Quote from: erik777 on September 24, 2015, 02:53:32 AM

Quote from: VeritasSapere on September 24, 2015, 02:05:09 AM

Quote from: erik777 on September 24, 2015, 01:28:04 AM

It's obvious you'll never come around as you are just here to promote XT. Nevertheless, I'll address this tired underlying assumptions that "people would still need to choose to download and install the new implementation" somehow magically protects Bitcoin for generations to come:

False assumption #1: Everyone who downloads the program completely understands the code.

The vast majority of people do not understand the code, and most are not programmers. Most of us, even those of us who program and can read code, downloaded and ran Core without having any idea about the details of what it does. (e.g., does it log IP addresses?) The reason I take time to post on this forum about the risks of XT is so people can make a more informed decision, and if they decide to install XT, will at least hopefully be wise enough to question each upgrade. Ditto for Core upgrades. Hopefully, discussion like this help broaden awareness of the risks of new code.

False assumption #2: Those who do have a better understanding of the features of the program they download are guaranteed to understand the implications of those features.

The reality is that unless they do understand the implications, they are likely to take the purpose of the code at face value for what its author claims it will do. Undoubtedly, there are some who installed XT who feel protected from DoS attacks because of the patch Mike put in it. Unfortunately, had they read the thread with the Core devs when they rejected it, they would of known that (a) the only claim of a Tor attack was on Gavin's node with no evidence that any other node ever had an attack via Tor, (b) neither Gavin nor Mike provided logs or other evidence of the attack when the Core devs requested it, (c) the Core devs listed many reasons why most DDoS attacks are likely to come from non-Tor sources, (d) even Mike acknowledged that innocent Tor users would be harmed by it and (e) when the attack came from likely non-Tor sources, only Tor would effectively be blocked, not the DoS sources it was supposedly intended to counter. Additionally, (f) this is intended to evolve, and could evolve with enough critical mass combined with XT whitelisting to make it very difficult to unseat XT as the primary code for Bitcoin nodes.

Does the XT website discuss any of this? No. Not only does it ignore all concerns raised by the Core team, Mike has even publicly made statements about the process in which this patch was rejected by Core that very deceptively try to make the Core team look like the bad guys but conveniently leave out the truth of why Core rejected it. Fortunately, the dialog between the Core devs and Mike on his pull request is very public for those interested in reading it.

https://github.com/bitcoin/bitcoin/pull/6364

I am sorry to oversimplify your argument but you are essentially saying that people can not be trusted to make the best decision in terms of what implementation to run. This is however where I believe the fundamental choice and source of the voluntarism of Bitcoin should and does lie. Not governance structures build on top of an implementation or on top of the Bitcoin blockchain, since then we would run into some of the same old problems that large organizations and states run into. The consensus mechanism for resolving such disagreements already exists and it depends on what code people choose to run and where the miners direct their hashing power. Maybe you are correct and the economic majority will not make the best decision, however this freedom of choice is so important and fundamental for maintaining the freedom and decentralized nature of Bitcoin that this would still be the best path to take. There is an irony in accusing XT of dictatorship while simultaneously saying that people should not be trusted with the freedom of choice.

Trust?!? I'm simply pointing out the obvious that applies to ALL OF US -- myself included. Nearly every one of us who uses software runs it without thoroughly understanding what it is doing under the hood, even those of us who have been programming for many years in many languages, and have the best capabilities to break it down -- IF WE HAD ALL THE TIME IN THE WORLD.

Heck, yesterday I just debugged a production system where code I wrote relied on JSON4J, which is open source, because I didn't know about its undocumented behavior of a constructor I was using. Could I of looked at the source beforehand to predict it behave different with different numeric data types? Yes. But, as general practice, we use a lot of open source libraries without ever looking at any of their code because no one has the time to inspect it all or discuss it on a forum.

Now, I do believe there is a good percent of Bitcoin users, particularly today in its early stages, who are largely informed, in large part, not because they read the code, but because they spent time on this forum and read up on how it works elsewhere, truly interested in learning. It was some time after I first ran Core before I really understood what it did and didn't do under the hood, and as a programmer still have more questions than answers, and in large part because of this forum, despite having over a decade of C++ experience since I will never have the time to peer review all the code.

And, while I would like to believe that nearly everyone on this forum understands Bitcoin, and am encouraged by the many people who do, I'm at the same time amazed at how many people defend XT and defend the blacklists based on the claimed intent of the person who submitted the code without even trying to understand the implications of it, both in the present and the future (Mike clearly states he plans to evolve it.) Yet, many who defend XT haven't even taken the time to read the discussion on Github when Core reviewed his pull request to see what his peers thought -- and, for most of the discussion, they held back quite a bit, trying to be nice about it, and not being too blunt about how incredibly bad of an idea it was. Yet, this incredibly bad idea is in the download from the nicely marketed XT website. While miners appear to be rejecting it for now, the best news of all, I'm shocked that 10% of nodes are actually running it (setting aside the impossible to calculate spoofing on both sides).

My point still stands, that you do not "trust" people with the freedom of choice in the form of alternative implementations of Bitcoin.

To be clear I do not like how the DDOS prevention feature has been implemented, and if you do want to support BIP101 it is better to run the version of XT that only implements BIP101 or just run Core with a BIP101 patch. I am not an unreasonable person, I would support most proposals to increase the blocksize it is just that BIP101 is the only proposal that has actually been implemented now, and as soon as a new proposal is implemented I would most likely support that instead especially if it represents a middle ground between these two extreme positions. Regardless of what you think of XT and what the future holds for BIP101, it has been a catalyst for change, this in itself should be seen as being a positive thing.