Should Platform run on all nodes or should Platform run only on High Performance nodes ?

[virgile]

www.dash.org/forum/threads/distributed-platform-storage-avoids-high-performance-nodes.53400/#post-232615

Please see latest comments of krilen there

thank you! i'd have missed that one for sure.

And i'll try to not get my button pushed too much, thank you for the advice. I do not like to leave comments unanswered in generals as it gives off the impression of talking to a wall for the community, otherwise.

 

[qwizzie]

DTP elections happen without a 10% majority. The funding governance system is for funding. Consensus is whatever 60% of nodes (currently) say it is. To that extent since we need a winning proposal we will go with the majority. Would you prefer a proposal that got less votes wins?

This directly invalidate your upcoming decision proposals. It is not what i want, it is how we use our governance system. You are changing the rules of our governance system, to push a winner through.

I will have no part in this. This is a gross violation of our governance system and you are by passing any no voter just to declare a winner.
Find another way or another system to find your winner. This Dash governance system is not be be used for that, it never was.

 

[kot]

It is absolutely false that such types of data are the only ones covered by NDA. Do you know that dimensions of screws upon a good bunch of vehicles are also covered by NDAs, that in fact a alot of design points are also covered by NDAs? It's mostly about technical details. Do you remember robert, that i am a cryptographer before all else? you publish vulnerabilities after they are patched not before. Or maybe you'd have liked the vulnerability of DIP 24 to be patched post platform release so that millions can be stolen from Platform? should i also have published that too?

But sure i mean if you all agree that it should be out, i won't go against the will of the network. I won't be that guy who fight you. Enjoy ruining DASH.

I find it insane that laypersons such as you robert are asking me publish security related issues and other sensible data before they can be dealt with. Introspection please. Subject closed do not even bother answering on that topic. IF i see a strong feeling from the community to look at the roadmap i will oblige, but as a security person i can tell you that would be madness.

You are mixing many things and understand it as a roadmap.
Roadmap is not a technical design document like (from your example) dimension of the screws.
Roadmap is not a vulnerability report, with a description of vulnerabilities or problems.

Roadmap is a visual representation of the strategy.

And by design the roadmap document should be used to present very high level information to the public.
If prepared and presented properly, it could even cover highly sensitive information without revealing any details. It is just a matter of preparation - not a rocket science.
So stay calm and professional, and don't be dramatic.

 

[QuantumExplorer]

This directly invalidate your upcoming decision proposals. It is not what i want, it is how we use our governance system. You are changing the rules of our governance system, to push a winner through.

I will have no part in this.

You don't seem to be able to listen to reason.

The majority wants platform out. A decision needs to be made. How do you make a decision without going with the majority? Answer this question.

 

[qwizzie]

You don't seem to be able to listen to reason.

The majority wants platform out. A decision needs to be made. How do you make a decision without going with the majority? Answer this question.

You introduce decision proposals, you keep the 10% treshold intact. If the network decides that none of your start options are acceptable (below 10% treshold) you will need to find a start option, that the network does find acceptable. That is how all decision and budget proposals work.

You can't abuse the governance system (by ignoring the 10% treshold), just to find a winner.
I feel like i am explaining Dash governance to a newbee here... this is getting weirder and weirder by the minute.

Either you use our Dash governance system and play by the rules or you (DCG) take a decision on how to start Dash Platform yourself and take full responsebility for that decision.

 

[QuantumExplorer]

You introduce decision proposals, you keep the 10% treshold intact. If the network decides that none of your start options are acceptable you will need to find a start option, that the network does find acceptable. That is how all decision and budget proposals work.

You can't abuse the governance system (by ignoring the 10% treshold), just to find a winner.
I feel like i am explaining Dash governance to a newbee here... this is getting weirder and weirder by the minute.

I feel like you are trying to find technicalities (incorrect - I might add) because you strongly believe that all solutions are bad. I must admit that I do admire how much you care about the project, but please understand that I do too, I have been pouring my life into this, and have every incentive to want this to succeed. We just have different views on how to see this success. I want a vote on the matter, you seem to not want a vote, and now you are saying that only a vote with 10% supermajority can be recognized. The DTP elections clearly show a different approach than 10% supermajority.

The issue is that the majority do want Platform out, delaying it because we can't decide on how to launch it is in no one's best interest. If you dislike DCG going with what the majority of the network want then there's no way for either of us to win in each other's eyes, and debating will be useless.

I have to get to more important work than arguing here on the forum.

 

[splawik21]

When there was a decision which firm to chose between Oglivy and the other one, can't remember now, in terms of logo change how was the argument of winning one worked? This is a decision proposal and imo the one which gets most yes votes wins even if not getting 10% threshold. Just my 2 duffs.
I don't comment much but I liked the idea of 1k most especially that many MNOs were waiting for the Evolution to come out knowing that they will be able to keep it running too. If there are security reasons behind which would make the platform release failed I'm more forward the HPMSs with higher collaterals.
I'm not a cryptographer neither a mathematician so what is best for the platform success I'd go this way.
ps. I have to admit that this is a quite hot and sensitive argument we have to bother with.

 

[stan.distortion]

The majority wants Platform done properly, not something hastily lashed together for the sake of meeting a release date that's already several years overdue and makes unnecessary sacrifices to decentralisation. The descision that's being asked for is no better than US politics, a choice between 2 (or in this case, 3) evils is no choice at all.

As for NDA's, this started out as an open source project with an active development community and has been turned into a veil of secrets that shuns open source. That has been tolerated so far but we've already been screwed over with the "other projects will steal our innovations" excuse, DCG was receiving millions for work that wasn't getting done. This has to end.

 

[qwizzie]

The DTP elections clearly show a different approach than 10% supermajority.

The DTP elections are completely different from our Dash Governance system. They vote differently, they have a different system of auditors (Dash Incubator team for example). You can not just point to DTP elections and then just say this is how we are going to vote on the upcoming Dash governance decision proposals for this specific topic.

 

[virgile]

You are mixing many things and understand it as a roadmap.
Roadmap is not a technical design document like (from your example) dimension of the screws.
Roadmap is not a vulnerability report, with a description of vulnerabilities or problems.

Roadmap is a visual representation of the strategy.

And by design the roadmap document should be used to present very high level information to the public.
If prepared and presented properly, it could even cover highly sensitive information without revealing any details. It is just a matter of preparation - not a rocket science.
So stay calm and professional, and don't be dramatic.

I will keep that in mind! indeed it seems there was a mishmash in the use of terms here indeed, thanks for pointing that out. I was talking of a group of sensible data yes.

 

[QuantumExplorer]

shuns open source

WHAT??? Every piece of code we write is open source. Please state facts.

 

[vazaki3]

WHAT??? Every piece of code we write is open source. Please state facts.

If every piece of code that DCG writes is open source, then where the Non Disclosure Agreement (that @virgile claims) applies?

And by the way, who demanded a Non Disclosure Agreement for someone to become a Dash developer?
Do we know his name, or is this also a Non Disclosure Agreement?

 

[stan.distortion]

WHAT??? Every piece of code we write is open source. Please state facts.

How many blockchain related thoughts and ideas do you see being discussed in here? 5 years ago there where still frequent discussion, often related to Evolution/Platform (how it could work was pretty much laid out in its entirety at one point) and they've dwindled away to nothing because none of it was taken on board. DCG had zero interest in ideas from the community and chose to do their (lack of) development behind closed doors because "someone might steal our ideas". There's no sign of that changing, sharding etc. isn't open for discussion, we have three choices varying between highly inefficient and highly centralised and that's it, end of discussion.

 

[QuantumExplorer]

How many blockchain related thoughts and ideas do you see being discussed in here? 5 years ago there where still frequent discussion, often related to Evolution/Platform (how it could work was pretty much laid out in its entirety at one point) and they've dwindled away to nothing because none of it was taken on board. DCG had zero interest in ideas from the community and chose to do their (lack of) development behind closed doors because "someone might steal our ideas". There's no sign of that changing, sharding etc. isn't open for discussion, we have three choices varying between highly inefficient and highly centralised and that's it, end of discussion.

Look, I have been CTO for a year. And now we are releasing Platform, when 3 other people in charge just couldn't make that happen. A vision was made, I got on board, and I had to finish it.

The ideas were mostly created in 2016. DAPI, Decentralized API, meaning that all nodes running platform could function to get queryable data from the blockchain. It's great to have an idea. Now build it. Other projects just don't really have this.

For the first time in a very long time, DCG is asking the network what it wants, presenting facts as best we know them. And a lot of people seem unhappy with this.

 

[QuantumExplorer]

If every piece of code that DCG writes is open source, then where the Non Disclosure Agreement (that @virgile claims) applies?

And by the way, who demanded a Non Disclosure Agreement for someone to become a Dash developer?
Do we know his name, or is this also a Non Disclosure Agreement?

It's in people's contracts that they adhere to not hurt Dash, for example if they find an exploit they don't exploit it for themselves, or make it public before it is fixed. Virgile misspoke when he used the term NDA. No dev has signed a NDA to my knowledge, just this part is in everyone's contract.

 

[qwizzie]

For the first time in a very long time, DCG is asking the network what it wants, presenting facts as best we know them. And a lot of people seem unhappy with this.

And why are a lot of people unhappy with it ?

Because of the very limited number of options that you plan to provide masternode owners on how to start Dash Platform, the low quality of those options (they either lead to centralization of Dash Platform or lead to starting Dash Platform with a much higher unsafe / risk factor), and the way you intend to use the Dash Governance system to select a winner (disregarding the 10% treshold, just focusing on most yes votes, pointing to DTP elections as precedent).

Maybe those very limited number of options on how to start Dash Platform change / get extended, due to the backlash of all those unhappy people. Maybe not.
But yeah, not much to be happy about so far.

Also it stings to me personally, knowing that there is a fix to the option of starting Dash Platform with a much higher unsafe / risk factor, which was already put on the internal agenda of devs, but never further developed.

 

[QuantumExplorer]

And why are a lot of people unhappy with it ?

Because of the very limited number of options that you plan to provide masternode owners on how to start Dash Platform, the low quality of those options (they either lead to centralization of Dash Platform or lead to starting Dash Platform with a much higher unsafe / risk factor), and the way you intend to use the Dash Governance system to select a winner (disregarding the 10% treshold, just focusing on most yes votes, pointing to DTP elections as precedent).

I never got a reply if you would like the Trustless Masternode shares option, like that people with 1K Dash would be able participate in Platform (and if you disagree with the market equilibrium theory well this would solve your worries).

The amount of MNOs that host by themselves right now is very low, hence the argument of "but people want to participate themselves" would be satisfied, because all those currently self hosting would be able to host a shared masternode.

To be clear this is a trustless solution akin to delegated proof of stake.

 

[qwizzie]

I never got a reply if you would like the Trustless Masternode shares option, like that people with 1K Dash would be able participate in Platform (and if you disagree with the market equilibrium theory well this would solve your worries).

The amount of MNOs that host by themselves right now is very low, hence the argument of "but people want to participate themselves" would be satisfied, because all those currently self hosting would be able to host a shared masternode.

To be clear this is a trustless solution akin to delegated proof of stake.

At this point i think i prefer the Distributed Platform Storage solution and further working out Proof of Service as mentioned by krilen, over the Trustless Masternode Shares option.

Provided that the Distributed Platform Storage solution is indeed viable as Krilen claims and the Proof of Service can indeed be developed in a far more simple way as Krilen claims (well, claims is perhaps an exaggeration .. he has some thoughts on that matter).

 

[vazaki3]

...the way you intend to use the Dash Governance system to select a winner (disregarding the 10% treshold, just focusing on most yes votes, pointing to DTP elections as precedent).

I also disregard the 10% treshold. I think that the 10% threshold method was never voted by anyone. Evan initialy decided about it, arbitrarily.

Lets put a governance question about it in the budget system . The "10% threshold method" should bypass the threshold of 10%, in order to become a legitimate status quo. Otherwise alternative decision methods should be put in the budget as candidate methods , including the decision method that @QuantumExplorer proposes. And let the most popular method (that also satisfies its own prerequisites) become the status quo.

It is a one Dash question. Who is gonna ask it?

 

[qwizzie]

I think the 10% threshold method was never voted by anyone.
Lets put a governance question in the budget about it.
The 10% threshold should receive 10%, in order to become a legitimate status quo.
Otherwise all methods should be put in question , including the decision method that @QuantumExplorer proposes.

I was hoping in time devs would develop a more distinct seperation between budget proposals and decision proposals in our Dash governance, dividing them in seperate assets and classifying them in code as either a budget proposal or a decision proposal.

Its a discussion topic that never really emerged. Currently in code a decision proposal is no different then a budget proposal and has the same conditions as a budget proposal.

 

[vazaki3]

Its a discussion topic that never really emerged.

It is a discussion topic that I repeatedly suggest since 2016 , but nobody takes my suggestion seriously.

 

[xkcd]

I don't comment much but I liked the idea of 1k

haha! Of course you do, because you run a hosting service, so you are the one that will get the most profits when every MNO has to carry the burden of higher hosting costs and lower ROI with the 1K solution, meanwhile nothing but premium vacations for you in the south of France!

 

[uniuniverse]

MICE TALKING TO MICE:

You've got the cart in front of the horse! Everybody is worried about centralization and nobody is worried about marketing! Unless Dash utilization increases dramatically, all this centralization argument means nothing!

 

[amanda_b_johnson]

With regard to @virgile 's and @QuantumExplorer's comments about having to choose either A) low fees, or B) network-wide distributed storage--is that perhaps a false dilemma if we consider than we can (and should) subsidize the storage costs? In other words, can't Platform storage be a "loss leader" for us?

For example, does Google/Gmail/YouTube not allow anyone and everyone to upload basically as much crap as they want to their platform--for free?

And they can do that, because they're not in the business of selling storage (not primarily, anyway). They're in the business of selling ads and enterprise software packages (and more).

Similarly, the Dash network is not in the business of selling storage. It's in the business of selling Dash tokens.

"Loss leading is a common practice when a business first enters a market." - Investopedia

Dash is looking to enter the Web3 market. Surely we can afford to have storage as our loss leader, even if just initially?

Also, I'll hear no more of the wording of a network-wide Platform upgrade as being "forced." As though every single network-wide upgrade we've done since 2014 hasn't also been "forced"? We always hard-forked away from those who didn't upgrade, didn't we? So let's stop this unnecessary labeling of Platform-on-all-nodes as an act of "force."

 

[splawik21]

haha! Of course you do, because you run a hosting service, so you are the one that will get the most profits when every MNO has to carry the burden of higher hosting costs and lower ROI with the 1K solution, meanwhile nothing but premium vacations for you in the south of France!

If you think I keep the hosting service running for the money you are definitely wrong Michael.
edit: my arguments behind the 1k idea was described but cutting only a piece of the txt is not nice man, your behavior really sounds like newspaper/photographs agencies cutting only a piece of the photo and showing it for the world and making it in bad view.
edit2: personally taking all arguments on the table I'm for HPMNs 4k collaterals though.

 

[rion]

@amanda_b_johnson is exactly right. That's why I say the main purpose of platform is to acquire Dash users (not sell storage, computation, etc).

Dash platform is a user-acquisition tool. It's a means, not an end. The purpose is to get people using Dash as money/currency (let's not lose sight of the main goal). Companies, for example banks, pay hundreds, if not thousands of dollars to acquire users. We can afford to sell block/tree space below cost if that means our main product (Dash the asset) goes up in value due to increasing our active users and transaction volume, which is best form of organic marketing there is.

Fortunately @QuantumExplorer and team have designed a system that allows for this, with the "cost multiplier". This will be a consensus variable that MNOs can set from time to time. To achieve below-cost storage pricing we'd set it to something less than 1 (if I'm not mistaken about how it's designed).

 

[vazaki3]

Also, I'll hear no more of the wording of a network-wide Platform upgrade as being "forced." As though every single network-wide upgrade we've done since 2014 hasn't also been "forced"? We always hard-forked away from those who didn't upgrade, didn't we? So let's stop this unnecessary labeling of Platform-on-all-nodes as an act of "force."

Of course it is an act of force.
The masternodes were always supposed to be servers that participate in a financial-numismatic scheme.
And now you are forcing them to become pornhub and wikileaks databases.
It is a tottaly different job, an unexpected job, so it is outside the pre-agreed unwritten masternode "contract".

So you assume that the platform WILL be abused and WILL be censored. Correct?

Why someone who has 10000 Dash in his wallet, he would like to face such a danger?
I think this risk can more easily be taken by someone who has 500 Dash in his wallet, or even less.
The bigger the risk is for the platform to be abused and censored, the less collateral fee should be asked to those who will host the platform.

 

[rion]

Those of you who really value "decentralization", do you value that primarily as a means to:

a) "fairness" (participation accessibility, not changing the rules as much, etc)
b) security (censorship resistance, attack resistance, etc)
c) some combination of both of the above.
d) some other reason.

I ask because some systems can be as (or even more) secure, despite (or even because of) some degree of strategic centralization:
- I consider my rough proposed solution both more "fair" and more secure than the high collateral systems.
- I also consider it better than the original plan of running platform on every node, even though it's more "centralized". The lower fees and other economic advantages far outweigh the minimal advantages of storing data on every single masternode.

 

[amanda_b_johnson]

@vazaki3 , don't be silly. If you are running a node of basically any blockchain, you likely already are hosting data related to porn/WikiLeaks/etc and have been for years.

 

[vazaki3]

@vazaki3 , don't be silly. If you are running a node of basically any blockchain, you likely already are hosting data related to porn/WikiLeaks/etc and have been for years.

Many blockchains may facilitate hidden financial transactions (also may publish url links) related to porn/wikileaks/etc, but most of these blockchains dont have the database that contains the raw, non fragmented data of the porn/wikileaks/etc. There is a big difference on this. If you facilitate the transactions or you publish the links to illegal content, you may argue in front of a court and claim plausible deniability, and be found not guilty. But If you store the raw/non_fragmented illegal data, the plausible deniability argument weakens, especially in case these illegal data are unencrypted.