Quote from: fluffypony on September 27, 2014, 05:55:52 PM

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

If you're not incapable of developing, why you are making random git pulls from BBR and inserting them into XMR without an idea of what the code actually does? Semi-capable might of been a more appropriate word.

I was responding to "slapper" who said: "we all know you are not a developer". There was no "more appropriate word" - he said something, I responded, fin. I have no idea why you're jumping on the bandwagon and misinterpreting my response to him, but hey, it's a free world;)

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

Actually, How did that gaffe slip past extensive peer-review by a team of no less than seventeen experienced XMR developers?
Unless it wasn't actually reviewed and you have unfettered access to the repo.

The decision was indeed taken by the core team to merge that. As you say, it was a long day, and we were fried.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

Why did it take a developer of a competing project to point out your mistake so you can revert it?

It didn't at all. His post on Bitcointalk and the revert are not related. It was caught on #monero-dev (as one would expect when something is pushed to staging and causes an issue) -

[2014-09-24T23:02:33+0200] 2014-Sep-25 00:01:56.528495 [P2P8]coinbase transaction spend too much money (14.336009159350). Block reward is 14.332489999244(14.130220173955+0.202269825289) [2014-09-24T23:02:36+0200] 2014-Sep-25 00:01:56.528601 [P2P8]Block with id: <450775fe7c413c32a6869eac58b074c4b93ae7597d0f11b071a88fc77465ca5e> has incorrect miner transaction [2014-09-24T23:07:52+0200] <@fluffypony> what distro, otila ? [2014-09-24T23:08:21+0200] I am lagged.. fedora 20 [2014-09-24T23:10:16+0200] it was maybe waiting tcp/53 [2014-09-24T23:11:11+0200] <@fluffypony> that's weird, it should fall back hardcoded seed nodes if it can't get dns seeds [2014-09-24T23:11:17+0200] <@fluffypony> good use case to test for, though [2014-09-24T23:12:25+0200] at least it does not notice if tcp/53 is connection refused [2014-09-24T23:16:39+0200] stuck in do_handshake_with_peer when I exit, I get only "[node] Stop signal sent" [2014-09-24T23:19:59+0200] <@fluffypony> welp [2014-09-24T23:20:02+0200] <@fluffypony> I see it otila [2014-09-24T23:20:32+0200] <@fluffypony> guess I'm reverting 014708fe71c1379af281ca9ac17e82c159e98e6d

Caught and reverted within 30 minutes of being pushed to staging. It's a mistake, it happens, move on.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

However it's a point made by zoidberg that if the XMR developers don't fully understand the commits they are implementing and just add features willy nilly from other project some damage could of been done. Malicous backdoor could of been inserted into the XMR this way.

We've only had the codebase for a few months. A slocount of just the src folder (excluding tests and epee) is 16310 lines of just cpp code. That's over and above the cryptography. It's a lot to grok and come to grips with. Nonetheless, we do not merge commits "willy nilly". This was done in a rush because we *do* respect CZ (which is apparently not mutual), and if he says that there's a major CN bug and is urgently contacting exchanges we are going to see the commit (which appeared to fix a varint overflow issue) and assume that's what he's talking about. You cannot take this single incident in isolation (given the circumstances) and extrapolate out from there.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

I see you making not so subtle insinuations that crypto_zoidberg is a glorified code monkey, If it wasn't for CZ there could potentially have a mtgox like situation on our hands with poloniex, you should be more respectful about background works. He has done some hand-holding on a number of occasions and you know it well. I actually checked out your git commits. Majority of them are version bumps, string replaces & single line additions. seems if zoidberg role is code monkey your role is secretarial duties- you are assigned to brush up on the verbiage and misc typos, But I suppose you prefer the role 'architectural visionary' or 'creative director' and consider pushing out code something mostly below you, but something you do to keep you occupied whilst you are having a morning croissant or on the shitter maybe?

I'm pretty sure the majority of my more recent commits are CMake related, so I have no idea how you reached that conclusion.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

Someones been practicing their best Jony Ive impression Grin

. I agree with what you said about code volume being a useless metric. Anyone can hire 10 freelancers from bangalore. The only reason it was added to infographic was I heard someone ask if the project was dead due to lack of activity?? So the clear numbers are shown to demonstrate neither projects are stagnating, clearly the fact BBR has less cooks in the kitchen doesn't mean the broth is being cooked slowly, nor is the quality suffering.

That person's obviously an idiot - I don't think BBR is stagnant on any level. I understand why you included it - my response wasn't to you or to the information in the infographic, it was to "slapper" bordering on hero worship because he thinks LOC is an indicator of anything.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

True innovation starts with: Taking the time to develop an intimate understanding of the codebase you are working with, so you are unrestricted to expand freely without worry.

When have we had the opportunity to do so? We didn't decide to fork Monero, thankful_for_today did. We ended up with it because thankful_for_today wanted to do stuff the community rejected, and we sort-of fell into it. Had we come together as a group and said "oh hey guys, we should create our own CryptoNote coin" I can fully agree with what you're saying. But that isn't the case and that isn't the situation, so we're incrementally refactoring completely undocumented code, building up an understanding of the cryptography (see the Monero Research Lab publications, for instance), and still adding functionality.

Quote from: dvdrewritable on September 27, 2014, 04:36:24 PM

Example: auditing the Proof-of-work so it's not blatantly crippled upon launch. A third party good samaritan had to step in to fix Monero's initial borked code, Hardly inspired confidence in the 'real' team. He had this to say:

Some valid points, fair enough

Quote from: fluffypony on September 27, 2014, 05:55:52 PM

The good samaritan that is a member of the Monero core team? I even mentioned he's a member of the Monero core team in my previous post - you must've glossed over that:) In other words, thankful_for_today gave us the blatantly crippled code. When we (the core team, although very much in its infancy and only loosely affiliated at that stage) realised that we fixed it.

Didn't completely gloss over it

It was the good samaritan who wrote this:

Quote from: NoodleDoodle on May 07, 2014, 05:53:50 PM

Like I stated in IRC, I am not part of the "dev team", I never was.

IIRC, it was only later he was employed/affiliated with the core team

Quote from: fluffypony on September 27, 2014, 05:55:52 PM

On the contrary, BBR's aliasing has a host of problems (such as: what if you lose you private key? how can you reclaim your alias, or are you expected to now tell everyone it's changed? and once there are a hundred variants of "Bob", I still need to have an address book that tells me that Bob Simons' alias is bob1982_waffle and Bob John's alias is bob-the-builder...how is that better than just having their actual addresses in my address book? and in both of these events, how is it *any* better than just having the address?) Instead of going with something that had been done, we developed an aliasing system that *every* cryptocurrency can use, including BBR and Bitcoin, that does not have the problems described. Perhaps you'll want to familiarise yourself with it first before assuming we've just "added in aliases" - https://openalias.org

If you lose your private key (seed) to your Monero wallet how are you supposed to spend your funds? or are you expected to tell everyone not to send you more money to that address? - exactly the same argument, So losing your keys seems a little straw grabby. The nature of decentralized protocol like this is there is no customer services to call up to issue a password reset. A simple backup will take care of that

There may be some variations of bob, bob-the-builder and bob21912. It's exactly the same on twitter- in fact any other service which allows users to choose their own names. The same 'problem' with your DNS aliases. if you propose users will be so confused between @bob-the-builder and @bob1982_waffle why wouldn't they be confused between bob-the-builder.com and bob1982_waffle.com?. caveat utilitor as always.

Additionally BBR sensibly limits allowable characters in their aliases. See if you can spot the difference between these two:

Code:

[wallet 48NGTd]: transfer 0 donate.monero.cc 10
For URL: donate.monero.cc DNSSEC validation FAILED!

Code:

[wallet 48NGTd]: transfer 0 donate.monero.cc 10
For URL: donate.monero.cc DNSSEC validation FAILED!

I followed your hangout where you discussed that. It's pretty neat but comes with it's own set of concerns. eg Malicious resolvers, spoofing/poisoning attacks- -Most resolvers aren't using DNSCrypt, setting up alias requires registering/managing a domain and keeping it renewed so it's not decentralised or optimal from a privacy standpoint arguably, although it's at least cool if you imagine it will take off and you escape worrying about name disputes.