Guys, again you are doing a very good job keeping us informed and especially working on so many important things.
The size needed to run the blockchain - decreased by using RocksDB
The problems with bandwidth - in progress
Hard Fork 20 (which I personally thought it got canceled, looking at its status on github
SMTs
Major changes like scalability, which can't be seen directly, but it is one of the most important.
I guess that @andrarchy was right when he said that your team is working continuously to make improvements on the Steem blockchain as they do.
You have my sincere congratulations for this!
Now the bad part...
What I personally am concerned about, when you have time to address it, is the fact that 11 days ago there was a downtime of Steem blockchain, if I remember well. It was a 504 error when trying to access steemit.com. The problem was also present on busy.org or when trying to run condenser locally.
My concerns are really big about it, since I am not so sure why this happened or if the blockchain was still running well at that time. I remember that steemd.com was also not responding so I think about this as a potential centralization problem of the blockchain. If this affected the whole blockchain, then can we really rely on Steem as completely safe and resistant to attacks?
PS: sorry for the self-vote, I am trying to make this visible...
There was no downtime of the steem blockchain, but there was a downtime of the steemit RPC node that the majority of sites and services built on the steem blockchain use to get their data.
This is exactly the type of problem these developments will help to solve. With Appbase and RocksDB it will be much cheaper and easier to run a full node so hopefully there will be many more available that the sites and services can use if one of them goes down.
Okay, thanks a lot for the reply, apparently it is not that easy to find somebody who knows what happened with that let's call it downtime of steemit.com.
So the apps that are connecting to the Steemit RPC node could somehow check if this node is down and switch to another one during the downtime? Is that possible or I am talking nonsense?
It would be up to the individual application developers to support that. It is possible though, and would be a smart way to design a more fault-tolerant app.
Couldn't have said it better myself @yabapmatt