Quote from: bitcrab on January 06, 2017, 02:05:39 pm

today a man made margin call happened again for bitCNY, about 50K bitCNY margin called, someone did price manipulation just by selling a little quantity of BTS in yunbi.

proxy bitcrab decided to unvote below witnesses, until they implement professional enough bitCNY price feeding.
spectral delegate-clayop bhuz bue delegate.ihashfury spartako fox elmato rnglab datasecuritynode

don't tell me btc38 always fail to provide price, don't ask me how to generate good price feed, you are witnesses, you should find ways to make that done.

Note - datasecuritynode wasn't being voted for by your proxy before anyway.

I would suggest we create a worker proposal to gain access to more price feed data rather than continuing to rely on bottom of the barrel free data. Vote for it?

I  don't understand why worker proposal are needed here.
if some witnesses have proved that they can provide professional price feed for free(no more than the block generation reward), then why should I vote for the worker propossal?

你的逻辑很奇怪，参与ICO得到的筹码跟上线后二级市场买来的筹码本来就没区别的好吗？以太，小蚁，都是如此，比特股并不是例外，这并不是生态低迷的原因。

I've put some thought into a bitcoin sidechain peg. One major challenge: it turns Bitshares into a high-trust environment. Here's why:

1) We create the two-way peg
2) Various users transfer a large amount, for example 25000 BTC, onto the BTS chain
3) Those bitcoins are held in the BTC chain, controlled by Bitshares witnesses in case of "withdrawal request"
4) This creates a huge bounty for witnesses to go rogue and steal the bitcoins.

Any large-enough group of witnesses could claim to be hacked and steal all the sidechained bitcoins. This isn't possible on the network today, I mean, witnesses can't steal your BTS. Using this would be very scary for me, just because I'm trusting that 20 or 30 people won't be tempted to steal 25 million dollars that they collectively control.

Another potential attack: a majority-by-votes BTS holder can vote in new witnesses for himself. Before, it was scary enough, as he could mess with our blockchain somewhat, with double-spends, etc. But with this sidechain, now he can steal all the bitcoins.

Thoughts?

I feel what you are talking is still multisig gateway, not trustless sidechain, however I am not 100% sure.
I just heard steem already began their sidechain work(maybe also peerplays?), big change to bottom infrastrature will be introduced, but I am not clear about the detail.

as btc38 officially said, due to "server logic error", they lost more than 10M BTS, more than 10M NXT and some BTC/LTC from the hot wallet, which totally worth about 1.5M CNY.

it seems below is the hacker's BTS account:

@alt has worked on his private project btsbots https://github.com/pch957/btsbots for long time, he adopt some special archtecture in this project for high throughoutput, simply speaking, at server side one new layer is added to the api node, this layer subscribe transaction data from lower layer, preprocess it and generate a new database which suit for clients' query, and provide these data to client through meteor's ddp protocol.

while the "new api node" provide data such as open orders, transaction history, encrypt memos there's no need for api request to witness-node,only when client need to broadcast new transaction an api request to witness-node is needed. under this design the server side can handle higher throughoutput.

we are now considering one possiblility: is it possible to build general rest API for DEX based on this architecture? if yes we can have general http api for DEX with good throughoutput capacity, it is definitely good news for current and potential bot runners and will attract more users to DEX.

as alt said he hasn't design general APIs while develop btsbots, however he can develop them if there is clear demand, if he works 5 hours per day, he can finish the work in no more than one month.

I'd like to get more inputs on this idea, will this help BTS a lot? is the archtechture suitable for public usage? if the feedback is positive enough we can begin the next step.

whether force settlement or margin call happen or not, the feed price should reflect the market price exactly enough, the price info should be collected from multi main exchanges, should not depend on a single exchange, especially when the exchange has very low depth.

and also, instead of directly adapting the last price,  can we develop some advanced algorithm that can reflect the market status more exactly to generate the feed price? as I know alt customized some algorithm in his price feed script, which behaves well, maybe the alogithm can be referenced.   

when China community say no to worker proposals, the key point is not "we are poor, we do not want to pay, we hate dilution", but "we need better planned,organized and estimated development".  users hate that the worker proposal reduce to just a way for someone to make money by doing thing with little sense, especially when some developer said that "if I can get money from worker proposal by just typing some codes, why would't I? what sense does the codes make? who cares?"

Bitshares is now in a stable stage that most of the features work well and no one feel big pain at any settings, however do we really need not to do further big development? I am not sure, regarding the planned but unfinished features such as bond market and KYC/identity system, have any plans come out from any developer's brain ? or even is it possible for bm to come back to lead the development of these features?

on the other side, we also need some development which do not touch the core codes but focus on the scalability/advanced user experience, for example roadscape's cryptofresh, xeroc's python library.

I feel we still need worker, however the worker proposal process need to be redesigned, this is a big topic, which may include but not limited to:

1.  add rating part for worker proposals, users can rate a worker propsal in a certain period after it is finished, suppose 60 stand for "pass" and 100 stand for "perfect" , the contributer can get 100% payment only when the rating score is above 60, and say 5%+ payment with score above 80, 10%+ for higher than 90 etc.

2.add a "tip" button to each worker, contributor work not only for money, but also for appreciation, in many scene tha latter play a more important role. these feature will help to discourage the workers that "just for money"

3.update the rule to avoid the case that the proposal is voted on and the work started for long time, then the worker is voted down and the worker get nothing.

the idea sounds nice, would like to see it can move forward.

regarding the worker pay advice, in the case that there is no enough bitUSD can be bought from market and finally the worker account borrow bitUSD and pay to worker, how will the worker account be finally settled? waiting someone to force settle?