Why you should not built on AE now

Don’t get me wrong, i’am an AE believer, lover and holder, but decentralization with POW by now is just wish thinking in my opinion.
Just a look at the mining pool hashrate distribution shows that it’s everything else than decentralized. ~99% on 3 chinese pools, far away from decentralization, leave alone the attack vectors this generates.

Offtopic mostly, but maybe a look into this paper will show up different attack vectors which comes with such mining centralization. Not too many think about this:

The Looming Threat of China: An Analysis of Chinese Influence on Bitcoin
Ben Kaiser, , Mireya Jurado, Alex Ledger
1 Princeton University, Princeton, NJ 08544, USA
2 Florida International University, Miami, FL 33199, USA

Yes, we are well aware of that threat and doing everything in our power to prevent centralization - we are currently working on Stratum support and enabling any æternity node to become a mining pool.

However, it’s worth noting that we are still in the early stages and the blockchain is young, so we’ll see how it’ll develop. Bitcoin-NG is the best solution in terms of code we’ve found but doesn’t really work without mining decentralization.

Thanks for the analysis, I’ll take a look at it.
And thanks for your support :slight_smile:

Best,
Albena

1 Like

The analysis is a must read when it comes to security aspects not many think about it as most relate only 51% and double spend attacks to security but there are way more nasty attacks even mining pools can generate on other mining pools and so all.
As said, it’s a must read in my opinion and as we all know, learning never ends :slight_smile:

Good luck and all the best as well.

1 Like

This is interesting. However, I have seen numerous opinions on how PoW is “bad” and despite having shortcomings, I still believe that currently it is the only way to secure a blockchain network. There is really no viable alternative. I am pretty sure that a similar document devoted to PoS will be much longer :slight_smile:

If you consider China a threat, then PoS is hardly any better, most probably it is even much worse. Blockchain systems are created to be as resilient as possible. There are always a number of theoretical attack vectors, but so far Bitcoin has persevered. You can consider attacks as viruses that the system must find a way to cleanse itself from. Adaptability is an important trait for public blockchain systems. The expression " what does not kill me makes me stronger" is fully applicable in this case.

Generally, a system can always be devised in a “better way”. However, “better” is subjective and sometimes becomes clear only after a period of time in which the system has been “in the wild”.

I am completely with you on the “Chinese mining pools constituting 99% of mining power” issue. The entire team is and we are working heavily to address this. However, mining is a competitive business and follows the lowest electricity price. Developing countries will always be a better place to run a mining operation for this and a number of other reasons.

Best,
Vlad

3 Likes

Interesting arguments, no doubts.

First of all i think POS is not POS, there are many different POS designs to fix different problems. While i agree that many POS designs aren’t good either some seem to be good, at the price of some centralization of course, see the trilemma problem in general.

Than again, my point on mentioning POW not being fair was as an answer to unfairness in the AE POW described by an other user some posts above. I still think that compared to POS, leave alone for a moment the security aspect, it would be more fair compared to POW mining. 1 AE costs just right now 42 cents, no matter if you buy that AE in China, USA or wherever. In contrast mining 1 AE is quiet different . A miner in a low cost region like inner mongolia might mine (fictive!) 1000 AE with an initial investment of 100$ including electricity while the german miner for example will mine just 50 AE with the same initial investment.
(Didn’t make a true calculation, just fictive, but i guess you get the point). There is no fairness at all and as said allready it’s leading to mining centralization, especially at lower coin prices.
There is one more factor, the higher the initial investment in a low electricity region the higher the gain. At least in POS this is reverse, the higher you want to get your stake, the more you have to pay as the price rises in that case with every coin you buy for your stake. Only comparing here fairness and decentalization, not the security.

About: If you consider China a threat. It doesn’t matter what country it is, it’s a problem of centralization in my opinion. If for some reason all the mining pools and mining facilities would be in the USA i would adress the USA of course, but as it’s currently china with a somehow repressive and controlling system i have to adress just … china.

About “what does not kill me makes me stronger”, sure, true, enough you don’t get killed and survive it. That’s thin ice here in my opinion. I think we all have witnessed enough examples where a given project didn’t recover from a serious attack, missconfiguration, bug, centralization and so on. There is no garantee someone WILL survive it. The higher the coin/project is by market cap the more responsibility is needed by the dev team.

What i absolutly do not agree is that by now mining is a competive business. This might have been the case until 1 year ago. It doesn’t hold true by now, even less in some months/years. It’s simply concentrated to some chinese regions with electricity access in the range of 1.5-2 cents per kw/h, some former soviet republics with electricity access in the range of 2-5 cents and that’s it all over. Having in mind that the world average price for electricity is 13 cents it’s no rocket science to realize that most parts in the world are not competive in the mining business, hence, there is no competition by now, hence this argument is truly wrong in my opinion.
Not only have chinese mining facilities the advantage of low electricity costs but as well have the advantage of less investement into hardware as they don’t need to pay for shipping, tax and custom fees, VAT, and like in the case of asics the producers run their own mining facilities at lowest hardware self costs. That’s not competition when you have ALL hardware manufactures in china & lowest electricity in china. That’s actually a monopoly scenario.
I made several researches on geographical mining centralization allready, there is no doubt that this happens right now. Even worse, there is even mining centralization within china itself to 2 regions, inner mongolia and shenzen, what’s left for the reset of the world … Just as a side note on how POW is flawed today from a decentralization point of view. Using the word decentralization while relying on POW is allready missleading in my opinion, no matter your personal/projects effords and goals are different and aim for decentralization.

However, i again, suggest to read this research, especially the attack vectors that apply for censorship, mining pool attacks and internet attack which most don’t give any attention too. Not that it will change anything, but at least it gives a good picture what future at some time will bring. Better aware than surprised, not?

Even i’am critical on POW in general nowadays i have full faith in the AE team and hope you all make the best out of the situation, now and in future.

2 Likes

Nice! Thank you for taking your time sharing your thoughts and being so supportive. I will make sure your comments get shared with the team.

Best,
Vlad

2 Likes

The Blockchain trilemma is so 2017. Have you not been paying any attention? AE scales exactly because they added Microblocks. The way they were implemented is a mitigation of lowered security. Attack surface can be argued - check the threat model on GitHub.

So your post is answered, in full, in the readmes on GitHub and possibly the Whitepaper. And there is this: Bitcoin-NG: A Secure, Faster, Better Blockchain

1 Like

Thx for the link and education :slight_smile:
I have readed too many white papers, gifthub channels and whatever not too take everything as real and given.
This doesn’t mean bitcoin-ng isn’t a good solution and i admit i haven’t digged deeper so far.

I argue that the blockchain trilemma is soooo 2017, it’s still an issue even now. If it wasn’t just show me an absolute secure, absolute decentralized blockchain project with 10,000+ transactions per second on POW.
Let me write the (your) answer here, there is none, right? Too bad, this means the trilemma problem is more than valid even today.

Now if you call 100 transactions per second fully scalable, so it be, for me it’s a minor improvement (just as an example as i had no time to read into Bitcoin-NG!)…

One more thing, is there any link to a stress test on Bitcoin-NG with all results? Did AE make allready such tests and where can i find the results please?

I think 100 tx per second is enough for on chain. 10000 tx per second better run on a layer 2 network like state channel or lightning network.

On-chain is for settlement. Blockchain is the settlement system for crypto currency. Only need to record all the results, not need to store all tx on chain, which require nodes to be data centers.

1 Like

I agree that it might enough, at least for now, but the question wasn’t if it’s enough or not but the trilemma problem a user claimed to be old-fashioned 2017 thinking by now, which in my opinion is NOT true and still fully applies…

I refer you to Emin Sirer, inventor of Bitcoin-NG. He should.have Simulations: Information for E. Gün Sirer :: Hacking, Distributed

2 Likes

This is the reason why you should not build on AE for now.
For a currency, numble of users and amount of capital are key factors to be considered.

I suggest you go to sleep and come back in two years. We will show you then that you were wrong. Closing this thread now. Further FUD will not be tolerated @xiahui135 . Not here and not in telegram.