Bitcoin Forum
May 03, 2024, 02:36:24 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 ... 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 [402] 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 ... 610 »
  Print  
Author Topic: [XCR] Crypti | Dapps | Sidechains | Dapp Store | OPEN SOURCE | 100% own code | DPoS  (Read 804603 times)
This is a self-moderated topic. If you do not want to be moderated by the person who started this topic, create a new topic.
GreXX
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500



View Profile WWW
October 24, 2014, 02:20:06 AM
 #8021

For the 3 or 4 of you talking about CfB, including Wulf, it is not an option. He doesn't have the answer either and we have at various times talked to him quite a bit over an extended period of time. We even offered to hire him on the board to help evolve Crypti months ago but he is too busy with his own projects.

For those claiming we had no algorithm or that there is nothing to "steal", that is completely and categorically false. We had all 3 algorithms and the entire system working perfectly in ideal circumstances. The issue is with creating the system the way we described and making it scalable.

We had a working and functioning version that did everything we described and we had actually moved on to coding custom block chains (which were more than half done). We did the crowd sale to get additional funds to hire developers and so we could hire a marketing team, re-build the web properties, etc to be ready for the release of Custom Chains.

We didn't see any issues until after the pre-sale and we started allowing people to install the client and start forging outside of cloud servers and a more controlled environment. An extensive beta period would have helped and if we could do it all over, we would have tried to have a larger test base, but we can't change that now.

We aren't bringing in outside help to write an algorithm from scratch or to code anything. They may provide pseudo-code at some point, but Boris is an exceptionally skilled programmer. Coding the solution isn't the issue, it's determining how to solve the forking and scalability problems. This may require us to develop a new solution, which is why we are bringing in an expert who has done code and security audits in the industry since MasterCoin, because they have a great understanding of how other coins have solved complex problems and also insight that we might not have. You asked us for action so we are taking it. It can never help to have extra eyes on.

Theres a reason no one has built this before. It is a complex problem to solve. There ARE answers, but we were trying to come up with a solution that allows us to keep the system exactly as described but fix the problem set that we are seeing. We have several other options and ideas on how we could build out a network that would serve the same function, some radically different, some more minor tweaks.

Either way, we wanted to consult an outside source who hasn't been blinded by our 100s of discussions and arguments and could approach it from an outside perspective, but with extensive experience and the resume to assure us that they will add something to the conversation.

You can call us incompetent, or incapable, or whatever you want. That is your prerogative and while we obviously don't like it, we get that there are people who are frustrated. There are always things any company could have done better. I am sure those car manufacturers with recalls right now wish their air bags weren't exploding. I'm sure Apple wishes Apple Pay wasn't double billing people right now, but it is. That is the nature of the business.

We will come up with a solution that solves all of the issues that caused us to create Crypti, we will code and implement it, and we will bring it to market. It may or may not look exactly like what we initially set out to build. Keep in mind, when we are done, it may look better.

"If you don't want people to know you're a scumbag then don't be a scumbag." -- margaritahuyan
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
Litoshi
Hero Member
*****
Offline Offline

Activity: 700
Merit: 500

Member of the Crypti Foundation Board of Directors


View Profile
October 24, 2014, 02:22:59 AM
 #8022


If you want to play this PoT game with large (apparently above 50 node) networks, then you have two choices.  Choice one, you can maintain one giant network where some nodes are "super-peers" in some kind of hierarchical tree structure that communicate directly with each other and only the normal peers that are under them, thus cutting down the number of peer-to-peer packet exchanges required.  Choice two, you can maintain numerous small networks (say 50 nodes each?) of equal peer nodes that are rotated in as appropriate to forge blocks when their turn comes - but somehow they've still got to get the blockchain results from the blocktimes the were not eligible to forge, an additional complexity.   


It's interesting to see that you had a very similar line of thinking to what I did. Both of these are possible solutions that I had written up and proposed as potential ways to build out a 2nd gen network. I was particularly fond of the second scenario and doing a rotating mini block for forging determined by a master node block that logged active peers and what subnet they were assigned to. It would be much easier to manage and could be organized similar to how something like DNS works in regards to having a consensus subnet peer list to prevent gaming. Unfortunately it was met by a lot of accusations of being too centralized.

There is also the problem of an attack by launching super-peers that send paying forges only to specific nodes, and empty forge requests to others.  With an open source node, this would be too easy.

Litoshi
Hero Member
*****
Offline Offline

Activity: 700
Merit: 500

Member of the Crypti Foundation Board of Directors


View Profile
October 24, 2014, 03:46:33 AM
 #8023



Read my reply months ago, before IPO and you will find me asking for more info on the algo and white paper - reason was to see what sort of technical solutions devs had in mind for the problems they wanted to solve, it became clear devs were not going to release an exact or specific outline of their algo and I bought in anyway. It now appears that there WAS NO defined algo at the time, only an idea of what the algo should accomplish.



They did have a working algo for the PoT.  It worked just fine as long as there were not more than 50 nodes up.  Once there were more than that, a scaling issue surfaced.

That resulted in most of the nodes not getting any forged blocks, and the nodes with the least latency were getting lots and lots of generated blocks.  There were also forks and "forks of forks" as the longer latency nodes were not updated quickly enough.  Many transactions showed as confirmed, but were never received, including 1200XCR I sent to my windows wallet/node from Cryptsy (which still has not arrived and not shown in the balance on Cryptsy).

They then reworked the wallet/node to correct any forking at every block, but it did not fix the scaling issue.  Several fixes were tested, but were unsatisfactory.  Now we have hired a programmer who is very experienced in altcoin programing to write the algo, incorporating the PoT, PoI, PoP, and PoS in the algo. 

As a temporary fix, and to get the network up and running, we are now using a random method of picking forgers, just like NODE. 



Hate to quote myself, but here is the latest news on Cryptsy:

http://www.marketwatch.com/story/silver-law-group-files-lawsuits-against-cryptocurrency-exchanges-bitcoin-savings-trust-and-cryptsy-2014-10-23

5000Bitcoins
Full Member
***
Offline Offline

Activity: 182
Merit: 100


View Profile
October 24, 2014, 04:19:15 AM
 #8024



Read my reply months ago, before IPO and you will find me asking for more info on the algo and white paper - reason was to see what sort of technical solutions devs had in mind for the problems they wanted to solve, it became clear devs were not going to release an exact or specific outline of their algo and I bought in anyway. It now appears that there WAS NO defined algo at the time, only an idea of what the algo should accomplish.



They did have a working algo for the PoT.  It worked just fine as long as there were not more than 50 nodes up.  Once there were more than that, a scaling issue surfaced.

That resulted in most of the nodes not getting any forged blocks, and the nodes with the least latency were getting lots and lots of generated blocks.  There were also forks and "forks of forks" as the longer latency nodes were not updated quickly enough.  Many transactions showed as confirmed, but were never received, including 1200XCR I sent to my windows wallet/node from Cryptsy (which still has not arrived and not shown in the balance on Cryptsy).

They then reworked the wallet/node to correct any forking at every block, but it did not fix the scaling issue.  Several fixes were tested, but were unsatisfactory.  Now we have hired a programmer who is very experienced in altcoin programing to write the algo, incorporating the PoT, PoI, PoP, and PoS in the algo. 

As a temporary fix, and to get the network up and running, we are now using a random method of picking forgers, just like NODE. 



Hate to quote myself, but here is the latest news on Cryptsy:

http://www.marketwatch.com/story/silver-law-group-files-lawsuits-against-cryptocurrency-exchanges-bitcoin-savings-trust-and-cryptsy-2014-10-23


I've always found exchanges to be useless to the ecosystem. Also ridiculously centralized as in one point of error, Mark Karpeles or Moolah.
Peer to peer sounds much better trading from wallet to wallet instead of keeping it on exchanges.

That being said I think any registered exchange in the US is going to be drowning in law suits since what american lawyers can do don't amaze me anymore.
I spell out the impeccable doom for all of them except maybe Coinbase. The rest of US exchanges do best by staying anonymous.
Not to say most of these exchanges are doing something shady.. their own ponzis, ipos, or not caring about customers losing their funds.
They tend to think they'll get away with it because it's "unregulated" but funny thing is that things stay unregulated until they're not anymore and someone always has to be the example.

Isn't C-cex and Bittrex also US exchanges? Expect them to be sued too for anyone who lost funds there. I'm just amazed who would be want to be first on registering an exchange in the US under your own name for cryptocurrencies. People sue Starbuccs if their coffee is too hot, or sue artists for songs they allegedely plagiarized 50 years ago.

Best keep stuff in your wallet nowadays anyway.

jabo38
Legendary
*
Offline Offline

Activity: 1232
Merit: 1001


mining is so 2012-2013


View Profile WWW
October 24, 2014, 05:35:57 AM
 #8025

coinbase is smart to stick with one coin and one coin only.  that is why I said they wouldn't ever add litecoin or any other coin.  makes law suits less likely. 

jiuge
Sr. Member
****
Offline Offline

Activity: 280
Merit: 250


View Profile
October 24, 2014, 06:36:07 AM
 #8026

For the 3 or 4 of you talking about CfB, including Wulf, it is not an option. He doesn't have the answer either and we have at various times talked to him quite a bit over an extended period of time. We even offered to hire him on the board to help evolve Crypti months ago but he is too busy with his own projects.

For those claiming we had no algorithm or that there is nothing to "steal", that is completely and categorically false. We had all 3 algorithms and the entire system working perfectly in ideal circumstances. The issue is with creating the system the way we described and making it scalable.

We had a working and functioning version that did everything we described and we had actually moved on to coding custom block chains (which were more than half done). We did the crowd sale to get additional funds to hire developers and so we could hire a marketing team, re-build the web properties, etc to be ready for the release of Custom Chains.

We didn't see any issues until after the pre-sale and we started allowing people to install the client and start forging outside of cloud servers and a more controlled environment. An extensive beta period would have helped and if we could do it all over, we would have tried to have a larger test base, but we can't change that now.

We aren't bringing in outside help to write an algorithm from scratch or to code anything. They may provide pseudo-code at some point, but Boris is an exceptionally skilled programmer. Coding the solution isn't the issue, it's determining how to solve the forking and scalability problems. This may require us to develop a new solution, which is why we are bringing in an expert who has done code and security audits in the industry since MasterCoin, because they have a great understanding of how other coins have solved complex problems and also insight that we might not have. You asked us for action so we are taking it. It can never help to have extra eyes on.

Theres a reason no one has built this before. It is a complex problem to solve. There ARE answers, but we were trying to come up with a solution that allows us to keep the system exactly as described but fix the problem set that we are seeing. We have several other options and ideas on how we could build out a network that would serve the same function, some radically different, some more minor tweaks.

Either way, we wanted to consult an outside source who hasn't been blinded by our 100s of discussions and arguments and could approach it from an outside perspective, but with extensive experience and the resume to assure us that they will add something to the conversation.

You can call us incompetent, or incapable, or whatever you want. That is your prerogative and while we obviously don't like it, we get that there are people who are frustrated. There are always things any company could have done better. I am sure those car manufacturers with recalls right now wish their air bags weren't exploding. I'm sure Apple wishes Apple Pay wasn't double billing people right now, but it is. That is the nature of the business.

We will come up with a solution that solves all of the issues that caused us to create Crypti, we will code and implement it, and we will bring it to market. It may or may not look exactly like what we initially set out to build. Keep in mind, when we are done, it may look better.
I choose to trust you.Hope you keep your promise  Smiley


░░░░░░░░░░░░░░░░░░░░░████████░░░░░████████░░░████████░░░░░░░░░░░░░░░░░░░░░█████████░░░░░███████████████████░░░░░░░░░░░░░░░░░░░░░░░░░████████░
░░░░░░░░░░░░░░░░░░░░████████░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░█████████░░░░░░██████████████████████░░░░░░░░░░░░░░░░░░░░░████████░░
░░░░░░░░░░░░░░░░░░░████████░░░░░████████░░░░░░░████████░░░░░░░░░░░░░░░░░█████████░░░░░░░███████████████████████░░░░░░░░░░░░░░░░░░░████████░░░
░█████░░░░░░░░░░░░████████░░░░░████████░░░░░░░░░████████░░░░░░░░░░░░░░░█████████░░░░░░░░███████░░░░░░░░░░██████░░░░░░░░░░░░░░░░░░████████░░░░
░█████░░░░░░░░░░░████████░░░░░████████░░░░░░░░░░░████████░░░░░░░░░░░░░█████████░░░░░░░░░███████░░░░░░░░░░███████░░░░░░░░░░░░░░░░████████░░░░░
░█████░░░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░████████░░░░░░░░░░░█████████░░░░░░░░░░███████░░░░░░░░░░███████░░░░░░░░░░░░░░░████████░░░░░░
░░░░░░░░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░████████░░░░░░░░░█████████░░░░░░░░░░░████████████████████████░░░░░░░░░░░░░░████████░░░░░░░
░░░░░░░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░████████░░░░░░░█████████░░░░░░░░░░░░███████████████████████░░░░░░░░░░░░░░████████░░░░░░░░
░░░░░░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░████████░░░░░█████████░░░░░░░░░░░░░████████████████████░░░░░░░░░░░░░░░░████████░░░░░░░░░
░█████░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░████████░░░█████████░░░░░░░░░░░░░░███████░░░░█████████░░░░░░░░░░░░░░░████████░░░░░░░░░░
░█████░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░░░████████░█████████░░░░░░░░░░░░░░░███████░░░░░█████████░░░░░░░░░░░░░████████░░░░░░░░░░░
░█████░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░░░░░████████████████░░░░░░░░░░░░░░░░███████░░░░░░░████████░░░░░░░░░░░████████░░░░░░░░░░░░
░░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████████░░░░░░░░░░░░░░░░░███████░░░░░░░░█████████░░░░░░░░████████░░░░░░░░░░░░░
░░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░████████████░░░░░░░░░░░░░░░░░░███████░░░░░░░░░░████████░░░░░░████████░░░░░░░░░░░░░░
░░░░░░░████████░░░░░████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████░░░░░░░░░░░░░░░░░░░███████░░░░░░░░░░░█████████░░░████████░░░░░░░░░░░░░░░

Wulfcastle
Hero Member
*****
Offline Offline

Activity: 546
Merit: 500



View Profile WWW
October 24, 2014, 06:56:21 AM
 #8027



We aren't bringing in outside help to write an algorithm from scratch or to code anything. They may provide pseudo-code at some point, but Boris is an exceptionally skilled programmer. Coding the solution isn't the issue, it's determining how to solve the forking and scalability problems. This may require us to develop a new solution, which is why we are bringing in an expert who has done code and security audits in the industry since MasterCoin, because they have a great understanding of how other coins have solved complex problems and also insight that we might not have. You asked us for action so we are taking it. It can never help to have extra eyes on.


Thanks for clearing that up GreXX!

Just a few more questions, hypothetically, in the event that both the team and the outside expert can't resolve the forking and scalability problems, what does it mean for the Crypti network as a whole?

- In that event is there a Plan B or does the temporary solution (random forging), become permanent?
- How would it affect features down the line? Would Custom Blockchains support still be a possibility?
- Are there any features that would have to be removed, or wouldn't work, with the temporary solution or if the forking and scalibility problems aren't resolved?
- Is it still possible to achieve all of Crypti's goals with the temporary solution?
- How does the temporary PoT solution affect the other two algorithms (PoI & PoP)

Please don't take this as an insult or anything like that, I just want to take all possibilities into consideration.
thelittlemanbtc
Newbie
*
Offline Offline

Activity: 19
Merit: 0


View Profile
October 24, 2014, 01:25:19 PM
 #8028

i bought in 2660 this is where the road ends for me



i have to pay my stuff in real life and suffer hard loss here its not money i cant afford to lose its ok. i get money in few week but need pay this monday  Smiley


held as long as possible .sold some last month



gg hate selling this cheap kkkkkk is make me mad i know you will make it


anyone can put up 1.5 btc wall in 750  Cry



good luck /)
amanai
Member
**
Offline Offline

Activity: 350
Merit: 10


View Profile
October 24, 2014, 01:37:44 PM
Last edit: October 24, 2014, 01:53:22 PM by amanai
 #8029

i bought in 2660 this is where the road ends for me



i have to pay my stuff in real life and suffer hard loss here its not money i cant afford to lose its ok. i get money in few week but need pay this monday  Smiley


held as long as possible .sold some last month



gg hate selling this cheap kkkkkk is make me mad i know you will make it


anyone can put up 1.5 btc wall in 750  Cry



good luck /)
u need 1,5 btc buy wall at 750?
thelittlemanbtc
Newbie
*
Offline Offline

Activity: 19
Merit: 0


View Profile
October 24, 2014, 02:03:52 PM
 #8030

i bought in 2660 this is where the road ends for me



i have to pay my stuff in real life and suffer hard loss here its not money i cant afford to lose its ok. i get money in few week but need pay this monday  Smiley


held as long as possible .sold some last month



gg hate selling this cheap kkkkkk is make me mad i know you will make it


anyone can put up 1.5 btc wall in 750  Cry



good luck /)
u need 1,5 btc buy wall at 750?

yes thanks
Wulfcastle
Hero Member
*****
Offline Offline

Activity: 546
Merit: 500



View Profile WWW
October 24, 2014, 02:51:51 PM
 #8031

i bought in 2660 this is where the road ends for me



i have to pay my stuff in real life and suffer hard loss here its not money i cant afford to lose its ok. i get money in few week but need pay this monday  Smiley


held as long as possible .sold some last month



gg hate selling this cheap kkkkkk is make me mad i know you will make it


anyone can put up 1.5 btc wall in 750  Cry



good luck /)

Was the 100K dump you?
thelittlemanbtc
Newbie
*
Offline Offline

Activity: 19
Merit: 0


View Profile
October 24, 2014, 02:55:51 PM
 #8032

i bought in 2660 this is where the road ends for me



i have to pay my stuff in real life and suffer hard loss here its not money i cant afford to lose its ok. i get money in few week but need pay this monday  Smiley


held as long as possible .sold some last month



gg hate selling this cheap kkkkkk is make me mad i know you will make it


anyone can put up 1.5 btc wall in 750  Cry



good luck /)

Was the 100K dump you?

no i can't sell below 700,someone see and sell before me  Undecided

take 80% loss and walk away
or keep and wait 2 month and get delay payment...


blue_wave
Newbie
*
Offline Offline

Activity: 40
Merit: 0


View Profile
October 24, 2014, 03:07:56 PM
 #8033

Hi everyone,
just to Let You Know i have changed my entire profile on BTT. I just didnt like the name. I have same one on Cryptitalk now.

(Its me... Let me Know)  Cheesy
GreXX
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500



View Profile WWW
October 24, 2014, 03:09:55 PM
 #8034



We aren't bringing in outside help to write an algorithm from scratch or to code anything. They may provide pseudo-code at some point, but Boris is an exceptionally skilled programmer. Coding the solution isn't the issue, it's determining how to solve the forking and scalability problems. This may require us to develop a new solution, which is why we are bringing in an expert who has done code and security audits in the industry since MasterCoin, because they have a great understanding of how other coins have solved complex problems and also insight that we might not have. You asked us for action so we are taking it. It can never help to have extra eyes on.


Thanks for clearing that up GreXX!

Just a few more questions, hypothetically, in the event that both the team and the outside expert can't resolve the forking and scalability problems, what does it mean for the Crypti network as a whole?

- In that event is there a Plan B or does the temporary solution (random forging), become permanent?
- How would it affect features down the line? Would Custom Blockchains support still be a possibility?
- Are there any features that would have to be removed, or wouldn't work, with the temporary solution or if the forking and scalibility problems aren't resolved?
- Is it still possible to achieve all of Crypti's goals with the temporary solution?
- How does the temporary PoT solution affect the other two algorithms (PoI & PoP)

Please don't take this as an insult or anything like that, I just want to take all possibilities into consideration.

Good questions Wulf and ones I would like to answer.

1. Is there a plan B if PoT isn't going to be solvable in the near future?

A. Yes, we have 4 additional possibilities for how to build the network out. Some involve the same PoT mechanic but incorporate master nodes and subnetworks for forging and consensus. I have proposed 2 possible plan B's, the current random algorithm is another, and Mike has also proposed 1 solution. So we have 4 solvable solutions ready that we have been discussing.

2. How would this change Custom Blockchains or future features?

A. It wouldn't. Custom Block Chains is actually something that we can build around which ever system we use. This is a feature that other coins could actually rip-off and implement down the road, which is why we have been sort of tight lipped about our code and how it works for now. As I mentioned, we were over half done coding custom chains prior to the forking issues.

3. Would there be any features that would not work with the temporary solution or an altered fix?

A. Well, this is a hard question to answer. All of the secondary market features, custom chains, etc will all still work. The synergy however might be a bit different. Let's call the initial perfect world PoT solution as the "Synergy System" because that was it's strength. You have a self replication and motivated network being build that rewards people for using and you build system around it that encourage use and increase transaction fees which in turn makes the network builders stronger and have higher earnings.

With the temporary solution and random rewards, or some of the additional solutions, the PoT algorithm and reward system will obviously be different. The key problem here is finding a way to make the algorithm and system equitably reward forgers so they can create calculators and understand the affordability of building out the network, and so we can predict how big the network will be at certain transaction volumes based on RoI. We should be able to predict at what point it becomes less profitable and how many nodes will likely exist at a given level. The only downside is that if the network is limited in size due to reward and we have say 200 nodes, does it become affordable for someone to build out 200 nodes of their own, provide the 1000 XCR, and fork the network. While we have precautions that we have built into place to avoid forks (which on a side note, the fork protection was causing some of the issues with people not forging because it didn't trust their requests due to timing issues), we still need to make sure that we take into account all possible attack vectors as we build out.

Also keep in mind, that in 1 of my solutions, much like BTT user MalReynolds discussed, if we switch to focusing more on the commerce side and to limit the ability to have rewards for individual users to just PoP rewards (similar to cash back), and focus on merchant point-of-sale terminals running the network, it would be a radical change and would still reach our goals, as initially described, but would change our system. The features in that type of situation would change because we would not be rewarding "forgers" and would instead be focusing completely on limiting any form of tx fee to merchants to truly create the first completely commerce oriented coin (with potentially 0% Tx fees).

So TL;DR, all features can still be implemented, but some of the intended Synergy could be altered.

4. Is it still possible to achieve all of Crypti's goals with the temporary solution?

A. See the 2 previous answers. I think that all of the goals are possible, but there is the possibility for a small hit to the synergy of the network as originally intended if we have to use say, a random algorithm as opposed to a predictable one. However, being random and unpredictable also has it's benefits and intrigues.

5. How does the temporary PoT solution (random) alter the PoP and PoI algorithms?

A. It doesn't change their functionality. We can still build them exactly as planned and the algorithms and capabilities don't change. They can interface in the same way with any of the possible PoT implementations we are discussing. While our primary goal is to have the network work as originally intended, we have spent considerable time designing, arguing, and diagramming possible solutions (which is part of why we have been so busy and a little bit more absent from the boards). The PoP and PoI algorithms and system are tied into one another, but the PoT is kind of a separate calculation with the result of PoP/PoI added in. We could probably combine PoP and PoI into a label like "Proof-of-Commerce" to make it a little easier to describe. So you take the calculation and weight from PoT and then add in the calculated weight added by PoC (which is like our cash back equivalent) and then you determine who will forge.



I hope that helps clear up your concerns and provide some valuable answers.

On a side note, we are assigning Eric the task of re-doing and improving our FAQ as well as getting it up to date (it still talks about the pre-sale!). These questions and previously answered questions will be great additions and if anyone comes up with other, relevant and answerable questions, I will try to take the time to sit down and give the best answers I possibly can.

GreXX
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500



View Profile WWW
October 24, 2014, 03:12:20 PM
 #8035


If you want to play this PoT game with large (apparently above 50 node) networks, then you have two choices.  Choice one, you can maintain one giant network where some nodes are "super-peers" in some kind of hierarchical tree structure that communicate directly with each other and only the normal peers that are under them, thus cutting down the number of peer-to-peer packet exchanges required.  Choice two, you can maintain numerous small networks (say 50 nodes each?) of equal peer nodes that are rotated in as appropriate to forge blocks when their turn comes - but somehow they've still got to get the blockchain results from the blocktimes the were not eligible to forge, an additional complexity.   


It's interesting to see that you had a very similar line of thinking to what I did. Both of these are possible solutions that I had written up and proposed as potential ways to build out a 2nd gen network. I was particularly fond of the second scenario and doing a rotating mini block for forging determined by a master node block that logged active peers and what subnet they were assigned to. It would be much easier to manage and could be organized similar to how something like DNS works in regards to having a consensus subnet peer list to prevent gaming. Unfortunately it was met by a lot of accusations of being too centralized.

There is also the problem of an attack by launching super-peers that send paying forges only to specific nodes, and empty forge requests to others.  With an open source node, this would be too easy.

This wouldn't be a problem in the way the doc I wrote describes. The internal master nodes would keep a ledger of who is authorized and added to the network as well as what subnet they exist and are allowed to communicate with. Anyone outside of that ledger trying to report information that doesn't match up would simply be dropped. The other subnets would only accept broadcast information from certified peers in their subnet and would drop outside direct comms from "super-peers" that weren't assigned to their equivalent master node.

Passion_ltc
Hero Member
*****
Offline Offline

Activity: 714
Merit: 500


Crypti Community Manager


View Profile
October 24, 2014, 03:17:00 PM
 #8036

Hi everyone,
just to Let You Know i have changed my entire profile on BTT. I just didnt like the name. I have same one on Cryptitalk now.

(Its me... Let me Know)  Cheesy

Good that you let us know that. Wink

Litoshi
Hero Member
*****
Offline Offline

Activity: 700
Merit: 500

Member of the Crypti Foundation Board of Directors


View Profile
October 24, 2014, 03:23:04 PM
Last edit: October 24, 2014, 04:14:45 PM by Litoshi
 #8037

Hi everyone,
just to Let You Know i have changed my entire profile on BTT. I just didnt like the name. I have same one on Cryptitalk now.

(Its me... Let me Know)  Cheesy

Good that you let us know that. Wink

Maybe I missed something?  Who is blue_wave?

If I guess correctly, do I get a prize?

Wulfcastle
Hero Member
*****
Offline Offline

Activity: 546
Merit: 500



View Profile WWW
October 24, 2014, 03:34:53 PM
 #8038

Good questions Wulf and ones I would like to answer.

1. Is there a plan B if PoT isn't going to be solvable in the near future?

A. Yes, we have 4 additional possibilities for how to build the network out. Some involve the same PoT mechanic but incorporate master nodes and subnetworks for forging and consensus. I have proposed 2 possible plan B's, the current random algorithm is another, and Mike has also proposed 1 solution. So we have 4 solvable solutions ready that we have been discussing.

2. How would this change Custom Blockchains or future features?

A. It wouldn't. Custom Block Chains is actually something that we can build around which ever system we use. This is a feature that other coins could actually rip-off and implement down the road, which is why we have been sort of tight lipped about our code and how it works for now. As I mentioned, we were over half done coding custom chains prior to the forking issues.

3. Would there be any features that would not work with the temporary solution or an altered fix?

A. Well, this is a hard question to answer. All of the secondary market features, custom chains, etc will all still work. The synergy however might be a bit different. Let's call the initial perfect world PoT solution as the "Synergy System" because that was it's strength. You have a self replication and motivated network being build that rewards people for using and you build system around it that encourage use and increase transaction fees which in turn makes the network builders stronger and have higher earnings.

With the temporary solution and random rewards, or some of the additional solutions, the PoT algorithm and reward system will obviously be different. The key problem here is finding a way to make the algorithm and system equitably reward forgers so they can create calculators and understand the affordability of building out the network, and so we can predict how big the network will be at certain transaction volumes based on RoI. We should be able to predict at what point it becomes less profitable and how many nodes will likely exist at a given level. The only downside is that if the network is limited in size due to reward and we have say 200 nodes, does it become affordable for someone to build out 200 nodes of their own, provide the 1000 XCR, and fork the network. While we have precautions that we have built into place to avoid forks (which on a side note, the fork protection was causing some of the issues with people not forging because it didn't trust their requests due to timing issues), we still need to make sure that we take into account all possible attack vectors as we build out.

Also keep in mind, that in 1 of my solutions, much like BTT user MalReynolds discussed, if we switch to focusing more on the commerce side and to limit the ability to have rewards for individual users to just PoP rewards (similar to cash back), and focus on merchant point-of-sale terminals running the network, it would be a radical change and would still reach our goals, as initially described, but would change our system. The features in that type of situation would change because we would not be rewarding "forgers" and would instead be focusing completely on limiting any form of tx fee to merchants to truly create the first completely commerce oriented coin (with potentially 0% Tx fees).

So TL;DR, all features can still be implemented, but some of the intended Synergy could be altered.

4. Is it still possible to achieve all of Crypti's goals with the temporary solution?

A. See the 2 previous answers. I think that all of the goals are possible, but there is the possibility for a small hit to the synergy of the network as originally intended if we have to use say, a random algorithm as opposed to a predictable one. However, being random and unpredictable also has it's benefits and intrigues.

5. How does the temporary PoT solution (random) alter the PoP and PoI algorithms?

A. It doesn't change their functionality. We can still build them exactly as planned and the algorithms and capabilities don't change. They can interface in the same way with any of the possible PoT implementations we are discussing. While our primary goal is to have the network work as originally intended, we have spent considerable time designing, arguing, and diagramming possible solutions (which is part of why we have been so busy and a little bit more absent from the boards). The PoP and PoI algorithms and system are tied into one another, but the PoT is kind of a separate calculation with the result of PoP/PoI added in. We could probably combine PoP and PoI into a label like "Proof-of-Commerce" to make it a little easier to describe. So you take the calculation and weight from PoT and then add in the calculated weight added by PoC (which is like our cash back equivalent) and then you determine who will forge.



I hope that helps clear up your concerns and provide some valuable answers.

On a side note, we are assigning Eric the task of re-doing and improving our FAQ as well as getting it up to date (it still talks about the pre-sale!). These questions and previously answered questions will be great additions and if anyone comes up with other, relevant and answerable questions, I will try to take the time to sit down and give the best answers I possibly can.

Thanks for the answers GreXX.  Smiley

With regards to a solution similar to MalReynold's plan, I think it would be very risky. Essentially you will be limiting Crypti to commerce only and shutting off doors to other potential avenues of use. The beauty of Bitcoin is that it has virtually limitless uses, as it is so open. The Bitcoin eco-system has been built mostly by third-party developers (Coinbase, Circle, BitPay), to provide a whole range of services and uses.

However with Crypti, by going with MalReynold's plan you would be limiting the types of services that can be provided by Crypti to commerce-specific ones. I realize that one of Crypti's goals is to become he first crypto-currency that is widely used in commerce, but on the other hand I don't think it would be a good idea to essentially lock Crypti (as a platform) down to just commerce.

With Custom Blockchains support, Crypti's uses could go far further than just commerce, you could have File Storage, Messaging Apps, Asset Exchanges, P2P Marketplaces etc. Instead of having a niche-based currency that is predominantly used for e-commerce you could have a whole range of uses for Crypti other than simply e-commerce.

Currently it is very hard for niche-based currencies to succeed  in the crypto-space. I believe that for Crypti to be really successful, you should look at it not as just a niche-based currency for commerce, but as an open currency with endless possibilities (or as the popular saying goes, one currency to rule them all).

GreXX
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500



View Profile WWW
October 24, 2014, 04:22:51 PM
 #8039


Thanks for the answers GreXX.  Smiley

With regards to a solution similar to MalReynold's plan, I think it would be very risky. Essentially you will be limiting Crypti to commerce only and shutting off doors to other potential avenues of use. The beauty of Bitcoin is that it has virtually limitless uses, as it is so open. The Bitcoin eco-system has been built mostly by third-party developers (Coinbase, Circle, BitPay), to provide a whole range of services and uses.

However with Crypti, by going with MalReynold's plan you would be limiting the types of services that can be provided by Crypti to commerce-specific ones. I realize that one of Crypti's goals is to become he first crypto-currency that is widely used in commerce, but on the other hand I don't think it would be a good idea to essentially lock Crypti (as a platform) down to just commerce.

With Custom Blockchains support, Crypti's uses could go far further than just commerce, you could have File Storage, Messaging Apps, Asset Exchanges, P2P Marketplaces etc. Instead of having a niche-based currency that is predominantly used for e-commerce you could have a whole range of uses for Crypti other than simply e-commerce.

Currently it is very hard for niche-based currencies to succeed  in the crypto-space. I believe that for Crypti to be really successful, you should look at it not as just a niche-based currency for commerce, but as an open currency with endless possibilities (or as the popular saying goes, one currency to rule them all).



Well I consider those uses you discuss as commerce. Companies that function like Coinbase, BitPay, etc can register custom block chains or work with the Crypti network itself, and provide those same services as registered merchants (they are providing services after all). This, in my mind, is the type of commerce we will be pushing as far as e-commerce goes. We would also be utilizing the hardware for point of sale systems for merchants, but that isn't the only form of commerce.

I guess what I should have said was the first crypto currency built from the ground up for "business". Maybe commerce is a limiting term in that sense. We want to encourage all forms of business on the blockchain, to include investment vehicles, storage, video, etc. However there is something to be said for focusing solely on building the 100% best possible system for person to person commerce and letting others build out and develop additional uses and services. I imagine Satoshi didn't foresee all of the services and things that have been built around the capabilities of Bitcoin, so I think focusing on a strong core network, easy fast payments from anyone to anyone, and allowing others to be innovative and build out on top of those 2 features easily, and using those as strong backbones, but without the limitations of Bitcoin and it's weaknesses, is the ultimate goal for all of us.

blue_wave
Newbie
*
Offline Offline

Activity: 40
Merit: 0


View Profile
October 24, 2014, 04:47:20 PM
 #8040

Hi everyone,
just to Let You Know i have changed my entire profile on BTT. I just didnt like the name. I have same one on Cryptitalk now.

(Its me... Let me Know)  Cheesy

Good that you let us know that. Wink

Maybe I missed something?  Who is blue_wave?

If I guess correctly, do I get a prize?

You can try Cheesy
Pages: « 1 ... 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 [402] 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 ... 610 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!