem of how to initially distribute coins in a project like NxT or Ripple.  Since the market has already encoded its best estimate of the “most efficient distribution” into the unforgeable global ledger known as the blockchain, why not use this?  Since all bitcoin users can cryptographically prove ownership of their share of bitcoins, the code-base of any alt-coin can be modified in a trivial way to allow bitcoin users to claim a share of any pre-mine in direct proportion to the percentage of bitcoin’s market cap they control.  This method (a) bootstraps alt-coins with a very large potential user base, (b) places all alt-coins on equal footings thereby allowing them to compete on their own merits, (c) automatically piggybacks bitcoin investors in any financial gains that may result due to community innovation, (d) fairly rewards innovative alt-coin developers as they can scoop up spin-offs dumped cheaply on the open market, (e) makes it difficult to “pump and dump.”

Our first case study is a clone of Ethereum called æthereum scheduled to launch shortly after the Ethereum IPO.  æthereum is functionally identical to Ethereum, but rather than purchasing ether with bitcoins, you can freely claim your share of æther by signing an æthereum address with your bitcoin private keys.  This process bootstraps æthereum by giving it a larger potential user base than Ethereum could hope to achieve with an IPO process.  A broad user base is important as recent work has shown that the market cap of a coin is proportional to the square of the number of users [2].  What’s even nicer is that you get your æther for free simply based on the investment in cryptocurrency that you have already made by holding bitcoins (that you can prove to the æthereum network by producing signatures with your private keys to claim your æther).

If you are interested in joining the discussion about æthereum, please take a look at the æthereum RFC thread [3].

References:
[1] http://themisescircle.org/blog/2014/03/14/the-coming-demise-of-the-altcoins/
[2] https://bitcointalk.org/index.php?topic=400235.msg6025866#msg6025866
[3] https://bitcointalk.org/index.php?topic=563925.msg6148297#msg6148297


Useful Resources for Spin-Offs

Simplified claim verification (SCV) proposal
This post describes a method that can be used to allow 100% of the valid unspent outputs (at the time the snapshot was taken) to be claimed.

Snapshot file format specification
This post describes the most up-to-date format for the blockchain snapshot files.

Example Snapshot file
This post contains a diagram that illustrates an example snapshot file.  

Example Merkle Tree
This post illustrates the Merkle tree construction for a given blockchain snapshot.  

Example Merkle branch claim proof
This post illustrates the Merkle branch technique used to prove one's claim.  

UTXO Distribution
This post summarizes the % bitcons by wealth and by claim number that be classified as pay2PubKeyHash (99.86%), pay2ScriptHash (0.14%), native multisig (0.0001%) and raw script (16 claims).  100% of valid outputs can be classified in one of these four bins.