In the interest of fairness, and not just a Steem cheerleading session, here's some info on the problems with Proof-of-Stake, Delegated or not.
Also, when the ENTIRE CHAIN froze and locked out people, I'd consider that a rather big "oopsie" that should be mentioned, in addition to the abysmal retention rate -- 13%.
Reference:
A topic not emphasized enough with PoS protocols is their lack of resilience in dealing with WORST-CASE SCENARIOS. For examples: extraordinary events that could knock offline or partition a significant portion of the network, or even the entire network. Or the risk of stolen/purchased private keys.
I encourage people to look into the technology deeper than a "everything is great" post. And I'm not even getting into the bot-infested problems Steemit has, either. That has been covered elsewhere.
In my view, what happened was dealt with appropriately, and I saw the incident as more favorable than unfavorable in terms of the way it was dealt with.
In terms of technical details, it was much less about DPoS vs. PoW, and much more about a corner case in the consensus logic built on top of the DPoS technology.
As blockchain consensus logic starts to become more and more complex (as many of the functions on Steem are) the possibility for unexpected scenarios and glitches also increases. Incidents such as this are likely to become more prevalent as the blockchain industry advances.
Having the blockchain freeze up is actually what you want to have happen in a situation like this, because it prevents the unexpected scenario from turning into something more serious like a loss of coins.
I have severe reservations about a system whose acceptable failure mode is "stop everything", especially when it keeps getting touted as the "best" solution and the "most transacted" chain.
I'm sure you can see how both of those goals diverge?
Also, why don't more exchanges carry Steem to begin with? You've obviously been around for a while, what is the reason for severe lack of fungibility?
To be honest, I'm not holding my breath. The reason you need complex "solutions" to these problems is because the underlying PoS forces you to -- and by doing so, they're just bandages over the problems Proof-of-Stake introduces -- either directly or tertiary.
Its just like Ethereum and their gradual realization that they can't scale, so they come up with increasingly complex answers instead of refactoring their base assumptions.
I'm sure you can see how both of those goals diverge?
Yep. Obviously never ever having a single software glitch would the the goal.
Also, why don't more exchanges carry Steem to begin with? You've obviously been around for a while, what is the reason for severe lack of fungibility?
We are a top 40 coin (not a top 20). The integration with STEEM is more complex than other chains. Many exchanges also want $1M+ listing fees. We have been getting added to more and more exchanges though. Higher marketcap would help.
To be honest, I'm not holding my breath. The reason you need complex "solutions" to these problems is because the underlying PoS forces you to -- and by doing so, they're just bandages over the problems Proof-of-Stake introduces -- either directly or tertiary.
I'll have to disagree with you here on a technical level. The glitch we are talking about would have happened on a PoW chain too, if the consensus rules had been written on top of a PoW chain. It had nothing to do with PoW vs. DPoS.
Its just like Ethereum and their gradual realization that they can't scale, so they come up with increasingly complex answers instead of refactoring their base assumptions.
Can you elaborate on how the issue we are talking about is an instance of STEEM running into a limitation that needs complex answers, and what the base assumptions are?
While I can empathize with your exasperation with the sort of empty optimism that often replaces deeper analysis, I'm a bit confused with your outrage that all or most posts about STEEM on the STEEM blockchain are generally optimistic about its future. Most people with a more scathing outlook tend to leave the platform or become relatively inactive eventually.
Also, though an unprecedented and rather scary event (many blocks being missed while the chain froze), what would be a better way of dealing with a similar anomaly than halting things so that everything remains safe until a solution is found and adopted by a supermajority of witnesses, in your view?
"Most people who hate us, leave us, so what's the big deal?"
It matters because given the abysmal 13% retention rate, when you only have a bunch of hooray-henries populating your platform - then there's something amiss.
The best question would be - "Why does Proof-of-(mis)Stake have this kind of edge case where we shut the whole thing down?" Here's the answer -- because its an inferior solution that requires trust of other nodes by its very design.
Everything bolted on top is because of this base flawed assumption. It adds complexity and gives "edge" cases a huge leg up in happening more than just "once in a while."
I'm here to watch the PoS fail-train come to its logical conclusion. Someone needs to witness it without being blinded by unicorn-vision - if only to write the sober post-mortem the day after.
Yeah, the terrible user retention rate is one of Steemit's biggest problems, at present. To be clear, I would prefer to have both "hoorah-henries" and "hellno-hanks" around, because it makes for more enriching discussion.
I think you have me wrong though. Just because I invested heavily in STEEM doesn't mean I believe PoS will be a sustainable system in the scale of decades. I'm too uneducated on the underlying technology to conclude that with confidence. I also don't think it will keel over tomorrow, next week, or next year, though.
You still didn't answer my question:
what would be a better way of dealing with a similar anomaly than halting things so that everything remains safe until a solution is found and adopted by a supermajority of witnesses, in your view?
but I'm guessing you would say there is no real solution because DPoS is an inherently dysfunctional system in your view.
To be clear, I would prefer to have both "hoorah-henries" and "hellno-hanks" around, because it makes for more enriching discussion.
Hey @d-pend. I suspect that through the same premonitory steps and walkarounds of @talltim, there are yet a few coherent and eloquent "hellno-hanks" here with some interesting and illustrative stuff worth a read to enrich healthy discussions. Maybe everything is not lost and the "hoorah-henries" won't run away with the trophy so easily yet. };)
I'm kinda curious what would you think about that?
That was far from a cheerleading session. He actually started off the article listing the problems. Perhaps yours could have been mentioned as well but what platform or technology doesn't have those.
As to Proof of Work vs proof of stake opinions may differ but I know where I stand and we'll see where things end up.
The only thing mentioned were early start-up issues, not the entire blockchain seizing up like a suburban dad having a heart attack from shoveling too much snow off of his driveway.
And no, the PoS vs PoW isn't an opinion - its technical merits and drawbacks that aren't touched upon, because every post about Steemit is rah-rah go team - instead of taking a hard look at why it functions the way it does.
Steemit is going to have to do something radically different to change that 13% retention rate, that's for sure. But I doubt they have the will to refactor a system they're obviously emotionally invested in.
I think that Steemit will fade into the background as other projects emerge on its blockchain. The seizing up that I've seen has been on other blockchains.
In the interest of fairness, and not just a Steem cheerleading session, here's some info on the problems with Proof-of-Stake, Delegated or not.
Also, when the ENTIRE CHAIN froze and locked out people, I'd consider that a rather big "oopsie" that should be mentioned, in addition to the abysmal retention rate -- 13%.
Reference:
https://medium.com/@hugonguyen/proof-of-stake-the-wrong-engineering-mindset-15e641ab65a2
I encourage people to look into the technology deeper than a "everything is great" post. And I'm not even getting into the bot-infested problems Steemit has, either. That has been covered elsewhere.
The incident was fully explained here: https://steemit.com/steem/@steemitblog/steem-blockchain-patch-issued
In my view, what happened was dealt with appropriately, and I saw the incident as more favorable than unfavorable in terms of the way it was dealt with.
In terms of technical details, it was much less about DPoS vs. PoW, and much more about a corner case in the consensus logic built on top of the DPoS technology.
As blockchain consensus logic starts to become more and more complex (as many of the functions on Steem are) the possibility for unexpected scenarios and glitches also increases. Incidents such as this are likely to become more prevalent as the blockchain industry advances.
Having the blockchain freeze up is actually what you want to have happen in a situation like this, because it prevents the unexpected scenario from turning into something more serious like a loss of coins.
I have severe reservations about a system whose acceptable failure mode is "stop everything", especially when it keeps getting touted as the "best" solution and the "most transacted" chain.
I'm sure you can see how both of those goals diverge?
Also, why don't more exchanges carry Steem to begin with? You've obviously been around for a while, what is the reason for severe lack of fungibility?
To be honest, I'm not holding my breath. The reason you need complex "solutions" to these problems is because the underlying PoS forces you to -- and by doing so, they're just bandages over the problems Proof-of-Stake introduces -- either directly or tertiary.
Its just like Ethereum and their gradual realization that they can't scale, so they come up with increasingly complex answers instead of refactoring their base assumptions.
Yep. Obviously never ever having a single software glitch would the the goal.
We are a top 40 coin (not a top 20). The integration with STEEM is more complex than other chains. Many exchanges also want $1M+ listing fees. We have been getting added to more and more exchanges though. Higher marketcap would help.
I'll have to disagree with you here on a technical level. The glitch we are talking about would have happened on a PoW chain too, if the consensus rules had been written on top of a PoW chain. It had nothing to do with PoW vs. DPoS.
Can you elaborate on how the issue we are talking about is an instance of STEEM running into a limitation that needs complex answers, and what the base assumptions are?
While I can empathize with your exasperation with the sort of empty optimism that often replaces deeper analysis, I'm a bit confused with your outrage that all or most posts about STEEM on the STEEM blockchain are generally optimistic about its future. Most people with a more scathing outlook tend to leave the platform or become relatively inactive eventually.
Also, though an unprecedented and rather scary event (many blocks being missed while the chain froze), what would be a better way of dealing with a similar anomaly than halting things so that everything remains safe until a solution is found and adopted by a supermajority of witnesses, in your view?
"Most people who hate us, leave us, so what's the big deal?"
It matters because given the abysmal 13% retention rate, when you only have a bunch of hooray-henries populating your platform - then there's something amiss.
The best question would be - "Why does Proof-of-(mis)Stake have this kind of edge case where we shut the whole thing down?" Here's the answer -- because its an inferior solution that requires trust of other nodes by its very design.
Everything bolted on top is because of this base flawed assumption. It adds complexity and gives "edge" cases a huge leg up in happening more than just "once in a while."
I'm here to watch the PoS fail-train come to its logical conclusion. Someone needs to witness it without being blinded by unicorn-vision - if only to write the sober post-mortem the day after.
Yeah, the terrible user retention rate is one of Steemit's biggest problems, at present. To be clear, I would prefer to have both "hoorah-henries" and "hellno-hanks" around, because it makes for more enriching discussion.
I think you have me wrong though. Just because I invested heavily in STEEM doesn't mean I believe PoS will be a sustainable system in the scale of decades. I'm too uneducated on the underlying technology to conclude that with confidence. I also don't think it will keel over tomorrow, next week, or next year, though.
You still didn't answer my question:
but I'm guessing you would say there is no real solution because DPoS is an inherently dysfunctional system in your view.
Hey @d-pend. I suspect that through the same premonitory steps and walkarounds of @talltim, there are yet a few coherent and eloquent "hellno-hanks" here with some interesting and illustrative stuff worth a read to enrich healthy discussions. Maybe everything is not lost and the "hoorah-henries" won't run away with the trophy so easily yet. };)
I'm kinda curious what would you think about that?
That was far from a cheerleading session. He actually started off the article listing the problems. Perhaps yours could have been mentioned as well but what platform or technology doesn't have those.
As to Proof of Work vs proof of stake opinions may differ but I know where I stand and we'll see where things end up.
The only thing mentioned were early start-up issues, not the entire blockchain seizing up like a suburban dad having a heart attack from shoveling too much snow off of his driveway.
And no, the PoS vs PoW isn't an opinion - its technical merits and drawbacks that aren't touched upon, because every post about Steemit is rah-rah go team - instead of taking a hard look at why it functions the way it does.
Steemit is going to have to do something radically different to change that 13% retention rate, that's for sure. But I doubt they have the will to refactor a system they're obviously emotionally invested in.
I think that Steemit will fade into the background as other projects emerge on its blockchain. The seizing up that I've seen has been on other blockchains.
Interesting - care to list any? Also, were they PoS based?
Inquiring minds want to know...