Obyte: Totally new consensus algorithm + private untraceable payments

[2weiX]

We have a new #auctionroom channel in slack:

Second auction of the day will take place approximately @ 20:45 UTC today (not earlier but maybe a bit later). Amount for sale is 48GB. Bids start at 0.45BTC.

If you want to take part in the auctions, you can get your slack invite here:
http://slack.byteball.org/

plus-upping this, push push!

[tonych]

I looked through the transactions in the block explorer and couldn't understand the technical terms. In this example transaction what are Unit, Authors, Children, and Parents? Is the Unit like a Bitcoin transaction ID and the Authors like a Bitcoin sending address?

https://explorer.byteball.org/#zD17W8iMLjb4Uxtt66NIcdhiCv9MCAtQFaBDY4jSXNo=

Unit zD17W8iMLjb4Uxtt66NIcdhiCv9MCAtQFaBDY4jSXNo=
Authors
UGP3YFIJCOCUGLMJUFKLXLNYNO4S7PT6
Children
V1UF7jRtqFUDWsOxmSLz23rTs7s0azPbpiNrXncsH74=
Parents
EjR+BnqoiGywB7WIEpELtxjyLku2bVylF8EcAjVGteE=
KSy9G0vl4PdpqZb9VCLJ2pnZyDtcUT+gE30O3uaX9uE=

Unit is hash of data unit, similar to bitcoin transaction ID, but unit can have more than just a transaction.
Authors are addresses who signed the transaction, similar to bitcoin sending address.
Children and parents are pointers to later and earlier units in the DAG (see the arrows between units).

[SatoNatomato]

Hello tonych

Im running a witness and it crashes with this

events.js:160
     throw er; // Unhandled 'error' event
     ^Error: write EPIPE
   at exports._errnoException (util.js:1022:11)
   at WriteWrap.afterWrite [as oncomplete] (net.js:804:14)


And how come it says this "distance above threshold, will witness" in the logs, but I havent told anyone about the witness ID/public-key, how come it is trying to witness?

Does it make any sense to run a byteball-relay as well as a witness, the logs seem to be the same, is the witness also doing relay?

[HI-TEC99]

I looked through the transactions in the block explorer and couldn't understand the technical terms. In this example transaction what are Unit, Authors, Children, and Parents? Is the Unit like a Bitcoin transaction ID and the Authors like a Bitcoin sending address?

https://explorer.byteball.org/#zD17W8iMLjb4Uxtt66NIcdhiCv9MCAtQFaBDY4jSXNo=

Unit zD17W8iMLjb4Uxtt66NIcdhiCv9MCAtQFaBDY4jSXNo=
Authors
UGP3YFIJCOCUGLMJUFKLXLNYNO4S7PT6
Children
V1UF7jRtqFUDWsOxmSLz23rTs7s0azPbpiNrXncsH74=
Parents
EjR+BnqoiGywB7WIEpELtxjyLku2bVylF8EcAjVGteE=
KSy9G0vl4PdpqZb9VCLJ2pnZyDtcUT+gE30O3uaX9uE=

Unit is hash of data unit, similar to bitcoin transaction ID, but unit can have more than just a transaction.
Authors are addresses who signed the transaction, similar to bitcoin sending address.
Children and parents are pointers to later and earlier units in the DAG (see the arrows between units).

Thanks

Also, in the advanced section of the wallet there was a string starting with xpub if I remember correctly. Is that like a Bitcoin public key, and is it safe to let others know it?

<edit>

This is the xpub string I referred to, and it is an extended public key for an HD wallet.
https://github.com/byteball/byteball-merchant

The merchant runs as a daemon and it doesn't hold any private keys. Instead, it knows only the extended public key and uses it to generate a new receiving address for each order.

xPubKey: extended public key of the home wallet that is to collect funds. To find it out, open the GUI wallet, click gear icon -> Advanced -> Wallet Information, look for "Extended public keys", double-click to select it (it starts with "xpub"), copy and paste into the conf.

I only know of one problem for Bitcoin HD wallet extended public keys. You can safely reveal an extended public key, but if you also reveal one private key from any generated in the same HD wallet an attacker can calculate all the other private keys from that wallet.


https://bitcoinmagazine.com/articles/deterministic-wallets-advantages-flaw-1385450276

The problem is this: although you certainly can securely hand out child keys with no risk to the parent key, and you can hand out master public keys with no risk to the master private key, you cannot do both at the same time. The exploit for when that situation does arise is actually quite simple

Does Byteball have this same problem with extended public keys and private keys? Is it safe to reveal your extended public key provided you don't reveal any of your private keys?

[tuvok007]

Where can I buy some byteball?

[HI-TEC99]

Where can I buy some byteball?

At this exchange

https://cryptox.pl/

In the Byteball slack

http://slack.byteball.org/

Or in this trading thread (make sure you use escrow)

https://bitcointalk.org/index.php?topic=1728405.0

[DiCE1904]

Hello tonych

Im running a witness and it crashes with this

events.js:160
     throw er; // Unhandled 'error' event
     ^Error: write EPIPE
   at exports._errnoException (util.js:1022:11)
   at WriteWrap.afterWrite [as oncomplete] (net.js:804:14)


And how come it says this "distance above threshold, will witness" in the logs, but I havent told anyone about the witness ID/public-key, how come it is trying to witness?

Does it make any sense to run a byteball-relay as well as a witness, the logs seem to be the same, is the witness also doing relay?

Is running a witness the only way to collect fees from transactions? Is there a guide on setting one up that you used?

[SatoNatomato]

Hello tonych

Im running a witness and it crashes with this

events.js:160
     throw er; // Unhandled 'error' event
     ^Error: write EPIPE
   at exports._errnoException (util.js:1022:11)
   at WriteWrap.afterWrite [as oncomplete] (net.js:804:14)


And how come it says this "distance above threshold, will witness" in the logs, but I havent told anyone about the witness ID/public-key, how come it is trying to witness?

Does it make any sense to run a byteball-relay as well as a witness, the logs seem to be the same, is the witness also doing relay?

Is running a witness the only way to collect fees from transactions? Is there a guide on setting one up that you used?

There is no guide, but I can help anyway, despite being possible competitors,

I cloned this https://github.com/byteball/byteball-witness

then edit the conf.js and run node start.js (or maybe I ran npm install before that)

So simple. Much wow. Also you could setup a reverse proxy and point its wss://you.com/bb to your desired port, open firewalls, and receive incoming connections, as specified on byteball/byteballcore or byteball-relay github.

[kola-schaar]

...
As we can see with only $10 billion market cap (less than the current market cap of bitcoin) storing 1 MB file in the Byteball system becomes expensive ($10). It shows that the design of the system (1 quadrillion cap of internal currency) precludes any widespread use case regarding data storing. Claiming that the system is scalable with regard to data storage is therefore unsubstantiated.

Please check if I did these calculations correctly as this looks like a rather serious issue.

Maybe you are right. But as I understand, byteball is not primarily intended as a mass storage:
•   page 33 Assets
•   page 37 Bank issued assets
•   page 38 None-financial-assets
•   page 38 Bonds
•   page 39 Commodity Bonds
•   page 40 Funds
•   page 40 Settlements
•   page 40 private payment
•   page 42 Fixed denominations assets
•   Page 45 Arbitrary structured data  (..this message type can be used to post Ethereum code for the subset..)
•   Page 45 Text
•   page 46 Voting
•   page 47 private messaging
Byteball is the wet dream of every banker. It is the Swiss army knife for the financial industry and / or Business economist, Accountant and Tax Consultant..
It‘s more like a very safe, secure, indestructible, tamper-proof paper. You can also use it to write books or letters, but it's actually too smart and expensive for this kind of use. But as tony explains (page 45 Text) you can store hashes.. - For me it looks flexible enough..

[davidoski]

Great calculations.

I do not expect the system to be a general purpose data storage though, storing 1MB is already too much for what people are storing in various blockchains already - which is certificates, and hashes of public keys or other such small things which are in the 4kb range.

If you are really really really rich then you can use it to store your mp3 files or your 4K movies in it.

So to me, this says, the design is solid, we dont want your 4k movies as transactions, but your 4kilobyte rsa keys are perfect, hell throw in a 128kilobyte message for fun.

System scales well.

Ok, so "data storage" is rather "just a little text storage".  This is not to criticize but to make things clear. "Data storage" and "scalable" sounds like god knows what.

[yvv]

Great calculations.

I do not expect the system to be a general purpose data storage though, storing 1MB is already too much for what people are storing in various blockchains already - which is certificates, and hashes of public keys or other such small things which are in the 4kb range.

If you are really really really rich then you can use it to store your mp3 files or your 4K movies in it.

So to me, this says, the design is solid, we dont want your 4k movies as transactions, but your 4kilobyte rsa keys are perfect, hell throw in a 128kilobyte message for fun.

System scales well.

Ok, so "data storage" is rather "just a little text storage".  

It is called metadata. DAG is great for storing this type of data. Of course, terabytes of audio/video/etc should be stored somewhere else.

[davidoski]

The other question that comes to mind is if a full node stores the whole history of DAG data. I couldn't find a clear answer in the white paper. In one place it is stated that the node stores "a current chain" which may be different than "a current chain" of some other node. It suggest that a node stores only a part of the whole DAG data but I'm not sure if this is really the case. If the full node was storing the whole historic DAG data it would be another issue with regard to scalability. It wouldn't be that much different than a blockchain with no blocksize limit (like Monero for instance).

[piebeyb]

Where can I buy some byteball?

You can buy it through their slack, maybe you can go there soon

[SatoNatomato]

Great calculations.

I do not expect the system to be a general purpose data storage though, storing 1MB is already too much for what people are storing in various blockchains already - which is certificates, and hashes of public keys or other such small things which are in the 4kb range.

If you are really really really rich then you can use it to store your mp3 files or your 4K movies in it.

So to me, this says, the design is solid, we dont want your 4k movies as transactions, but your 4kilobyte rsa keys are perfect, hell throw in a 128kilobyte message for fun.

System scales well.

Ok, so "data storage" is rather "just a little text storage".  This is not to criticize but to make things clear. "Data storage" and "scalable" sounds like god knows what.

Any takers on implementing byteball-bittorrent, that is popular these days with webtorrent and gittorrent, so now we need, bytetorrent, or torrentball?

Torrentball is a fancy name.

[SatoNatomato]

I wonder if it would be possible to modify the wallet, to have a list of witnesses, which are all owned by me.

Then sending transactions from my wallet to my other wallet.

The fees would end up with my witnesses? They would see it first, the would sign it.

Then my 12 witnesses would be modified to actually ask/pass on the transaction to the other real witnesses for confirmations. Eventually the transactions would be in the Main Chain, but the real witnesses would have no chance to collect the fee?

Would this work tonych?

[davidoski]

I wonder if it would be possible to modify the wallet, to have a list of witnesses, which are all owned by me.

Then sending transactions from my wallet to my other wallet.

The fees would end up with my witnesses? They would see it first, the would sign it.

Then my 12 witnesses would be modified to actually ask/pass on the transaction to the other real witnesses for confirmations. Eventually the transactions would be in the Main Chain, but the real witnesses would have no chance to collect the fee?

Would this work tonych?

No, the list of witnesses can't differ more than 1 from the rest of the network. (the White Paper - page 11)

[ArabMist]

Agreed. Encrypted messaging and transfer of value anonymously as byteball's best chance of success. It does those two very well

[davidoski]

Agreed. Encrypted messaging and transfer of value anonymously as byteball's best chance of success. It does those two very well

I wondered about this too To make a private transaction you need to first pair your device with your counterparty. Doesn't it reveal your ip?

[SatoNatomato]

Agreed. Encrypted messaging and transfer of value anonymously as byteball's best chance of success. It does those two very well

I wondered about this too To make a private transaction you need to first pair your device with your counterparty. Doesn't it reveal your ip?

You could do it over tor onion services or i2p hidden "eepsites" I guess?

On the other issue:

Does the transaction include the list of witnesses in an immutable way? So the first witness could not act as a proxy and rewrite the witness list?

[davidoski]

Agreed. Encrypted messaging and transfer of value anonymously as byteball's best chance of success. It does those two very well

I wondered about this too To make a private transaction you need to first pair your device with your counterparty. Doesn't it reveal your ip?

You could do it over tor onion services or i2p hidden "eepsites" I guess?

Obviously but that doesn't sound like anonymous payments if I have to hide behind Tor to make it anonymous.