[HN Gopher] Why Decentralization Matters (2018)
___________________________________________________________________
Why Decentralization Matters (2018)
Author : janandonly
Score : 40 points
Date : 2022-06-15 10:06 UTC (12 hours ago)
(HTM) web link (onezero.medium.com)
(TXT) w3m dump (onezero.medium.com)
| XxiXx wrote:
| Laziness is the enemy of decentralization. All those crypto
| shenanigans won't change a thing about this.
| criddell wrote:
| This article was written four years ago. Web3 wasn't a
| punchline yet.
| janandonly wrote:
| Paywall-less version here: https://scribe.rip/why-
| decentralization-matters-5e3f79f7638e
| ross-sec-audio wrote:
| This was written by a venture capitalist who makes a tonne of
| assumptions about the future:
|
| "The internet is still early in its evolution: the core internet
| services will likely be almost entirely rearchitected in the
| coming decades. This will be enabled by crypto-economic networks"
|
| There are definitely a lot of problems with the current climate
| of tech monopolies, but the issue they describe is not related to
| them being centrally managed (see why decentralisation section).
|
| I'm glad the author admits that web3 doesn't currently work due
| to fundamental issues with performance and scalability.
|
| Fix that, and then lets see what you have to offer. Right now,
| people are being duped out of their money on a daily basis. Any
| gains made through specalative crypto investments are someone
| elses loss.
| jamesliudotcc wrote:
| Says someone writing on Medium.
| abnercoimbre wrote:
| I appreciate a good argument supporting decentralization, but
| promoting cryptonetworks weakens it. From Adrian Colyer's paper
| [0]:
|
| > The top four Bitcoin miners combined have more than 53% of the
| average mining power. In Ethereum the top three have 61% of the
| average power. Miners do change spots in the rankings over the
| observation period, but place is contested by only a few miners.
|
| [0] https://blog.acolyer.org/2018/02/16/decentralisation-in-
| bitc...
| tromp wrote:
| The article confuses miners with mining pools. Each pool can
| have hundreds of contributing miners, who are free to switch
| mining pool any time.
| smarx007 wrote:
| Hear, hear! I see no reason why can't we go back to sharing
| information across sites on many domains hosted on a number of
| small servers, like in the 90s. For example, having a Raspberry
| Pi in every home. This sounds like decentralization to me.
| api wrote:
| I really hate how cryptocurrency has sucked all the air out of
| the room with decentralization. The two are only somewhat
| related.
| cmckn wrote:
| Federation is more fun IMO than the antagonistic, "trustless"
| crypto ecosystem. It's also generally cheaper to participate
| in, whereas crypto nodes and miners will only get more
| expensive as it scales.
| guywithahat wrote:
| Well it's the only one that's really taken off. The fediverse
| started to get bigger until mid-sized nodes started making
| ban lists and the largest nodes left, meanwhile nothing new
| is happening with torrents and chrome killed off native peer-
| to-peer libraries when they started being used by YouTube
| competitors
|
| As much as I personally dislike how crypto has been pitched
| it's the only field where decentralization has really entered
| the mainstream
| PeterisP wrote:
| The things you list seem actual examples where
| decentralization _did_ take off, but faded once people
| recognized the tradeoffs of other features implied by
| decentralization and the mainstream audience chose that
| they do not really want decentralization - i.e. it 's not
| that it did not yet get a chance to take off, but rather
| that we-the-peopled tried it out and practically found out
| that for most use cases we prefer centralization.
| splitstud wrote:
| TurningCanadian wrote:
| > Centralization has also created broader societal tensions,
| which we see in the debates over subjects like fake news, state
| sponsored bots, "no platforming" of users, EU privacy laws, and
| algorithmic biases. These debates will only intensify in the
| coming years.
|
| This doesn't make any sense to me. It sure looks like motivated
| reasoning. What's the connection? The fewer sites there are, the
| more fake the news there will be? Why? State sponsored bots
| benefit from there being fewer platforms? When everyone had a
| blog, were blogs ever known for keeping out comments from bots?
| EU privacy laws only apply to large centralized sites? C'mon. I
| can see "no platforming" of users and algorithmic biases being
| related to centralization at least.
| kareemsabri wrote:
| Perhaps they mean the _debate_ over these subjects (fake news
| etc.) is exacerbated by the existence of a centralized
| authority to appeal to to remove it. Before Facebook I don't
| recall much debate over removing something from the broader,
| decentralized internet.
| TurningCanadian wrote:
| That makes sense, thanks. There were DMCA takedowns before
| centralization, and there was certainly debate over SOPA/PIPA
| though. Other illegal content was taken down in the past too.
| (Libel, exploitative media, etc)
|
| I think the difference now is that we've widely experienced
| actual real-life impacts of unfiltered & unregulated but
| widely circulated speech. Attacks by incels and
| dis/misinformation about COVID come to mind. Many people now
| realize that they and/or society at large been exposed to and
| tricked by malicious actors, or at least seen their effects.
| There have been several protests stirred up by foreign
| operatives for example. IMO that shouldn't be legal, but in a
| way it's another form of yelling "fire" in a crowded theater
| that we have to grapple with defining and regulating.
| cuteboy19 wrote:
| Are they saying that the GDPR doesn't apply to you if you're
| "decentralised"? And that's somehow good?
| Barrin92 wrote:
| Can we talk about the use of the word 'decentralization' in
| context of crypto? I think it's deeply misused because these
| systems are the opposite. They're global state, single sources of
| truth. _Distributed_ , not decentralized. This is not just
| semantics but explains a lot of their features, like the reason
| they can be a currency, or why they're so slow. (pushing
| everything through a global state machine is a lot of work).
|
| Decentralized systems matter but they do away with a single
| source of truth, that's what makes them responsive. They have
| local consensus, not global consensus. Torrent networks are
| decentralized, protocol based chat apps are (riot/matrix),
| Mastodon is, and so on.
|
| When articles like this say that performance of crypto networks
| is a 'technical problem' to be solved I don't think they
| understand that it's not a problem but a necessary and intrinsic
| feature because to maintain global consensus in the absence of an
| authority you need to impose costs on the users of the network.
| That's why crypto systems aren't comparable to the decentralized
| web.
| cyanydeez wrote:
| If they were just a state machine, they'd be useful.
|
| They're meant to be a pseudo physical state machine where
| changing the state takes real work. If you decoupled the
| needless work aspect, you'd probably end up with a useful value
| proposition (which no one would adopt)
| mad44 wrote:
| Replying to "Why Decentralization Matters"
|
| https://muratbuffalo.blogspot.com/2018/03/change-my-mind-abo...
___________________________________________________________________
(page generated 2022-06-15 23:01 UTC)