371–380 of 515 posts
Plenty of technical topics where i get better results from duck duck go (and i dont even like it)
Everytime one of these threads show up, I find myself wondering why someone doesn't launch a technical and research focused competitor. There seems to be such broad agreement that Google is now terrible for research that there's for sure a market for it. Edit: I'd love to see a some-of-the-web search engine like this. Start just with university sites, prepress archives, quality forums, public dev Slacks, etc.
Neeva is worth looking at. I recently switched to it exclusively for a week and found it as good as Google for 90% of what I do. (FWIW, previous attempts doing this with Bing or DuckDuckGo last about a day before I get so mad I switch back.) Right now Neeva seems very good at navigational queries, which I do 90% of the time. It's still not as good at deep research queries for obscure things. Probably related: Neeva i…
The time has come for a distributed, decentralized search engine powered by the immensely powerful devices in our pockets, with rankings determined by our peers and local network, not by a single global ranking. I'm not saying a distributed search engine with localized results will replace Google (you still want the global standardized search results in some cases), but rather we need a better alternative to simply asking your friends/communities like Reddit. The combined spare computing power of every smartphone on the planet is probably greater than AWS or Google Cloud could ever be. If people were incentivized to sell their spare battery/CPU/network capacity for web crawling, perhaps by earning money, this would solve the advertising/privacy problem with Google's business model as well.
Earlier quoted context omitted.
Yes, because it sux, in general. Maybe as an option.
If something changes within ~300ms before your click, the only way you'd be intentionally clicking whatever's there now is if you're amped up playing some online game
Gaming is exactly one of the problems and all those sites that lie on spectrum between news to gaming and utilize some aspect of dynamic presentation.
The results keep getting "refined" so as to suit the popular 80% of queries, while getting much worse for any technical or obscure queries. Forced synonyms and "people also searched for" are typically useless and almost infuriating. Once you get off the first or second page, the results get even worse-- with pages entirely unrelated to the query (e.g. not even containing the searched phrases). They are probably testi…
I often see this line of thought come up whether topics like this float around and wonder: how can we demonstrate /quantify /prove this. Especially since I'm already partial to agreement.
Earlier quoted context omitted.
This is a reply to myself... but does anyone have ideas about how this can go away? Have we lost the internet to eternal marketing and data-mining arms races? Is this why people are flocking to search-opaque places like Discord? Sometimes I feel like that internet isn't for me anymore, and that's a little distressing.
profit optimized is the inevitable result of any system that has profit motivation banning advertising would help for awhile, but other profit streams would be optimised and expand to fill void, such as data collection/mining
- ad revenue would drammatically decrease
- buying ads would actually be affordable by local sellers
- the market would self regulate against monopolies
My take: Github Copilot and similar products will replace many google queries in the near future, for code, but also for other specific domains. When I type a query, I'm essentially trying to summarize my context, and I can't compete with an advanced deep learning model that reduces my code to a 300-dimensional context vector. The same is true for a lawyer writing a legal document or a scientist writing an academic p…
Almost all personalized information retrieval systems are some variant of this type of recipe since the late 90s no matter what mathematical model they use.
Perhaps the problem for Google is that there is strong incentive for the content providers to play a cat and mouse game (aka SEO) with google. Depending on business model, anyone sufficiently popular in this space will probably have similar problems once companies have to either pay for content or revenue share. Github is a situation where people (open source developers) "donated" their code to Microsoft, but most of the world is not that egalitarian.
The results keep getting "refined" so as to suit the popular 80% of queries, while getting much worse for any technical or obscure queries. Forced synonyms and "people also searched for" are typically useless and almost infuriating. Once you get off the first or second page, the results get even worse-- with pages entirely unrelated to the query (e.g. not even containing the searched phrases). They are probably testi…
Yes, that is also my observation. They will show content that completely does not contain your query words if that content is just popular enough. The result is that niche technical topics get drowned out by related popular discussions. Also, I am by now 100% sure that Google has just stopped indexing the long tail. Like if I search for function names of public source code that I downloaded from GitHub, Google won't…