Think a bit more about what it means for Bitcoin to be a distributed computer.

Also, the word "Bitcoin" doesn't only have one meaning.

Treating Bitcoin like an electronic payment system is a very artificial limitation for a distributed computer.

If you can't get Armory working, why would you expect raw transactions to be easier/safer?

You're just exponentially increasing the odds that you'll permanently lose Bitcoins.

Everybody talks about "Satoshi's coins" as if the only one he has were the ones he mined.

Why would that necessarily be the case?

Suppose Satoshi was independently wealthy before he invented Bitcoin. Maybe he'd already had a pretty successful career before he decided to finish the great work of the cypherpunks that nobody had quite managed to get right yet.

If that were the case, he could have purchased the majority of his coins in the nearly days instead of mining them, so that he could spend them freely without everybody freaking out.

If you want to help the network, have one node and make it accept incoming connections via ipv4, ipv6, and Tor.

You're trying to do raw transactions yourself instead of just using Armory.

I clicked on thread expecting a false dichotomy, and I was not dissapointed.

The block withholding attack might actually end up being what limits the size of pools in the end.

If it's no longer safe to accept random members of the public into a pool, then the pool membership needs to be limited to trusted hashers.

This means a larger number of pools for a given number of hashers.

BitFury/CEX/GHash are all the same. They are three branches of the same vertically-integrated operation.

Rather than sell their ASIC chips wholesale, they use CEX to attract retail buyers and point all that hashing power at their own pool.

Austin:

http://www.meetup.com/BitcoinAustin/

btcd is already open source, and Conformal Systems has a fairly large team working on it. Their actual business is Coinvoice - btcd was developed for their own use and they're giving it to the community as a gift.

Armory could work with btcd now, except that it requires direct file access to the blockchain and btcd uses a different file format than bitcoind uses. Alan has talked about making Armory work as a network peer (no file access needed), and if they ever do that you could use Armory with btcd then.

I think we're being played for fools.

Gavin does a good job of playing the "good cop" by telling the community what we want to hear, knowing that very few people will wait around and see if the speeches turn into effective action.

While he was giving a State of Bitcoin speech extolling the need for alternate implementations, Jeff Garzik and Mike Hearn (the bad cops) were busy on GitHub to blocking every effort developers of those alternate implementations were making to try to make what he was talking about a reality.

See:

https://github.com/bitcoin/bitcoin/pull/2900
https://github.com/bitcoin/bitcoin/pull/3982

Wait a minute, I thought he retired from that position? https://github.com/bitcoin/bitcoin/pull/2900#issuecomment-45892726

Let's review:

They are going to be out of a job shortly after btcd finishes getblocktemplate and bloom filter support.

The "Core Developers" won't be able to filibuster any more.

They have CEX, where "investors" who apparently can't do math will buy mining shares for twice their NPV.

The insanity at CEX is a direct result of them not having effective competition because all their potential competitors are legally shut out of the market.

As a result, you get a situation where people will pay 1 BTC to buy a fixed amount of hashing power with a net present value of 0.5 BTC, because apparently nobody buying mining shares at CEX has any fucking idea what they are doing (or if they do get it, they just don't care because they think there will be a greater fool they can sell to).

Oh that's rich.

I can't believe I used to waste so much time over at Tickerforum back in the day.

Maybe some are, but they're not seeing the real issue.

The problem is that many people would like to compete with CEX and ghash but they are hamstrung by the SEC.

It does seem to be the case that building things that work in the real world, not jus in laboratory conditions,  takes time.

False flag trolling.

The dangerous way to get your coins re-mined is to create transactions that send 100% of the inputs to transaction fees, privately send that transaction to a cooperative miner (don't broadcast on the network!) then trust the miner to credit you (minus some commission) in the coinbase transaction.

Downsides are that you have to trust the miner (mining pool operator), and with getblocktemplate you also need to trust all the participants in the pool to not spot your transaction and forward it to another pool who could put it in their block instead. Also, smart blockchain analysis tools could look for that kind of thing and correlate your "clean" coins to the "dirty" coins based on the output size. And the miner might keep records intentionally or unintentionally. Oh, and if that block gets orphaned you could lose your money.

Making this practical means solving each one of these problems, and maybe some more that I didn't think of. Any single flaw in the scheme means you don't actually get any privacy.