This would be a huge advantage of bitshares over other exchanges, if it was not implemented a dumb way. How does this dumb implementation add more safety?

It seems I'm not the one who lacks understanding here. Good luck with your Keynesian view of what drives the free market. I hope it doesn't come to dominate and corrupt the governance of BitShares.

Also note that clayop is also a committee member. I recommend his role there also be looked at.

The other committee members can chime in here if clayop has been contributing and participating in those activities adequately.

If not he should be voted out.

@xeroc?

I agree @sudo, we need to replace witnesses that don't do their job well and producing reliable, frequent and accurate feeds is part of a witness' duties.

His dereliction of duties is bad for BitShares. We need to get good witnesses in here to replace those like clayop, people like@rolandp who is a dev and has proven his capabilities for the Steem community.

Rolandp just posted his witness proposal here on the forum. I recommend anyone reading this take a look and vote for him.

@yvv what are you talking about mate? I think you confuse bitshares DEX purpose... Every bitasset created in bitshares is backed up by min 2.5 collateral value in bitshares. This is why in bitshares if you want to short sell you need to place at least 2.5 collateral value. This is why bitshares is safe and will avoid bankrupt. Because every bitasset created is backed up by 2.5 times collateral value!! This is why bitshares is superior to any bank and to any centralized exchange..Because you know that the " Bank" actually holds a lot of collateral for your "Money"...This is not a flaw in bitshares!! This is the superior market selling point of bitshares!!

Especially now that we have more than $1 mil in bitassets is bitusd and bitcny and we all know that for these bitassets the blockchain holds at least x2.5 collateral in bts value, THAT is the selling point of bitshares. I know that now I feel very comfortable keeping my bitusd in bitshares Dex. What we need to do now is just do the same for biteur, bitgbp, bitgold, bitsilver, bitbtc and reduce the premium paid for those bitassets as well and be close to peg as bitusd and bitcny are currently are...

Bitshares DEX is supposed to be your own bank. Not your own trading house! I want to feel safe keeping bitassets. I do not want bitshares blockchain to keep fractional reserves so you can margin trade with leverage!! The people who designed bts knew exactly what they were doing!!

Going back to margin trading of centralized exchanges that you mentioned...All exchanges that give leverage are supposed to keep balanced books, take opposite positions to hedge and keep only the spreads charged. Well I have news for you..Most of them don't do that! Not even very big regulated exchanges that offer leverage to their customers. This is why when the CHF - EUR peg broke a couple of very big exchanges with margin trading went bankrupt and others almost bankrupt..

I hope that you do know that most banks just keep only fractional reserves of peoples money right? Well I want bitshares to be exactly what it is..A bank that keeps more than 250% of collateral of peoples money!

As a final note, I hope you do not learn the hard way about margin trading in Poloniex as I did when I lost 80% of my bts holdings when I used them to margin trade (going long of course)..Poloniex and all these exchanges most probably manipulate prices in their favor. They see all the order books and they have really sophisticated bots to drive margin calls and screw you in their favor. So be careful when margin trading in Poloniex. They can screw you even if you are in the right direction of the trade..For example after too obvious pumps you go short and all of the sudden their servers don't work until prices recover and you do not get your entitled profit, or get margin called when you wanted to close your position etc etc..

^^THIS^^THIS^^THIS

The difference of perspective between BitShares and Poloniex is essentially the same as Kenesian vs. Austrian economics.  Traders will ALWAYS be a much smaller subset of everybody else who can benefit from BitShares. If you appeal to people who want a safe place to store their wealth that same safe place is not going to be a hot trading platform - the two objectives are contrary to one another.

This is yet another example of a dichotomy within the BitShares ecosystem, like Stealth vs. open ledger. Savers and Traders want opposite things out of BTS - one to be a stable store of value the other wants volatility so the trading action is profitable.

Why not market to both groups? Appeal to savers by offering a dividend on their BitUSD, and to traders via a campaign of "Safe Trading"? The issue with the system of trading yyv seems to think so highly of is the leverage isn't safe. Haven't you learned from all those examples of over-leveraged derivatives that collapse when one key investor with huge stake gambles big, looses, then all the smaller fish that relied on gains from trades of that big guy go down, and it's just a house of cards waiting to fall. That can't happen with adequate collateral. Sorry to burst your bubble but you're looking through rose colored glasses and they blind you to the real dangers of the mainstream financial system. Sadly, people aren't willing to see those dangers until they are personally faced with them. It strikes me as a weakness of a gambling addict.

The leverage you like so much is merely an illusion of safety made attractive by potential profits. That safety will evaporate faster than a bot can place an order when the inevitable collapse begins.

Has anyone looked at @wackou's plans to implement this configuration?

I know he was waiting on a few pull requests to be merged before it was possible but I'm fairly certain those are now in place.

I suspect there will be increased risk of attack as with the upsurge in traffic. This scheme could help shield the witness nodes by only connecting to seed nodes and letting the public his the seeds and not witness nodes directly.

Wackou and I talked about this late 2015 early 2016, but he has been extremely busy, not to mention until the last month things were pretty quiet.   

IOn the other hand, most if not all Chinese feed sources has a DDOS protection that blacklists your ip with *many* less API calls than the specified.

Feeds can be published from a non producing witness with the active key.

Interesting. Would their DDoS protection kick in from 1 request an hour? I presume when you say "feed sources" you're talking about exchanges?

I don't believe your last point is correct. The active OR owner keys can be used to publish feeds, but the API call will fail unless the account calling the publish_feeds API is voted in (active). Don't get the state of being "active" / voted in confused with the type of key used on the account calling the API.

I would like to see some type of method for non-active witnesses to publish "informal" or "unofficial" price feeds to show shareholders they are capable and what their accuracy would be. It would also provide potential witnesses a way to perfect their scripts.

Glad to see your desire to improve the quality of the witness role.

These are not new topics, but it seems to be very difficult to get traction. I believe it's gonna require coming up with a BSIP and then getting community to recognize it.

There don't seem to be many witnesses willing to discuss such details, let alone take the initiative to write a BSIP to formalize the conventions. I also don't recall if there is a formal approval process BSIPs must go through by committee or whoever. I think they're more a matter of formalizing proposals in a document. BSIPs are good standards to have.

Some reform / evolution of the witness role will require coding, for example to implement standards that can be measured and automatically removing or disabling witnesses when those standards are violated, such as when X number of blocks are missed, or fails to publish feeds at a minimum frequency, or the feed accuracy slips outside an certain deviation from other witness feeds.

We have discussed a worker proposal to get professional feed sources, but they are quite expensive from what I gather. There is also the issue of centralization. That's one reason why no single feed script should be relied upon. To my knowledge there are 3 primary scripts most witnesses use, however the parameters each witness uses for them can be different based on that witness' influence. There are few standards on setting those parameters.

There is a Telegram bot created by witness spartaco to notify of missed blocks, set time slots for when during the hour a witness should publish their feed (use /slots in Telegram to list which minute is assigned to what witness. There are 3 groups that bot works in: BitsharesDEX, Bitshares_Witnesses and BTS Witness Alerts). Here are the commands the bot responds to:

/help
/price [ASSET]
/missed WITNESS
/monitor WITNESS
/listmonitor WITNESS
/stopmonitor WITNESS
/status
/feed [ASSET]
/slots
/setslot WITNESS MIN
/rmslot WITNESS
/locations
/setlocation WITNESS LOCATION
/rmlocation WITNESS

Please use the BTS Witness Alerts channel for urgent matters only. It's kindof like the "Hotline" to all witnesses. Witness alt recently posted in the forum saying several witnesses were broadcasting double blocks back to back. If true that would be a major issue and could cause forking of the blockchain. It turned out to be a false claim however the best place to communicate such things is via this real-time Telegram channel.

I would like to hear more about your offer to buy a subscription to a professional price feed source. Your offer is generous but I don't think we should continue to rely on altruism to support necessary blockchain operations to publish accurate price feeds.

Price feeds and block production require 2 different skill sets and I suggested these should be separate roles. The general attitude I see expressed however is that "producing a price feed isn't very complicated" and thus need not be split out. I have a different perspective which I've expressed in more detail elsewhere but it seems I'm alone in it. There does seem to be some agreement on the idea of automating some type of witness monitoring, but when it comes to saying what corrective action to take mostly I hear crickets.

It's also important to note these issues are of interest primarily to witnesses, and many of them are not interested in them enough to contribute to the discussion.