Topic: Lightning Network Observer (Read 14122 times)

if you brush away how the devs have made their software and look at the function and process of how LN works. you will see its actually simple that it allows people to play around with unconfirmed parent-child transactions and outputs of.. so fundamentally simple

however looking at the software and gui layout and how they use silly new buzzwords, they have over complicated the interface and connectivity and explainers of how it works. even the back-up has been made to be harder then just copy/paste
they have done many things to make it complicated

put it this way
LN years ago was its own fresh new concept. meaning possibilities were endless and it was a fresh idea that could become anything, because it started from nothing. however within days of release, instead of making the network easily connectable to bitcoin/any blockchain scheme. they set silly policy and rules that then inferred that any blockchain scheme had to change to be LN compatible. rather than LN be a fresh network finding the simple way for LN to be blockchain compatible

analogy
family have a child
parents speak english. but they chose that the child should get a tutor and the child only learn/speak spanish from a young age.. and now the parents need to also learn spanish just to talk to the child. and then when its too late to turn back and teach the child english. the parents are now having to talk more and more spanish and void themselves of ever speaking english themselves on a daily basis
..
LN is not complex under the hood, using seeds to create keypairs and then making mutually agreed raw transactions is not complex.. but its made to look more complicated than it is, so much so that even the fangirls that idolise it for years still dont understand it

and if you think that devs dont make it complicated.. that is a laugh
devs know they are replaceable, they are not gods.. so if they make things complicated where only their expertise/deity status can be used to evolve it. they get to keep their job and keep receiving sponsorships to continue updating
its a known function of devs to never complete software and require their involvement to keep updating it, otherwise they would only have work for like 6 months per project if they did just build things the simple way.. its their way of having a career/job security, by doing things a certain way to keep their contract/job/grant/sponsorships flowing for years instead of completing in months

I doubt that lightning network is being made overly complicated on purpose or that more straight-forward solutions are available that are not getting traction due to some conspiracies to make matters more difficult.

I don't see anything wrong with having fee incentives to build, and of course, if fees are not competitive, then people will search for alternative ways to transact... so if there are ways to transact without having to go through third-parties who are controlling  your abilities to transact, then that is a good thing.

I also cannot see that lightning network is being promoted as being better than the bitcoin network, yet it seems to be another kind of transactional method opportunity that is available for bitcoin holders to use, and yeah no one wants to lose money by holding value in lightning network channels - which truly had happened to some folks running nodes, and perhaps more in the earlier days of lightning network. 

In one of my earlier posts in another lightning network thread, I had mentioned that I had one of my channels forced closed in my Breez wallet, and so I still have not reopened that channel or put money (BTC) back into that wallet.  I have been using Phoenix for nearly a couple of years, and so far I have never lost any money through that wallet, although maybe I am currently using an outdated wallet since they chose to move their App out of the Apple store in the USA, so probably I am not getting the updates to their wallet software.. so I might have to stop using it at some point. 

I recall once around August 2023, I had some issues with my phone and I was not able to access the Phoenix wallet until accessing through a new phone, and so I did not have any problem recovering to a new phone through the seed words, even though I was worried that I needed to make sure that I was not disconnected from running the wallet for several weeks, so I think that I was without access to the wallet for nearly a week before  I finally was able to reactivate the wallet with the back up seed words.  In August 2023, I had done that seed back up restoring with my Breez wallet too, but my loss of my channel with the Breez wallet was in October 2023, and I think that I did not know the channel had been force closed for around a month after the force close had taken place.

what if i told you that making it complicated is exactly the devs game
first they make bitcoin more annoying by not allowing tx count scaling and also wanting fee's to increase and then adding bloat just to really make bitcoin burdensome
then they make LN be the promoted "only solution".. but then make LN burdensome so that people then are made to pay middlemen to 'hire a hub' or 'rent a channel' .. because as most do know.. the whole point is that bitcoin does not offer ways for middle men to make commission processing payments so the big institutions needed a system to make people pay middle men.. and so yep all bitcoin bottlenecks and tailgates lead to offramps they hope push traffic to LN so institutions that sponsor the devs can make ROI on the funds they paid to make crypto more annoying


but with that said, someone else in this topic insinuated LN is popular far more then other subnetworks.. yet i just done a 5 second search and found that the avalance bridged network has more liquidity than LN.. 'wrapped btc' has more liquidity
but it is funny that people try to promote LN without doing the research to back up their insinuations of how magical and perfect their silly subnetwork is
silly due to it still having 8year old bugs, yet foolishly promoted as the evolution and better network than bitcoin.. pfft its not

I like the idea of potentially regularly showing lightning network statistics, and surely new channels open can be helpful, yet seem to need the other statistic of channels closing or active channels to see if the channels are increasing or not.  Also, I understand that some of the complications with lightning network statistic is the difference between public channels and private channels, so if the channels are private then they become challenging to measure.

Surely ease of use has become a barrier by a lot of folks, and many folks do not want to have to figure out how to run a lighting network node unless it close to plug and play, and I had some troubles with some of my wallets, which are more difficult to have what is thought to be both open source and self-custodian for normies (like me) who so far are not running their own lighting network nodes.

Ay franky ... this is simply my posting style. I'm not afraid of errors in my calculations and thus I tend to formulate my posts a bit cautiously.

But in this case I'm about 90-95% sure that my analysis is correct and there's indeed a low five-digits renewal of LN channels per year (or a high four-digits number), even if many of them are relatively short-lived.

Perhaps this could indeed been another function for a possible "LightningBuddy": to scrape also the percentiles. If they're relatively static (i.e. the median age of the first and the fifth percentile don't increase significantly) then my assumption has been proven. But of course this is a bit a "niche" application. Perhaps I'll scrape these values privately and also analyze the archive.org archived numbers if they exist ...

And yes, Lightning could indeed be more popular. I have however never been one of those who thought that LN is the ultimative and only scalability solution. You may confuse me with other forum users ... IMO it is excellent as a micropayment solution but has to be complemented by other strategies, but not L1 big blocks

you make alot of probably's, assumptions, approximations and estimations in your numbers and then say your numbers are "a fact that"... but even then your assumptive estimated numbers are still not numbers you like...

You're not wrong, but it's probably also because it's merely what the Lightning Network currently needs? In fact, it might be too much because there's very low demand for Lightning transactions. There's an opportunity cost for keeping capital locked in those channels, but those users will probably be just be HODLing them in cold-storage. Plus it's good to have some altruism to keep the network liquid and ready to be used if users need it.

While what you post is interesting, it doesn't contradict my interpretation of the "new node" stats.

Let's take the newest 1% which are on average 10 days old. That's about 120 nodes. Only this percentile already makes up for a renewal of around 4000 nodes per year, because we can assume safely that each 10 days 1% and thus around 120 nodes are created.

Percentile 5 has 2 months of average age, this means that 500 nodes are newer than (approximately) 2 months, and we can assume that this number was similar during the last year.

Let's make a rough estimation about the first 5 percentiles:

- Percentile 1: 120 nodes: 10 days => 4380 renewed (120 * (365 / 10))
- Percentile 2: 120 nodes: 20 days => 2190
- Percentile 3: 120 nodes: 30 days => 1460
- Percentile 4: 120 nodes: 45 days => 973
- Percentile 5: 120 nodes: 60 days => 730

=> Result percentiles 1 to 5: 9733 nodes

Note: Channel/node stats were slightly falling in the last year, so up to this point perhaps we get only 9000-9500, but not much less.

You have to sum some more percentiles (those that are less than a year old, i.e. up to percentile 25). So in reality there is a renewal of probably more than 12000 nodes per year. It doesn't contradict your findings because indeed there are a lot of old nodes, but the fact that there are a lot of short-lived nodes leads to a renewal of a lot of nodes during the year.

It's still not the numbers I'd like to see. But it's also not exactly "dead".

nope
because if you observe, 50% of nodes are not new but over 2 years old, only 5% have reset in the last 2 months, only 25% have reset in the last year

Average Node Age  1,078.2 days  2 years
Percentiles

1    10 days
5    2 months
25    a year
50    2 years
75    4 years
95    6 years
99    6 years

observing real stats is fun, but you making up stats is just funny
in no way have 14,600 nodes of 12,000 reset in the last year

based on 12000 nodes, only 3,000 reset in the last year

also, and heres the killer
https://1ml.com/node?order=capacity

you can look into the top capacity nodes and realise they have been around for years.

oh, last fun fact
lets take okx (top cap)
couple years ago it used to have more channels than coin so for instance each channel had 0.5btc average/channel  in 2022
so under YOUR ASSUMPTION the channels should have less than 0.5btc per channel now due to your fiat price assumption

HOWEVER they now have more capacity than channels. meaning they have more coin per channel
thus they are not taking out coin due to fiat price adjustment

have a nice day

It's not necessarily "dead". I would rather qualify it as currently "stagnant", but even that might not necessarily be true (see below). Very likely some of the early Lightning hubs and nodes have dismantled, while new ones are appearing. In this list from 1ml.com, for example, about 40 nodes were detected as "new" in the last 24 hours. This is not exactly a lot, but we can thus calculate that at least 10,000 new nodes pop up per year (365 * 40 = 14,600). The current node count is of about 12,000. This seems to hint that many nodes/channels are "renewed" regularly.

In this context, a relatively stable Bitcoin "capacity" hints even at a growth. When someone creates a new Lightning node, he will very likely calculate its capacity according to the USD/EUR/other fiat value, not the BTC value. He'll not say "I'll create a LN node with 0.01 BTC" but "a LN node with a capacity to be able to transact regularly for about $1000" (except for some BItcoin maximalists/early adopters). As the BTC price has grown from late 2022 to now, this means that probably people are planning either more channels or have more confidence to increase the capacity of their channels.

if you observe, you will find its been HOVERING around 5.2k btc for 2 years.. also average node age stats are over 2 years(over 1000days).. so basically the network has been dead for 2 years with not much up or down movement on average.. just value remaining locked and stuck in limbo

5210.42 Bitcoins, that's more than $500,000,000 with the current price of Bitcoin at $97,000. That's actually relatively good liquidity if it's compared against some DeFi applications in Ethereum. Plus its possible "function" as a "privacy" layer is not well-known yet. If it becomes more common to use it as a "privacy" layer, demand for liquidity could increase, which should also cause an increase for network capacity.

by the time Libra was shut down, it had to change so many features due to regulatory demands that the codebase would have made banks' COBOL systems blush.

Ah, thanks! I didn't know this feature of mempool.space, and it seems it also shows some historic charts for channels and nodes, at least for the last year or so. It's not as intuitive as BitcoinVisuals' charts though and a bit more limited. But maybe then the LightningBuddy isn't that necessary as I thought. Of course it would be a nice addition to this thread anyway. I think a weekly interval would be best, so the thread is regularly bumped but not daily.

My source for the last post was https://1ml.com/ , which seems to be a quite established service with a lot of indicators to choose from. Only the historic chart feature is missing.

I can take a look at it.

What is the best source of this information?

There is something of lightning in mempool.space

https://mempool.space/lightning

Looks like I can do basically the same I do with feebuddy, as info is easy to get in mempool.space

Libra/Deim is a dead project.  Facebook tried to push such a project and it got attacked, so they had to stop the project.   It is not possible to work on a dead project, is it?

It seems the BitcoinVisuals chart is indeed not updating anymore. It keeps ending on September 26.

It could be an interesting idea to create a "LightningBuddy" bot for the forum which scrapes the data from 1ml, at least once per week or so (not so frequent like the ChartBuddy in the WO or the FeeBuddy in the Mempool Observer).

Or is there any other publicly accessible website with historic chart data about nodes, channels and Bitcoin capacity?

We could also do it manually of course, if the OP has nothing against it. The current data of November 30 (around 18:30 UTC) for the three most important stats are:

- Number of Nodes: 12,137
- Channels: 46,654
- Bitcoin Capacity: 5,210.42 BTC

Even if of course LN stats are not as perfectly accurate as on-chain stats, I think this would give some insight in how the network is evolving.

There's an actual company that's currently building something on top of the Lightning Network. LightSpark, https://www.lightspark.com/

I was told that one of the people behind LightSpark is David Marcus - one of the founders of Libra/Deim. 👀

John Carvalho, in the Bitcoin Takeover Podcast, said that it's very hard to build on top of Lightning because, according to him, the network has its "design flaws". David Marcus, as an alternative blockchain/network developer, then why would he leave Libra/Deim and pivot to Lightning?

This is the most challenging part: adapting to the new metric.
And, of course, it depends on the bitcoin price, as it is used to do fiat denominated payments.
I bet this will change when payments will be sats- denominated!

Thanks also to Franky for being so Franky!

If all that's true then I'm going to be in big trouble if I don't open self-sustaining channels before the fees shoot up. 


You know what, screw it. Time for me to spin up my own image host. (That includes an image proxy which caches talkimg images of course.)