elective-stereophonic
elective-stereophonic
Nxt 2.0 design
Please login or register.

Login with username, password and session length
Advanced search  

News:

Latest Stable Nxt Client: Nxt 1.12.2

Pages: 1 ... 25 26 [27] 28 29 ... 51

Author Topic: Nxt 2.0 design  (Read 217035 times)

MrV777

  • Hero Member
  • *****
  • Karma: +115/-4
  • Offline Offline
  • Posts: 988
    • View Profile
Re: Nxt 2.0 design
« Reply #520 on: February 18, 2016, 09:30:15 pm »

I agree, if you ask 'what is the biggest problem nxt has today?' it is adoption. The amount of transactions is only 2000 per day :( Many of the new features (like shuffling, MS) need much more work to get user friendly enough to reach adoption.

Scalability is not our problem today. Finding new users is our challenge.

True, but it's a little of a chicken and egg thing  :)
With more adoption, we need more scalabilty.
To take advantage of more scalability, we need more adoption

...What if 2.0 could have both  :o
or 1.8 and 1.9 could be focused on user friendliness and 2.0 scalability

Just me 2 NXT again  ;D
Logged
NXT: NXT-BK2J-ZMY4-93UY-8EM9V
NXT nodes: 209.222.98.250, 216.155.128.10

cc001

  • Hero Member
  • *****
  • Karma: +68/-4
  • Offline Offline
  • Posts: 829
    • View Profile
Re: Nxt 2.0 design
« Reply #521 on: February 18, 2016, 09:36:16 pm »

Meanwhile http://www.coindesk.com/researchers-redesign-scaling-decentralized-blockchains/
"... The increasing popularity of bitcoin as a digital currency has made scalability a "primary and urgent concern" for the bitcoin network, the authors say, touching upon a topic that has been hotly debated for months in the bitcoin space."

yes, but Bitcoin has 100 times more transactions per day. If we don't find new users we will never need to think about scalability. even if we get traction, it will take a long time until scalability is a "primary and urgend concern" for Nxt. It is ok to think in advance about future problems, but scalability it is not our primary problem right now. It doesn't make sense to build a factory that can produce 100 Lambos [hey Marc ;) ] per day, if you can sell only one per month.
Logged
cc001 Personal - NXT-8RXS-2SSK-RNF2-HSNL8
NxtReporting.com - The Nxt Asset Exchange Portfolio Manager with Profitability Tracking - Donations are greatly appreciated on NXT-5W4G-GAR6-JHJP-H8ZTW

Jean-Luc

  • Core Dev
  • Hero Member
  • *****
  • Karma: +816/-81
  • Offline Offline
  • Posts: 1610
    • View Profile
Re: Nxt 2.0 design
« Reply #522 on: February 18, 2016, 09:44:24 pm »

Scalability is not our problem today. Finding new users is our challenge.
I agree, but this is besides the point. It is not like the work on 2.0 will conflict with finding new users. For the next few months, it would be mostly me alone that will work on 2.0. I need to put a basic framework in place, before development can be parallelized and tasks split between more developers. And then still, it will continue to take development resources only. Attracting new users, marketing and adoption is not the job of the core developers.

If you wait until scalability becomes a problem, it will be too late to start work on 2.0.

Logged
GPG key fingerprint: 263A 9EB0 29CF C77A 3D06  FD13 811D 6940 E1E4 240C
NXT-X4LF-9A4G-WN9Z-2R322

lurker10

  • Hero Member
  • *****
  • Karma: +168/-33
  • Offline Offline
  • Posts: 1334
    • View Profile
Re: Nxt 2.0 design
« Reply #523 on: February 18, 2016, 09:50:17 pm »

Meanwhile http://www.coindesk.com/researchers-redesign-scaling-decentralized-blockchains/
"... The increasing popularity of bitcoin as a digital currency has made scalability a "primary and urgent concern" for the bitcoin network, the authors say, touching upon a topic that has been hotly debated for months in the bitcoin space."

yes, but Bitcoin has 100 times more transactions per day. If we don't find new users we will never need to think about scalability. even if we get traction, it will take a long time until scalability is a "primary and urgend concern" for Nxt. It is ok to think in advance about future problems, but scalability it is not our primary problem right now. It doesn't make sense to build a factory that can produce 100 Lambos [hey Marc ;) ] per day, if you can sell only one per month.

If NXT reaches the transactional volume of Bitcoin making these huge architectural scalability changes will be completely out of the question. Too many users would then be affected, there is no chance in hell they will come to an agreement. Have a look at Bitcoin, a small change of the block size is debated for half a year.

NXT can change while it's small, being small is an advantage. If NXT doesn't evolve to let businesses have their own sidechains, it will not be popular, it will be just another altcoin with a few bells and whistles. Yes, AE, MS, Marketplace - all these features can be done with Bitcoin, in a roundabout way through third parties, but they can be done. NXT must do something Bitcoin can't do because it's already too big for that. This is NXT's competitive advantage, it must be used if we want NXT to survive at all.
Logged
Run a node - win a prize! "Lucky node" project jar: NXT-8F28-EDVE-LPPX-HY4E7

NxtSwe

  • Hero Member
  • *****
  • Karma: +124/-9
  • Offline Offline
  • Posts: 657
    • View Profile
Re: Nxt 2.0 design
« Reply #524 on: February 18, 2016, 10:07:44 pm »

If you wait until scalability becomes a problem, it will be too late to start work on 2.0.
This!
Logged
Check out the NxtLib, the .NET Framework API for the Nxt platform.

Jean-Luc

  • Core Dev
  • Hero Member
  • *****
  • Karma: +816/-81
  • Offline Offline
  • Posts: 1610
    • View Profile
Re: Nxt 2.0 design
« Reply #525 on: February 18, 2016, 10:18:30 pm »

I am proposing again to make the childchain migration process prolonged over time instead of brute splitting of the coin.
Here is how it works:

- Leave the NXT as the mother currency. It will be unprunable blockchain.
- Create sidechains infrastructure.
- Create the NXT replacement sidechain. Let's call it NXT2.

Now start making people migrate to NXT2 chain. To start with make higher TPS (10x higher is a good start). Motherchain won't get any new features so it will be frozen forever with the current featureset. Now start raising fees on the motherchain eg increase them 2x every month.
Finally most of the activity and capital should flow into NXT2 without forcefully taking away the value from asset and NXT holders. Also we will be able to avoid financial shocks and asset dumps, have better backwards compatibility with current services and more time for people to adapt.

Also we can eventually recreate the Genesis again and remove the bloat of the outdated motherchain. Finally NXT2 will be as lightweight as proposed 2.0 design.

I don't consider this a viable approach from software engineering perspective. It would require maintaining code for two different types of blockchains, prunable and not, but unlike the split fNXT design, without a clear separation of transaction types between the two. Every single transaction type will have to be supported on both, and inevitably with small or not so small differences in behavior, resulting in code full of special cases in very fundamental parts of the code. This is not a system I am willing to start developing, let alone maintain in the future.

And in the end, assuming someone goes this route, you will have a system where the Nxt child chain is unwanted baggage, not being prunable itself. Why would anyone want to create a child chain on such platform, and have to carry multiple gigabytes of legacy NXT transactions? As soon as the situation is explained to any business potentially interested in a child chain, they would just ask for a clone instead, without those extra Nxt data. And we are back at where we are now. The Nxt child chain itself must not be a burden, it must be architecturally equal to the other child chains, in order for 2.0 to be considered as a usable child chain hosting platform.
Logged
GPG key fingerprint: 263A 9EB0 29CF C77A 3D06  FD13 811D 6940 E1E4 240C
NXT-X4LF-9A4G-WN9Z-2R322

Marc De Mesel

  • Hero Member
  • *****
  • Karma: +228/-83
  • Offline Offline
  • Posts: 643
    • View Profile
Re: Nxt 2.0 design
« Reply #526 on: February 19, 2016, 04:32:39 am »

dear DEVs, you don't need to proof something to the blockchain industry or the crypto-currency world because you have already something! you have made an innovative all-in-one concept really unique and i am sure that under the wood there is a wonder as code and software architecture but that is not what the user sees, uses and needs ...
if only you invest the same effort to create this childchains architecture in rewriting the marketplace it will restart the economic dynamics... and bring a bit of life to this ecosystem...

Thank you and @++

thumbs up

Marc De Mesel

  • Hero Member
  • *****
  • Karma: +228/-83
  • Offline Offline
  • Posts: 643
    • View Profile
Re: Nxt 2.0 design
« Reply #527 on: February 19, 2016, 04:46:28 am »

Will you stop hardcode fees ? Will you let decide the forgers ( in the fNXT motherchain ) which transactions to include or not even with 0 fee ? If no why ? Since everyone speak about free market here.

Same for the childchain NXT. Everything will be prunable so are you going to let the market decide of the fees ? If no why ?

My brother also talks about this often. He thinks it would be a great improvement. Allowing forgers to also choose which childchains to support would also make childchains possible without the nxt blockchain becoming too big. Childchains would be forks only carried by those nxt forgers that choose to support it/validate transactions for it.

Making nxt more scalable is good, but the cost is way too high if it means splitting nxt into 2 coins.
« Last Edit: February 19, 2016, 04:52:10 am by Marc De Mesel »
Logged

Marc De Mesel

  • Hero Member
  • *****
  • Karma: +228/-83
  • Offline Offline
  • Posts: 643
    • View Profile
Re: Nxt 2.0 design
« Reply #528 on: February 19, 2016, 05:28:30 am »

The initial cost of creating a child chain should be very high, to protect against abuse.
Unfortunatly that will result into low usage as well.
And that's fine, we don't want anyone and their brother creating a child chain just to see how it works and then abandoning it. To create a child chain, you must be an established business and the token you are going to use for your child chain must have market value. If not, use the Monetary System.
But as I said, at first creating child chains and changing their properties will be a manual process, in which case if there is any fee it would be to cover the development efforts for it. Until the system is working smoothly enough to think about automating it.
I think it would be better if that would be possible that everyone can just try to create his coin. Test it around, notice that their coin is only being supported by the last 2 nxt nodes for the last 200 blocks and realise they need to start giving value to their coins if they want it to survive it to survive longer time they create one. This all without any nxt blockchain bloat.

Just like in my proposal.
https://nxtforum.org/core-development-discussion/aternate-nxt-2-0-design/

I would love to explain this proposal to you verbally, if you are open to it.

Very good post.

On first impression I liked nxt 2.0 as I thought it was MS done well but if you can't cheaply launch your coin, it's not done well and adoption will be - again - poor.

In this alternate proposal I like very much that people can launch their altcoin cheaply (as they are basically forks that nxt forgers can individually decide to support or not) while also not bloating the nxt blockchain but still giving value to nxt as you need it to forge the altcoin.

And they can pay fees with their altcoin directly!! No need even for the altcoin to buy/have nxt.  This is a great improvement over the current Monetary System while not having to split up nxt in 2 coins.


Back to bed now, this nxt 2.0 thing has kept me up. I'm very worried.  :-[
« Last Edit: February 19, 2016, 05:51:35 am by Marc De Mesel »
Logged

blackyblack1

  • Hero Member
  • *****
  • Karma: +165/-82
  • Offline Offline
  • Posts: 1764
    • View Profile
Re: Nxt 2.0 design
« Reply #529 on: February 19, 2016, 05:56:37 am »

I am proposing again to make the childchain migration process prolonged over time instead of brute splitting of the coin.
Here is how it works:

- Leave the NXT as the mother currency. It will be unprunable blockchain.
- Create sidechains infrastructure.
- Create the NXT replacement sidechain. Let's call it NXT2.

Now start making people migrate to NXT2 chain. To start with make higher TPS (10x higher is a good start). Motherchain won't get any new features so it will be frozen forever with the current featureset. Now start raising fees on the motherchain eg increase them 2x every month.
Finally most of the activity and capital should flow into NXT2 without forcefully taking away the value from asset and NXT holders. Also we will be able to avoid financial shocks and asset dumps, have better backwards compatibility with current services and more time for people to adapt.

Also we can eventually recreate the Genesis again and remove the bloat of the outdated motherchain. Finally NXT2 will be as lightweight as proposed 2.0 design.

I don't consider this a viable approach from software engineering perspective. It would require maintaining code for two different types of blockchains, prunable and not, but unlike the split fNXT design, without a clear separation of transaction types between the two. Every single transaction type will have to be supported on both, and inevitably with small or not so small differences in behavior, resulting in code full of special cases in very fundamental parts of the code. This is not a system I am willing to start developing, let alone maintain in the future.

And in the end, assuming someone goes this route, you will have a system where the Nxt child chain is unwanted baggage, not being prunable itself. Why would anyone want to create a child chain on such platform, and have to carry multiple gigabytes of legacy NXT transactions? As soon as the situation is explained to any business potentially interested in a child chain, they would just ask for a clone instead, without those extra Nxt data. And we are back at where we are now. The Nxt child chain itself must not be a burden, it must be architecturally equal to the other child chains, in order for 2.0 to be considered as a usable child chain hosting platform.
If you are talking about business let's look at the problem from a business point of view.
NXT started with a business proposal of 1000 bytes permanent storage for 1 NXT payment. Great deal! Later you have decided that this kind of market is not interesting and raised the fees 30x together with limiting the maximum message size to 160 bytes. This way we have lost a cut of the permanent storage market. Now you are proposing to totally remove the business for the permanent storage from the NXT services. Instead you are moving forward the business of temporary ledger backed with archival nodes. Those are 2 kind of different markets with different businesses interested in them and different fees.

My proposal is to stick with both markets and let people use permanent and temporary ledgers together. Permanent ledger should cost much higher but it is still very attractive for people. You are talking about bloat but this is not a bloat. This is a permanent record which we promised to people and this 1Gb database is a proof to the businesses that we are sticking with our promises.
Logged

LocoMB

  • Hero Member
  • *****
  • Karma: +101/-37
  • Offline Offline
  • Posts: 751
    • View Profile
Re: Nxt 2.0 design
« Reply #530 on: February 19, 2016, 06:54:51 am »

Hard coded fees are serious nonsense anyway. I proposed quite a while back to adjust the fees to a real world, non-inflatable utility, like a scaled basket of several surface mail costs - where *physical* activites actually take place like *physically* moving an envelope over *physical* distances.
Or electricty costs, or even bitcoin mining costs, because those include a basket of electricity costs.
 
A process that cannot be spoofed, something that is not pulled out of thin air!

But simply hardcoding fees is as stupid as having the price of bread fixed in a central bureau of prices- and we all know what that leads to. And how often does any sage have to adjust them in order to make it work?
In any case, this discussion is very harmful to NXT, and there are plenty of non-disruptive changes and amendments to the NXT platform to increase adoption.

Increasing adoption? Who is going to build a project on a platform that is in LIMBO for at least a year, because the rules are totally unclear?

Golly Gee, I wonder how much adoption we will see until NXT2.0 is out?
And how long after that until any projects feel secure enough that it is viable and stable? And for the predominantly technologically minded people, I should emphasize: ECONOMICALLY viable and stable?

So- putting myself into the shoes of a potential project manager seeking a platform:
Quote
'NXT may be a nice platform, but I have no idea what it will be like in 15 months from now and I can't jeopardise my project so I'll wait until then... or go somewhere else, becasue there are other similar platforms around where the conditions are a bit more clear'



« Last Edit: February 19, 2016, 07:02:48 am by LocoMB »
Logged
TOX
90E54E5B5213290EE616D425CADC473038CFABFA53C913271AA8559D1937DC4AF3A354A9E4E5

martismartis

  • Hero Member
  • *****
  • Karma: +73/-10
  • Offline Offline
  • Posts: 1238
    • View Profile
Re: Nxt 2.0 design
« Reply #531 on: February 19, 2016, 07:08:09 am »

Very rough idea to eliminate fnxt from the proposal:

Quote
- A new main chain will be created, on which NXT becomes a token used for forging only, "forgingNXT". The current NXT ecosystem will become a child chain, preserving all features and holdings except the ability to forge. At the hard fork block, each NXT owner will have his NXT converted to both tokens in 1:1 ratio, and all other holdings migrated to the NXT child chain.

- It will always be possible to exchange NXT to fNXT, so small stakeholders not interested in forging may decide to sell their fNXT to large stakeholders running forging nodes. This would lead to some centralization, but also to a higher percentage of the (f)NXT stakeholders forging and thus securing the complete Nxt ecosystem.

1. NXT remains as a mainchain with its forging
2. There is button in client "clone NXT", which let to create sidechain with it own parameters and wishful features (MS, AE, DGS, etc) as it is now by creating currency or asset.
3. Everything works as proposed in OP, except fnxt stuff, sidechain creator pays fees in NXT, while sidechain inside works with its native token, sidechain prunable every 1440 block, etc
4. Prune data from NXT mainchain which we can see could be prunable (ask/bid orders, messages or whatever) every month or half a year, storing just snapshot or whatever on mainchain.

Does this approach solve scalability and bloat problem? Is it possible to do technically (I think yes :) )?
If this could be possible solution, IMO we could solve these asset dump/not dump, fnxt/nxt ratio, fnxt distribution discussions, having one mainchain with NXT instead of two fnxt and not NXT chains, and working sidechains stuff :)
Logged

Cassius

  • Hero Member
  • *****
  • Karma: +207/-18
  • Offline Offline
  • Posts: 2459
  • Rather be a pirate than join the navy
    • View Profile
Re: Nxt 2.0 design
« Reply #532 on: February 19, 2016, 08:27:03 am »

BCNext's original plan was that MS coins would come to be widely used as actual currencies. How did he foresee the transition from using NXT as a currency to using something else? Anyone know?
Logged
I head up content for BitScan, crypto business hub.

testdruif

  • Full Member
  • ***
  • Karma: +71/-1
  • Offline Offline
  • Posts: 223
    • View Profile
Re: Nxt 2.0 design
« Reply #533 on: February 19, 2016, 09:11:41 am »

Something is still not really clear to me. (forgive me if the below text is non intelligent)

Right now when I want to get something published on the blockchain I create the transaction and pay the nxt minimum fees to have it included.

With 2.0, in order to get my transaction processed, someone needs to create an "nxt block" which in turn needs to be picked up by forgers that create an "fnxt block"?
fnxt forger will get the fnxt, "nxt block creator" will get the nxt.

Who is going to create that childchain block? will there be "nxt forgers" who basically trade the nxt transaction fees (they recieve nxt from the nxt block) for fnxt (the pay fnxt to get it included in the fnxt blockchain)?

are there scenarios where the "price/worth" of fnxt is higher than the "price/worth" of the transaction fees of the childchain block?

example:

Lets say that I want to send 1000 nxt to someone. it will cost me 1 nxt in transaction fees.
The blockchain creator (lets assume there is only one transaction going on) will need to pay 1 fnxt to get the block included and receives 1 nxt

Can it be possible that 1 fnxt costs 1002 nxt to trade (the nxt block creator needs to get his fnxt from somewhere and it's still a user driven economy)?
Would it then not be cheaper to transfer 1 fnxt to the person I was first intending to send 1000nxt? (and thus defeat the purpose of nxt as a currency at that point)

And if the price for fnxt would be greater than what the nxt block creator get in return (the nxt) it will probably not be worth it for the block creator to even consider creating a block
« Last Edit: February 19, 2016, 09:14:40 am by testdruif »
Logged
**Necessity is the mother of invention**
NXT-NNGD-V8TN-3MZR-DWWBE
https://arguseyes.net

icoin

  • Newbie
  • *
  • Karma: +2/-0
  • Offline Offline
  • Posts: 9
    • View Profile
Re: Nxt 2.0 design
« Reply #534 on: February 19, 2016, 09:27:24 am »

This looks like compression of blocks? Do you want to consider this in Nxt 2.0 design?
_https://www.reddit.com/r/Bitcoin/comments/46gtjm/thin_blocks_early_results_messages_are_on_average/
Logged

Windjc

  • Sr. Member
  • ****
  • Karma: +59/-17
  • Offline Offline
  • Posts: 360
    • View Profile
Re: Nxt 2.0 design
« Reply #535 on: February 19, 2016, 10:50:30 am »

The more I have studied the more I have a concern about this issue - once Fnxt is created Nxt is relegated to "Child Chain Status." THINK about that for a moment. NXT is no longer a MAIN CHAIN coin/token.

So we expect the price of a child chain to have the same speculative attraction as all the full chain competitors out there? And we expect the SECURITY of a child chain to be as strong as a main chain?

We are devaluing Nxt, if not in real world terms, at the very least in perception. And perception = reality. As we know.

Can someone address these issues please?
Logged

Hachoir

  • Full Member
  • ***
  • Karma: +12/-13
  • Offline Offline
  • Posts: 113
    • View Profile
Re: Nxt 2.0 design
« Reply #536 on: February 19, 2016, 11:16:04 am »



We are devaluing Nxt, if not in real world terms, at the very least in perception. And perception = reality. As we know.

Not true at all, and stupid
« Last Edit: February 19, 2016, 01:20:34 pm by Hachoir »
Logged

TheWireMaster

  • Sr. Member
  • ****
  • Karma: +27/-0
  • Offline Offline
  • Posts: 356
    • View Profile
    • NXT Folks
Re: Nxt 2.0 design
« Reply #537 on: February 19, 2016, 11:17:10 am »

I have a feeling that the core developers honestly have a vision that is not related to making money, but to make a great decentralized platform.
But if that is is the case, and no agreement is found with big investors in nxt1.x, why not create a new chain for the nxt2.0 and maintain the current one improving the existing functionalities and keeping an eye on backwards compatibility?
It could be backed up by a nxt1.x asset and distribution of that would determine the initial distribution for nxt2.0.
In that way nxt2.0 will be future oriented and nxt1.x will be the one that people can be sure will not change too much except for improvements of the current implementation.
I would prefer 1 NXT with a wide consensus on how to go forward.
Logged
NXT-5WW2-XQ63-CFGM-G7YAJ

LocoMB

  • Hero Member
  • *****
  • Karma: +101/-37
  • Offline Offline
  • Posts: 751
    • View Profile
Re: Nxt 2.0 design
« Reply #538 on: February 19, 2016, 11:31:04 am »

I have a feeling that the core developers honestly have a vision that is not related to making money, but to make a great decentralized platform.
But if that is is the case, and no agreement is found with big investors in nxt1.x, why not create a new chain for the nxt2.0 and maintain the current one improving the existing functionalities and keeping an eye on backwards compatibility?
It could be backed up by a nxt1.x asset and distribution of that would determine the initial distribution for nxt2.0.
In that way nxt2.0 will be future oriented and nxt1.x will be the one that people can be sure will not change too much except for improvements of the current implementation.
I would prefer 1 NXT with a wide consensus on how to go forward.

yes, if you are a java dev you can do that, no one is stopping you...
Logged
TOX
90E54E5B5213290EE616D425CADC473038CFABFA53C913271AA8559D1937DC4AF3A354A9E4E5

lurker10

  • Hero Member
  • *****
  • Karma: +168/-33
  • Offline Offline
  • Posts: 1334
    • View Profile
Re: Nxt 2.0 design
« Reply #539 on: February 19, 2016, 11:33:18 am »

I have a feeling that the core developers honestly have a vision that is not related to making money, but to make a great decentralized platform.
But if that is is the case, and no agreement is found with big investors in nxt1.x, why not create a new chain for the nxt2.0 and maintain the current one improving the existing functionalities and keeping an eye on backwards compatibility?
It could be backed up by a nxt1.x asset and distribution of that would determine the initial distribution for nxt2.0.
In that way nxt2.0 will be future oriented and nxt1.x will be the one that people can be sure will not change too much except for improvements of the current implementation.
I would prefer 1 NXT with a wide consensus on how to go forward.

yes, if you are a java dev you can do that, no one is stopping you...

The NXT 1.x chain should be functional after the hard fork. It makes sense to forge on both chains.
Logged
Run a node - win a prize! "Lucky node" project jar: NXT-8F28-EDVE-LPPX-HY4E7
Pages: 1 ... 25 26 [27] 28 29 ... 51
 

elective-stereophonic
elective-stereophonic
assembly
assembly