A lower rate is definitely what we want for the reasons you listed.  One thing this opens up is more elaborate circumvention of the limiters we currently have because it removes the capcha entry time from the equation.

We could add authentication to a social media account and do verification on account age. I did some research on this a while ago and none of them seem forthcoming with an "account created" time but there are some heuristics that can be applied.

It also seems like some people aren't keeping what they get from the faucet, they turn around and sell it so it goes to whoever can find the most contacts who will trade kind of defeating the usefulness to anyone not doing this.

Yea that'd be excellent.  We'll put together a script with the purpose of being drop-in code.

I think the biggest advantage of the freemium ad concept is if people don't understand it, they can just ignore it and they'll just get ads like normal.

Ahh I see, the representatives do very little different than a plain wallet, since they're supposed to be long term things it's probably best to run a command line version on a server rather than a GUI on your desktop but I run them on cheap 3$/mo VPSs with no issue.

Setting one up involves running a node that just has a modest amount of disk space and network bandwidth, CPU usage is almost non existent.  You can then pick an account and tell people, "this is my representative account, set me as your representative" and they can do that under Settings -> Change representative.  When they do that it'll take the representation away from their old rep and give it to you and your vote strength would increase by the amount of their balance.

What your node will be doing is watching the network for transactions and signing them.  This lets the nodes in the network know quorum to make sure they haven't been disconnected.  The representative also watches the network for forks and if it sees them, it participates in the vote to pick which block survives.  Forks are also an almost non existent thing since they're only created by people who changed the code to create them.

I think this method is an incredible benefit to scalability and that's one of the things I'm most excited about and hope people share in that.  Proof of work verification is several million times easier than generation, despite the fact that I had a 24core system being saturated running the faucet because the faucet was creating transactions, the VPS 1 core servers that are the representative nodes were sitting at less than 1% CPU utilization.

I really liked your trade thread, thanks for doing that I think a lot of people wanted it!

The trade thread had a side effect of creating that secondary faucet mining market but that's to be expected, my only miscalculation was in how many people were willing to sit clicking captchas for hours for essentially no money.  So I created an artificial scarcity that's being satisfied by people willing to spend time for xrb which they in turn sell for BTC to people who don't want to spend time clicking the captcha.

The initial distribution is my absolute least favorite part of this whole thing because no matter what happens, someone sees it as unfair.  I would much rather be spending my time on the technical aspects and helping people setting up systems that make use of rai nodes.  There are some doing setting them up and unfortunately my time has been saturated by a mountain of faucet requests, which I don't see as providing long term value.

It's off because we've been over target the last couple days and we're focusing on making sure it works going forward.

Everyone concerned with the long term health of the project is primarily concerned with the health of the network itself, first and foremost.

For the PoW, 1 proof is generated on the send and 1 proof is generated on the receive, this is why people will see CPU activity when receiving from the faucet, the faucet itself also generating its own work.  Other than that there is no PoW in the network, only when transactions are being added and the work is only applied to those doing the transactions.

Taking the example of RFCs and how they're maintained, how do you see their incentive structure being similar or different than something like this?  From my perspective, groups maintain the TCP standard without TCP giving direct in-protocol incentives for its use, people build on top of it which is where their incentives lie.

There's been some interest in where I see the long term position of RaiBlocks, does it have a business plan or a way to generate funding.  I don't view RaiBlocks as something that should have or needs a business plan; it's more of a technical specification that people can build other real revenue generating businesses on top of. 

The goal is to have minimal ongoing expenses so any cost of operation could be rolled in to a normal cost of business like bandwidth or storage investment.  Ideally I think RaiBlocks is something that deserves an IETF RFC that's maintained by by the industry but needs almost no change, maintenance, and no investment.

The last week of faucet distribution, despite being an enormous burden to core development and a distraction to issue fixing, has been a great network stress test.  All of the representative nodes run on VPS servers with 1 shared CPU core and had no problem processing the load.  The network is designed so nodes that experience load are those that are sending or receiving balances in proportion to their individual activity.

Today we doubled our participation rate again which is great news for popularity.

We let the faucet stay at it rate we set a few days ago even with the increased traffic so it's close to empty again and will be for a bit while we calculate rates.

I'm going to take some time this afternoon to discuss some of the technical questions and roadmap goals asked in the last few days.

Thanks for all the feedback everyone.  I'm glad to see in general things are working with a few issue reports.

I'll be taking time tomorrow to write up some of my thoughts on next steps.

There's a lot of people at it today.  I'll be back later to look at it closer but people are getting handouts fairly consistently.

I'm glad everyone is excited and I want to set the expectation that the faucet serves the distribution schedule and the purpose of getting it to a variety of people.

From time to time this may mean taking it down and changing how it works.

Even after uninstall it leaves the data in the folders at the bottom of the page in that link.  You can save those backups though they're encrypted.

If you want to clobber it you can delete those directories and start the wallet up again and it'll regenerate.

When the wallet starts up it tries to unlock the wallet with an empty password.  If this doesn't work it shows "Wallet Locked".

You can load your wallet seed https://github.com/clemahieu/raiblocks/wiki/Wallet-Backups

Did you enter your password?  Settings -> Password -> Unlock

Either tonight or tomorrow for the faucet.  The RaiBlocks network itself won't be affected and will remain up.

Today we have twice as many nodes connected as ever before and we reached the daily distribution target.

At 11pm GMT ~20 min we're going to take down the faucet for a combined update to that and the site.

The faucet isn't meant to be a job, that's why it's low.

I know everyone has experience with cryptocurrency but it's worth reiterating caution.

Right now this network has been actively used for 1-2 weeks and it needs to continue to undergo more rigorous peer and code review.  I have confidence in the system but it's still software and software can have flaws.

Please make sure you're being conservative with your decisions.

We only have a basic block explorer right now.  https://raiblocks.net/#/block-explorer and in the wallet you can get a more complete listing of the whole ledger.

We're definitely looking to add these.

It should be pretty straight forward. Once the wallet has synced and it's unlocked, your balance should increase.

If it says "wallet locked" or "synchronizing" your balance won't update. Synchronizing shouldn't take more than 10 minutes depending on your connection and drive speed.