What we did
We did it! We minted our first block! Four days ago, @lux-witness has validated the transactions, signed and appended block number 22 682 862 to the steem blockchain, and the other witnesses accepted it!
We are from now on a true witness of the steem blockchain! Thanks to all those who voted for us! We won't disappoint you!
Let's dispel straight away potential questions: yes, the "block production reward" is 0.97 SP. That is, at current prices, about $2.45. So if we wanted to use this reward to pay the bills for the server, we would be able to "Power Down" the wonderful sum of $0.19 (19 cents) per week for the next 13 weeks.
We need to do much more if we want to begin recouping the costs of running a node. So what have we done during the last week?
Hardware side
Thanks to @pstaiano, we have been able to heed the call of @gtg and have added a new seed node: seed.steemit.lu:2001 to the network. Pablo has also set up a backup witness server and used the tools to failover them successfully. Next week when we'll be up for the next block we'll be ready ! :-)
"Software" side
A lot has been going on on the "software" side. No, not that software. I'm talking here about "spreading the word" in order to advance toward our stated objectives: more quality content creators, more business interest for the steem blockchain and ultimately STEEM going mainstream on fiat exchanges such as Bitstamp.
Steem is, without contest, the best blockchain software currently in production mode as I have explained, alongside @roxane, @alexvan and @hightouch to a local public during the First Steem Meetup which I organized last week in Luxembourg. Check the post to see the slides I have shown.
Not only I know of no better blockchain software but it would appear that there really aren't that many contenders out there. At the recent conference in Malta on "Blockchain, Credentials and Connected Learning" the projects were all about registering diplomas and qualifications on either Bitcoin, Ethereum or (seldom) use complex Hypeledger Fabric bespoke designs. Basically no other technology has reached the necessary level of familiarity for people to start imagining projects around. I naturally introduced steem and its features at several occasions.
I participated in two other events at political level, as detailed in these posts:
- Blockchain4EU
- Poker Champion Tony G turns MEP Blockchain Champion!
My idea is pretty simple, I want to leverage the power of the steem blockchain and its many excellent features (as explained in the meetup presentation) to advance the European project. I know it's a (very, very) long shot. Considering I'm "Mr. Nobody", it's ambitious ... on the borderline with madness ...
What we'll do
What are the plans for the near future?
Hardware side
Not much on the "hardware" side. As we explained, keeping the blockchain running is paramount but (fortunately) at this point in time the first 20 witnesses are doing a great job and there are 100 other backup witnesses in front of us.
However we can't help but notice that steemit and busy do not always work quite so smoothly. As it has been explained several months ago, the anatomy of the steem ecosystem has several server tiers
And this layered model is currently not aligned with the economic mechanism of the blockchain.
That is, the block production rewards go to witnesses. But what makes the applications chug along are overwhelmingly the RPC Nodes. They are a lot fewer (8 or 9 if I'm not mistaken, compared with 135+ active witnesses) and they cost a lot more to operate.
A witness node needs about 45 - 50 GB of RAM. For an RPC node it's more like 256 GB or more !
Yet if I'm not mistaken there are no economic incentives to run an RPC node. The only things that push someone to run one are:
- love, pride, belonging for/to steem
- within the context of running a witness node, use it as an argument to gather more votes for the witness
We perceive this situation as a weakness of the current model and believe it should be addressed. Ideally, someone running an RPC node should be directly incentivized with SP, in proportion to the cost of running such a node. This would have the advantage of offering a lot more scalability and a much smoother user experience and prepare steem for the next 10 000 000 active users!
Software side
On the "software" side (no, not that software, I'm referring here to PR outside the platform), next week should be a great one but I don't want to say more right now in order to avoid jinxing it.
So I'll stop here (but remain available to answer questions from the curious)
Check previous updates:
the cost of the full RPC node should indeed come down drastically so that even witnesses in the top 75 ish should be able to pay for them out of the revenue at a steem price of roughly 3.
We started a bit before you and are also doing a lot of "software" and it seems to help, slowly... we are hitting roughly 1 block per day now and need 2 to pay for it all... so getting close
and I also added my vote to your pool... its not that much but its a few MVests more...
Thanks @felander, you guys have my vote as well. Maybe we'll also do some (computer) software in the future as well, let's see. You are three I understood, we are only two :-)
We are slowly getting noticed, I have hope :-)
Qué bien, interesante proyecto. Buena suerte.
Muchas gracias @mariajuliany, si te interesa lo que estamos haciendo nos ayudaría muchísimo tu voto a nuestro witness, esto lo puedes hacer muy fácilmente desde el siguiente link :
Nuestra idea es hacer crecer la comunidad de Steem y al mismo tiempo mantener la red estable. Saludos
Congratulations on minting your first block, @lux-witness!
You made me curious about next week's PR events. ;) But I don't want to spoil your surprise.
On the full RPC nodes costs, if I read steemitblog's recent post on scalability correctly, I remember they are attempting to fix the issue by lowering the need to run a full node or to have the full blockchain in memory instead of on fast disks.
This makes sense to me, since the blockchain will only grow in size and keeping up with the growing memory requirements for a full node is not enticing, even if additional rewards would be offered.
I liked the Steve Jobs quote!!! Keep up the good work @lux-witness!!!
Hi.
You should be just fine running on a 32 GB server, as long as you have the memory mapped file stored on a SSD drive instead of a 'traditional' disk. Dropping to a 32 GB should help cut down on costs, and it should not affect the reliability of your node. Details here.
Technically a 16 GB will work too, but it is quite a bit slower for replay.
Thank you Tim. It also depends on how fast the SSD is and we running, for the moment, on VPS, we do not quite control that ... Pablo has run some tests and it appears the SSD provided by the host is of the sluggish type so we take advantage of their "guaranteed 50GB of RAM" for the time being
Ah, ok. that makes sense :)
welcome to the witness family.. congratz with the block
Thank you!
You got a 20.74% upvote from @upme thanks to @lux-witness! Send at least 3 SBD or 3 STEEM to get upvote for next round. Delegate STEEM POWER and start earning 100% daily payouts ( no commission ).
Sneaky Ninja Attack! You have just been defended with a 18.61% upvote!
I was summoned by @lux-witness. I have done their bidding and now I will vanish...
woosh
A portion of the proceeds from your bid was used in support of youarehope and tarc.
Abuse Policy
Rules
How to use Sneaky Ninja
How it works
Victim of grumpycat?
You got a 11.16% upvote from @postpromoter courtesy of @lux-witness!
Want to promote your posts too? Check out the Steem Bot Tracker website for more info. If you would like to support the development of @postpromoter and the bot tracker please vote for @yabapmatt for witness!