The Internet Is a Potemkin Village: Proof of Dead Internet Theory? | Hacker News Reader

The Internet Is a Potemkin Village: Proof of Dead Internet Theory?

forum.agoraroad.com

38 pointsermantrout20 comments

I remember "going on the Internet" (with a capital "I" back then) as an activity. It felt like walking through a forest with its vast ecosystem; a colored, rich variety of life, an intriguing mixture of things both alien and familiar, tiny niches all around and a delightful surprise never too far away. Nowadays browsing the web feels more like a walk in the concrete jungle; bland, bleak and teams of people paid to promptly rectify anything improper

You can still find lots of intriguing people and interesting stuff, you just have to look harder for it in the current landscape.

And to be honest, as a person in the 20s sometimes seeing what the “old web” actually was through survived sites and archives… I really wonder if you guys were actually that better than the Internet today, I think a lot of positive thinking about the past is just nostalgia. I saw people were always fighting for Internet clout since the beginning, and all kinds of bland stupidity and toxicity happened through IRC, forums, and blogs. (And yes, the web was always plastered by awful ads since its inception.)

It is not nostalgia. There is a definite odor now emanating from big tech. The stink is on everything.

There was also a LOT of awful stink in the dot-com boom era. Right now the stink is more prominent because the bubble isn't bursting as fast as expected (or, the power of a few investors became too strong because of increasing financial inequality and rampant money printing, and they are all backing on all these shitty tech stuff that doesn't even improve anything, and nobody can do anything about it. Techno-feudalism, ho!)

Two things are different now:

1. commercialization. Most properties that people engage in are monetized and thus algorithmically manipulated.

2. things were smaller. communities were smaller and had little spill-over into other areas. What someone said on some forum wouldn't get picked up by media[1] to amplify whatever it was.

[1] with the exception of "leaks" new CPU, VGA, Apple device, etc.

1) The Internet was quite monetized with ads from almost its inception, and things like targeted advertising, ROI tracking, and sponsored searches were already rampant in the 90s, though it expanded even more in the 21st century.

2) You can still find lots of smaller communities and websites (that don't get reported to the mainstream) today, probably even more than before. An incredible number of blogs, forums, niche subreddits, Twitter groups, fandom communities, Discord servers, you name it.

There were ads but content wasn’t manipulated algorithmically for engagement and “eyeballs” as mr Grove put it or outrage as we see today. Moreover the monetization was by the website so the content creator wasn’t optimizing for engagement themselves.

With most services having some kind of API “journalists” set triggers or scour for particular things so they can write about something scandalous two people said on some service somewhere so they can present it as zeitgeist.

Google search has been noticeably worse than the old days for some time now, and others aren’t much better. It seems that they’ve solved the problem of crawling and indexing the infinite graph of the internet by limiting it to smaller and smaller portions of the graph, and continuously pruning old content. The small sites of the internet are nearly extinct. Personal sites, self hosted blogs, forums, etc have all gone by the way side, but how much of that trend is due to algorithmic reinforcement? Maybe we hit a point where they just got too difficult to distinguish from spam sites and we just didn’t notice as more and more of the internet was filtered out.

> Personal sites, self hosted blogs, forums, etc have all gone by the way side, but how much of that trend is due to algorithmic reinforcement?

I'm not so sure it's aggressive pruning or just Google's algorithm driving it. The internet has matured a great deal, to the extent that any search term with possible commercial intent will have been targeted and optimized for.

I still regularly see blogs, forums, and personal sites in low commercial intent hobby searches.

Out of curiosity/skepticism I tested their search claims and DDG actually gave me 7 pages of results. Google gave me 21 pages until the "omitted results" link, and 427 total results (out of 694 million claimed) after clicking that. Searching for results before the same date in 2016 gets another 40 pages. Messing with the URL reveals that Google will never return more than 1000 results, pages after that give an error message stating so. Bing claims 18 million results but ends at 633, weirdly showing links for additional pages after that.

I don't think it's proof of a "dead internet", rather search engines cutting corners where they can to deliver a more performant product at the cost of losing more obscure results.

The internet might not be dead, but we sure as hell don't have good ways of navigating the oceans of spam and clickbait.

I've been thinking about that. Can we leverage proof of work to sign meaningful content? I'm thinking of a voluntary HTML tag, containing a cryptographic signature, that has to match the page content, to be eligible for placement in a "spam-free" search engine.

For an actual creator, a few seconds of publishing delay is nothing. To an army of spam bots, having to waste time and energy calculating hashes could be crippling.

What if we apply this to every link on the page?

I'm trying to imagine how that could work, would we need to encode the English language to a has or something?

All it is is a standardized signature that means "my computer spent time publishing this data packet", encoding agnostic. Given such proof, we at least know it cost something to put on the web.

Why not skip the 'proof of resource expenditure' abstraction and just have a platform where posting requires a deposit of 0.1usd or crypto equivalent for low friction, retrievable after 6 months of no spam reports? It would at least limit to the more dedicated spammers.

Because that requires centralization and payment processing. You're describing a search engine it costs money to appear on. Proof of expenditure is the low friction crypto equivalent.

Interesting post indeed, The writer was a bit aggressive lol, But he made a great point

To think that google claims millions of results for each search but in reality it will only ever display a couple hundred results if you try to pursue it.

Very cool site too !

Loved the theme

I found there seems to be a hard limit of 1000 results, so claiming millions truly is a joke.

It feels like in childhood life was a vast and colorful ecosystem; In the process of coming closer to adulting we develop algorithms to filter stuff we look for. Soon enough it becomes dull and full of spam and clickbait, with a few well known seen-as-reliable sites that have become our go-to-places. If we solve the problem of navigating through life in a mature way without losing what it's all about, maybe we can solve navigating the internet as well.

Is there a good round-up of the best science/math blog entries?

Reply on news.ycombinator.com