Changes

From Ravencoin Wiki
Jump to navigationJump to search
7,357 bytes added ,  22:27, 6 August 2018
no edit summary
__TOC__
 
== 05 Aug 2018 - Telegram ==
'' Tron ''
 
Tron:
Ravencoin is a protocol platform for assets and will work with anybody's infrastructure. Additional requirements can always be added by exchanges, or issuers, but are not native to the asset protocol.
 
----
 
Tron:
You’re absolutely right. We are watching the issue closely. To put it in perspective though, our block times have been roughly 1.5x the speed of BTC during the worst of it. We have already switched to DGW on testnet and Ravencoin is slated to move to DGW on mainnet at the end of October. Our primary concern is not the code, that’s already done. Our primary concern is not allowing the chain to split. A chain split would cause far more damage and confusion than 5 minute blocks. The diff swings are a result of being profitable, then not, then profitable again, in an undamped oscillation. DGW will fix it, but it requires a hard fork where everyone must upgrade the software. This carries some risk.
 
----
 
Tron:
x16r is the hashing algorithm, and is different than the difficulty adjustment algorithm. It will still be x16r. The difficulty adjustment tries to make it so it is 1 minute blocks no matter how many cpus/gpus are mining. Currently it adjusts by looking back 2016 blocks and making a percentage adjustment to bring it inline. There is a 4x limit, which we are close to hitting. It isn’t good but it shouldn’t get much worse.
 
----
 
Tron:
RVN will still be RVN. The problem is that anyone that doesn’t upgrade when the algo switches over, will be on a different chain. Imagine the problems if CryptoBridge doesn’t upgrade, or Cryptopia, etc. and they have a different view. A hard fork is just a software upgrade, but it is really important that certain nodes upgrade. We have a managed upgrade already scheduled for the end of Oct that lets everyone decide on their own whether to upgrade and when enough mining nodes have upgraded, it seamlessly switches over at a set percentage.
 
----
 
Tron:
No, shorter block times aren’t the problem. Rentable hashpower, and multipools which auto-redirect hashpower to the most profitable coin makes it more difficult to calibrate. That’s what changed.
 
----
 
Tron:
The longest chain is the winner. We just need to make sure it is crystal clear which one that is. If we have 1/2 mining power on one chain and 1/2 on another, it gets murky. If that ever happens, just don’t transfer RVN during that time and you’ll be fine. And, if you’re mining, make sure you’re on the longest chain.
 
----
 
Tron:
DGW will go a long way towards that. We actually want increased hashing, we just don’t want it coming and going in multi-day intervals.
 
----
 
Tron:
I don’t think miners using a pool will need to change anything. Just make sure the pool you’re using has upgraded whenever the software is upgraded. Make sure you’re using the latest version to view/store/transfer your RVN.
 
----
 
Tron:
My understanding is that there was a Pull Request for RVN on Ledger. I don’t know if/when that will make it into production.
 
----
 
Tron:
Yes, unix (linux), Mac, and Windows. We need to make sure the client cross-compiles for all of them.
 
----
 
Tron:
Yeah, there were some awful asset names. I imagine that some of the web explorers and asset marketplaces will want to run names through a filter.
 
----
 
Tron:
There’s going to be lots of interesting opportunities. I think someone will build a marketplace for asset names.
 
----
 
Tron:
There’s really only one chain. The problem is that when you change the diff algo, then the version of the software using the old algo sees only old algo blocks as valid, and software using the algo only sees new algo blocks as valid. The “real” one SHOULD be the new algo chain, but there’s no BOSS that says you can’t run older software. So it is really up the users to upgrade and all agree to use the newer software. I can’t compel anyone to upgrade, but it would make economic sense for everyone to use the new software. We can write software that doesn’t change the algorithm until it sees that most miners have upgraded because we can put a version number in the blocks and count the mined blocks. That’s what BIP9 does — Google it if you’re interested in how it works.
 
----
 
We’re hiring like crazy. Most of our hires are for tZero, Medici, and Bitsy. If you’re a great developer, you should apply. Bruce is in NH and he visits Utah and we visit him in NH.
 
----
 
Tron:
Yes. I don’t know how trademark rules would apply. I fought two trademark battles where I felt I was in the right. I won one, and folded the other against a giant “perfect” word processing company.
 
----
 
Tron:
There is support for that. I’m not in favor of that, in general, but it did make sense for our asset hard fork.
 
----
 
Tron:
I hope not, because I think it would devalue RVN, but it is certainly a possibility. Interestingly enough, anyone who had RVN before that fork would have RVN and RVC, but I still think it would would be a net negative overall.
 
----
 
Tron:
Sometimes it is cheaper to mine, and sometimes it is cheaper to buy. Right now it is cheaper to buy — until the diff adjusts. I wrote a paper on how the two are interlinked. https://letstalkbitcoin.com/blog/post/bitcoin-value-and-mining-difficulty
 
----
 
Tron:
Bruce and I agree that it doesn’t need shilling. Eventually, we’ll need to let more people know it exists and that it is a platform they can build on, but that isn’t even true until the end of October when assets activate on mainnet.
 
----
 
Tron:
If there’s a Raven Classic, it probably wont have asset support.
 
----
 
Tron:
Yes, we would like RVN to continue to be viable cash. We hope it gets used for rewards (paying all holders of your token in RVN), and for on-chain atomic swaps (paying for assets with RVN where the trade happens all at once).
 
----
 
Tron:
I’m not 100% sure. It seems like we’re in a bear market — similar to 2014-2015. Some of it is being sold to buy into RVN, EOS, and ADA, or into ICOs which are still going strong outside the US.
 
----
Tron:
I agree - kinda dumb. It is a completely arbitrary line. Enforce rules against outright fraud, but let everyone invest their money any way they choose.
 
----
 
Tron:
I think there will be sites listing assets for sale. We’ve put “forsale” and “forsale_price” keys in the metadata spec for this reason. Brokers could scan the chain and the meta data to build a website for exchanging asset names.
 
----
 
Tron:
https://github.com/RavenProject/Ravencoin/blob/master/assets/asset_metadata_spec.md
 
----
 
Tron:
I will share the scripts to allow anyone to register any names they want. I’m excited to see what gets registered.
 
----
Tron:
Sending assets will require a small RVN fee. I’d love to get rid of the fee, but it opens up too many attacks that don’t cost anything.
 
----
 
Tron:
It’s been wonderful chatting with all of you. Thanks for your support of RVN. I’m working on some python scripts that’ll let you create batches of assets (with associated meta data) from a Google spreadsheet or csv. I’ll make them available next week for everyone to play with. It’ll make it easier for anyone wanting to register multiple assets.
 
== 03 Aug 2018 - Telegram ==

Navigation menu