It is follow-up of https://bitcointalk.org/index.php?topic=194078.40 (I can't post reply on that topic).

Proof-of-Work for maintaining "distributed consistency" in DECENTRALIZED PEER-TO-PEER network is the most innovative part of original Bitcoin paper.
Block-chain with bigger combined work amount is the winner, period. No single person or group of people should make decisions which branch is legal and which is not - it is against fundamental principles of Bitcoin.
If there is centralized regulation - then do not call it DECENTRALIZED.

Current version of Bitcoin has list of embedded checkpoints - https://github.com/bitcoin/bitcoin/blob/master/src/checkpoints.cpp
As stated in follow-up thread one of purposes of such checkpoints is to help new client initialization to reject potential spam from malefactor.
Same purpose can be achieved via hard-coded MAIN_CHAIN_MIN_POW as TierNolan proposed ( https://bitcointalk.org/index.php?topic=194078.msg3008155#msg3008155 ) - and this approach does not ruins whole Bitcoin foundation as checkpoints.

You can use different Bitcoin client, so far Bitcoin-qt is fair and dont misuse the checkpoints

Wow, great piece of information. I have never known about it

How do you know that there is no other block-chain with bigger work amount?
Perhaps fork started right from genesis block, and was mined all the time at private cluster at super high-speeds (now on cluster of asics).

How we would select right chain? Well, Bitcoin is decentralized. And no small group of people should decide which branch is correct and which is not.
Checkpoints hardcoded in code were selected by small group of people.
But if Bitcoin is still claimed to be decentralized, not controlled by government, small group of people (which can be blackmailed, or forced by thermo-rectal cryptanalysis) - then only majority of users should have right to make decision.

Users should vote on which branch is correct and which is not. Bitcoin already has such voting mechanism - vote by computing power.


Of course hidden fork crunched in shadow is unlikely, and maybe checkpoints do not harm due low probability of such fork.
But they harm in other way - they compromise Bitcoin decentralization, which is in my opinion is the biggest attractive feature of Bitcoin.
Today they add checkpoints, tomorrow they could freeze BTC on your address, by adding small tweaks to validation routine (just like checkpoints validation).

Very important observation and issue. I agree that Bitcoin as an algorithm is much more integrated with the developers, the community, etc. than most want to believe. So a lot of trust is required for some very few individuals to run the network. There is very little awareness of these issues. The comment in the sourcecode reads

    // What makes a good checkpoint block?
    // + Is surrounded by blocks with reasonable timestamps
    // (no blocks before with a timestamp after, none after with
    // timestamp before)
    // + Contains no strange transactions


The vote by computing power / economic majority is something very different from consensus of users and consensus of developers.  

In the end you are always free to improve the ecosystem (algorithm + mechanisms). Indeed here lie some major weakness in terms of the pure P2P aspect. "contains no strange transactions" is not exactly a principle one can rely on.

Yes - this is nature of Bitcoin Proof-by-Work, decision of majority is determined by computing power. It has both advantages and disadvantages.
If there is stronger power than computation capacity (checkpoints), which belongs to small group of people - then Bitcoin is centralized currency. You should either accept centralization (and then do not call Bitcoin decentralized) or vote by computing power.

Moreover, if somebody has such computer power he could ruin bitcoin in other ways even with checkpoints. Having such computing power it is possible to reject all new transactions, and produce only empty blocks - checkpoints do not protect from this.
Checkpoints help only in very small number of rare cases, but same purpose can be achieved by other approaches ( https://bitcointalk.org/index.php?topic=194078.msg3008155#msg3008155 ).

You should discuss this topic in Development & Technical Discussion section only. In Newbie section this duscussion is useless, no dev reading here I bet.

Well, it is quite likely than these issues have been raised before and thought of by smart people. This is consensus by authority I guess. Improving the system is a very hard problem.

https://bitcointalk.org/index.php?topic=437
https://bitcointalk.org/index.php?topic=1647

You are misunderstanding what this solves. Believe me, if some easy solution would have existed, it would be there already. Essentially a fork before such checkpoints would destroy the network. Although the probability is low, this is a very non-trivial effect. Checkpoints eliminate the possibility. Yes, the pure algorithm in the paper and reality can diverge rather strongly. The downsides of this approach are in a way trivial, as long as enough people ensure the codebase is not corrupted. Centralization is a problem, but it is not at all binary. And yes, Bitcoin as an ecosystem is much more central than one could hope for. But there are very good reasons for that, and changing this is an extremely hard problem.