Make Master Protocol harder to censor

[ripper234]

The 1Exodus marker address is making it easy to people to censor Master Protocol transactions.
This spec issue is a placeholder for discussion on how to upgrade the protocol in a way that will make such censorship harder.

@petertodd, @dexX7, @LOLLOLOOLOL, can you share your thoughts on the matter here?

@CraigSellars FYI

[ripper234]

See also confirmation that Eligious is censoring Master Protocol.

[luke-jr]

Forcing people to mine transactions they don't want to mine is bad. Don't bother trying, because it's impossible to prevent anyway, just makes more work for the good guys who should be able to spend their time improving Bitcoin instead of battling spammers.

P.S. I recommend investing in a good dictionary, since you don't seem to know what "censorship" means or how to spell. :|

[ripper234]

Fixed op for spelling, tnx.
Let's agree to disagree Luke.

[luke-jr]

Are you willing to agree to disagree? Or are you still planning to try to force me to agree with you?

[ripper234]

You don't need to agree with me, just like governments don't need to agree to let Bitcoin operate.

[ripper234]

To put it another way, Bitcoin doesn't need permission from governments to work, and Master Protocol should not need permission from Core Devs or Pool Operators to work.

[luke-jr]

Nor should you try to force core devs or pool operators (I think you really mean miners) to do your bidding against their will.

[ripper234]

As most actual mining is done by pool operators, the two groups are the same.
The so called group of miners, who do contribute hashing power to a pool, don't actually mine Bitcoin, they are just renting out their hash power. They have zero control over protocol decisions, don't validate transactions, and are at the mercy of the small group of pool operators (that is until they decided to leave centralized pools and join P2Pool).

I'm not going to engage on ideological debates with you Luke.

[luke-jr]

So which part of that (not that it's accurate) do you assert justifies you forcing others to do your bidding against their will?

[dexX7]

Nor should you try to force core devs or pool operators (I think you really mean miners) to do your bidding against their will.

Without transparency (e.g. published blacklists), I don't see how this is a choice made by miners.

Nevertheless, what one could call censorship, can be called freedom from the perspective of the pool operator as well.

---

Getting rid of the Exodus marker is the first target, moving away from m-of-n multisig encoding altogether the second, given that m-of-n is almost exclusively used by metacoins. A marker (in form of an Exodus output or an unobfuscated prefix within the data package such as XCP uses) has it's benefits though, namely the ability to identify Mastercoin transactions really fast. This is good in terms of faster validation, but comes at the cost of being exposed to censorship. The question is probably: is there something that does not create additional cost for Mastercoin, but creates work and cost for those who'd like to filter the transactions, but are not a user?

My personal opinion and priority: 1. Get at a point where we don't need to rely on the merit of pool operators at all, e.g. by moving parts off-chain. We are probably far away from this, but it can start with any meta-data that is nice to have, but not strictly needed for the transaction to justify storage within the blockchain - such as lengthy asset descriptions or whatsoever.* 2. Improve the product, get away from being considered as spam, but worth to mine. The initial guesture of slightly increased fees, fees as basis of DEx spam protection etc. was a good idea, but unfortunally not fruitful. 3. If it's not critical, avoid cat and mouse games which come at the cost of both parties.

[luke-jr]

@dexX7 Thanks for your input. Eligius uses my public spam filter branch, and miners are capable of seeing what they are mining at any given moment in full detail using GBT. I think your strategy sounds like a good solution to aim for; please feel free to ping me if there's anything I can do to help out in that area (I can't guarantee anything with Eligius since wizkid057 runs it now, but I can certainly recommend things to him if they make sense).

[petertodd]

@luke-jr "Forcing people to mine transactions they don't want to mine is bad."

Hey, just so you know I'm going to be starting a charity to help low-income pregnant teens pay for their abortions, provide sex-positive sexual education, and also operate a sunday school praising the word of God.

Should I have separate donation addresses for each of those categories so you can blacklist funds going to just the causes you dislike, or is that too much trouble and you'd rather block them all in one go?

[petertodd]

@dexX7

The Exodus marker is only truly useful if you have SPV mastercoin clients; we don't so dropping it for now until there is a secure way of querying blockchain data with fuzzy matching is reasonable. As for the encoding, blockpop does solve this stuff in a solid, forward-compatible, way.

Get at a point where we don't need to rely on the merit of pool operators at all, e.g. by moving parts off-chain.

Metadata sure, but remember that you can't replace the core data with off-chain stuff without a severe reduction in security.

If it's not critical, avoid cat and mouse games which come at the cost of both parties.

Honestly, that Eligius is the only pool blocking stuff is telling you know... The other 95% of miners realise that politicising transaction acceptance makes Bitcoin less valuable for everyone. Equally, this isn't a cat and mouse game, there's a tiny number of moves in the game that make sense, and you can very quickly force miners into adopting heavy-weight blacklists. Heck, that 95% aren't even adopting light-weight blacklists.

[dexX7]

Thanks @luke-jr, I assume that's the one: spamfilter? In general I think it's sad to see that there is a killswitch to block all bare multisig transactions which even made it's way into bitcoin:master while Eligius in particular (and I give you much credit for this) remains to be the only larger pool (that I know of) which supports exotic transactions. I'd rather wish to see more options than the only one which is intended to block "metacoin spam".

Especially when put into relation where the overall footprint on the chain is absolutely insignificant and providing simple tools such as an endpoint to retrieve unspent outputs of m-of-n transactions resulted in a complete remake of Chancecoin which is now basically build around the idea of spending dust instead of creating pollution - to name one example.

The Exodus marker is only truly useful ... As for the encoding, blockpop does solve this stuff in a solid, forward-compatible, way.

I actually see one argument to keep the marker as it is: the available infrastructure build around "addresses". Just think about Electrum - you may right now fire up your local client, fetch all Exodus transaction references within seconds via it's console with the command getaddresshistory("1EXoDusjGwvnjZUyKkxZ4UHEf77z6A5S4P") and reconstruct the current state within a very reasonable amount of time.

Maybe I missed your point or the capabilities of blockpop, but I assume you were talking about obfuscation of transaction data and suggest to drop the Exodus output. Say we drop the Exodus output - how are transactions identified?

Metadata sure, but remember that you can't replace the core data with off-chain stuff without a severe reduction in security.

Yes, that for sure. In this context I was thinking about meta-data like asset name, category, subcategory url, extra data and a memo field. Since not even the name provides any reliable information, but introduces name squatting and scam, I'm not really a fan of right it from the start. Going one step further and moving order publishments/negotiations would probably be acceptable, too. It's really too bad we are not at a point of output bound transactions - I dream about a decentralized digital asset exchange with honest pricing and market depth on steroids.

While I think without a doubt some form of secondary network for those things would be very awesome and should be aimed for, it's questionable, if the gain outweights the cost of establishing such a network - at least from a short term perspective. If we can come up with something that indeed offers actual benefits besides being "blockchain friendly", like faster transactions or whatsoever, that's a different story, but I don't see a light right now.

@luke-jr: I'm really interested in your input, not because I consider Eligius' block as game changing event, but as tip of the iceberg and the need for a solid solution which might offer other benefits has probably consequences far greater than the scope of Mastercoin in general.

Honestly, that Eligius is the only pool blocking stuff is telling you know... The other 95% of miners realise that politicising transaction acceptance makes Bitcoin less valuable for everyone. Equally, this isn't a cat and mouse game ...

I was actually surpirsed as well and expected worse. Nevertheless, isolated this doesn't justify much, but it can't hurt to think one step ahead or to act proactive instead of hoping this remains to hold true - better be safe than sorry.