Although in your definition, organizing for a proposal to be heard, IS part of the Democratic process.
I totally agree that bitcoin is not democratic , that's why i put quotes . In bitcoin you only vote with your associated hashpower as a mining pool . Full nodes don't vote/play part in consensus . They are just observers of the outcome ( block creation ) that mining nodes do . I think we agree on that ?
I agree, but it would depend on the node. If it's an economic node, then it plays an important role. Because what secures Bitcoin? The Miners or the Full Nodes? It's a long debate, BUT if Bitcoin requires consensus to "define" what "Bitcoin" is, which is a system/protocol of key assignment and re-assignment of values, then what secures Bitcoin? The nodes. They enforce the rules.
The article is based on one argument . That bigger blocks will create problem to propagation and decentralisation . What it should do is first define propagation and decentralisation .
Propagation with nodes that are on what bandwidth ? And with what specs ?
The average download speed globally is more than 50 Mbps and upload is close to 30 ( source https://www.broadbandsearch.net/blog/average-internet-speed-around-the-world ) . Let's see how fast a 1 MB block is downloaded with 50 Mbps ( that's a speed an average user has currently ) . We will use https://www.omnicalculator.com/other/download-time .
So for 50 Mbps the time to download 1 MB blocks is under 1 sec ( website says 0 sec ) .
For 10 MB blocks download time is again under 1 sec ( website says 0 ) .
For 100 MB blocks download time is 2 sec .
Time to propagate these blocks to it's connected nodes will probably be the double . All that process is under 10 seconds for 100 MB blocks .
Offcourse nodes need time to validate blocks . If i try to do it with hand it will take much time , if i try with rapspi that's the most common for btc it will be faster , and if i do it with high end machines it will be really fast . The problem is that the author doesn't consider the third option at all . So does most of the btc community . The belief is that even a user with a low specs machine and an awful bandwidth should be able to validate transactions . And all the other users should suffer high fees , accept adoption limits , wait for years for L2 solutions that doesn't work etc because someone who might never use or use occasionally the network wants to validate it's own transactions .
Now let's get to decentralisation . Who defines what is decentralisation ? And how much decentralised or centralised a network is ? Common question i make , can a 5 nodes network be decentralised and a 10000 nodes network be centralised ?
Decentralisation doesn't come from the number of nodes . Decentralisation comes from the economic incentives participants have . Being an active node offers to that , and that's achieved only through PoW .
At some part the article says that soft forks are inclusive while hard forks are exclusive . Why ? Because he thinks so . That's the whole article , his opinion without any scientific facts .
Soft forks ( easy path ) makes certain that no one is out of the network , even if he doesn't update . Who do you think would be interested more than anyone to stay on the network ? Those who profit from it , not by speculation on prices but from providing real services . Miners would do it . Listening nodes that can profit would do it . But , we want to protect the most lazy , the most cheapskate user and i don't know what else definition i can give " because decentralisation " .
Decentralisation is not a new concept . Tocqueville wrote his book " Democracy in America " close to 1830 . In that book he writes about the importance of participation , voluntary structures , local government and more . If all these were done by following what the most weak person wanted then there couldn't be decentralisation ( and that's why current "democracy" sucks ) . This would be a race to the bottom as everyone should follow what the worst person wants and usually the worst person wants to do nothing and have anything . Bitcoin is a race to the top . If you want to be a participant in the network that provides work you have to learn to cooperate . Look how pools were created . You have to be honest . Look how many dishonest pools dissapeared . You gotta learn to trust . Look how miners get paid for their shares because pools earned their trust .
Listening nodes would be more expensive , yes . But that doesn't mean that there can be collaborative nodes . In an environment such as btc has become where no one trusts anyone that is a no option . And the fun part is that this forum has a trust evaluation system . And taking that a little further the one of the key elements in finance , trade and much more is trust .
That article mentions about the internet rules and if you change the rules you create a new network , so no one would ever do a hard fork to change it's rules . Imagine an internet that only 7 users per second could use it . There would be no online commerce . There would be no fast innovation as information/knowledge would be limited . And i can't even think what else wouldn't be possible . In summary that author provides examples without any reasonable facts behind . He just express his opinion which is pure nonsense .
That's with the presumption that every node would have access to above-average internet connection speeds.
Plus for the question, how many nodes are required to be "appropriately" decentralized. I believe there's no right number, BUT I could tell you that the MORE full nodes = MORE security assurances.