Topic: Bitcoin round table (Read 2889 times)

maybe Bit_happy should have said 7 days..  because i dont think Lauda realises that 1000 blocks is just 7 days..

i said it ages ago 10,000 block trigger(70 days) PLUS X months grace period for the remaining 25% laggers to finally move over seems better than 7days+28

If only someone warned us about this blocksize thingy 1 year ago & maybe proposed "raising the block size limit"...

A wise man accepts he doesn't know everything.
A fool believes he knows everything.

You're asking a very tough question. Nobody in this world can tell you the right solution nor is there a perfect solution. I'd say the only thing that we could decide to do within 3 days (16/02/2016) is a HF to raise the block size. However, I'd say a 1.5MB block size limit with the 'workaround' that Gavin used in order to prevent blocks that take too long to validate. Albeit, this is where thingss becomes tricky. You can't rush a HF and choosing a grace period that is too short could cause a negative effect. Even Garzik said that the minimum grace period should be 3 to 6 months. By this time Segwit would be ready for deployment. One could add a HF as a backup plan if Segwit fails to deliver on time, however the developers are certain that everything will be properly coded and tested in time.

---

---

Tl;dr: I'd say that you just can't solve it in '3 days' and especially not right now. If you had asked me this 1 year earlier, then raising the block size limit would be my final answer (most likely).

...there is no way to implement those upgrades within 3 days
Sorry that wasn't clear:
What if the block size debate had to be solved in ~3 days or less?    ...implementation is ASAP.

>newfangled breadboard
>not manstyle wire wrap
Peasant.

This is always a valid concern since banks can always print billions of dollars and build thousands of farms and take over 90% of hash power. But I think in the end, people have a choice in selecting which version of software they run, bank can not force bitcoin on you like they did with fiat money. So it is ultimately decided by the people's awareness of the code that they run

In another word, if you understand the candidate you can make better decision, obviously that is not even possible for presidential election  

like you're always right? right? i think you need to stop being so cocky, really....

i dunno anymore who is worse, those that ignore the other while they think they are always right and actually post no-sense, or sig spammer

So 1 wire = 1 component, good to know. Stop with the nonsense. It's quite easy to spot the people who won't admit that they're wrong. You're not even trying hard enough.
Again, how is this relevant to OP? Because it isn't.

http://www.threadabort.com/?p=9

Even if we had very good solutions to deploy right now, there is no way to implement those upgrades within 3 days via soft/hard fork without causing losses. Albeit one could argue that if the miners really wanted it you could push a soft fork really quickly (so far, it took weeks to activate one), but this is not the case with a hard fork (as everyone matters). There is no way of solving it in such a small amount of time without causing significant loses/confusion/chaos.

If the block size debate had to be solved in ~3 days how would we get it done starting right now?

The average joe that acts irrationally? I guess that might be true.

I don't need to. There are better ways to learn boolean algebra than "from nand to tetris". I've yet to see how this is related to the OP.

People usually worship something they don't understand. I recommend you read some fundamental computer knowledge, that battery, wire and bubble experiment is a good start, then followed by "from nand to tetris". Even a teenager can build a computer using 3000 components, it is not rocket science, far from it, just layers and layers of translation

Very bad analogy. You're comparing a computer language to a human language. Learning another language is pretty much useless (aside from being able to communicate with more people). With computer languages the possibilities are 'infinite' (limited by imagination, skill and the technology though). Apparently you think that CS majors are only about computer languages, which is very wrong.

Great to know. I used to run Google in my free time.

---

---

As always: personal incredulity.

I used to manage thousands of these so called experts, they just can master computer language like 000101010100000101111, but it doesn't mean they can think better than others. Just like you don't understand chinese/japanese does not make you less intelligent than chinese/japanese people

“今者臣来，见人于大行，方北面而持其驾，告臣曰：’吾欲之楚。’臣曰：‘君之楚，将奚为北面？’曰：‘吾马良。’臣曰：‘马虽良，此非楚之路也。’曰：‘吾用多。’臣曰：‘用虽多，此非楚之路也。’曰：‘吾御者善。’此数者愈善，而离楚愈远耳。

A fine case of personal incredulity. Just because you are cognitively limited and can't understand those concept that does not mean that they are confusing/bad/similar to FED or whatever nonsense you're going to spit up next. People with relevant degrees (yours is not) understand the underlying infrastructure and code. Various experts have and will review these technologies. Lightning Network and Sidechains are the way to scale Bitcoin unless someone comes up with something better. Scaling via the block size is inferior.

---

---

I do wonder when your contract ends.

People's choice  

It's simple, if it is a complex and no-one-can-understand scheme supported by a large group of decision makers, then it is definitely centralized decision making. But if it is a 1+1=2 like simple and easy-to-reach-consensus scheme supported by a large group of decision makers, then it is just they align with major consensus

If you start to use QE/FRB like terms like SW, LN, SC, then you are getting more and more close to Federal Reserve way of working, trying to confuse people with strange terms so that they don't understand what is going on

In worst case scenario you still have some people running dial-up network and GSM phone (in fact a lot of people), so the bitcoin data traffic should be reduced by 100 times following your logic

If you prefer settlement system, then I'm thinking about a settlement layer for networks. You do the data settlement on backbone internet between thousands of full nodes with 1GB bandwidth, while the rest of the nodes are running SPV nodes and require much less bandwidth. This model is much simpler to handle than a financial settlement system


Semiconductor technology has almost reached its physical limitation, so the current trend is just adding more chips, like you did in GPU mining instead of CPU mining

However the theoretical limit of Optical Fiber has not been reached yet, and average home are not using Fiber yet (Some part of the world is deploying fiber to home) a fiber in theory can at least handle TB level data. And today's commercial solutions already can provide 100GB bandwidth. If you have 5000 private companies running nodes on 10GB network, we already can handle the traffic for the next 10 years


It is obvious that Moore's law does not apply with user growth either, so you won't have exponential traffic volume growth. In fact I think in one decade there will be maximum 10 million users, which you already can handle with today's infrastructure

Yeah, that "solution for the unbanked" was just a sales pitch, didn't think anyone took that seriously. But fine, if you need things spelled out: We lied, now what?

To be fair tho, the great unwashed don't need to run a node: a web wallet (for the well-to-do hoi polloi in possession of a smart phone/3G/4G interweb) would be an upgrade to the tin beggar's cup their substantial life savings are otherwise secured in.