• Forum has been upgraded, all links, images, etc are as they were. Please see Official Announcements for more information

OFFICIAL DASH ROADMAP - Version 2.1.1 // Delivery Milestones

Status
Not open for further replies.
Soon the Dash bug bounty program will be up and running, first in private, then later open to the public. If there are any specific tests you would like the program to address, contact me and we can consider it.
 
There is something I don't understand. The MN will be run on specialized hardware to keep up with transaction demand... so that means regular supporting nodes (non-MN) will be excluded simply because they won't be able to keep up with the network?

Anyway, regarding livenet stress tests, hmmm, I'm not entirely sure how MNOs will be able to compensate if half the network drops off. On the face of it, @akhavr idea of running a backup node over 3g seems a bit too ambitious, though a home fibre connection might work? In any event, it seems to me, the performance tests that are placed upon MNs should be self-adapting... perhaps then, it could recognise that half the network has disappeared and compensate accordingly.

When first responders are put through emergency simulations, one of the things they are being trained for, repeatedly, is to keep calm and follow procedure. Perhaps that's where we need to be. Is that over the top? - I don't think so because dash's reputation would be on the line and it's not like we are talking hobby.. we are supposedly transitioning to a more responsible position.
 
There is something I don't understand. The MN will be run on specialized hardware to keep up with transaction demand... so that means regular supporting nodes (non-MN) will be excluded simply because they won't be able to keep up with the network?

Yes unless they spend the money to upgrade their node to the level of a MN they will likely not be able to keep up.
 
My current understanding is that masternodes will take over the job of the miners in the long run.
So, ideally if a non-MN, regular node can't meet the current requirements of the masternodes (collateral, static IP, maximum tx/second, DDoS stress-test, protocol version etc.) they will not get paid, but they would still act as a useful source of the blockchain data for example.

The Dash bug bounty is great, and I think it's a very good way to find security holes.
On another level of quality assurance, I could imagine that the masternodes occasionally switch to stress-test mode for a few minutes, and ask every other masternode/regular node to flood it with requests, simulating extreme traffic/DDoS attacks. The cooperating masternodes would report the target's ability to respond, measuring it's toughness. If this value is lower then the current requirement, the masternode temporarily would loose its status (and profit potential). It's just a raw idea, and it might be too expensive to accomplish in a fair way, I don't know... A decentralized self-testing mechanism would risk only one masternode at a time and it would also continuously monitor the network health.
 
After some though... I wanted to revise my statements regarding MN hardware.

If special MN hardware were a PCIe card, and it were cheap enough I could fire-and-forget it... No concern for the loss... I could see the advantage.

Hosts are not going to allow the CPU to do the privileged mining. But an ASIC on PCIe, that doesn't crank up the power bill... And includes a few other massively parallel enhancements to chain and DAPI processing... I could see that.

But only if it's cheap enough that it's disposable.
 
But only if it's cheap enough that it's disposable.

Cheap not just because of "disposable", but because some countries have braindead customs. And it's not the theft of 44% price that concerns me, but the issue that the "administration" of the import can easily eat workweeks. And, of course, you'll be flagged in "the database".
 
Cheap not just because of "disposable", but because some countries have braindead customs. And it's not the theft of 44% price that concerns me, but the issue that the "administration" of the import can easily eat workweeks. And, of course, you'll be flagged in "the database".
Another solid point.

Special MN Hardware makes the matter of being an MNO exceptionally cumbersome. Well, we expect a lot of MNOs. Sure. But you can't expect every MNO to start a war with the State. It is fine to ask much of MNOs fiscally, resourcefully, technically...

But, asking the impossible will only hurt DASH. You're essentially asking "MNOs need to find a solution to global government corruption or they don't get to run a masternode." This is a problem as old as the Earth... No one has ever solved this problem. If you make running a MN contingent upon it, there simply won't be any masternodes.

Dodging government is about dodging it's corruption. There is nothing in all of recorded human history which existed, had government added to it, and was improved. DASH simply will not be able to function with that burden.

I'm not seeing a balance on hardware that would be so cheap it would be disposable, but still be useful...

I understand the notion, but I find another aspect of DASH that makes me wonder if the nerds in charge poke their heads out and look around at the real world anymore...

Yeah, I'm going to bring it up: IX is STILL broken... I understand the delay in gearing up the major codebase transition for Evolution. I do. But, hamstringing real-world efforts during that delay was an obnoxiously bad move. There could have been parallel work going on, laying a foundation... But, nope... Leave IX broken for 2 years... I realize it wouldn't be efficient. Sometimes life is messy and you have to do shit the hard way, or you end up doing nothing at all... Doing nothing, and holding back everyone else, just because it's not perfect, or might result in a bit of re-work... I cannot overstate the absolute foolishness of the decision NOT to fix IX BEFORE the long delay...

And we see this special hardware thing, again, not considering the real-world implications...

We're way, way better off reducing collateral to increase node count. If collateral were 500, that would put DASH waayyy beyond even the weak "full node" argument. Of course, we really don't care about positioning against legacy crapcoins, but the argument would be moot on it's face and not able to distract... One must also consider that a lowered collateral would likely have a more than doubling effect, as all the whiners who talk about how unfair it is that they can't afford a money hose, would be able to run a money hose... After complaining at length about how they lack the technical qualifications and want the job dumbed down instead of improving their resume...

I identify as a MasterNode Operator! You elitists stop oppressing me! The Usefularchy is keeping me down! I refuse to think! Run my MasterNode for me! I'm entitled!

The only upside I see is that the notion of MN hardware is very, very far future-looking. It will be quite a long time before this bridge need be crossed. Decreasing collateral/increasing node count, could put that off until I die of old age and let my son worry about it.

But, so was the budget/DGBB implementation, and a total crap version was instituted even though much better ideas were tabled...
 
Last edited:
Anything regarding Masternode hardware and other physical requirements should be done very gradually and cautiously. Imagine if all of a sudden there was a special hardware requirement -- a lot of MNOs might simply decide that's too much trouble and cash out their collateral. What would happen to Dash's price and marketcap if a mere one hundred MNOs (~2% of total) dumped their 100,000 Dash on the market?
 
Anything regarding Masternode hardware and other physical requirements should be done very gradually and cautiously. Imagine if all of a sudden there was a special hardware requirement -- a lot of MNOs might simply decide that's too much trouble and cash out their collateral. What would happen to Dash's price and marketcap if a mere one hundred MNOs (~2% of total) dumped their 100,000 Dash on the market?

Evan said there would be a transition period. And besides, I think a few other coins will go down a similar route.
 
I seriously doubt that the human race is capable of producing enough transactions to need the capacity that special MN hardware will bring.

Look at the blocksize ambition. We'll need more hard drive space well before the population grows so large that we'll need more processing power. Transparent hardware compression/encryption? Sharding will be an inevitability because there's simply nowhere to store a blockchain that big. Wimbly Bimbly? "Blockchain forever" is simply untenable. The need for a chainless crypto is inevitable, and privileged mining gives you a hint... 25MB blocks at 1 block per 2.5minutes... How fast will your 8TB drive fill up? Storage arrays? When you expect more from a masternode than you do from any other server, you're being ridiculous. And we're only talking about storage space... Demanding more is fine. Demanding more than the entire technology industry can deliver because you won't rethink the basics of an experiment gone too far...

The future brings government blockchains/crypto. At that point, you are a fool if you don't believe non-government cryptos will be banned. Special Hardware means those "illegal currency terrorists that are undermining the nation's economy (a.k.a., MNOs)" will have bull's eyes painted on them. Freedom is so unpatriotic! The overwhelming majority will always prefer being Big Brother's pet, and willingly abide any and all propaganda spewed.
 
Last edited:
I seriously doubt that the human race is capable of producing enough transactions to need the capacity that special MN hardware will bring.

Look at the blocksize ambition. We'll need more hard drive space well before the population grows so large that we'll need more processing power. Transparent hardware compression/encryption? Sharding will be an inevitability because there's simply nowhere to store a blockchain that big. Wimbly Bimbly? "Blockchain forever" is simply untenable. The need for a chainless crypto is inevitable, and privileged mining gives you a hint... 25MB blocks at 1 block per 2.5minutes... How fast will your 8TB drive fill up? Storage arrays? When you expect more from a masternode than you do from any other server, you're being ridiculous. And we're only talking about storage space... Demanding more is fine. Demanding more than the entire technology industry can deliver because you won't rethink the basics of an experiment gone too far...

The future brings government blockchains/crypto. At that point, you are a fool if you don't believe non-government cryptos will be banned. Special Hardware means those "illegal currency terrorists that are undermining the nation's economy (a.k.a., MNOs)" will have bull's eyes painted on them. Freedom is so unpatriotic! The overwhelming majority will always prefer being Big Brother's pet, and willingly abide any and all propaganda spewed.

Reading your recent posts about scaling is like reading my own posts from 1 - 3 years ago haha.

But things have moved on since then.

I don't doubt what you say but I think you've allowed yourself to jump from A to C without going through B.

I don't see dash's roadmap as being set in concrete. For example, it's entirely possible dash will run multiple chains simultaneously; some for practical purposes (a treasury chain) and others for the purpose of live testing e.g. a blockless chain that you refer to. In such a setup, it's entirely possible that dash will evolve into something very different.

The way I see it, there are currently four major approaches to scalability:
  1. Payments channels (litecoin, digibyte etc)
  2. Interesting variations of Delegated Proof of Stake (antshares / neo is notably different / interesting)
  3. Directed Acyclic Graph - blockless payments e.g. iota, byteball etc.
  4. Real-time on-chain Proof of Work e.g. dash
Future cryptos might use a combination of the above e.g. payment channels + DAG. But regardless, they won't all attain the status of digital cash, for they will find their own niche. Remember, the best crypto analogy is that of programming languages e.g. they don't all have to be as fast as C+

And finally, imo, the threat level against MNO's has remained constant to where we were two years ago. Yes, there are bigger numbers involved, but that's relative to where we are in crypto as a whole. And I don't just see MNOs moving into data centers, it's easy to imagine a whole bunch of cryptos moving in when you consider distributed apps and file storage e.g. filecoin etc.
 
Sorry if I missed that, but when are CoPay and DashCore 12.2 expected to be released on the test net?

DashPay Wallet Alpha (September 2017)
  • Dash version of CoPay, pre-Evolution functions, on test-net
  • ...
  • DashCore 12.2 Release (September 2017)
 
Sorry if I missed that, but when are CoPay and DashCore 12.2 expected to be released on the test net?

this is all a bit behind
there should be an announcement / update in the coming week / soon

testnet is not open / public yet
 
Last edited:
22: Tier-2 conceptually is able to handle much larger block propagation, due to having much higher quality network connections. We plan on having large farms of co-located servers, eventually connected using ethernet hand-off. Masternodes located in the same facility can be used for local syncing, allowing more rapid propagation of blocks to catch new servers up. Tier-1 remains permissionless, but also becomes vastly more expensive. Processing of blocks uses custom hardware, allowing for us to process transaction/transition signatures in a parallelized way.

I am in agreement with @akhavr @TheSingleton, I am confused as to why we need further centralization in data centers? I am also amiss as to why we need specialized hardware? Maybe, the Dapi's processing overhead which is 500-1500 tx/s extra load on top of the functions already performed by the masternodes?

Maybe we do not fully understand Dapi and its use cases? DashMail or Dmail could compete with google and solve the following problem with email? If this is true maybe we are thinking of masternodes in the wrong way. Is there a future tense of the word, futurenodes? In the future maybe they are fully decentralized and/or distributed clusters. Do we look more like a blockchain powered CDN as powerful as Netflix or Cloudflare?
 
I am in agreement with @akhavr @TheSingleton, I am confused as to why we need further centralization in data centers? I am also amiss as to why we need specialized hardware? Maybe, the Dapi's processing overhead which is 500-1500 tx/s extra load on top of the functions already performed by the masternodes?

Maybe we do not fully understand Dapi and its use cases? DashMail or Dmail could compete with google and solve the following problem with email? If this is true maybe we are thinking of masternodes in the wrong way. Is there a future tense of the word, futurenodes? In the future maybe they are fully decentralized and/or distributed clusters. Do we look more like a blockchain powered CDN as powerful as Netflix or Cloudflare?
DAPI doesn't currently exist. How it will effect MN load is entirely theoretical.

That said, 2 and 3 generation old server hardware is already more than capable of even the worst case scenario. Even if everyone in the world started using DASH exclusively, Evolution, as roadmapped in June of next year (other than blocksize), is more than capable.

The only uses I can see for hardware are privileged mining, which would be very constrained instead of the landscape it is now, and transparent compression to storage. The blockchain cannot, and will not, go on forever. Perpetual blockchain cannot be. NOTHING lasts forever. Something else has to occur before it collapses. When, not if. I could see a transparent compression SSD interface. But, why not let the market develop this for you?

I've already outlined the many detriments and faux-benefits and it falls on deaf ears.

The only remaining idea for the hardware, is expressly "dongle permission." You have the hardware, or you're not an MNO. Like pirate-paranoid software companies made you plug into your printer port in the 1990s-2000s.

I'm confused as to the real-world utility that could be conceived by the pursuit of this path. What little there could be, is minor to the point of why bother. Maybe Evan just want's a nerd fun money sink for tax deductions? Whatever...

As for the perpetual blockchain... It seems possible, given Evolution's feature set, to perform a "stagnant VIN sweep-forward" type of operation...
 
Last edited:
Status
Not open for further replies.
Back
Top