Greetings,

The implementation has been reviewed by interested developers, some bugs were found and closed, the discussion of cryptography/schemes happened on this forum before the launch. Correct, it is not trust-less, requires to trust 12 honest/well-behaved/incentivized/fee-collecting nodes/entities called witnesses on the network. There is no probabilities of confirmation times or similar, a transaction can be in two states, unconfirmed or confirmed.

There has been lower amount of technical/design discussion last week due to high price and everyone clapping, its about time someone got to the technical/criticism parts again.

In the Byteball wallet you have "seed backup" which gives you about 12 words to remember/write down. Then you can on another computer enter these words and it will restore your bytes balance.

However, that does not restore any blackbytes, so easiest and best way is to, on old computer (A), pair it with new computer (B) and require "1 of 2" signatures to spend. Then you can verify with small transactions from new computer (B), and then you can delete/trash the old computer.

Byteball wallet has multi-signature support built in. Which acts as a good backup.

Check in "Global settings" options for backups, you can also take a backup on a usb-memory, or encrypted full backup and save it on Google Drive for example.

The flight oracle is run by tonych so there is too much incentive for him not to lie, and he has very much already earned street cred.

Most oracles should not be anonymous though, unless you think shady deals.

You trust the oracle, the oracle knows because it is asking/requesting the data from flightstats.com it would be possible to construct the oracle so it uses more sources, but even an normal insurance company has to trust someone to tell when a flight has landed or not.

If the oracle lies but flightstats.com does not, you may be tricked, but in that case its open to the public and you can show that the oracle lied - hence the oracle can go to hell and nobody will use it again.

Its better than an insurance company because even me and you can be the insurer now, and more transparent than dealing with them.

Thanks for the frequent upgrades, better those than crickets and promises. 

Remember to take rest, do not overwork, chill and relax.

Technically, all the bytes are generated/created in the genesis transaction unit, for tonych to create more, would require breaking same crypto as breaking bitcoin to create more coins. The same cryptographic library is used, secp256k1.

This is not a bitcoin clone. Its not another altcoin.

He can do other bad things, since he is now in control of 12 witnesses, he could stall or destroy the network, or double spend - but such actions would be obvious to everyone looking.

I, for one, prefer to trust one main developer, one person, rather than a group with their dirty dynamics always going on, for this significant yet fragile phase of the project. As the project matures, so will many more witnesses appear and community gain their trust and revoke tonychs control of 12 witnesses, when the project is stable, then we can have drama and invest in popcorn.

Of course its free and open source software, https://github.com/byteball/byteballcore

You can also see the source of the transition bot.

---

Notice, if you use a different hub, the default GUI client will pull the witness list from the hub, so the byteball.fr hub will push its witness. Just a friendly reminder.

Hopefully people and the community will stay sane, as the price increases more trolls will appear here and other places to troll and FUD.

Well, you can send 1 byte to someone, you will just pay the transaction fee which is on average 520 bytes.

Upgrade to latest version 1.7.1

Yes, you can connect many btc addresses to the same Byteball wallet.

The only good iota project is this https://github.com/froggey/Iota

We cant FOSS the prototype as parts of it are used in the current iteration, the most meaningful parts of course, which is the porting of secp256k1 library, thats basically all that was missing on current esp8266 sdks.

The other problems faced was actually understanding and working through the Byteball light-client js code, and testing.

Now for the next iteration, see the problem with ESP8266 and similar chips, Realtek has a newer almost similar but slightly faster, is the power/energy consumption.

In IoT, especially for things which are sensors, they arent usually connected to power-grid, and run on batteries. ESP8266, despite being weak ass shit in general computing, 80Mhz, if you overdo it can do 160Mhz, with 96kb ram, it still uses on average a whooping 300mAh and peaks at 800mAh. Optimize all you want but you aint gonna get a usable sensor lasting more than 6 months on 4 AA batteries, if it will report something even each 5min.

And if you put 10 000 sensors somewhere you aint gonna go replacing them all every 6 months. We need to go deeper, lower in energy used, to last years, and yet still make secure WebSocket connection and perform Byteballs ecdsa (secp256k1) and sha256, this is the current challenge and it is very fun.

ESP8266 is like a fat cow compared to whats needed.  

Well, you should stop telling false information to your imaginary clients, as obviously you know nothing about Byteball, you should be a good sport, a good person like your mother tried to teach you to be, and direct people to ask in Byteball slack or Tony.

You cant win with troll.

Now you are just changing your original question to be "what is the result of a stresstest", and if tonych actually makes a stresstest you would just say "but oh you run the stresstest with AMD Ryzen and more powerful motherboard than what is available for normal people living in Moldova".

Byteball is the first DAG-coin and that hurts your little ego since Iota is still not on any exchange and has the retarded idea of using Proof-of-Work in the space of IoT.



Use this https://bitcointalk.org/index.php?topic=1608859.msg18291320#msg18291320 as tonych gave you answer to your "scalability" question previously.

As more transactions are posted to the network transaction times decrease - throughput increases.

Just pathetic CfB being pathetic.

Witnesses make transactions and that costs just as any tx, if the witness isnt chosen by many in the network, itll loose more than it earns.

I see commits which add a dependency to google service, for the push notifications, not good for privacy, but have to check more.

First DAG-coin. First.

It is nice to have our own dedicated troll, especially an enterprise-grade Iota-coin developer.

Ill write a short script to check any time troll is writing in this thread, then it means Byteball has done something very good, time to buy more.

First DAG coin to actually be in livenet/production without people having to ask "developer please include my balance as it was on this date in the next milestone, look in your own books I do indeed own that much tokens" form of  deciding consensus.

Great work!

Which witness in the default 12 list would you like me to replace?

He answered you already. Several times.

By the way, your IOTA-scam still sucks, very shameful project, its implementation is weak and concept is retarded. Proof-of-Work in IoT space is beyond retardation, only a simple narrow-mind could have thought that scam would fly. Your IOTA project has a "consensus by developer keeping a database of everyones balances", has only censored forums, and is a failed scam.