[SKY] Skycoin Launch Announcement

[chompyZ]

I have yet to see an IPO mechanism that 'solves' the equal distribution issue, or the dumpers issue.
The best long-term solution is imho relaying on POW, where most of the coins are mined in the future, thus even the most greedy actor can consume that much coins in the present. This is of-course in complete contrast to PoS coins that suffer from very high and growing concentration of wealth between a few.

In any case, I also think dev is putting too much emphasis on the IPO mechanism, where what really matters is the coin technology, innovation and smooth launch.
At the end of the day, it is the market that decides about the price, as well as about the coin's viability, rather than the early adaptors and the short-term actors.
my 2cents.

[e1ghtSpace]

4> The server running bitcoind for the IPO, is taking over four days to download the bitcoin blockchain. I keep deleting peers.dat and restarting it and it goes for a few hours and then stops. Slow peers are slowing down the download.

I used the blockchain torrent to combat this issue and it only took 1 day. However I'm not sure if it works with bitcoind but since the qt essentially is bitcoind with a GUI it probably would work.
Or maybe you could put the torrent into another computer (if you can'to put it straight onto the bitcoind server) on the same network and make the bitcoind use that computer to download the blockchain.

[lovely89]

I have yet to see an IPO mechanism that 'solves' the equal distribution issue, or the dumpers issue.
The best long-term solution is imho relaying on POW, where most of the coins are mined in the future, thus even the most greedy actor can consume that much coins in the present. This is of-course in complete contrast to PoS coins that suffer from very high and growing concentration of wealth between a few.

In any case, I also think dev is putting too much emphasis on the IPO mechanism, where what really matters is the coin technology, innovation and smooth launch.
At the end of the day, it is the market that decides about the price, as well as about the coin's viability, rather than the early adaptors and the short-term actors.
my 2cents.

Core technology work = 3 years.
Distribution method ~~ 2 months?

Also, distribution will be done over years but it will be much more difficult for anyone who doesn't contribute to the project to gain a large stake, excluding the developers.

The extreme distribution problems seen with nxt and bitcoin will be minimised. Instead the people who designed and built the platform will be in control. Unless a fool proof 'fair' distribution method is realised, this seems like the best solution for a long term fair spread of the coin.

[dance]

IPO Update

The IPO cannot work the way we originally intended. There is no way to cap the number of Bitcoin per person that can participate in the IPO.

Browser finger printing does not work. It cannot reliably detect which users are bots and which are not. Bot nets and even proxies still get through.

Then we were going to do IPO over Bitmessage, but with validation of Bitcoin talks username. However, there are people who have several thousands Bitcoin Talks accounts registered. These users would be able to participate in the IPO multiple times, but normal users would be unable to.

You can do facebook page check and phone number check. It will make sockpuppeting much harder.

The other frustrating thing,
1> The bitmessage library for golang we were using does not actually work. We had to rewrite the IPO script into python. Ideally, we would have used JSON/HTML requests on Bitmessage and use golang, but the Bitmessage API implements XMLRPC... (There is a massive effort underway to rewrite Bitmessage on C++ and create language bindings for each language, so these problems should decrease over time)
2> When building the exchange for doing the IPO, we had to use bitcoind. There is no RPC in bitcoind to check the balance of the unspent transaction output set for an address. We originally intended to generate a list of private keys and addresses on a computer never connected to the internet and move the address list to the server running the IPO. These offline addresses were to be used for receiving IPO funds and their balance queried by the server. This would prevent bitcoin from being stolen if the IPO server was identified and hacked. However, this is impossible under the Bitcoin API, for the same reason that Bitcoin makes building mobile thin clients very frustrating. We have to do the IPO with a hot wallet and move the coins into the cold wallet over time instead of just receiving the coins into a cold wallet address. We thought there was some way to get the unspent outputs for an address with bitcoind, but this is apparently very difficult.
3> Some git accident happened and the deterministic wallet input feature in the Skycoin wallet GUI seems to have disappeared. This feature is used in the IPO for importing the wallet the addresses coins are received into.
4> The server running bitcoind for the IPO, is taking over four days to download the bitcoin blockchain. I keep deleting peers.dat and restarting it and it goes for a few hours and then stops. Slow peers are slowing down the download.

Maybe use NXT instead of Bitcoin?

[coin@coin]

Thanks for the update and looking forward to the the IPO, no rush, get it well organized and ready first. 

[asnghost]

Thanks for the update and looking forward to the the IPO, no rush, get it well organized and ready first. 

Yes, I agree with above.

[ern]

If only 2% to 10% will be sold during the IPO, how will the rest be distributed?

[skycoin]

Update:

Two separate people are independently working on the IPO script. Should get done soon.

Another dev is trying to fix the merge conflict and error that is appearing on the github repo.

The main dev doing the design work is disorganized and prone to burn out. We are forcing him to get an engineering project management and to write things down, so other people can do them.

Documentation and descriptions of software libraries that need to be written, will be placed on Github immediately. We are not waiting for Aether/Merkle-DAG to be completed to release documents. Document releases will be smaller, more frequent.

Tutorials for developers are going on a github wiki.

Initially we thought, the darknet would be working fairly quickly and we could run the whole client on top of it. However, now parts of the coin could not be finished because it required infrastructure that was not implemented yet. We are moving away from relying upon infrastructure that is not completed and focusing on what we can get done now

In particular
- the darknet will be in Golang, not C
- the darknet will be separate from the daemon for the coin, until its ready
- the coin will use existing modules and will not transition to Merkle-DAG for block storage, until Merkle-DAG is working

We will release a prototype consensus algorithm and run the chain on top of that, with some safe guards (developer signing key) to deal with attacks/errors that arise. This will protect against loss of coins, while allowing us to launch. The consensus mechanism will be deployed in stages, instead of trying to make it perfect and 100% implemented at launch (which may delay launch another year).

[LemonAndFries]

Update:

Two separate people are independently working on the IPO script. Should get done soon.

Another dev is trying to fix the merge conflict and error that is appearing on the github repo.

The main dev doing the design work is disorganized and prone to burn out. We are forcing him to get an engineering project management and to write things down, so other people can do them.

Documentation and descriptions of software libraries that need to be written, will be placed on Github immediately. We are not waiting for Aether/Merkle-DAG to be completed to release documents. Document releases will be smaller, more frequent.

Tutorials for developers are going on a github wiki.

Initially we thought, the darknet would be working fairly quickly and we could run the whole client on top of it. However, now parts of the coin could not be finished because it required infrastructure that was not implemented yet. We are moving away from relying upon infrastructure that is not completed and focusing on what we can get done now

In particular
- the darknet will be in Golang, not C
- the darknet will be separate from the daemon for the coin, until its ready
- the coin will use existing modules and will not transition to Merkle-DAG for block storage, until Merkle-DAG is working

We will release a prototype consensus algorithm and run the chain on top of that, with some safe guards (developer signing key) to deal with attacks/errors that arise. This will protect against loss of coins, while allowing us to launch. The consensus mechanism will be deployed in stages, instead of trying to make it perfect and 100% implemented at launch (which may delay launch another year).

+1

[provenceday]

still watch this.

[Korean]

How well does the Meshnet work right now?

Can it get pass the great firewall of China?

Is there a standard hardware that you recommend?

[SomethingElse]

How well does the Meshnet work right now?

Can it get pass the great firewall of China?

Is there a standard hardware that you recommend?

+1 I would really like to know if the meshnet could break the great fire wall

[coin@coin]

How well does the Meshnet work right now?

Can it get pass the great firewall of China?

Is there a standard hardware that you recommend?

I don't think the meshnet is close to be ready, I believe it'll come at a later stage after the rest is up and running and stable. But I may be wrong.

[dance]

As evidenced with Cloakcoin, the infrequent updates, erratic messages, refusal to accept community support, and difficulty establishing communication with the developer destroyed what was a promising coin; I would hate to see Skycoin go down the same path.

Cloakcoin never was a promising coin. Only as P&D on technologically ignorant people.

[skycoin]

Update

We have library now, for checking the unspent outputs for addresses. We are using obelisk tool kit and it is working well. The bitcoin blockchain synchronization has been running for a week now. It is on block 285,100 of block 323,354. It is 70 GB on disc and we dont know why. It has slowed down periodically, but has not stopped completely like bitcoind has. It is downloading a block a second. Blockchain synchronization should be finished within a hundred thousand seconds in the worst case (hopefully) and then we can test to make sure the IPO software works.

A basic exchange will run over bitmessage and possibly IRC. Bitmessage will have a few problems. You have to download 20 GB of messages before the message response appears. The server will also only be able to send out a few messages per minute because of the PoW requirement on the message. Another solution, might be a web server with cloudflare and it might be enough, depending on the DDoS conditions. A tor hidden service is another option. The Skycoin darknet is not ready to host this application yet because required components have not been implemented yet.

We want to get the IPO over and get back to the darknet and encryption utilities.

[nonlinearboy]

Update

We have library now, for checking the unspent outputs for addresses. We are using obelisk tool kit and it is working well. The bitcoin blockchain synchronization has been running for a week now. It is on block 285,100 of block 323,354. It is 70 GB on disc and we dont know why. It has slowed down periodically, but has not stopped completely like bitcoind has. It is downloading a block a second. Blockchain synchronization should be finished within a hundred thousand seconds in the worst case (hopefully) and then we can test to make sure the IPO software works.

A basic exchange will run over bitmessage and possibly IRC. Bitmessage will have a few problems. You have to download 20 GB of messages before the message response appears. The server will also only be able to send out a few messages per minute because of the PoW requirement on the message. Another solution, might be a web server with cloudflare and it might be enough, depending on the DDoS conditions. A tor hidden service is another option. The Skycoin darknet is not ready to host this application yet because required components have not been implemented yet.

We want to get the IPO over and get back to the darknet and encryption utilities.

so still have chance to join in the IPO?
just waitting for IPO.

[skycoin]

How well does the Meshnet work right now?

Can it get pass the great firewall of China?

Is there a standard hardware that you recommend?

I don't think the meshnet is close to be ready, I believe it'll come at a later stage after the rest is up and running and stable. But I may be wrong.

Once the IPO is released, I was wondering if we could have more direct leadership from the Skycoin developers.

We need a direct line of communication with the developers so that we can discuss strategies on how to increase adoption, marketing, and coordinate other efforts to make this coin a success. We need the community to work on how to simplify the technology to the masses and how to channel their hatred towards the big internet companies to increase the adoption. A constant communication with the developers will ensure that our efforts are in line with their vision and that we will be progressing forward not backwards.

Furthermore, when we do launch the meshnet, there will obviously be a lot of technological hurdles to overcome. We need technical support from the developers to help overcome any of these problems that our users may experience. Otherwise frustration from deployment will lead to many users giving up.

While these sporadic updates are fine for now, this cannot continue once the coin is released. A more constant presence is needed.

As evidenced with Cloakcoin, the infrequent updates, erratic messages, refusal to accept community support, and difficulty establishing communication with the developer destroyed what was a promising coin; I would hate to see Skycoin go down the same path.

Thank you

Yes. Right now development is disorganized. Trying to get all developers in one place. Skycoin has a large number of active developers, but very few of them are working at one time, because no one knows what needs to be done. There is not enough project documentation and technical specifications.

I think there needs to be a developer chatroom. Marketing and most everything outside of the core cryptography and blockchain, will be handed over to the community after launch. After the darknet is up, communication will increase significantly.

The meshnet will be a massive coordination effort. It requires people on ground and several different hardware teams. It will probably require ASICs eventually and its an undertaking far larger than the coin implementation. Right now the focus is finishing the coin IPO, then getting the darknet working at basic level. This will enable file sharing (distributed file system), websites and communication tools. The darknet will still be useful, even if it is merely running over traditional internet.

Then we can add mesh, peer-to-peer connections to the darknet topology. Then eventually tunnel normal internet traffic over the darknet/meshnet to a peerage point and transit back to the normal internet (similar to a VPN, but running over meshnet to exit point). Just getting those simple steps working, would be a great achievement.

The good news, is that after much frustration and coding, we discovered that the meshnet/darknet implementation is actually much simpler than imagined. The internet will face palm at the power and elegance of the resulting architecture.  The implementation is simple enough, that there can be no unanswered questions about its security and the performance and privacy are beyond Tor. Tor is over 180,000 lines of code and this method is clearly superior with significantly less complexity and higher performance.

[skycoin]

How well does the Meshnet work right now?

Can it get pass the great firewall of China?

Is there a standard hardware that you recommend?

+1 I would really like to know if the meshnet could break the great fire wall

Yes. It was designed explicitly to defeat existing state-full packet inspection hardware. This includes the great firewall.

There is no fixed port, no protocol headers. There is not a single byte of plain text traffic in wire protocol, which can be used for identification of the protocol type. The traffic is complete noise. Disguising the traffic to be recognized as SSL or HTTPS is very easy.

The darknet is designed to run over IPv6. If you are tunneling over clearnet, you can potentially create thousands of different IPv6 addresses for your server, have each connection use a different IP address and rotate out IP addresses. If you are only connected to a fixed set of hand chosen nodes, there is no way to trace your node public key back to an IP address.

The traffic can only be identified through a statistical analysis. Blocking the traffic effectively requires installing hardware close to the end-user and for controlling traffic within the data center at peerage and cross-connect points.  Detecting and blocking access at the edge of the network, is much more practical than blocking it at the center. However, doing this requires installing large amounts of very expensive hardware.

Blogs, twitter feeds and other static content on the darknet would be stored in Merkle-DAG format and peer-to-peer replicated. So syncing data in china only requires the data leave border once (once a copy of the data is outside of the border, its replicated peer-to-peer between the subscribers and each chunk has signature checked against public key of the publisher). A satellite uplink, or few clandestine connections to Hong Kong might be enough.

China is focused on blocking traffic entering and leaving the country (at border), which is tightly controlled. The focus is on blocking access for end-users. We have to do test to see if they are able to block traffic internal within China (which we doubt). We also think there is minimum ability to block traffic within and between data-centers within the country. Most of the blocking will be at the border between China and rest of internet and at the ISP level on the individual user.

[coin@coin]

Try using Slack for now before the DarkNet. This is pretty good, I hear.

https://slack.com/is/team-communication

Do you have plans to make a version of FireChat? Mobile Meshnet on Android or iOS?

I agree with JessicaMILFson, Slack is the way to go if you want to keep things going within a "distributed" team.
Web app, desktop app and Android + iOS apps.

I use it myself for other projects and is one of the best tools.

I see you guys use Trello already there is an integration for Slack too, together with GitHub and GoogleApps, SupportFu, etc, etc.
You can have as many tools as you use integrate with it and keep things organized in one place.

There is a free version for unlimited users and 4/5 integrations but less options, and a $7 per user per month with full features like
"restricted channels", more integrations, restricted users, etc... Give it a try but I guess you guys have looked into it already.

Also a Forum would be nice, it'll get the Community involved and one place where everyone interested can join the discussion.
I'm happy to contribute with Community, Social, Support, Publishing, Events and Slack, sent you a PM with some references.

[statdude]

Try using Slack for now before the DarkNet. This is pretty good, I hear.

https://slack.com/is/team-communication

Do you have plans to make a version of FireChat? Mobile Meshnet on Android or iOS?

I agree with JessicaMILFson, Slack is the way to go if you want to keep things going within a "distributed" team.
Web app, desktop app and Android + iOS apps.

I use it myself for other projects and it's one of the best tools.

I see you guys use Trello already there is an integration for Slack too, together with GitHub and GoogleApps, SupportFu, etc, etc.
You can have as many tools as you use integrate with it and keep things organized in one place.

There is a free version for unlimited users and 4/5 integrations but less options, and a $7 per user per month with full features like
"restricted channels", more integrations, restricted users, etc... Give it a try but I guess you guys have looked into it already.

Also a Forum would be nice, it'll get the Community involved and one place where everyone interested can join the discussion.
I'm happy to contribute with Community, Social, Support, Publishing, Events and Slack, sent you a PM with some references.

I was just about to say: Slack + Trello is the way to go.