@ODELL, I love you but you turned my pleasant nostr timeline into a twitter feed of “I love devs and you hate devs!”

Please consider having a long form podcast discussion somewhere on the deeper issues. 

A few on my mind:

1) Are the core devs and the development process too centralized? Is there anything that can be done to reduce risk? Do we want large companies to fund devs? How do we avoid a situation like big pharma funding the FDA?

2) The community fought hard for small 1 MB blocks. Why were we surprised by ~4 MB blocks and spam jpegs. Was this a failure of software architecture? Communication? 

3) What should be our software design philosophy for Core? Should we add powerful new technologies where the full use and abuse cases are unknown? Or rather should we select only the most important use cases and then tightly enabling only those narrow purposes? 

4) We all saw how a contributor almost got malicious software into Linux. Are we funding enough security auditors?

5) Is there a way to improve communication between the devs and the community? If we need a site perhaps we should build it. 

What else am I missing?

Thank you. 🙏 

 Re: #3

I was told Core devs had announced they will not be doing any more soft forks.

I think something like Core but a layer downstream could implement things like OP_CAT or various demanded soft forks so curious plebs could run more powerful (but potentially more risky) versions of the protocol. Seems like a way to have our cake and eat it too in tinkering more safely. 

I’d love to see something like this arise (it could be called Mantle). As long as it was run and vetted by credible technicals (not me) I would contribute sats and run the code for more “cutting edge” expressive versions of Bitcoin. 

I’d like to see the envelope pushed in a safe and reasonable way. Core should stay conservative imo. 

 Yes, absolutely. It’s like how to create extremely reliable software. Create a bulletproof micro kernel and then add layers. (Eg. QNX, sel4)

My understanding is that Liquid was intended to be used as you described. A “playground” where stuff like smart contracts and Monero-like confidential transactions were run but in a sidechain where it could experiment but not harm the core L1 network. 

Perhaps these experimental features could either be added to Liquid or into a similar sidechain where you would peg-in and -out some sats into a network that looks like the core L1 but with experimental features.

I think it’s a great idea. 😀 The only issue I see is that people would probably want to experiment with it and wouldn’t use a large amount of bitcoin. With high L1 fees, the peg in and out transfers would be expensive on the base chain. Perhaps there is a way to do it to peg in and out with Lightning. 🤔 

 That’s a really good idea to run new implementations on liquid but the federation would have to agree to it and then choose to run the OP code(s) or whatever changes they wanted right?

There needs to be another place to run this stuff and see how it works out in the world and to begin to stress test it. Liquid would be an awesome sand box, I would peg in some sats to play immediately. 

 Yep, check it out. I appreciate Blockstream’s forward thinking with it and how it could be used to improve security and confidence. 

 What do you use Liquid for? 

 Now, just as a replacement for Lightning. I used to make small ‘zaps’ to friends and family kids using BlueWallet. 

That worked well (fees were low) when they were custodial. When they went non-custodial we all got rugged. 

The exit transfers required base chain transactions (I think for channel liquidity) and the fees were extreme (~$30). In many cases consuming the balances. 

Anyway, long story short, I wanted to find the next best service, as non-custodial as possible. Lightning doesn’t work for our use case because of the channel opening fees. So I moved everyone to Liquid. 

 So you peg in to Liquid and since it the 1 minute blocks it’s adequate as a fast tx system? 

 Yes, I pegged into Liquid and it’s fast from there. Liquid has a mempool so transactions show as pending quickly and finalize shortly thereafter. 

I’ve used Boltz (Lightning) and SideSwap (BTC) for peg ins. 

 Also are you using Aqua to access Liquid? 

 Yes both Aqua and Green. Green seems to be much cheaper than Aqua but I use both to support Aqua. 

Fees still aren’t as cheap as custodial Lightning (eg Primal). I pay 0.1 sat / vbyte with Green which works out to 15 cents per transaction. 

All things considered, I’m mixed about both systems. Liquid is easier for newbies, but with higher fees making it difficult for zaps but pretty good for everyday transactions. Non-custodial Lightning has lower fees but with periodic and expensive L1 chain transactions. 

 No perfect solution exists but we have good optionally and substantially better functionality right now vs when I started in Bitcoin in 2019. 

I’m hopeful Lightning can be developed further. If it was just a bit easier to setup channels and receive LN payments non-custodial I think it would be a-near perfect fast payments solution. Very good as is imo but high-fee environments make it less economical if you haven’t set up a channel. 

Did not ever think I’d use Lightning daily a few years back. Now I do and it’s non-custodial and payments are successful 99%+ of the time. Really promising imo. 

I will start to use Aqua just to play with Liquid. I have it setup already just no funds added bc I haven’t had a real reason to use it yet. 

 Interested in your experience of using non-custodial Lightning day to day. 

Which app is your favorite?

How often do you need to make L1 transactions? 

Also, do you think it could be used by a young niece / nephew to receive small payments? 

 I use ZeusPay as my receive wallet and it’s a 7/10. It works reliably to receive but it’s usually slow and takes a bit of finesse on my end daily. 

The operations (as far as UX) of the software aren’t always consistent which is weird to me but in the end my funds do always settle. If they could make it 20% more efficient it would be a 9/10. But I understand that they’re doing a lot of “hacking” to get an embedded LN node to receive funds. So it is what it is. 

ZeusPay also just won’t work reliably for sends. So that pushed me to setup a Phoenix embedded node for sends, and it work really well. I love that wallet and it feels like peak Lightning. If they could make it receive as well as all of the other functionality they offer it would be 10/10 software. 

I think a 12 year old could use ZeusPay with some help setting it up for sure, but they’ll have times of being frustrated with it. But in the end it will work for them. Phoenix for sends is highly recommended.