The End of the Googleverse
theverge.com
theverge.com
Yes, we can most of us name real websites that we still read & rely on. When did those sites start publishing? How many of the creator-controlled, non-commercial websites/blogs you read began less than say 5 years ago? I bet the number rounds down to 0.
Google search sure as hell played their part in creating this world, but fixing search isn’t going to bring back an internet worth searching.
IMHO, the thing that changed was the signal:noise ratio.
When Google started, it was, what? 1:10? 1:100? Interesting pages : crap.
Now it's... 1:10,000? 1:1,000,000?
That turns a challenging problem into a possibly impossible one.
Granted, Google bears culpability in this, because at some point they realized: we can make as much or more money if we optimize for ad delivery than user search intent. And once they showed those cards with a wink, content optimized to this for its own monetization.
FFS, look at National Geographic these days: https://www.nationalgeographic.com/
However, if these platforms exist, there's no telling how many multitudes more people wouldn't even attempt to put stuff online. Life is busy and spending even an hour setting up an online presence on the "open web" is too much for most people.
Many years ago, I wrote my own homepage in PHP. With every version of PHP it needed to be fixed again, so after a few years I gave up. I tried to install some existing software, but you need to update that regularly, too often. So now I am happy that there is Substack I can write on (despite complaining about its functionality all the time), because otherwise I wouldn't have a blog.
So they prioritize ‘known good’ sites (big platforms, major companies, etc.) over unknown, individual sites. Then, since it’s now harder to be seen as your individual site (and less work), this pushes people to move to higher ranked platforms. This increases the ratio of noise (the ratio of spam/malicious sites increases compared to good ones) making it worthwhile to skew more towards results from platforms, etc.
Platforms also have the benefit in that it’s a lot harder for a YouTube video to be malicious, giving them more freedom to take chances recommending unknown content. It still could be hate mail, but at least it’s not going to ransomware you.
I still reflect when I go back on tours of 70s / 80s / 90s media, and while I get some of the "oldies tour" is also kind of a "greatest hits" effect, but there seemed to be a lot more creativity, especially wacky creativity, in the past.
Some of that is callous mass market media calculations by "big media": multi-hundred million dollar movies/games aren't going to be off the beaten path.
What is strange is the ARMIES of "content producers" and all they do is produce things in basically the same way.
So the "SEO" optimization seems to clip the wings of all grassroots content production. Google should have enabled the long tail. It ... sort of ... has, but man its hard to find it drowned out by all the mediocre crap out there.
People don't seem to be isolated from the internet and let some path of creativity percolate for years or even decades like what used to happen. Internet monoculture really seems to be destroying interesting creativity more than it delivers novel creativity from around the world.
"Comorbidity" with this is the death of patience and attention for wacky stuff, people click away too quickly.
About the only actual evidence I can deliver is the oppressive nature of the cinema sequel. For some reason, I just mark the first Matrix movie (1999) as the death of creativity in Hollywood.
However, of course, that may be the point in the 18-40 demo that I aged out of the great popular culture recycling machine. I think a lot of that 18-40 demographic number comes from the fact that by the time you hit 40, you've seen all the tricks, and the repackaging of the tricks/tropes just seems cheap compared to the "original" experience you had with the trick/trope.
"index of:" AND private AND .jpeg
It sure seems like the opposite is true to me. There's more content, but 95+% of it is garbage. There seems to be less good content than ever before.
But how true this is probably depends on what sort of things you enjoy.
But at the same time I understand that a large number of people choose convenience over managing their own hosting.
Like, say, 2 hours of Battletech lore: https://m.youtube.com/watch?v=c71x68uWd5k
Or an 8 hour serialization of the early-90s Aliens comics: https://m.youtube.com/playlist?list=PL3Js5pJZN4Mkl-n7Ktk6h6A...
Because at some point? Do what you love. Screw engagement metrics. People who care will find you.
There are lots of channels that manage that, sometimes it's fun to watch a knowledgable person explain something or review a product purely out of entertainment even if you will never buy it.
If people are informative an audience will find them, if they're entertaining and audience will subscribe to them.
Whatever downsides youtube has for a creator is vastly outmatched by the upsides for most creators.
Nobody⁺ gets 'famous' writing these days.
A smaller, tighter knit community will encourage the sharing of more intimate details. If, however, all the action is happening on larger stages that pull people's attention out of the phpbb's and into the tiktok reels, those conversations no longer happen, even if the volume of communication has increased 1000fold
Outside of blogs, I personally know people in non-technology fields who have their own personal website, to make it easier for people to contact them and view their portfolio of work. They made their through page builders to avoid the need to learn how to program, and they personally found some collaboration opportunities due to their personal websites. Searching for someone's name and possibly finding their personal website still has value.
Though I do admit that there are far more heavily commercialized or clickbait articles nowadays (such as padded-out listicle-style articles that bury the important information, which may also have questionable accuracy), I personally believe there are still good websites out there to find and learn from, which I've discovered through personal recommendations or searches on niche topics of interest.
A major source of well-written personal websites for me have been those that focus on the use of specialized software (such as Anki and SuperMemo) in spaced repetition: I've learned a lot from excellent recently-published, non-commercial blogs found directly through Google searches. I've also found similar very high-quality and non-commercial personal websites and blogs when searching about other specialized topics, such as workflows to use the Vim text editor to create technical documents with LaTeX typesetting.
I’d love to hear any recommendations you have. I’ve found several good articles on SRS, but I’m always interested in reading some more.
I’m sure I’ll have questions, but I’ll read through some of your posts firsts.
I plan to re-read the post as I gain a better understanding of higher-level mathematics over time, but from my first reads, I could relate and better put to words the idea that using spaced repetition to study technical concepts (though I personally always make sure it's not the main method of study, versus practice exercises) can help you better grasp connections between concepts over time.
A statement from the author's article that I still think about from time to time is: "I can’t emphasize enough the value of finding multiple different ways of thinking about the “same” mathematical ideas."
---
From a completely different angle, Nicky Case's "How to remember anything forever-ish" is a fun introductory post for giving an overview of what spaced repetition is, in case a friend or colleague is curious about the study method (2018): https://ncase.me/remember/
---
For a much older article that contains the fundamentals of making good elements/items/flashcards, Piotr Wozniak's "20 rules of knowledge formulation" is an article that I've referred to time and time again. I treat the rules more as guidelines, as I personally don't use images or cloze deletion, but I've improved my retention and understanding of topics by avoiding questions that ask for lists, and by always trying to understand a topic before making a flashcard about the subject (though this can be flexible to a certain extent, depending on the subject): https://supermemo.guru/wiki/20_rules_of_knowledge_formulatio...
---
As a last one that focuses more on the technical aspects of using the SuperMemo article specifically, I applied the information from this article published in 2010 to create a special Google Sheet to make it easier to import spreadsheets into SuperMemo: https://thesupermemoblog.wordpress.com/2010/05/15/exporting-...
Specifically, to import dozens of vocabulary flashcards at once, I open a spreadsheet file (aka workbook) and begin with a main spreadsheet with the English in column A and the translation in column B. A second spreadsheet then contains a copy of the first spreadsheet's information, but appending "Q: " to the column A entries and "A: " to the column B entries.
I then export this as a .csv file, replace the extension to .txt, and then use Vim (though Sublime Text or another text editor equally works well), to replace each instance of `,A` (that is, each instance of a comma that separates each item) with `\rA` (that is, a line break), to fit the import formatting required by SuperMemo.
This process helps a lot for picking up new phrases from newspaper articles and storing them in my records, and then importing them into my spaced repetition software for review without too much typing. I would guess that Anki's import method is easier and more straightforward, but I still do prefer SuperMemo despite the complexity (and lack of mobile access for the version I use), as I prefer the item scheduling. My impression is that in the long run, I have fewer reviews with the same amount of retention, which anecdotally seems to be true so far.
Because what is the alternative? Use the sterile platforms provided by Silicon Valley mega Corps? Fuck that.
I will continue self publishing my half baked thoughts on my personal website like it’s 1998. link is in my profile
I don’t think there has ever been a better time to learn on the internet than today.
It's also not realistically possible to auto generate spam video content like you can for blog posts.
It definitely is. If you look for videos just a bit longer then 10 minutes (something to do with YouTube recommendation algorithm) you will notice how much unnatural en spammy conversation and “information” is being told (backed by meme or stock images or just a talking head) before getting to the point of the video (which often is barely 1min of content). Sure it takes more time to actually speak the text and edit the video. But it really feels like generated garbage to fill time which can be easily churned out. Now with the dawn of AI I can only image this getting worse and worse.
This thing about video length is mostly a myth. "The algorithm" juggles many variables but I've seen no evidence content length is one of them.
There was a very common pattern of rambling to meet the 10m mark because this was a threshold to add a midroll ad to your video and thus make more money.
So your point still applies, it's just purely out of creator greed and not chasing youtube's heuristics.
Regardless, you're correct that watch time may be a motivating factor in making them stretch content longer than it needs to be though. I don't think they get paid for watch time (it's all about ad impressions or CPMs) but it would positively impact their channel statistics.
I think naturally practitioners/experts have better insider insight about the industry than full time content creators but they can't dedicate as much time to create content. Lots of youtube channels just try to cater to beginners, abuse clickbait titles, recycle content from other sources, fill the videos with low effort memes.
As an example, try to look up on YouTube how to replace your car fob’s battery (doesn’t matter what brand) and you will absolutely encounter an 8-10 minute video offered up for what’s probably a 15 second value and something the manufacturer could have provided in a decent 3 panel illustration hosted on a static .html page, but won’t.
Right, because you're not supposed to do that yourself. You're supposed to go to your dealership and pay them $50 to change the $0.20 battery in 15 seconds. Do you really think you're qualified to do such a difficult operation by yourself?
I actually got into an argument on Nextdoor once because someone was promoting a battery store that charged "only" $20 for this "service", and felt it was a "good value" and got mad that I was pointing out that you can just buy the battery by itself for less than a dollar and do it yourself.
People always wanted to make money on the web, but the sites of people that didn't still appeared in search results. Now search result positioning depends on the amount of ads you're prepared to host. So, the people who are optimizing for revenue are listed way above the people who aren't optimizing at all and are just creating.
There are times when I've searched for a tumblr page literally by the exact and totally unique name and still not found it on the 3rd page of results, all of which were unrelated garbage.
They've decided that they'd rather piss you off than lose money sending traffic to sites that won't pay to play. It's toxic.
Btw, I’m not dogmatically against ads. I believe they serve a purpose and are necessary. Just pointing out the side effects.
Now everyone wants to get paid for whatever they produce.
I broadly agree with your points, and yes we can fight back by spending most of our consumption time on personal web sites and blogs. Long form reading feels better than doom scrolling.
I'm hoping ActivityPub sorts out the discoverability somehow.
If you're still not convinced just look at Youtube search. There's no excuse for that to be terrible, yet it is.
Using Google to access the web now, is like using the Yahoo portal to access the internet in the early 2000s.
But this is great news for the conglomerates that have been around since early 2010s that can now enjoy being always #1 because Google itself has locked those sites in as "reputable" and "trustworthy". You can only imagine how those sites exploit this for personal gain.
Do you really think it is feasible/a good idea that the moment someone makes a new site google should send all the traffic to that site?
Please THINK how easily that would be abused.
Then last year I made a English version of that website. Traffic has been very very slow initially, but slowly and steadily increasing. I suspect if we don't do anything to it for a couple years traffic will just keep increasing just like the original.
I generally have better things to do than to jump onto the cargo cult that is SEO, but the GP's description of how Google throttles web sites is exactly what I observe anecdotally, and is exactly how I hypothesized what Google is actually doing (without doing any SEO research).
Of course, there's not going to be any conclusive "proof" of such claims unless they admit it themselves, or unless you start an antitrust lawsuit against Google.
I definitely agree it's not a good idea to route all traffic to a new site that hasn't "proven" itself through some test of time, but that's more evidence that they're probably doing this throttling.
I thought the same until recently. Google doesn't appear to be indexing/prioritizing anything that's old, infrequently updated, or noncompliant with AMP, even if it's an authoritative source.
People blame SEO gaming but I'd sooner accuse them of alignment tweaking (all old/outdated beliefs are purged; my queries were religious and controversial in nature). For some topics, there is still an entire other internet out there they pretend doesn't exist...that starts on page 1 of fucking Yandex.
Not advocating for Yandex specifically (I don't trust them) but...try other options (Kagi, Bing, etc.).
And niches have always been the strength, the value of the open web. That you and the 14 other people on the planet with same obscure interest can connect.
Instead they focused on growing the network and making content creation and broadcast of content free.
So today less than .5% of content produced is consumed according to the UN report on the attention economy.
Whats the fix? Content production needs to be scaled down. A better system is required that decides Who gets to broadcast, how much and when. Bandwith Advertising consumes within such a system must have a cap. Now there is no cap. Ads can be shown on everywhere and all the time. "Scaling things up" should be a political/social decision not a business/technical decision.
The focus must be brought back to solving info overload.
If user targeting were banned, we’d be back to economic incentives for content creators that are similar to pre-internet media and publishing.
In particular, you’d be paid for having a consistent measurable audience, which means the money would go to distribution channels that had high quality editors and curation. Today’s ad netwoks funnel that money to clickbait farms.
Cuz One journo surfaces info. 6000 journos copy it and broadcast it and all expect to paid.
The solutions have to built around content reduction cuz right now we are over loaded with millions of people creating nothing new and expecting to be rewarded.
Today, they get paid the same as the original journalists, specifically because ad auctions are designed to transfer all wealth to the middlemen, and therefore commoditized content creation and distribution.
You can't fix search no more than you can fix social meda because search and social media has grown so big that they are politically important and hence state property.
Once search and social media reached critical mass, the state got involved. The state was probably always involved but they took control when it got 'too big to be independent'. When google and social media showed that they can overthrow governments and affect elections, the party was over.
Even if there was an internet worth searching, google would not showing us that. Google will be showing us what the state wants.
'Beyond a critical point within a finite space, freedom diminishes as numbers increase.' - Dune
An internet where all ordering of information is done via secret algorithms that are constantly being tweaked behind the scenes, optimised to prioritise popularity and other metrics useful to advertising.
Online advertising services companies formed from high traffic websites calling themselves "tech" companies.
In 1997, the founders announced Google in a paper where they vowed to offer a search engine that could be studied in the "academic realm".
http://infolab.stanford.edu/pub/papers/google.pdf
But we soon learned Google lacks integrity. Say one thing, do another. Fast forward. Outside of Google, who actually studies the Google search engine. People without any university education, or at least no need for it, calling themselves "SEO consultants".
When Google publishes academic papers about their internal operations is this done to push search engine technology into the academic realm. No. It's purely commercial, a recruiting tactic; Google is "showing off".[1]
1. https://news.ycombinator.com/item?id=37313826
High traffic websites incorporated as so-called "tech" companies have sought to bury noncommercial websites beneath all the advertising-driven garbage and/or to discourage web exploration and development outside their walled gardens.
However, it's still possible to use the internet for noncommercial purposes. High web traffic or market cap does not give any single website the authority to define how the internet can be used.
Arguably it's easier to use the internet today than it ever has been.
The web is not the internet. And the internet is not a handful of gigantic websites calling themselves "tech" companies. In practice, the internet is reliant on non-binding cooperation. As such, no one can own the internet, and no one can control it.
One can try. And that's what large so-called "tech" companies have done. It may have worked in the past and it may continue to work, for now.
But things can change, they will change, and Google could become just like AltaVista.
Contrary to most stories people tell about the mid 90's, I actually liked AltaVista better than Google. When someone first showed me Google I was not interested and I did not switch immediately. I liked comprehensive searching, combing through large numbers of results. Google seemed antithetical to that, focusing only on the "top" results.
One difference between then and today is that no one back then was patiently and desperately waiting for something better to come along, publishing articles such as "The End of the AltaVistaverse". There are people today who really want to get past Google and on to something better. They have been waiting for years. Progress has really stallled.
I tried it - it was unbelievable! A simple search box, type in your keyword(s) you're interested in, no query syntax and - voila! Good, relevant results! It was simply amazing! AltaVista essentially died overnight.
But really, Google's biggest technology was AdSense. It's how they make their money. It's why all their other services, save for gmail and google docs, have come and gone. Search has become less and less relevant in today's world. But there was a time when it seemed magical!
Search engines used to require you to use Boolean operators "+" "AND" "OR" "NOT" to filter search results. My high school library had a laminated card next to the internet connected PC (via dial up modem) explaining how Boolean operators worked and with suggested search engines to use (this was around 1996/1997).
The eye opener to me was my high school class was given an assignment to write about the upcoming "G8 Summit", Russia had just joined the G7 creating the G8 and it was a big news story at the time. I can remember how frustrating it was to find any results searching for things required constructing strings like "G+8+Summit OR Group+of+8+Summit"etc in sites like infoseek and still getting nonsense results. I found the information I needed to complete the assignment by going to public library which had copies of various daily newspapers archived I basically quoted newspaper articles to complete the assignment and concluded the internet was useless for finding information. I complained about my situation at home and my Dad Casually mentioned have you tried "google.com" next ay at school I typed "G8 Summit" into google and immediately got relevant results. I was a convert to google immediately it was so far ahead of every other search engine at the time it was like night and day.
God damn I wish I had any mentorship at all when I was a kid, just anybody who knew about computers.
In my more cynical moments, I feel like people got it backwards: DoubleClick acquired Google.
I am still bitter they cancelled Google Wave, and then the cool Apache open source version stopped being supported.
A lot of our digital lives fades away when Google and other companies sunset products. I am mostly OK with that, since new stuff gets developed also. We control our own personal domains and web sites, our own writing, and our code. I can live with enjoying tech Google, Microsoft, Meta, etc. produce, hopefully without having strong dependencies on any of it.
Eventually one of my friends forced me to try it and ... it was the coolest thing! It worked so well! I used it for note taking and collaboration but not for long because it was cancelled basically immediately.
I also remember telling people "No, google is not your friend" before 2007, though I don't remember why now.
That is not my recollection. I was blogging at the time they're talking about, and I think the first time I heard about SEO was in the context of trying to get your site ranked higher than everybody else. I grant that the techniques for early SEO came out of good, semantic-web intentions, but I feel like the time between them being implemented and the time they were misused for personal gain was like 3 minutes, max.
Version 1 SEO was to cram as many words as possible into the page.
There's a reason why meta keywords attribute has been useless. For years [1]
[1] https://developer.mozilla.org/en-US/docs/Web/HTML/Element/me...
This is always a strange claim for me, because it does not even remotely match my experience. And now we have the rise of Chat-AIs, and people who claim it's so much better than google, which also does not even remotely match my experience. And now I ask myself: am I just too stupid to see it? Or are the people too stupid to not see it? Do I search just too different things from what others search that I can't see it?
And then I also remember the claim that the youth is insanely uneducated about technology and the modern world, even worse than old people, and I wonder whether this is the reason for those claims. Do those claims all come from young people, who live in a very different world and are just very clueless about those things?
So at the end, does Google end, because it's not cool with the youth anymore? Not hipster enough and getting lost in the difference between the culture of the generations?
I was literally unable to perform the search that I wanted, and all Google suggested repeatedly was adding a dozen quotes on each side. Most frustrating search experience I've ever had.
Its unfortunate but theres more money to be made being less helpful. For example, google ads pay based on session length.
See youtube killing its short form videos with the 10 minute monetization cliff --> gap filled by vine and then later by TikTok
Once you change the usefulness of your product it's only a matter of time
That makes me want to shout. "Did I say that?"
Then I replaced 'Google' with 'Bing', and Google returned results without the quote which were irrelevant - but that makes total sense to me.
It’s weird. I suspect it’s more expensive to search this way or something because it’s such a basic feature that they haven’t done for a decade.
I'm not quite sure of the reason. There's a couple of possibilities:
1. DDG is relying on Bing for a lot of their search data, maybe bing's just got better, or DDG have configured their integration better. I don't feel that DDG's had a significant improvement, it definitely feels like they've caught google on the way down more so.
2. Google losing the SEO war. As it gets easier to run and monetise blogspam sites, more of them exist, and as time goes on, a longer tail of topics get "served" by these sites. Unlike the spam of years past which tried to convince Google that their viagra selling site was really about top ten movie downloads, these sites are nominally about the topic the user searched for, so maybe Google is having a harder time trying to evaluate "quality". LLMs are being given a lot of the blame for this, but this is an older phenomenon than that, there were plenty of sites doing e.g. templated articles fed product spec sheet, or pay a minimum wage intern to rephrase the wikipedia page before that.
3. I've cut down my usage of Google services to basically just YouTube and YouTube Music. I also run ublock origin pretty much universally. Maybe my de-Googling has just been successful to the point that Google doesn't have a accurate profile for me anymore, and this has more of an impact on search than I expected.
What are they losing, though? Isn't this more like a snake eating its own tail? Or the human-centipede of online advertising?
I'm not convinced serving quality results is really in line with Google's incentives. If they trick me into clicking some garbage and have to go back and try again, isn't that another shot at showing me ads?
Even in technical realms this holds true. One of the huge disappointments of learning JavaScript/typescript recently is how terrible and spammy searching for anything on it is. Coming from the golang world where I'm used to search results being highly relevant with few advertising spam blogs among them.
I look for recipes a lot. While what I find matches the "SEO-shaped recipe blog page" template mentioned in TFA, nevertheless the results still have the recipe(s) I'm looking for.
When I need to fix/build/repair stuff, searching just youtube almost always gets me what I need or want.
My hypothesis is that it depends on how you search. For me, Google has become unambiguously poor, and I've stopped using it because I get bad results.
But I think that the reason for that is Google's use of AI (not the LLM stuff, but the stuff they started using a number of years ago). Google is clearly trying to interpret "what I really mean" from my queries instead of just taking them at face value, and it's awful at doing that. For me. It's obviously good at it for other people. It may have to do with me actively doing my best to prevent Google from tracking me.
Google tracks analytics on the quality of its search results, simply by observing which results users click on, how long it takes them, or whether they don't click on anything at all.
The idea that Search is somehow worsening doesn't make any sense.
What has maybe happened is that Search has improved for the population as whole, but gotten worse for certain subsets of users -- but still a net gain overall. Or alternatively, that the web itself has changed -- that Search is doing better than ever, but some users just don't like what the web offers in response. (E.g. nobody writes straight-up recipes without tons of intro anymore -- but that's also so they have tons of space to insert ads, it's not about SEO. It's nothing to do with Google.)
The "search experience" however, has degraded since the mid 2010s, for me at least. I'm in a constant battle to filter through advertising and bad content that is "optimized" for the algorithm to find the information I need. The constant need to monetize the technology more and more has certainly reduced it's usefulness en-masse.
But I don't think chatbots or alternatives (beyond curated indexes) will solve this problem - as you say the web itself has morphed around ways to optimize one search engine's ranking algorithm.
Search is worsening. We've all seen it with our own eyes. I don't care how much analytics Google is gathering, that's irrelevant.
Just take on example: try googling for the release date of an upcoming video game or movie which you know for a fact hasn't had an announced date. Then view the top 5-10 results and let me know if you consider them to be of a high quality.
What do you get? Speculation? If so … I mean what did you expect? No one knows the answer so neither would Google.
Its job is to find websites talking about what you entered as your query. If all people are doing is speculating … that’s all Google will return.
The reason that I don't think it's this is because often there are sites that offer exactly what I'm looking for. Google doesn't surface them in my results, but other engines do.
What analytics are you thinking of that would show search is somehow MORE accurate? Because if it's just searches and click throughs and traffic and revenue...
More people with internet = more people searching for stuff. More mobile phones = more people searching for stuff out in the world. Worse results = user goes back to google and does ANOTHER search and search revenue just doubled again. Huzzah bonuses and vacations to Disneyland all around this year!
I think by any measure, growth and clicks and revenue are going to go up no matter what they do so I'm not sure exactly what metric you'd be tracking to see whether Google search results are actually qualitatively better? Short of the site going down entirely, there's probably nothing they can do to slow growth. But nowhere in that compounding money farm is the quality of results relevant.
The search is probably fine, but it's relegated to the bottom half of the page. It could also get worse if Google isn't able to keep up with the SEO and scammers. Personally I feel like Google has lost interest in providing a good search experience, they still have, what, 90% of the market. There's zero incentive to ensure it's a good experience and they can keep milking the advertisers.
Meanwhile the competitors have been forced to do better and especially Bing has. Any search engine that rely on Bing has become really good in the past five years. That might also be part of it, those of us who went to Bing, DuckDuckGo or other search engines for other reasons have learned to use those and when we then go back to Google it's worse, because we haven't used it for ten years.
But you're probably right, the search bit is fine, it finds the same results as Bing, so it's down to presentation and Google used to be the best and now they are the worst.
I've been using it for over a year, 100% of the time, and love it. Great results and awesome functionality like weighting certain sites, AI summarization, etc.
I don't work for Kagi, I just seriously love it.
Just look at Google Assistant. It has gotten significantly worse the last two years. Strangely, the only time it asks me for feedback is when it happens to work. Shouldn't be hard for it to detect when it craps out I talk frustrated at it, but they don't seem interested in getting feedback at that point. Ah, but it happens to turn one light on once (versus the 20 something went wrong times), prompt for feedback on that specific one time it worked.
Do you honestly think that when Google is presented with an improvement in search results that would reduce its revenue that they choose the improvement?
Enough of course until someone shows them better results and then they will switch search engines just as we switched from altavista to google all those years back.
Including one particularly panicked search because I sold the ceremonial weapons from Rosalyn Monastery.
ChatGPT and other LLMs are also absolutely horrible on accuracy, don't get me wrong.
But, here's an example of what's wrong with Google.
I search something like "Can mangoes grow in Washington state?" and at the top of my results is the condensed "People Also Ask" question answer result. These attempt to read and condense a webpage (of questionable accuracy) into an answer for my query, but they are often full of shit.
For example expanding "Where are mangoes grown in WA" shows an answer about Western Australia rather than Washington. Another answer tells me "yes" but when I read the actual article it clearly says "no, they won't survive".
Today, I think they are losing. Quality primary sources are often crushed by unusable websites, which understand google's analytics very well. If I make it hard to find the information, but I make it seem that it's the next paragraph down, my search results will improve!
Google itself is causing the enshitification of third party websites, many of which have paragraphs and paragraphs that are obvious spam. I'd take any videogame guide website from 2005 over the first page of google today
It does seem like they've finally removed github scrapers like gitmemory -- at least I haven't seen them in a while.
[1]: https://addons.mozilla.org/en-US/firefox/addon/ublacklist/
Perhaps the problem with Google is that it's trying too hard to convince people that it's smarter than it really is. I treat it as a stupid, mindless computer and it works fine most of the time.
You can tell who all the old people are that were using search engines since back in the day. LOL
All I ever expect from a search engine is it find pages with the words in my query (and maybe exclude certain words as specified), anything else is just gravy.
Still, I agree that keyword search remains practical.
Digging through search results to find what you need is work. Having something confident make you believe an alternate truth is not.
People who actually try to use code output from chatbots quickly realize it just only _looks_ like code of the right shape and quality. There have been many blog posts and comments here touting how great the code ChatGPT generated is while the given example response is complete nonsense once you look a little deeper.
I recently search for "raspberry pi uart hat" and Google responds with the following:
Not many results include the word "uart" so we eliminated it from your search terms. Here are thousands of irrelevant results about random raspberry pi hats you aren't interested in.
I run a fairly popular web game (Redactle https://redactle.net). Thousands of people play it and there are thousands of links on social media which Google doesn't seem to care about. When you search 'redactle' you should be getting my site or a clone of the original setup by a fan; This is how it works on Bing and DDG. But on Google the top results are ad-infested link-farm game sites, a domain squatter page at redactle .com and crap like that. They simply don't recognize the social proof that is out there. A thousand https://spammygames{1..N}.com sites linking to each other beats out all those people linking from reddit,fb,x etc.
Imagine trying to find videos on the Cyberpunk genre and getting nothing but videos on that buggy video game that came out a few years ago.
In my own personal experience, I think the younger generations care less about search. They just don't use it much, nor do they value the information & cultural pipeline in which search is valuable.
google still is king for every single type of search i do. specially code related.
i've tried again and again to use "Chat-AIs" and they get so much shit wrong while implying they are correct that i just gave up.
Yeah that claim doesn’t get made in any sort of widespread way. There are variants of it that do (“kids only know how to use their phones not keyboards”) but your specific wording here makes me think this claim came from some hyper specific bubble or from your own subconscious.
Care to clarify or explain what I’m missing?
And after thinking about, it makes sense. Technology today has reached a plateau of quality and comfort, has removed the struggles of the past, and is now cheap and widely used in society. That's very different from the previous decades, where tech was an uphill-battle, where computers were expensive and only used by those who had demand for it. Tech today has saturated society, and this also changed perception.
Now more people can evaluate the actual ability of people. And old people today have legit reasons to learn and use technology. And the most notable change is: the old people today, are the middle-aged and young people from decades ago, meaning they are the ones who had already in the past reasons to learn technologies. So comparing old with young today in those aspects, is very different from 20, 30 years ago. And it seems, the young people today have finally lost their benefit of early access to technology, and now they compete again with experience and knowledge.
So regarding Googles search this means, maybe the youth is just lacking good google-fu, which older users with more experience have. Or they generally have a difference stance on how the search should work, because they have no deeper understanding on what the Search is, what it should deliver and can deliver, and what not. Some of the comments here go very strong in that direction, that people have a very different approach on what they expect from Google. So again, it's a difference in perception.
The rest of my searches are going to ddg after Neeva shut down and Kagi hasn’t fit the bill (though I’m not remembering why).
But if there’s something I need to buy or do commerce- google is by far the best. So I try and use it the least.
Repeat once per generation, to infinity
And then some PE vultures do a hostile takeover and we're back to step 1.
Happened to the Greeks, the Roman Empire, Imperial China, the British Empire, IBM, Intel and AMD back and forth, Toyota, Boeing, Yahoo, Google, MySpace, Facebook...
Also presumably soon to happen for the United States, Nvidia, Mysterious Twitter X, TSMC, and others currently reigning as kings in their respective corners of the world.
Note that falling doesn't mean they will disappear into the history books, but they will all fall sooner or later.
If an entity wants to stick around into perpetuity, never go higher than #2 and maybe never go higher than #3 so when #1 falls you don't end up becoming the new #1 out of nowhere.
It may have certainly been friendly to a lot of users, yet it did not ring friendly to a lot of useful content.
Then also comes the geo segmentation. There was a time when quite a few results would be from foreign language sites or other countries. Sure, it created noise when the language was very much foreign, though later there often was a link to Google? translate.
I guess there may still be a way to have all that in the mix, but, generally the results are being streamlined.
I'm not sure anyone else really qualifies as a big very consumer-facing tech company in the modern sense from that era or before though, of course, there were many once popular companies like MySpace that just flamed out. Yahoo! might be an example.
The only thing that was ever revered about them was Bill Gates, who was more loved by business schools as a case study in how to run an extremely aggressive company.
OTOH Microsoft software's operation experience was far from happy. Bugs, crashes, byzantine installation and update procedures were all the daily reality.
To say the truth, few things were significantly better. And if they were (say, VMS, which was rock solid), they were also significantly more expensive.
Commercial Unix wasn't a hugely better; I personally crashed Solaris software, and the OS proper once. Bizarre installation and update procedures applied to it equally well.
MS started having a hit after a hit to its reputation in the tech community when it started to try to do something about the internet. They tried to muck with Web technologies so as to create its own silo, not compatible with Netscape's: custom HTML and CSS, JScript vs Javascript, etc. They also started to talk trash about competition, never a good sign. Eventually they won, partly via business pressure, partly via good technology, and many a web developer cursed the MSIE5 chokehold since then, for a decade at least.
Windows 95 was a target of many techie jokes, but, to my mind, MS have done a tremendously successful rework of desktop experience and its visual language, which was and still is widely emulated as "the common standard". Writing an OS as a DOS extender was a cool hack, which allowed to launch DirectX, and make windows a prime gaming platform.
Terminals connected to minicomputers--which is what I used at work through most of that period? DOS-based word processing and spreadsheets were pretty nice by comparson.
Macs? They were fine in the mid-80s although I didn't use them in school unless my group was using them for some project.
Unix workstations were getting pretty nice by the mid-90s but those weren't something for a home user.
But expectations were much lower then, computers were expected to be a thing for a 'hobbyist' to tinker with, and those who did generally did it because they loved it.
(I still don't think what were then called "PCs" running MS-DOS or (beginning in 1985) Windows were as beloved as Amiga or Commodore 64 though, Microsoft-OS PCs were for "business", and nobody really loved them).
(Macintosh was first sold in 1984, so mid-80s was the first Macintosh).
I know they’ve changed their tune on open source but it’s really hard for me to break from my 20+-year perception that they are the evil company that makes awful software.
It's a little embarrassing to think about right now when you realize that Microsoft has never poisoned a river or sent thugs with billy clubs to bust a union, and so all of us had bigger fish to fry. But there was never love.
Microsoft represented an existential threat to many, many parts of the IT ecosystem every bit as real as dumping toxic waste into salmon breeding grounds.
MS is like a boxer who doesn’t just want to beat you in the ring. He wants to outright kill you too so you will never fight again.
It was truly amazing. MSDN on CDs.
They were very beloved, back when they were a scrappy upstart in the 1970s.
"There is a growing chorus of complaints that Google is not as accurate, as competent, as dedicated to search as it once was"
The word "growing" is a link to hackernews
"Subscribers quitting Netflix in droves over new policy."
"No one uses Google anymore."
"AI replacing 80%+ of programmers at software companies."
"People outraged over [latest nothingburger]."
You can come up with an unlimited amount of catchy content, and most of it makes it to the front page of sites like HN and Reddit.
However, the last couple concluding paragraphs, and the comparison to AltaVista, I found to be extremely weak. It was as if the author had diagnosed all these problems, and felt like they couldn't just end the article but had to wrap it up with something they felt was a potential solution.
In my opinion, there is no potential solution, and that's both a good and bad thing. At worldwide scale, I find it hard to believe that any company wouldn't fall down the path Google has: they are pretty fundamentally a victim to Goodhart's Law [1], and that dynamic is inevitable.
I think the bad thing about that is it does kind of make me yearn for some of "the good old days" when not everything was just teeming with SEO nonsense and blogspam. For example, when looking for particular software solutions, I've found it impossible to use Google to do "ProductA vs. ProductB" type searches anymore - all the top links are just some sort of affiliate link spam, the analysis (if there is any) flat out sucks.
So the good news is that it's made me step away from the Internet more: e.g. for the example above, I don't even do a search: I'll ask for recommendations from friends and colleagues, post to a helpful employee alumni mailing list, etc. You know, the way we used to do things "in the good old days".
If anything the complete disgust and enshittification I find with so much online content and algorithms these days has actually led to me being mentally healthier - I just spend less time online (HN being the bad habit I'm trying to break) and more time in "the real world".
> More recently, there’s been a shift to entertainment-based video feeds like TikTok — which is now being used as a primary search engine by a new generation of internet users.
What does it mean to use an entertainment-based video site as a "primary search engine"?
Some people (probably in particular older people) see the internet as the sum of everything (eg: Reddit + Facebook + TikTok + etc). Other people, I think, have less of a concept of the internet as a mass data collective and see it as more silo'd. And within those silos they feel that certain ones will have results/content that is more inline with their viewpoints, desired results, etc. So the result is we see some of these social media sites like Tiktok becoming very much a micro-internet to these users, and they specifically DON'T want google-style results that reflect the internet at large (and they don't want to use search modifiers to say only search tiktok.com).
In some ways we may be trending back towards the days of AOL/Prodigy/Compuserve/The Well, where you virtually hang out with "your people", and content that doesn't exist within your group basically doesn't exist at all.
TIL
It seems more like a side effect of how everyone is trying to become an influencer than a new permanent state of being, but for the time being it can be interesting
"Best tool for x", "book recommendations" etc etc
Its such a shame the genuine good information is so hard to find and where you do its in some walled garden.
I'm old, but I would never search YouTube for how-to information. Videos are too light on solid information and are too hard to use as references.
Tiktok is worse, though, having even less information.
* replacing the steering rods on my Sprinter van
* restoring factory state on an ancient Garmin watch
* mixing adobe plaster
* tying a prince of wales knot in a necktie
Depends on the video.
I rarely use videos for how-to though. The written word is more convenient as it allows me to skim it and it’s easier to skip ahead. Video is good for demonstrating things where motion information is useful though.
I didn't know Tiktok could do this, but I guess life finds a way.
There are a few links scattered about on relatively minor topics , but nothing substantial showing that it's in question. Either that's clickbait, which I'm now starting to associate with the verge, or it's far too early days to tell.
Like other comments here my impression is not the same as the post. And the slivers of reasoning given are weak at best.
The analogy checks out.
Search engine spam isn't something that's happened to Google, it's a consequence of their dominance in directing web traffic. No matter which beautiful flower you find, point a power washer at it and all you get is dirt and mud.
Google is a traditional search engine. Traditional search engines run on keywords. You enter the keywords you want on the webpage and press “search”. Didn’t find what you were looking for? Add more keywords to make it more specific. Things you don’t want polluting the search results? Exclude a keyword by adding a “-“ in front of it to remove those specific results. Rinse and repeat until you find what you need or give up.
This, IMO, is actually a good thing. If Google stayed really great (and thus dominant), there might be less incentive for new info retrieval and knowledge/"cognitive" tech. But the decay of Google makes consumers far more likely to experiment with alternatives. It sort of feels like the circle of (tech) life.
1. SEO content farmers have filled it up with low quality results.
2. Google used ML to tune the system and it optimized for the lowest common denominator, incompetent searchers.
The average person never learned how to construct succinct search queries and so habitually enters too little or too much into the search box. Current Google Search wants to help the poor searcher so it adjusts their entry, or in some cases rewrites it completely, before processing it and returning results it has decided that they really want.
edit for clarity
I would love to compare people who passed the old interviews and worked at old G, to the new ones, and just see what kind of culture, personality, and technical competency each side had. Then you get to see how those people shaped the company and how public opinion changed.
They don’t understand branding or trust, or if they do, they simply don’t care. The reputation for cancelling products to the point of untrustworthiness demonstrates this.
Google is a classic case of the fish rotting from the head down.
It says "Illustrations by Michelle Rohn" but maybe it should say "Illustrations by Michelle Rohn and an AI"
Some years ago, i would occasionally have to prefix my query with !g to get the Google results, but these days, if DDG cannot find it, chances are Google cannot find it, but will happily return a bunch of spam pages instead.
It's a choice like Linux vs Apple. I choose to use something (DDG) that is a less good generally but better overall.
Privacy is better than search results.
But won't be surprised to see layoffs happen in ad hungry articles such as the Verge thanks to GPT-4 being a candidate to replace them.
Not insinuating anything, I just find it not weird