Quote from: hdbuck on January 30, 2015, 08:47:27 PM

I dont know if someone actually made a similar poll since search engine is down so here we go..

Bitcoin fork proposal by respected Bitcoin lead dev Gavin Andresen, to increase the block size from 1MB to 20MB.

Interesting read about it: https://bitcointalk.org/index.php?topic=919629.0 (locked Cry

)

Feel free to discuss it further.

Quote from: phillipsjk on January 31, 2015, 12:24:46 AM

Quote from: inBitweTrust on January 30, 2015, 09:41:16 PM

... and others that have very slow internet connections under 2Mbps wont be able to assist with being a full node.

You mis-interpreted my post. My point was with 20MB blocks, you will no longer be able to host a node on consumer Internet Plans. The important number was not the 2Mbps download, but the 4Mbps upload. BTW, to get those numbers, I simply multiplied my former node's bandwidth usage by 20.

My current ISP has reduced the speeds it is offering for consumer access. The fastest upload offered on a consumer plan is 3Mbps. Business plans have an option with 5Mbps of upload. (Experiments like 250Mbps down and 15Mbps up are being grandfathered)

Telus, the local phone company, is better with it's VDSL offerings. You can get 5 or 10Mbps of upload speed. For independent providers (using the same lines), like the one I was with, your maximum upload speed is 5Mbps. To get 10Mbps or more, you have to go to fibre, costing $$$$. VDSL and cable may technically be able to handle that, but the companies involved don't want to cannibalize their lucrative fibre offerings.

My overall point is that if we go ahead with this fork, we have to do it with the understanding that the era of running a full node on a consumer Internet connection will be over. Due to extra CPU overhead (due to transaction processing), the days of running on a VPS may be over as well.

Now, with 20MB, I expect dedicated hobbyists and small businesses will still be able to run a node, but it will be a considerable expense: hundreds if not thousands per month. Webhosting may be cheaper, but then you can't keep an eye on your box; or plug hashers directly into it.

In consideration of node concentration weaker areas will have fewer nodes and nodes may become geographically more centralized.
Analyzing current node geography a large concentration is in the USA and Europe with higher mb requirements limpacting the spread where there are a small amount of nodes at present and weaker internet service.
https://getaddr.bitnodes.io/

I would still say a variant of blockchain compression should be looked at as well but any change to the size limit should consider hosting capacity in the areas with the lowest node distribution, that way there will not be to many places left behind by any changes.

That said I agree Phillip has a point we are kicking the buck down the road but something should be done and concede that Gavin may be correct that this should be done sooner than later
https://bitcointalk.org/index.php?topic=941331.msg10315817#msg10315817
Roadstress as well
https://bitcointalk.org/index.php?topic=941331.msg10330786#msg10330786

It's been a while since a D@T thread had me read through it all

Other useful links
https://blockchain.info/charts/avg-block-size
https://bitcointalk.org/index.php?topic=941331.msg10315826#msg10315826
https://bitcointalk.org/index.php?topic=144895.0

Quote from: theymos on February 03, 2015, 12:36:30 AM

Bitcoin can be changed in a backward-incompatible way and still remain Bitcoin. It was done by Satoshi with the version checksum change, for example.

Hopefully there won't be any huge hard fork controversy in the future. It'd be a big mess if people had to actively decide between one fork or another. If this does happen, then I will endorse the most correct version of Bitcoin, and this version is what I'll mean when I say "Bitcoin". In particular, these are some principles that any potential hard fork must not violate:
- The network must remain substantially decentralized.
- The inflation schedule must be the same or lower/slower. (Though I'm not 100% sure whether lowering inflation would be OK.)
- No one should be allowed by design to steal your money.
- As much as reasonably possible, no one should be able to prevent you from spending your money.
- Anonymity should be at least possible.

I will oppose any unsafe hard fork, even if it's proposed by Gavin. I and the sites I have some hand in are independent of the dev group, the Bitcoin Foundation, and other companies/organizations. I don't know whether Gavin's current proposal is safe, so the only thing I'm doing now is recommending caution.

Quote from: Raize on February 02, 2015, 11:33:51 PM

It was wrong and I hope the separation of miner and development continues for at least a few decades before miners and developers are so embedded with each other we have a repeat of what led to Bitcoin in the first place.

I wasn't a fan of the whole idea of giving miners any special say on the issue. (Though it wasn't actually much of a vote, since miners could only confirm/reject P2SH.) Miners are basically employees of the network, and it should be the actions of users and businesses that influence what miners do, not the other way around. It would have been possible and better for users and businesses to (at a reasonable pace) force miners to accept the P2SH change.

----
Who is this Mircea Popescu and why is anyone giving this clown any time of the day?

https://bitcointalk.org/index.php?topic=608444.0

CTRL+F for his name.
----