Gold collapsing. Bitcoin UP.

[Richy_T]

Wooohooo, thanks for voting me in as Secretary guys. I'll try to do my best to foster a community spirit and promote Bitcoin Unlimited and the ideas it embodies which are desperately needed for bitcoin in these testing times.

Yay!

Now, coffee, black, two sugars. Hop to it.
[doublepost=1453488006][/doublepost]

@lunar The idea of separating node code from wallet code is a very good idea.

You're going to have a hard time of achieving that if you're starting point is a Bitcoin Core fork, as that code base was never designed that way.

I think this is a bigger goal that we should be working towards. Not BU, nor classic but something that all implementations that care to can build upon. Core stalled out with libconcensus, for whatever reasons (we can speculate) but I think it (or likely something like it) is essential as we move forward.

I have freed up some time to work in bitcoin project going forward so hopefully I'll be able to present something soon.

 

[cypherdoc]

Blocksize Poll Summary:

https://www.reddit.com/r/btc/comments/426wt7/blocksize_polls_summary_to_date/

 

[albin]

Gavin seems to be gradually warming up to BU-esque ideas continually:

 

[theZerg]

@lunar yes I have considered it and I think posted something about it before. I would like to see long term a wallet node split with the wallets running on laptop or android and having a special trust relationship with your node and the ability to fall back on SPV. I've been brainstorming on exactly what advantages trusting the node would give the wallet...

 

[molecular]

What about SPV light-weight wallets such as Mycelium, don't they already separate wallet code from node code in a sense? They are not a full node and do not validate blocks, instead my understanding is they simply are a client that connects to several full nodes as servers and request the data they require from the servers.

It's even "worse". In the case of mycelium, they client connects (through a proprietary protocol) to a mycelium server operated by mycelium. Similarly, electrum client relies on electrum servers. These are also specific to electrum, but anyone can run one and the users can freely select which one to connect to. Protocol is proprietary too, but not closed.

 

[rocks]

Core has tripled their support, they now have 3 companies backing them.

 

[Fatman3002]

Thought I'd slow down the FUD-train and post this here as well:

https://bitcoinmagazine.com/articles/f-pool-chinese-pools-will-stick-with-bitcoin-core-1453395328

Mr.Weirdum has updated his article.

 

[cypherdoc]

Core has tripled their support, they now have 3 companies backing them.

don't overstate it.

those guys and their small companies have supported Core from the beginning.
[doublepost=1453501600][/doublepost]

 

[chriswilmer]

Ledger too? Wonder why.

 

[YarkoL]

Ledger Wallet, not the journal... I think?

 

[solex]

It would be interesting to see a survey of full nodes owners finding out how many use the wallet. I suspect that quite a few still do, and I admit that I find it useful occasionally. Removing it will force some users away, so it is a decision not to be lightly taken.

@chriswilmer indeed: Ledger is some wallet provider, not your journal...

 

[chriswilmer]

Haha, guys, rest assured I know whether the journal supports Core and also that I know about the hardware wallet Ledger

I am STILL surprised that the HW wallet Ledger is showing support for Core.

 

[YarkoL]

Well, it was I who was confused for a moment...
but I take your word for it, Professor

 

[albin]

Hardware wallets are vulnerable PR-wise to really abstract FUD mainly pertaining to sidechannels and oscilloscopes, so avoiding antagonizing unethical furious handwavers is a good defensive business decision.

 

[solex]

@cypherdoc
Thanks for mentioning. I have updated BUIP016 to reflect Gavin's latest pull-req for sigops.

 

[bitcartel]

 

[cypherdoc]

[doublepost=1453507614][/doublepost]I have an ongoing theory which I've already elucidated here.

Bitfury is going to force a block size increase to help pay for their 3 mega liquid cooled mining centers that leverages superior BW despite certain public citations to the contrary. There will be nothing the Chinese, or core dev for that matter, can do to combat this. If anything they should just go with it as they need more TX fees in aggregate as well in the long term. What we are witnessing is all according to plan; increased mining decentralization driven by fixed currency supply free market economics.

 

[albin]

That relates to something that I've been thinking about: perhaps industrial mining firms / pools are reluctant to go into great detail about their blocksize preferences, for fear of potentially revealing proprietary business information? Someone very knowledgeable about the business could conceivably deduce alot from very little information.

 

[Richy_T]

It's even "worse". In the case of mycelium, they client connects (through a proprietary protocol) to a mycelium server operated by mycelium. Similarly, electrum client relies on electrum servers. These are also specific to electrum, but anyone can run one and the users can freely select which one to connect to. Protocol is proprietary too, but not closed.

I have dug through the Mycelium code and the design is fairly sensible. I don't like their protocol (unless things have changed, they serialize Java objects) but for what they are doing, it's much as you'd expect.
[doublepost=1453508625][/doublepost]

It would be interesting to see a survey of full nodes owners finding out how many use the wallet. I suspect that quite a few still do, and I admit that I find it useful occasionally. Removing it will force some users away, so it is a decision not to be lightly taken.

@chriswilmer indeed: Ledger is some wallet provider, not your journal...

The wallet should not be removed but be merged away into an external application or plugin-type solution.

There are many things in core that simply should just not be there. Changing these things then becomes a source of contention when it is not really even within the scope of what consensus should be.

We are at 0.11.something. We need to be looking to 1.0.0

 

[cypherdoc]

@albin

Or publicly stating one position only to be secretly planning the other?

C'mon, it's a no brainer that mining pools that large will need huge TX fee throughput increases to pay their fixed expenses along with profits. LN does nothing but steal those fees away. These guys aren't stupid.

No way they let a bunch of LN hub owners simply code up fee siphoning server software as the future of Bitcoin. .