Ark - New L2 protocol

[darkv0rt3x]

In @TheBitcoinConf, the enthusiast Burak, presented his work on a new L2 protocol, that as he says, is intented to solve some of the issues with LN that he found over time.

He made a post in Medium:
https://burakkeceli.medium.com/introducing-ark-6f87ae45e272

An also posted in Bitcoin dev mailing list
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2023-May/021694.html

For the most technical, what you guys think of this?

[1714556085]

1714556085

[1714556085]

1714556085

[n0nce]

Interesting; the blog post doesn't really explain much about how it is supposed to work, but I will find some time to read through the technical documentation on his webpage.

If I understand it right, 'routing nodes' are still needed; however since they become optional (as opposed to Lightning, where everyone's node can route transactions if needed), I fear centralization could happen.
Also not sure what incentivizes people to even run such a node.

But maybe this all becomes clearer through his 'deep dive'.

[darkv0rt3x]

Interesting; the blog post doesn't really explain much about how it is supposed to work, but I will find some time to read through the technical documentation on his webpage.

If I understand it right, 'routing nodes' are still needed; however since they become optional (as opposed to Lightning, where everyone's node can route transactions if needed), I fear centralization could happen.
Also not sure what incentivizes people to even run such a node.

But maybe this all becomes clearer through his 'deep dive'.

Thanks for that link. I will also go through it to see if I can understand the project better. He has some points in his arguments on the link from Medium site. I didn't have the time to read his post in the mailing list, but will read it too when I have some time!

[NotATether]

I like it! It has elements of my Ephemeral settlements and voting-based Layer 2 designs inside it.

Only roadblock is that it depends on opcodes that haven't been fully ratified yet (ANYPREVOUTANYSCRIPT in BIPs 118 or CTV in 119 I believe), so that's got to be solved first so that the network can function officially. But that should be simple enough for protocol devs to do.

The more pressing task is to design the Ark nodes software. This doesn't necessarily depend on LN which is good.

[DaveF]

3. If i understood it correctly, it looks like liquidity problem is moved from peer (pair of user) to service providers. On LN, we already see some "dominating" LN node which have lots of channel or big capacity. With Ark, i have concern we'll very few dominating service provider.

I think that is going to happen no matter how a L2 protocol is done. There's going to be some providers that are going to take a large percentage of whatever service there is. It's just the nature of how it's has to be done. Once money is involved, people are going to only slowly put their own funds in. However, there will always be people in the beginning that took a larger risk with more money and grew and grew and grew. The flip side is if this particular L2 solution goes away or fails or has other issues then those people will lose significantly more money then some guy named Dave who said what the heck I'll put a couple of $100 into this and see what happens.

As for the protocol in general. It looks interesting but as you and others have pointed out it's dependent on other B IP's being finalized and released so it's still a ways away who knows what will change between now and then.

-Dave

[n0nce]

3. If i understood it correctly, it looks like liquidity problem is moved from peer (pair of user) to service providers. On LN, we already see some "dominating" LN node which have lots of channel or big capacity. With Ark, i have concern we'll very few dominating service provider.

I think that is going to happen no matter how a L2 protocol is done. There's going to be some providers that are going to take a large percentage of whatever service there is.

That's true, but while in Lightning you do gain something from being well-connected even as a regular user (better / faster / cheaper routes for your own payments), I see no incentive at all to become a service provider in Ark. Routing fees (at least in Lightning) are still very low nowadays, so I doubt that is incentive enough.

I'm not too worried about the privacy aspects of such centralized providers, if the cryptography / anonymization is done right, but more about availability guarantees / resilience against DoS attacks from state actors and similar.

[dkbit98]

For the most technical, what you guys think of this?

Ark looks very promising and I like that side effect of using Ark will be improved privacy, similar like Coinjoin is doing but offchain.
Yesterday I listened the interview with main developer Burak from Turkey talking on Bitcoin Takeover podcast.
He worked before on Liquid and on Lightning Network, and Ark was created as byproduct of new Lightning wallet, but he introduced a lot of improvements.
His reason for creating Ark is to make Bitcoin available for everyone, and I agree with him that that is impossible to do on mainnet or with Lightning:
https://www.youtube.com/watch?v=iQ7TLBhh9r4

That's true, but while in Lightning you do gain something from being well-connected even as a regular user (better / faster / cheaper routes for your own payments), I see no incentive at all to become a service provider in Ark. Routing fees (at least in Lightning) are still very low nowadays, so I doubt that is incentive enough.

There will be incentive in Ark, same as in Lightning, and Ark will be compatible with LN from what I understand.
Listen the interview I posted above.

[darkv0rt3x]

These 2 threads could be merged:
https://bitcointalk.org/index.php?topic=5454229.0
https://bitcointalk.org/index.php?topic=5453928 (self)

[NotATether]

I asked Burak on the Telegram channel what language will the first Ark clients be written in. He answered "Golang and Rust" (much to my dismay at not knowing either of them).

Currently on the ARK github repository, there are only drafts of specifications, but a codebase will be started soon.

[darkv0rt3x]

I asked Burak on the Telegram channel what language will the first Ark clients be written in. He answered "Golang and Rust" (much to my dismay at not knowing either of them).

Currently on the ARK github repository, there are only drafts of specifications, but a codebase will be started soon.

Yeah, me neither! But even in the languages I'm a bit more comfortable, would probably need quite some time to learn only the protocol details! But would be interesting to be able to contribute in some way!

[darkv0rt3x]

There is this new video of Burak and a couple of friends in a journey explaining Ark protocol. I'm watching it right now! Seems to worth the 2h video.
Take a look at it!

https://www.youtube.com/watch?v=EocWax43QgQ

[Kryptowerk]

There is this new video of Burak and a couple of friends in a journey explaining Ark protocol. I'm watching it right now! Seems to worth the 2h video.
Take a look at it!

https://www.youtube.com/watch?v=EocWax43QgQ

Oh that's awesome. Giving it a try tonight.
Is it content that can be understood by not mega-tech-savy folks? Well, guess I'll find out by myself...

Are any of the ARK-folks actually active on the forum here?
In any case, exciting times. Just hope timeframes for implementation and adoption aren't as huge as they were/are for Lightning.

[darkv0rt3x]

There is this new video of Burak and a couple of friends in a journey explaining Ark protocol. I'm watching it right now! Seems to worth the 2h video.
Take a look at it!

https://www.youtube.com/watch?v=EocWax43QgQ

Oh that's awesome. Giving it a try tonight.
Is it content that can be understood by not mega-tech-savy folks? Well, guess I'll find out by myself...

Are any of the ARK-folks actually active on the forum here?
In any case, exciting times. Just hope timeframes for implementation and adoption aren't as huge as they were/are for Lightning.

I could only watch the 1st half hour. Maybe tomorrow I can watch a bit more. I think you people here can easily understand the protocol. Way easier than me. Pretty sure will be easy for you guys!

[Kryptowerk]

Yeah, I wish some of the OG (devs) would scoop these guys one or two BTC so they create a TED-talk quality presentation.
I also watched about 30 minutes or so. It could be structured a little better but what really made it hard for me to pay full attention was the varying audio quality. Guess I am just spoiled with high quality productions.
Will give this another go soon.