@irritant, I like your idea of customizing the RPi-version of the client / daemon to “sign” blocks in some way. I don’t think it has to be something that is personally identifiable, but it would be useful for network analytics.
Does anyone have concerns with adding this functionality, assuming it’s technically feasible?
Hi irritant,
I am no networking expert, nor am i a P2P expert and really I’m only a CryptoMoney enthusiast. Definitely, I am a newbie and trying to learn.
That said, I don’t think the bandwidth consumption issue is directly related to the blockchain size issue. Rather, so far as I can tell when one opens-the-port or port-forwards one’s node goes from one way communication with a maximum 8 other nodes to two way communication with up to I think 125 other nodes. Guys describe data rates in the megabytes per second continuous and their ISPs throttling them back, or they exceed their terrabyte data limits in only a short period of time, etc. So, some of these guys on home and even commercial connections have closed the port and gone back to 8 connections default. At least that is the best as I recall, it has been a while since I looked into this but here is one thread that I found influential:
http://www.reddit.com/r/Bitcoin/comments/1uh38f/why_has_the_amount_of_bitcoin_nodes_fallen_30_in/
If you come to a different conclusion, let’s discuss it.
Thanks for everyone’s interest, and again thanks to Tea42. I hope many of us can now get Peercoin 0.4.0 RPi going in the next week or two so we have the best experience base to go forward with.
[quote=“river333, post:59, topic:2152”]So I thought I’d post a possible plan for doing this. Please let me know whether it is ok or needs changing.
Sample person Bob wants to receive a tip.
Bob obtains a Raspberry Pi.
Bob follows Tea42’s guide to set up Peercoin on his Pi: Cryptoblog - notícias sobre bitcoin e criptomoedas!
Bob posts on the forum with a photo of his Pi, with his PeercoinTalk username visible in the photo. A recent newspaper is also possible (tabloids lead to an automatic disqualification though 
).
Bob sends a PM to the project organizer with a photo of his RPi screen with RPi serial number visible, and his PPC address.
The project organizer sets up a unique account on github for Bob.
The project organizer uses this account to make a commit on github, and inputs Bob’s PPC address on peer4commit.
The project organizer posts proof of the transaction on the forum, and Bob hopefully confirms receipt of the tip.
Bob can go on to turn his RPi into a full node or mint if he so chooses, but that is up to him and separate from this project.
Bob acquires a lifelong love of all things Peercoin and learns the importance of decentralization.
All good? If this is ok then a new repo for the project needs to be set up on github, and added to peer4commit. NewMoneyEra can send the PPC to be used for tips to the peer4commit address, if he is happy with this set up. I would recommend doing this in increments though, there’s no point having a bunch of PPC locked in the peer4commit fund if only a few people end up availing of the tip.
Please post your comments. I’m assuming that Tea42’s guide is well suited to this? I’m not an expert on this so I’ll need confirmation. Also open to other suggestions on verification.[/quote]
Wow. River333 you don’t let any grass grow under your feet. You get it done! I appreciate that 
Regarding your steps 3-7, I am wondering if it has to be this complicated?
3. Bob posts on the forum with a photo of his Pi, with his PeercoinTalk username visible in the photo. A recent newspaper is also possible (tabloids lead to an automatic disqualification though :P ).
Bob sends a PM to the project organizer with a photo of his RPi screen with RPi serial number visible, and his PPC address.
The project organizer sets up a unique account on github for Bob.
The project organizer uses this account to make a commit on github, and inputs Bob’s PPC address on peer4commit.
The project organizer posts proof of the transaction on the forum, and Bob hopefully confirms receipt of the tip.
I have the deepest respect for Sigmike and the enormous value he has created for Peercoin with peer4commit, but I am not yet convinced this software-improvement model best fits our purpose. Also maybe we don’t have to have github for this project?
This doesn’t have to be multisig but it would be cool and we would be showing we are up to speed at Peercoin and promoting the best security.
Imagine that we create a 2 of 3 multisig address to distribute tips out of.
(skip this section if you don’t want to be bored by the numbers;)
Suppose, I donate starter Peercoin that is enough for 100 tips of 10 full PPC each. Counting the transaction fee that would be 10.01 times 100 or 1010 PPC total. We recruit three trustworthy Peercoin tip verifiers here from the forum. The multisig tip requires any two signatures from the three verifying keys held by our three tip verifying forum members. (technical area: it would be nice for the first two verifying signatures to receive 1 PPC each automatically (help here please Sigmike) to reward them and motivate early prompt verifications). If this is possible then to fully fund 100 verified RPi Peercoin installations it would then be 1010 plus 201 or final amount 1211 PPC to make 100 RPi installations. I am totally good with this number. Sorry to bore with numbers here.
So, here is my imagination on steps 3-7 modified/simplified:
[s]strikethrough[/s] is delete, and [u]underline[/u] is newly added.
Bob posts on the forum with a photo of his Pi and his Pi screen with his PeercoinTalk username visible in the photo. A recent newspaper is also possible (tabloids lead to an automatic disqualification though
A. a window showing his RPi serial number; and
B. an image of his Peercoin Qt wallet with the green checkmark in the lower right hand corner indicating that his RPi is now a currently synced Peercoin node; and before taking the picture Bob has clicked the “Receive coins” tab so
C. Bob’s wallet “Address” to receive coins is visible in the picture; and to aid the verifiers sending him his tip
Bob must also paste the digital version of his receiving address into his forum post.Bob sends a PM to the project organizer with a photo of his RPi screen with RPi serial number visible, and his PPC address. Anyone of the three verifiers checks Bob’s RPi serial number against a list of numbers that tips have already been sent to and also checks items B for the green check mark, and C that the screen address matches the digitally supplied address, then signs a 10 PPC tip for Bob (actual transaction value 12.01 PPC = 10 tip 1+1 to verifiers and .01 transaction fee) and sends the signed transaction to the other two signers/verifiers. It would be nice if somehow this could be done publicly to show everyone how it is done. But whatever. Anyway, one of the two other verifiers signs Bob’s tip and transmits it to the network. The first two verifiers automatically receive 1 PPC each for their efforts and service.
The project organizer sets up a unique account on github for Bob. Bob receives his tip in his Qt wallet and anyone who wants to check on the blockchain can see that he did receive it.
[s]6. The project organizer uses this account to make a commit on github, and inputs Bob’s PPC address on peer4commit.
- The project organizer posts proof of the transaction on the forum, and Bob hopefully confirms receipt of the tip.[/s]
Would this work? With the multi-sig? Or, even simpler without multi-sig?
Also, how could this protocol be scamed?
Thank You for your help in this project.
NewMoneyEra
Edit: a few protocol clarifications
One caveat, I’m not trying to cheap out here as I’m going to donate the same amount no matter what. But, the 10 PPC tip was conceived when PPC was around $2 each and now is ~$2.50. I’m totally good with that as a RPi costs about $60 all up. If the value of PPC were to jump up I would think we should lower the tip amount, and conversely if PPC price were to plummet down we should raise the tip to a reasonable portion of the cost of an Raspberry Pi. Almost like we are subsidizing the RPi purchase so long as it serves at least initially as a Peercoin node. In other words, I don’t think we should buy the whole thing - the end user (tip recipient) should have at least some of his own skin in the game.
Thoughts?
Makes sense to me regarding value of tips
Regarding the tipping, agree peer4commit is not best suited for smaller fixed value commits. A while ago super3 launched a plan to build a social faucet, that would work for this. The advantage would be that it is more transperent assuming it works they way I think it should work. Anyone knows where super3 is at with this?
Don’t know much about the new multisig you are talking about, but it sounds good to me. Just needs some supporting links to the ledger in a thread.
Hi Cybnate,
Thanks for the positive feedback. Multisig is the hot new protocol which has been integrated into Bitcoin and is also in Peercoin protocol. Many of the most savvy think multisig will greatly enhance security against all kinds of theft of cryptos. In the most common version there are three keys to control the spending of the funds in one multisig address. When you make the multisig address you can set it up so that either one, or two, or all three keys are necessary to spend the funds. The most common arrangement is requiring two of the three keys to spend the funds.
In our case for giving tips to RPi verified Peercoin nodes three forum members who want to help verify nodes and send tips would each have their own private key. Only two of those keys are necessary to spend, that is to send the tip, so if one person gets too busy, or sick, or whatever the other two can still send a tip.
In more normal circumstances multisig can mimic many of the security checks we have already built into our fiat financial systems.
Here is a great video on multisig: https://www.youtube.com/watch?v=zIbUSaZBJgU
This guy does some really great Bitcoin videos. He recently did one on 51% PoW attack costs that still has the BTC forums reeling.
Multi-sig certainly looks to be better suited to this, good idea NewMoneyEra.
If it works it can also be hopefully used for other similar projects.
[quote=“river333, post:67, topic:2152”]Multi-sig certainly looks to be better suited to this, good idea NewMoneyEra.
If it works it can also be hopefully used for other similar projects.[/quote]
Hi river333,
I think multisig could be a cool way to do this and like you said may hopefully be used in other similar projects. That said it is a bit more complicated and new, so if you would want to get started by personally giving out the tips I will go with that as I totally trust you. You have clearly shown your good will on the forum in the many good acts you do.
Thank You for all the good things you do,
NewMoneyEra
[quote=“NewMoneyEra, post:68, topic:2152”][quote=“river333, post:67, topic:2152”]Multi-sig certainly looks to be better suited to this, good idea NewMoneyEra.
If it works it can also be hopefully used for other similar projects.[/quote]
Hi river333,
I think multisig could be a cool way to do this and like you said may hopefully be used in other similar projects. That said it is a bit more complicated and new, so if you would want to get started by personally giving out the tips I will go with that as I totally trust you. You have clearly shown your good will on the forum in the many good acts you do.
Thank You for all the good things you do,
NewMoneyEra[/quote]
Thanks 
I suppose it depends on how soon we want to get started with this, and how long it will take to figure out multisig.
If it doesn’t take too long to start with multisig, then it would probably be better just to use that for everything, rather than complicating things with two different options.
Also I’m not really sure if a 1 PPC payment to the private key holders for verifying is necessary, especially if it complicates setting it up.
I’m happy to verify for this occasion without payments as I’m keen to understand whether we can get multisig to good use with the current Peercoin implementation(v0.4). It seems to be different from Bitcoin (behind?) and we don’t have any supporting applications as far as I’m aware. So I guess we need some developer support on explaining how this might work with Peercoin.
Edit: just scrolled through some thread with the multisig subject. It seems that we have a different multisig implementation than Bitcoin and that we already had a few developers pleading to adopt the Bitcoin implementation. Guess something to put on the wishlist for next version of Peerunity?
Cybnate could you post that thread? I wasn’t aware of that.
I know Sigmike is totally up on multisig for bitcoin and peercoin. He mentioned to me that he couldn’t get multisig addresses to mint at that time, but we shouldn’t need to mint them anyway. I haven’t asked him about multisig for this project because I hate to bother him as I know he is so busy doing many important things for Peercoin/unity/shares etc. And, then there is also the problem that I am a procrastinator
My understanding is that multisig doesn’t work with minting right now because the Peercoin client (RPC or Qt), isn’t configured to require more than a single privkey to process the coinstake transaction.
I presume that it would attempt to mint with mature coins held in a multisig address, but if it found a block, it would fail when it tried to validate its own keys.
(Hopefully we’ll have a more technical answer, soon)
Ok, I did some digging beyond the rumours which started with the lack of createmultisig command.
It appears that multisig technically works, but that we are missing a link for a practical implementation.
Sigmike showing multisig works: http://www.reddit.com/r/peercoin/comments/1zd7gp/will_we_get_multisig_in_peercoin/
No minting, but that’s ok for our purpose.
And someone else: http://www.reddit.com/r/peercoin/comments/1p44qo/ which I think shows some problems to make this work in a practical way. At least I couldn’t get it to work, but have to admit my knowledge is limited. Even after viewing @nme’s video.
But this underway: https://github.com/Peerunity/Peerunity/pull/15 and would fix the problem?
I think we indeed need a technical answer or demonstration of the whole process with Peercoinwallet and Blockchain to clear up the muds.
[quote=“Cybnate, post:74, topic:2152”]Ok, I did some digging beyond the rumours which started with the lack of createmultisig command.
It appears that multisig technically works, but that we are missing a link for a practical implementation.
Sigmike showing multisig works: http://www.reddit.com/r/peercoin/comments/1zd7gp/will_we_get_multisig_in_peercoin/
No minting, but that’s ok for our purpose.
And someone else: http://www.reddit.com/r/peercoin/comments/1p44qo/ which I think shows some problems to make this work in a practical way. At least I couldn’t get it to work, but have to admit my knowledge is limited. Even after viewing @nme’s video.
But this underway: https://github.com/Peerunity/Peerunity/pull/15 and would fix the problem?
I think we indeed need a technical answer or demonstration of the whole process with Peercoinwallet and Blockchain to clear up the muds.[/quote]
So we should be able to get this started once Peerunity is released?
Ready to buy my first Raspeberry Pi
Which model should I buy? Model B I guess.
Are there some recommended additions: case, SD card, etc. ?
Thanx for your help.
[quote=“mably, post:76, topic:2152”]Ready to buy my first Raspeberry Pi
Which model should I buy? Model B I guess.
Are there some recommended additions: case, SD card, etc. ?
Thanx for your help.[/quote]
Model B is the most used at the moment. I’ve read that the new model has the same hardware, and only the form is different, but I have no experience with it.
As for an SD card, it’s best to buy a class 10 or better, that is 10MB/s minimum write speed. Don’t get fooled by the read speed they mention on the packaging, that is always much higher. http://en.wikipedia.org/wiki/Secure_Digital#Speed_class_rating
A case is not necessary but recommended because it protects against stuff falling on your pi, small coffee spills etc
Raspberry Pi ordered
I have a pi I ordered for playing round with and remember you will need a power supply to for it, that and the SD card u need to buy, its a micro usb charger same as many android phones or tablets so u can use one of those if it gives a high enough voltage. I just remember when I ordered mine the wait on charges was 3 weeks longer than the 5 weeks for the pi, but I hope they more readily available now.
Fuzzybear
Sent from my HTC Desire using Tapatalk 2
Thanx for your advice, Fuzzy.
I ordered the whole Raspberry Pi minting package from Amazon : Raspberry Pi, micro USB power supply, class 10 SD card (8GB might be a bit too small though, we’ll see) and a black protecting case for about 70 euros.
I should receive everything before may 13th.