I agree to let the competition machanisim solve the problem: obviously shareholders appreciate and would like to vote the witnesses that provide more and better service.
and AFAIK, to provide good price feed is also possible via free/open source script, for example, alt has published his script at https://pypi.python.org/pypi/btsprice/0.2.21, it's a good reference.
on the other side, I wonder whether we need to consider one question: suppose publishing price feed become what witness must do, shall we increase the block reward? and to how much?
now being a witness can get some reward, but the reward is not very attractive, we need to balance the responsibility and the reward, maybe we need to reestimate what a reward is more reasonable? maybe a more reasonable reward can lead to more active ecosystem?
kindly give your input on this topic.
@Bhuz sorry if I did things carelessly, I' recheck and take actions accordingly.
Competition COULD be part of the mechanism. Proxies are making choices about who to vote for for other reasons however. The way the system is now that is their right. However, we cannot claim this to be a competitive field so long as the voting is done by different rules for each proxy voter and stakeholders.
In other words, you cannot expect people to compete in a competition where the rules are not clear for being a witness. I can point to myself for example. Regardless of fulfilling what I and many consider the parameters of what is needed for a witness, I still do not get your vote. So if you want competition, you have to offer a fair playing field. It doesn't exist now. In Bitcoin mining as
@btswildpig likes to keep referring to, such competition is all just mathematics. Those that enter into that play field can simply do the math and calculate the reward/risk and invest accordingly. Comparing that to what happens is DPOS is utterly futile.
Sure there are plenty of free feed scripts, however the primary issue which this thread claims is the sources being the problem. Specifically Yunbi and its thin and easily manipulated market. I have chosen to remove Yunbi entirely as a feed source thus, and have enabled some other sources for publishing CNY. My price appears to be similar to abits now in its median.
If publishing feeds are to become 'officially' part of being a witness, then the process the way it works now would be reversed. Each witness would generate blocks at random intervals as they do now, and at random intervals the blockchain would query the witness for feed data. If it is not available, then there is no 'reward' for the generated data. This would help a number of factors over the way the current way works. Right now being completely voluntary you have some witnesses publishing at different intervals than others.. some every hour or every several hours, while others more frequent like every several minutes (as I do).
Now with feeds being 'unofficially' part of what witnesses do, witnesses are disincentivized to publish data feeds because they are an expense. Some of the TOP voted for witnesses (including ones you vote for) do not publish very often only for the sake of not having to pay as much BTS fees. This is doing nothing good for the median price in my estimation. I maybe wrong, but I think having as close to real time data as possible certainly makes the DEX look more attractive and impressive than a floating median that bears no resemblance to any current pricing due to a motley crew of variable price times.
To have it something they must do as part of the job description, but increase the price, might be helpful, but only if the 'rules of the game' are clearly defined. eg. a witness MUST publish feeds at a minimum of every 5 minutes. If a witness settings are shown not to follow the rules, then votes should be removed regardless of the political implications or nepotism. That's the only way to have competition.
Increasing the block reward would certainly make for raising the bar more attractive and bring in new blood. Again though I come back to the rules of the playing field, without that, we can never attract anyone to want to play. The current compensation makes being a witness a hobby at best. It is not monetarily attractive as current compensation for running optimal nodes instead of minimal ones is break even at best. The only saving grace now is that because there is zero development happening on the blockchain and thus no new version updates due to no worker proposals, it doesn't require a lot of time.
I already know the arguments against increasing the reward are going to be that we should be willing to settle for less and work harder to increase the value of BTS so that what we get becomes more valuable. Nobody in the crypto space is going to tolerate this line of thinking for long when they can go to other projects and get paid reasonably well for even less effort than what goes into being a witness (eg. dash master nodes). So this volunteerism fantasy really needs to be considered just that, a fantasy, and we need to start presenting BTS in a unified community manner that is no nonsense, rational, exciting, and profitable.
Proposal:
Worker proposal is made to pay out to witnesses for professional feed data. The worker proposal will be administered by a trustee who operates a script. The script will take all BTS and borrow bitUSD collateralized at 300%. It will then daily distribute the bitUSD to all the witnesses based on the number of publishings done daily to a certain maximum, and the number of markets that were published too. Ideally if all witnesses are participating they should all receive the same reward. However, if some are doing better than others, they will get more. Competition keeps everyone going for the greater reward, or at least keeps everyone putting MAXIMUM effort into getting the rewards that are available for publishing feeds.
It would be more ideal if we could know just how many data feed sources are being used, but the current system doesn't support that in any transparent way. If it didn't then I would suggest rewards be distributed based on number of sources.
That is my input on this topic.