Hacker News Reader

Comment by freyr | Hacker News Reader

Do people miss the Google of old, or the Web of old?

The Web used to be very information dense. It was built from the ground up by specialists and enthusiasts, educational and governmental institutions, and groups of like-minded individuals.

Then it became commercialized and commodified. Entire industries popped up to shovel mass-produced crap to masses of people through Amazon, Google, Facebook, etc. Forums became subreddits, blogs became tweets, homepages became Facebook profiles. And we’ve hyper-optimized the attention economy to keep us hooked on clickbait headlines and 30 second videos. The Web just bears no resemblance to what it was 10-20 years ago, and web search results are a reflection of that.

Maybe there’s a way to restore the Web of old. But that seems like putting the genie back in the bottle. More likely, IMO, something will eventually replace the Web, and Web search along with it.

View on HN

Google also has an outsized influence on the kind of content that's on the web.

At some point in the past, from what I remember, they started ranking blog content lower than more "authoritative" sources, hastening the decline of personal blogging. If you're wondering why every recipe on the internet has ten paragraphs on how this recipe is their grandmother's favorite and how it reminds the author of the freshness of summer produce, it's because that's the kind of content that Google will give a high ranking.

The web was a better place when it was quirkier, and there's probably a ton of quirky content that still exists, but it won't get surfaced by search engines that much these days, and it's buried under tons of garbage.

Time to get back to curated lists of interesting websites.

I unironically do that. Maybe I should publish? We could make a small project that just has interesting websites and reviews? Yelp for the interesting web?

And you could call it something quirky, like Yippee!

No, I'm not based in the valley :/

Or Yahoo

...

No, wait, some squatter sits on it.

I think some aggregators, like Jason Kottke, post lots of interesting things, but not usually the weird, unpolished, personal stuff that I associate with the 90s web. Something like early Yahoo would be cool.

Well i wasn't really alive back then, closest i can do is tumblr

Please do. One of my favorite things about Hacker News is that it exposes me to a treasure chest of interesting and informative blogs and other content (both archival/“historical” and current) that would be unlikely to show up in my search results.

www.readsomethinginteresting.com / www.thinking-about-things.com are both pretty good aggregators.

www.stumbleupon.com was a thing I liked back in the web 1.0 days - impressed to see it is still going.

There is https://curlie.org/ already.

I miss the non-commercial web of old. It's got into the point where I think my ultimate search engine is one that ranks pages by fewest ads.

And I get that people make a living by putting advertising on their pages. I'm not really against that. (Though I don't do that myself for my own reasons, I wouldn't presume to tell other people what to do with their websites.) But it's been abused to the point where number of ads on a page is a decent proxy for how crappy the content is.

When people decide to put ads on their content, the temptation is there to produce content to carry the ads. Before ads in the old days, you just produced content to produce content. Again, I'm not saying that ads are bad. But the purity of the old web is lost.

I don't think there's really a way to bring it back. I think the answer is that we just have to get better filtering on our ends to get the specific information that we want.

I can't for the life of me understand why people don't put static ads that are relevant to the consumers of their content. I was close to doing this once for a product I was CTO of, but I couldn't push it all the way through. We were so niche, and knowing exactly who our audience was. It's mind boggling to me that we chose to darkly monetize their data instead of showing them relevant domain ads.

> I can't for the life of me understand why people don't put static ads that are relevant to the consumers of their content.

It's kind of a matching and scaling problem. The ad publisher wants to have a well targeted audience that will convert well, while the website wants to generate ad revenue that won't drive people away.

The problem is that the ad publisher probably doesn't want to deal with the hassle of negotiating with dozens of niche websites for their audience, while the websites don't really want to do the same either. Hence, the rise of ad exchanges, which offers benefits for the ad publishers (deal with only a handful of exchanges, all kind of ad targeting criteria) and the websites (deal with only the exchange, add an ad snippet to generate revenue almost immediately).

The catch is that to extract the most money from ads, the exchanges do all kinds of tracking to figure out the most optimal ads to serve and we end up with the current adtech industry, instead of manually targeted static ads.

it works the other way around now. you make a business blog written around your product

Teclis has a neat approach to this:

> we count the number of uBlock Origin blocked requests on the page, and if too many, we kick it out, leaving only "clean" pages in the index.

Wow, that's great!

My blog uses a 3rd party comment system that does it's own tracking, I'm sure.

This makes me want to ditch that to keep it clean.

Much of that still exists. It's just harder to find than it used to be because the corporate stuff is so loud.

There are a lot of little things that could make the non-corporate internet more appealing to the average person. Why, for example can't I have a shared identity across several independent forums? OpenID came out long ago.

> "...he corporate stuff is so loud."

yah, money shouts, it doesn't merely talk. that's a consequence of (implicitly) deciding that money should have a voice in the first place. we explicitly sanction this via a gigantic industry, advertising, in which google has a prominent position. until we collectively decide to deemphasize this whole industry, at least in part by cultivating the small and interesting, the web and the internet at large will be dominated by money shouting.

I think we'll probably always have shouty corporate stuff on the internet, but perhaps we technologists can help bring the rest to a broader audience.

You're... right in that that stuff still exists. Many websites from 2004 are still online. But many are not. And new sites haven't been created. Instead, sites like reddit/facebook/twitter have captured all new content.

  Do people miss the Google of old, or the Web of old?

I miss both, but I miss the Google of old more. How can I be sure of that? Because much of the web of old still exists! There are plenty of decade-old pages out there. And a person could still find them on Google today, if only Google didn't fill its initial page of search results with advertisements and content-farmed dreck.

  > Maybe there’s a way to restore the Web of old.

Maybe search engines should also be able to link towards old cached content, internet archive style. Even if we can't rejuvenate the old internet without creating new economics to support it, we could keep the information alive.

The web of old was inaccessible to 99.9% of the world. Most people do not want what you're looking for. You can find it by diving deeper and searching differently.

Spam, clickbait and malicious content has always been around, yet Google somehow managed to filter it out just fine back in the day. As another commenter points out, good content from the past still exists even, yet is no longer surfaced.

Google can trivially detect all of the things wrong with the modern web and use them as negative ranking signals. Ads, analytics and affiliate links are a pretty good proxy for "shit site".

The problem is that Google ultimately benefits from spam sites and bad results in many ways:

* bad results increase the time you spend on search result pages and increase the probability the best "result" is actually one of the ads at the top * spam sites may contain Google ads and analytics

Ultimately it is not in Google's interest to give you good results.

> Do people miss the Google of old, or the Web of old?

Yeah, I wish I knew how to get back to it, but I think that's impossible without a big enough movement. Yes, I know about tildeverse and some other things but the countervailing network effect is just too big for it to be sustaining right now. The closest is Mastadon but it's not really that great.

What would make Mastodon more appealing for you or the hypothetical user for whom you think it's not that great? The main thing that comes to mind for me would be more people I know in person using it.

I think I agree with that but also I think it needs something like groups like Facebook has. I only want to see gardening stuff today, for example. I do like the whole mood though... it's not the battlefield that Twitter is.

It seems to me that use case is better served by forums, but it would be nice if I could use my Mastodon (or Diaspora, Frendica, Pixelfed, etc...) account as my identity on multiple forums. I don't think it's necessarily desirable for one piece of software to try to do everything, but it's great when different services can talk to each other.

Yeah, that would work too but it's kind of hard to find forums since reddit dominates all search results... I've been trying to train myself to look for them specifically though.

I can imagine paths to improvement that don't rely on Google improving its behavior or something displacing it as a search engine. For example, what if forum software spoke Activitypub so people could follow threads and boost insightful comments from Mastodon?

Many moons ago, someone on Reddit said "Facebook requires registration in order to view content, which tells you all you need to know."

I never forgot that remark. So that's a thing that I'd recommend to Mastadon: make it viewable by anyone, even if they don't have an account. Nothing smells quite like "you are the product" than requiring an account.

You do not need an account to view content on Mastodon. Posts can be set followers-only, but most people don't do that for most content.

Here's the profile directory on the biggest instance, for example: https://mastodon.social/explore

Nothing really. The core concept is let's say "wrong" (for my personal case). Mastodon is a stream of very short posts.

Problem a - shortness, means no longform texts there, so they are posted in different sites, externally. Short posts aren't very informative usually.

Problem b - stream. Currently almost all of the social media are structured like this and it is a bad format. It is not engaging, it is not possible to find something which flew by ten minutes ago. It's like a bad version of Facebook, if you didn't happen to see some post "in time" then it might as well don't exist.

Problem c - engagement, discovery, popularity and other similar issues. This is a common problem for all media, but for Mastodon it is even worse because of the fragmentation. It's broken even at the registration stage, e.g. just where should I register to find what I want? Impossible to find.

I've tried using it twice due to hype on HN, and abandoned it quickly.

A also bothers me. It isn't hard for an instance admin to change the character limit, and does seem like a way instances could stand out from each other, yet it's rarely done. Federated instances do seem to handle long posts gracefully (there's a "read more" expando link).

B seems inherent to social media as we know it. What would be a better design for you?

C is handled in part by relay servers synchronizing content between instances. Small instances often aren't well-connected though and I think both instance admins and relay operators could do better here.

It was a bit before my time, but peak usenet and mailing lists of the era looked awesome. The breadth and depth of the people who would participate in discussions (specifically in my field, e.g., comp.arch and the like) was quite amazing. Luminaries in the fields of computer CPUs, networking, programming, operating systems, etc. From academics to professionals to students and hobbyists.

That still goes on, but the gold seems harder to find, the signal to noise ratio lower. Might be just nostalgia.

I don't think it's just nostalgia. The barrier to entry for the early internet was much higher on a financial front and on a knowledge front. You had to really be a techie to get deeply involved in the internet in the early days.

> Maybe there’s a way to restore the Web of old

I don't think so, but the next-best thing might be to build an inverted search engine, where instead of crawling the web, people submit their sites to the index. I can imagine a company like Mozilla doing this to preserve the open web (if Mozilla were interested in preserving the open web).

Supposing it's just a youthful phase. Like the fashion industry for example. Where skinny jeans were popular in the 80's then baggy jeans in the 90's. Then skinny jeans are back in sync now. Don't you think personal blogs and forums will come in sync at some point in the future?

Big news old man: skinny jeans died a few years ago and now the mom jeans of the '90s are back in

I thought we moved on/back to baggy jeans

Nah, we back on ripped jeans now homie. Prob by the time I’ve hit submit on this we’ll be on 70’s flares again. Cycles are speeding up in this as everything.

How old? The commercial aspect was already inseparable from the rise of the web in the mid-90s...countless startups of the dot-com boom scrambling to find a niche to turn a profit from. Many an interesting site shut down upon failing to do so, that's certainly unchanged to this day.

HN reminds me a lot of the old web.

Yes, I miss it very much. I miss the days when people wrote RFCs for something they felt deserved to exist, rather than business plans for something they hope someone will invest in.

yep.. i even just miss the google reader days. why don't sites use rss/atom anymore? it drives me nuts

The evolution you described happened only because the internet is freely accessible to anyone, anywhere, except if you're a communist. Reverting it back to some state you deem worthy is saying the internet was broken. Trying to nudge the Web into some direction you like is exactly the thing that will destroy the Web.

If you don't like subreddits, twitter, or the mass produced crap - avoid it and focus on what you like. The early internet stuff has not been banned and banished. Some if it has been forgotten, maybe, but nobody is going to go through the trouble of keeping a VHS store just to tickle your nostalgia.

I still can find a community to play Re-Volt with. That is only possible because of the internet and the way it is.

> the Web of old No

Reply on news.ycombinator.com