Pages:
Author

Topic: We could be sitting at $10,000 if we solve the scaling situation with no drama - page 2. (Read 1360 times)

legendary
Activity: 1372
Merit: 1252
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'

Even Greg Maxwell admitted the other day that segwit is not required for LN.  

Sure it's not required, just like saying you don't require 2 hands and 2 legs to live. You live yes, but it sucks. Here's Greg Maxwell talking about how not having segwit sucks for pretty much everything:

https://www.youtube.com/watch?v=LHPYNZ8i1cU

Why not do segwit as a HF and make sure all tx have their witness data segregated?  Yes it would take time but there's no rush for it.  What IS urgent is capacity.  If we're going to agree to a HF for blocksize, why not allow that to be done first?

Because you can't do a hard fork urgently without fucking up big time. There's also technical reasons to do segwit first, hard fork later. There's not a single argument that makes sense about a HF before segwit, specially when you add the word "urgently" in the same sentence as "hardfork".
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'

Even Greg Maxwell admitted the other day that segwit is not required for LN.  

Sure it's not required, just like saying you don't require 2 hands and 2 legs to live. You live yes, but it sucks. Here's Greg Maxwell talking about how not having segwit sucks for pretty much everything:

https://www.youtube.com/watch?v=LHPYNZ8i1cU

Why not do segwit as a HF and make sure all tx have their witness data segregated?  Yes it would take time but there's no rush for it.  What IS urgent is capacity.  If we're going to agree to a HF for blocksize, why not allow that to be done first?
legendary
Activity: 1904
Merit: 1074
You might be missing the point here..... The battle is not really about the technical changes.... but rather who will be in control. Most of these parties use

the technical differences as an excuse, but it is actually about "control" The Bitcoin Core team can announce a 5mb block size and the most perfect

solution and the other side will still not be satisfied. The "control" is everything.... it includes "brag right" .... everyone wants to be the "winner" in this

battle.... so it is more about pride and self gratification, than the technical merit of this whole issue. This situation is so pathetic at the moment, it is not

even funny anymore.  Angry
legendary
Activity: 4410
Merit: 4766
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'

Even Greg Maxwell admitted the other day that segwit is not required for LN.  

i know this and so do many. but the guys making LN (rusty russell:blockstream) are using functionality based on elements:segwit for thier elements:LN and so for theirs it all becomes dependant on it.
legendary
Activity: 1372
Merit: 1252
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'

Even Greg Maxwell admitted the other day that segwit is not required for LN.  

@cellard -- Not that I believe bitmain is using it, but  i'm pretty sure we can kill covert asicboost in other ways than SWSF.

Sure it's not required, just like saying you don't require 2 hands and 2 legs to live. You live yes, but it sucks. Here's Greg Maxwell talking about how not having segwit sucks for pretty much everything:

https://www.youtube.com/watch?v=LHPYNZ8i1cU

We need segwit anyway, there's no reason to circlejerk about it, just activate it already as a SF to not keep wasting time, we solve two problems at once, then we get 2MB HF later, properly planned with enough time for everyone to upgrade (aka 1 year at least)
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'

Even Greg Maxwell admitted the other day that segwit is not required for LN.  

@cellard -- Not that I believe bitmain is using it, but  i'm pretty sure we can kill covert asicboost in other ways than SWSF.
legendary
Activity: 1372
Merit: 1252
if there is going to be a hard consensus then there is no need for the cludgy 2merkle upstream filter/stripping tier network workaround soft crap

the ultimate 'compromise' which removes the cludgy code of bad math manipulations, tier networks etc. is

1merkle block of 4mb (not 2merkle and not a 2mb Base 6mb witness, nor 2mb Base 4mb witness... just simple 4mb single area weight)
all keypairs. native, segwit, schnorr sit in the same single block area
tx sigops limit at or below 4k forever
tx kbyte limit at or below 50kb forever
POLICY (soft block limit) increments at 0.25mb(like it has dont in last 8 years)

new fee priority formulae - based not on value. but age of tx and bytes of tx

remove the 'average fee estimate' and return the 'reactive fee'
add in the other features everyone wants that can only happen as a hard fork

and then everyone gets to have their cake and eat it


going for a 2mb 2merkle (2mb base 6mb witness) is just pushing the cludge down the river.

if they want segwit first then do the 1mb base 3mb witness(4mb weight).
then as a hardfork remove the 1mb base so the '4mb weight' becomes the block limit. and all keypairs sit in the same area and all the cludge of going soft gets removed without causing more issues than necessary



lastly.. this is the most important thing. read it 3 times. have a coffee and let it really soak into your minds.
rushing to activate segwit is MEANINGLESS
segwit 'activation' only creates the tier network. yet its the need of people to move funds over to segwit keypairs AFTER (emphasis AFTER) that which is where the 'gestures' of promises that have been made for 18 months could occur.

there is no way of moving 46million outputs into new segwit keypairs without it causing atleast a year of further delay ant normal pace before seeing any noticeable benefit. or causing a megastorm of mempool bloat and tx fee war of everyon rushing to use the new keypairs. which then defeats the benefit of pretending to reduce tx confirmation time/cost issues.

take your time and really let the last 2 paragraphs settle in your mind and realise the activation alone is not the hallelujah moment of segwit utopia

We need segwit as a soft fork to kill the covert Asicboost scam from Jihad Wu which is a major cancer for the community. This is basically why he doesn't want it as a soft fork.

We need segwit as soon as possible to let lightning network tech to develop on bitcoin for actual global microtransaction scaling, otherwise we are asking for litecoin to take the lead. Litecoin is getting added on bitstamp too so it continues making moves.

Lets get segwit now, then a HF later, there's no other way around this, unless you want UASF nodes to skyrocket.
legendary
Activity: 4410
Merit: 4766
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

i have spoke to many that actually understand segwit flaws/limitations/realities (meaning the smarter crowd, not the reddit scripters) but still want it. and ultimately its the HOPE of them 'getting paid to run a LN node'
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
If we are able to activate segwit without no major drama, we can be sitting at $10,000 in No Time. As people realize bitcoin has solved the scaling solution, and lightning networks are coming with finally features to compete against existing payment networks, while retaining the features of a gold-like digital gold for those that couldn't care less about buying coffee with bitcoins and just wants to hold, the price would explode because everyone will be happy.

I say compromise. Throw the big blockers a bone, let's raise the blocksize to 2MB. But first, the compromise on the other side must be that segwit gets activated ASAP with hashrate agreement to avoid UASF.

As pointed out by anyone with a functional brain, the code must be developed, and peer reviewed by Core devs. 95% of people trust only Core, you can't pretend to do this without them.

Let's give time for the hard fork. One year, then we hard fork, safely, not rushed trash.

There's tons of great technology that require a hard fork, hopefully some can be included too, so the hard fork does not get wasted to only raise the block size:

https://bitcoinhardforkresearch.github.io/

But if there's no time after 1 year, oh well, then whatever, just do the fucking hardfork to get the 2MB blocksize, as long as we get segwit NOW without UASF I couldn't care less at this point.

Lets get our shit together and get this done, otherwise YOU ARE ASKING FOR UASF TO HAPPEN, and this could cause a crash due the confusion of the 2 tokens. Let's not kill the amazing downtrend with internal drama and let's find a fucking way to get segwit in already.

UASF is already working because BU is no longer on the picture in case you haven't noticed, now we just need the so called 80% hashrate to agree on a proper hard fork and not the frankeinstein segwit HF proposed by buffoons in suits.

We have still time, if not, then get ready because UASF will only continue gaining traction.

 
I'm not saying segwit is bad, but I don't quite get why a lot of people are so hungry for it.  Maybe you can explain why you want it so badly in the first place?

legendary
Activity: 4410
Merit: 4766
if there is going to be a hard consensus then there is no need for the cludgy 2merkle upstream filter/stripping tier network workaround soft crap

the ultimate 'compromise' which removes the cludgy code of bad math manipulations, tier networks etc. is

1merkle block of 4mb (not 2merkle and not a 2mb Base 6mb witness, nor 2mb Base 4mb witness... just simple 4mb single area weight)
all keypairs. native, segwit, schnorr sit in the same single block area
tx sigops limit at or below 4k forever
tx kbyte limit at or below 50kb forever
POLICY (soft block limit) increments at 0.25mb(like it has dont in last 8 years)

new fee priority formulae - based not on value. but age of tx and bytes of tx

remove the 'average fee estimate' and return the 'reactive fee'
add in the other features everyone wants that can only happen as a hard fork

and then everyone gets to have their cake and eat it


going for a 2mb 2merkle (2mb base 6mb witness) is just pushing the cludge down the river.

if they want segwit first then do the 1mb base 3mb witness(4mb weight).
then as a hardfork remove the 1mb base so the '4mb weight' becomes the block limit. and all keypairs sit in the same area and all the cludge of going soft gets removed without causing more issues than necessary



lastly.. this is the most important thing. read it 3 times. have a coffee and let it really soak into your minds.
rushing to activate segwit is MEANINGLESS
segwit 'activation' only creates the tier network. yet its the need of people to move funds over to segwit keypairs AFTER (emphasis AFTER) that which is where the 'gestures' of promises that have been made for 18 months could occur.

there is no way of moving 46million outputs into new segwit keypairs without it causing atleast a year of further delay ant normal pace before seeing any noticeable benefit. or causing a megastorm of mempool bloat and tx fee war of everyon rushing to use the new keypairs. which then defeats the benefit of pretending to reduce tx confirmation time/cost issues.

take your time and really let the last 2 paragraphs settle in your mind and realise the activation alone is not the hallelujah moment of segwit utopia
legendary
Activity: 1372
Merit: 1252
If we are able to activate segwit without no major drama, we can be sitting at $10,000 in No Time. As people realize bitcoin has solved the scaling solution, and lightning networks are coming with finally features to compete against existing payment networks, while retaining the features of a gold-like digital gold for those that couldn't care less about buying coffee with bitcoins and just wants to hold, the price would explode because everyone will be happy.

I say compromise. Throw the big blockers a bone, let's raise the blocksize to 2MB. But first, the compromise on the other side must be that segwit gets activated ASAP with hashrate agreement to avoid UASF.

As pointed out by anyone with a functional brain, the code must be developed, and peer reviewed by Core devs. 95% of people trust only Core, you can't pretend to do this without them.

Let's give time for the hard fork. One year, then we hard fork, safely, not rushed trash.

There's tons of great technology that require a hard fork, hopefully some can be included too, so the hard fork does not get wasted to only raise the block size:

https://bitcoinhardforkresearch.github.io/

But if there's no time after 1 year, oh well, then whatever, just do the fucking hardfork to get the 2MB blocksize, as long as we get segwit NOW without UASF I couldn't care less at this point.

Lets get our shit together and get this done, otherwise YOU ARE ASKING FOR UASF TO HAPPEN, and this could cause a crash due the confusion of the 2 tokens. Let's not kill the amazing downtrend with internal drama and let's find a fucking way to get segwit in already.

UASF is already working because BU is no longer on the picture in case you haven't noticed, now we just need the so called 80% hashrate to agree on a proper hard fork and not the frankeinstein segwit HF proposed by buffoons in suits.

We have still time, if not, then get ready because UASF will only continue gaining traction.

PS: Before some idealist claims bitcoin price doesn't matter, well get real, it does. Let's try to do this smoothly, to create a less disruption on the price as possible.
Pages:
Jump to: