8 ms·
The worst is the SEO pages, especially for popular how-to topics. Every single first page result will be a page that has expanded what used to be said in bullet
by kstenerud 4y ago
The worst is the SEO pages, especially for popular how-to topics. Every single first page result will be a page that has expanded what used to be said in bullet point form, into a 1500 word essay that you must now slog through in order to get the information you need. And this is by necessity, because Google values prose over bullet points.
The majority of things that come up on search nowadays consist of computer-generated essays. The only way around this is to limit your search to Reddit. And I fear that soon Reddit will be flooded with AI generated fluff to take advantage of the residual "real-user" SEO.
- Gigachad 4y agoLiterally anything google values will be abused and ruin the internet. If they valued shortness then everything would be trimmed to bullet points removing useful info. The problem is really not that google is doing a bad job, it’s that the incentives to get views on pages ruins the internet.
- inductive_magic 4y agoThat sounds like it makes sense - but if human readers know what's a good search result and what isn't, why would Google not be able to optimize their algo accordingly? If the results for a query don't match the "nature of the query", it's a bad result. If the algo can be gamed, it isn't aiming for the actual goal, i.e. it fundamentally fails at information retrieval.
- Gigachad 4y agoBecause anything google can detect, the spammers can generate to match. We will soon enter an era of GPT generated spam which not even humans can tell if it's real info or just garbage that looks real.
- thr717272 4y agoLuckily on Kagi I can just block sites the moment I realize it is nonsense. Or boost or even pin sites that I feel are extra valuable. Back in the old days I had a txt-file on my desktop containing domains with a minus in front, llke -example.com -example.org for all the worst offenders, so that I could post them on the end of every contested search. After a while I think Google found a way to ignore those as well as every other filter I had - at least I gave it up.
- trinsic2 4y agoI have never used anything but google, duckduckgo and startpage. I always end up coming back to google because the results are not as good. Well startpage has the same results as google, but i stopped using it for reasons I forgot. What made you chose Kagi?
- thr717272 4y agoTested it and found it realized doublequotes something none of the big ones that I can access (bing, ddg, google) does. That was it. And the first time I saw a reported bug in the search results getting serious attention and a fix I was sold.
- onion2k 4y agoif human readers know what's a good search result and what isn't That assumes there's a single, canonical idea of what a "good result" looks like. I imagine that's dependent on a lot of factors, from the context of why the user is searching, their demographic, and their biases. When I search for "chocolate chip cookie recipe" I want straight facts - ingredients and bullet point steps and nothing else. My mother would want some background about the recipe, where it's from, and what makes it different. You can't get that from the search term alone. You need a lot more context. It's exactly what the promise of 'personalized search' based on tracking everyone around the internet told us it would give us but it doesn't seem to work very well.
- Nextgrid 4y ago> That assumes there's a single, canonical idea of what a "good result" looks like Maybe we can't agree on what people want, but what I'm sure we can agree on is what people don't want. I don't recall anyone ever saying "I wish there were more ads in the content I'm consuming". So we could at least start with that.
- bootsmann 4y agoYes, and now turn that sentiment into a metric that can be measured as input to the model. At first it was click-through rate, but that turned SEO into sites into clickbait to get high click scores. Then they switched to time spent on the site, and as a consequence we get recipes that are filled with 2h of prose above them just so people stay longer on the site.
- ajsnigrutin 4y agoRecipes are a pain in general, even without google, doubly so for baking. Baking is a science, not art... so recipes calling for a cup of flour and two teaspoons of butter are useless to me, while recipes with ingredients in grams are useless to many american grandmas. On the other hand, a pinterest page with a photo and a large overlay asking me to register is useless to both of us.
- 4y ago
- owisd 4y agoI disagree that whatever ranking system google uses will just be gamed to the same extent. If google downgraded the rank of pages with lots of ads and trackers it would substantially reduce the incentive to game the system, the reason it won’t is because most of those pages are running Adsense so there’s every incentive for google to keep the status quo.
- frankzander 4y agoThat is exactly the point!
- yunohn 4y agoIf you provide less monetisation for content creators, you end up with less content too. This is not as simple as you think…
- piaste 4y agoYou could reward the type of monetization that is correlated to high-quality content and penalize the type that is associated with spam. Site offers extra paid content or services, or takes donations? Rank up. Ads? Neutral or rank down based on type of ads. Affiliate links? Rank way down.
- tazard 4y agoLess content is fine when the issue is too much junk content. Content would then be created by enthusiasts instead of businesses
- carlosjobim 4y ago> enthusiasts instead of businesses Those are more and more the same person - especially when it comes to online ventures.
- Loic 4y agoYou end up with less trash/monetization content and more content from passionate people writing about what they love/like/do for the joy of sharing. But this does not sustain the chocolate factory.
- ekianjo 4y agobullet points would be an improvement over bullshit useless prose
- thegginthesky 4y agoThe real problem is that Google holds an effective monopoly in search, so any digital marketer worth a salt will optimize the crap out of their blogs and pages to get higher and higher rank within Google alone. This will, in turn, lead to what we see today: a race to the bottom in terms of quality to get the best SEO results. I've found that a mix of search engines give me the best results, I'm trying my best to incentivize competition. If I want super local results anywhere outside the US, I'll use Google+uBlock because that's what it works, but everything else I use Kagi or DDG.
- matheusmoreira 4y agoWhy do people even care about views? Because of advertising. It is the root cause of all problems on the web today. All Google has to do to fix the web is put itself out of business by delisting pages with ads including its own. Search and the web at large would be fixed in days.
- gniv 4y ago> Literally anything google values will be abused and ruin the internet. That is true to the extent that the "anything" is only a proxy for quality. Keywords, reputation, etc are all proxies. But now with these AI models Google should be able to assess the true value of a page. Does it provide useful information? In what area? Etc. Then invert that in order to answer queries. I really hope this is where page ranking is headed.
- JohnFen 4y ago> But now with these AI models Google should be able to assess the true value of a page That'd be nice, but I am skeptical for a number of reasons. Not the least of which is that where Google has already applied AI, it has seemed to make things worse, not better.
- theclansman 4y agoYeah this, if I want to know how many chapters are there in a game, I have to real an entire article telling me useless stuff before I can get the info I want. Witj ChatGPT I can avoid this. I would say most of my Google searches were for answering questions, not to find articles related to a topic.
- echelon 4y agoEven if Google had been good stewards, they failed to diversify. AI AdBlock is going to kill whatever is left of Google within the year. AI content spam is absolutely going to kill social media and search results, but AI adblock will be worse to Google's bottom line. It's coming, and it's going to be brutal.
- greyman 4y agoIn what sense will it kill social media?
- echelon 4y agoUnless social media websites begin to require ID verification, they'll be flooded with LLM-powered bots. Humans will tire of this.
- regularfry 4y agoWe can't be far off the state where any captcha-style humanity test that can be implemented is immediately vulnerable to bots, too.
- drowsspa 4y agoNot captchas, OP probably means actual IDs. And to be honest, with some cryptography surely we can find a way to make it work and be anonymous
- JohnFen 4y agoI simply cannot imagine being willing to provide identity documents to any of the social media/advertising companies. They would have to somehow demonstrate they're trustworthy first.
- michaelt 4y agoThe bit of social media people use to talk to their IRL friends and family will be fine. But the bit of social media where people are listening to celebrities, getting their news, sharing memes, discussing common interests? It might get flooded with bots so convincing as to be indistinguishable from normal users.
- Farow 4y agoI'm guessing that monetizing comments is a lot more difficult on reddit as any links can be removed by admins/mods/automoderator.
- arnaudsm 4y agoSince word heuristics and pagerank are useless nowadays, which signals should Google prioritize instead? Bounce rate ? Engagement ? Which metrics can proxy user satisfaction without turning into clickbait-land?
- j_crick 4y agoAnd the worst of SEO pages are cooking recipes, where you have to scroll down screens of badly written meaningless text before you actually get to ingredients list. At least Google still keeps indexing recipe microformats.
- fnord123 4y agoI haven't used Google in ages. Do they convert stupid recipes using volume for flour to use mass?
- j_crick 4y agoI haven’t noticed, but I usually hang around King Arthur’s website for bread recipes and they use mass and bakers’ percentages, so do other baker blogs that I may find. Maybe it’s different for things other than bread, but I can’t say. Hail Eris!
- JohnFen 4y agoI use this pretty often: https://www.traditionaloven.com/conversions_of_measures/flour_volume_weight.html https://www.traditionaloven.com/conversions_of_measures/flou...
- carlosjobim 4y agoSearch your recipes on YouTube. It is a better way of getting the information and higher quality. Edit: To the friendly people downvoting me: Try for yourself and you'll be surprised how much good and informative content you can find on YouTube.
- ajsnigrutin 4y agoHi guys! Last month I was visiting my grandma... (3 minutes of this) ...so don't forget to like and subscribe this! Now, i usually use this kind of flour, but I decided to replace it with this special kind from my local farmers market, which is great, because it's sold by this old lady who works... (5 more minutes of that) ... so if you like putting ginger into your cookies, write it down in a comment (3 more minutes of this) ... now you have to fold in the cheese... i have a separate video (15 minutes) just explaining how to fold it in, so I won't show that here, but click the link in the description to see that ... Yeah... i prefer text. It's even wors with tech stuff.... do they expect me to retype the commands/code from the video?!
- lordnacho 4y agoThe computer generated essay and the fake stackoverflow clones are the worst thing ever. It's harder to code now than just a few years ago because of all the BS people put on the internet. I'm finding some value in live chats with the repo maintainers, but that's not really viable for everything, plus won't be easy for juniors to do either.
- mmcgaha 4y agoMaybe I am wrong about this, and maybe the SEO junk sites would find a way to evolve, but what I really want from google is a plonk button so I can have a personal kill file for sites that I never want to see again.
- Eddy_Viscosity2 4y agoWow, yes to this -- your own google blacklist.
- mcherm 4y agoAs a bonus, Google would have access to the list of who had blacklisted what sites -- surely that would be useful to them as a signal to detect "SEO spam sites".
- Eddy_Viscosity2 4y agoBut if they do that, then Goodharts gonna get his law involved and SEO will then involve gaming these lists to blacklist competitors. Arg.
- fnord123 4y agoGoogle allowing the blacklisting of sites with lots of ads on them and using google analytics? They have enough information when yiu clock back and select a different result. And this hasnt killed SEO spam.
- AnIdiotOnTheNet 4y agoThat's because Google is an ad company. They will never be part of the solution.
- squarefoot 4y agoIt would be possible to do that at browser level through extensions, for example there is one to filter out Pinterest results from search results. I wouldn't however expect that potentially great feature to be natively available in Google search anytime soon, unless they made it as a analytics filled paid service and in a way that gave their advertisers and investors the power of circumventing it, that is, both plonk and unplonk as a service.
- wslh 4y agoLimit your search to Reddit, HN and/or Stack Exchange! I really loved the first paragraph of the article about Reddit: "Reddit is currently the most popular search engine. The only people who don’t know that are the team at Reddit". I want to add this that just happened at my company, that shows really bad execution on the Reddit side. Bad as if the current Reddit management is surfing tiny waves in some paradise. Our company tried to use Reddit Ads for years. We are not US based and there was problems using every credit or debit card. The ones that work everywhere and. obviously, in Google Adwords. We retried a few months ago and it worked but our Reddit ads were not approved because they were in the crypto space where indeed we are in the security auditing and research side and also work for Web3 projects. Then, they assigned a higher level Reddit contact, they gave us several forms to fill to show that we were not in the wrong Web3 side. We completed every form over several weeks, until they say they don't have an account manager for our account. At this point I don't know if the right thing is to talk to the CXO level directly. Yes, Reddit management doesn't understand their own business and most probably loosing a lot of money. BTW: I specifically ranted about ways of challenging Google back in 2013 in the same vein [1]. [1] http://blog.databigbang.com/letters-from-the-future-challenging-googles-search-engine/ http://blog.databigbang.com/letters-from-the-future-challeng...
- Phemist 4y agoSo, assuming that this "mismanagement" is not just happening in the crypto space, but across the board in other spaces too, they are missing out on short-term money, but are more valuable for the searches that users are now doing (including me). So.. is it actually bad execution?
- wslh 4y agoThe mismanagement is not between quotes since all compliance was filled and acknowledged. Could you please elaborate on your question? Since our issue was supposely solved filling all the requirements there is no difference about the different spaces. This is bad execution by the book.
- carlosjobim 4y agoTo me all signs point to Reddit never having been a commercial venture, but rather an influence and propaganda machine for United States government agencies. So they already have their budget paid.
- carrja99 4y agoA few years back I heard a startup pitch for an application that would crawl all those recipe short stories and extract the recipe for you.
- treeman79 4y agoWhile nice. The copyright lawyers would have a field day.
- brookst 4y agoFun fact: recipes are not copyrightable. The text around them can be, but that’s what you’d be stripping out or summarizing. If you have an amazing recipe you want to keep control of, trade secret is the way. And then it can’t very well be posted to the internet.
- jvanderbot 4y agoBecause of a recent HN article on the subject, I was searching for "correlation of ACT scores and IQ", and the first 5 hits were the _same_ crappy 1500 word essay, where various section headers were programmatically changed to include exactly what I searched for. How challenging would it be to just check common searches and generate a page that hits all relevant wordings by including those words right in the page. I tried several different variations like "How to convert ACT score to IQ", etc. Same stupid page.
- foobarian 4y agoThis is a major reason I think LLMs could help with SERPs - presumably it will be a lot harder to do SEO effectively because now it involves the model training barrier.
- brookst 4y agoModel training is already trivial, just expensive. As LLM optimized HW appears the expense will drop. By 2030, I’d expect decent consumer hardware to be both faster and two orders of magnitude cheaper than what’s in use today for model training.
- makestuff 4y agoThe industry term for it is "made for advertising". It is a constant battle because some people in adtech want to increase impressions at all costs, and some people want to block these sites from their systems. Guess which one wins out when executives see how much money it makes?