Great news! I already use Brave Search and it's very good, a built-in image search was the one thing missing. (And maybe Maps integration too.)
Brave cuts ties with Bing to offer its own image and video search results
Brave Software, maker of the Brave web browser, has tuned its search engine to run on a homegrown index of images and videos in an effort to end its dependency on "Big Tech" rivals. On Thursday, the biz said image and video results from Brave Search – available on the web at search.brave.com and via its browser – will be …
COMMENTS
-
-
Friday 4th August 2023 18:50 GMT Anonymous Coward
Now we just need to get Duck to team up with them
Bing and Google are sniffing their own farts, and by that I mean their search team in ingesting too much of the AI screed that M$ is helping produce, and the rampant SEO garbage fire that Google created and help destroy the modern web. So AI models and web crawlers are ingesting ever more generated content.
Left unchecked it could hit a tipping point where the ML systems create a feedback loop with the crawlers, and the FAKE web web gains enough mass to flip the filtering systems and the mass hallucinations get flagged as the genuine content.
Better hope that lawsuit against the Internet Archive doesn't take out the primary public record of the pre-AI-circlejerk era.
-
-
-
Friday 4th August 2023 14:14 GMT BenDwire
Re: A pity about …
I'm with you on that, but after collecting about $100 'worth' of tokens I decided to donate them all when Brave withdrew their own token storage. I then turned all the crypto crap off.
I was, and remain, fearful for giving some crypto outfit a copy of my photo ID just for the ability to store coins / tokens. I'll stick to gambling my hard earned on the stock exchange ... I'm sure the outcome is somewhat similar!
-
Friday 4th August 2023 15:07 GMT Marty McFly
Re: A pity about …
>fearful for giving some crypto outfit a copy of my photo ID
Look up KYC & crypto. "Know Your Customer". That is not something any crypto outfit really wants to do. It is something they are required to do. Frankly any crypto exchange doing KYC is an indicator of them playing by the rules rather than a fly-by-night exchange that doesn't do KYC.
-
Thursday 3rd August 2023 18:00 GMT Tron
Other solutions.
You could bounce your search query off third party web page code, as long as you return the favour, creating an anonymising network. It would work for search queries just as it would work as a crowd VPN service. You could scrape/filter Google search results, allowing users to avoid specific results. This would be particularly useful as Google now ignores half of what you type it and sends you popular/trending crap instead of what you want. You can also implement stuff like persistent search and distributed search, allowing users to create their own links by order of usefulness for any search terms. This would effectively remove the issue of censorship of search. You could also allow endless results (as Google originally did), rather than the 3 or 4 pages it now limits you to. Lots of cool new stuff to do. Would be nice is someone actually did it.
-
Thursday 3rd August 2023 19:15 GMT chivo243
Brave search, not great yet
I use Brave everyday, but there are issues with search, and I often use google on Firefox to find what Brave Search fails to find for me. One fly in the ointment is that I live in a town in the US with the same name as a large Canadian city, even adding the state name to the search string doesn't help, Brave shows me shops for that city in Canada.
-
Thursday 3rd August 2023 19:22 GMT Rich 2
Google search
“… Google Search is getting worse. Part of the problem is AI content generation,….”
I don’t buy this at all. Google search was good back in the day. But it’s been gradually getting worse for a long time and has been rubbish for years now; long before “AI” became a thing.
I think the reason for this is that they are no longer interested in actually providing useful search results any more, and haven’t been for ages. They are only interested in serving up whatever it is they need to serve up to make money out of you, which is why the first 20 results of any search point to Youtube and Amazon (presumably because Amazon pay Google boat-loads of cash), with the remainder pointing to stack overflow. If Google happens to provide a useful search result these days, it’s purely by accident
-
Friday 4th August 2023 19:08 GMT Anonymous Coward
Re: Google search
Yeah, double edged problem, as it is both a BS excuse Google uses to cover up changes they have made in the search functions, and a legitimate technical challenge they are fighting.
The fact that none of the technical or advanced search options work anymore was a choice they made, not a side effect of the whack-a-mole game with the content farms and SEO companies. And the parasitic/symbiotic, love/hate relation Google has had with those A-holes predates LLMs by a long while. So a big chunk of the content farm is a leopards ate my face whinge from a company that figured out they could shove more paid links in front of your search results if they took away most of the ability to restrict the search terms. They happily let content farms operate as long as the got a cut from ads running on the content farmed pages, and the users would end up re-googling their search more often then not, giving them an extra slice of sponsored content to serve before the actual search results.
Now Foogle has buried results as far back as page 3, and is starting to fail returns for literal searches copy and pasted from saved pages.
Bing and Google suck now, and bad enough that homebrew is yielding better results. They are also not agile or nimble at all anymore, so the market is ripe for a disruptive entrant. Let them burn, few will miss them in their current form.
-
Sunday 6th August 2023 16:42 GMT Jonathan Richards 1
Re: Google search
> technical or advanced search options
Yes, this. If one learned searching long enough ago then a search term like "(BOFH OR PHY) AND (QUICKLIME OR SHALLOW)" comes naturally. The Fine Article gets a +1 from me for the mention of Mojeek, which had flown under my radar until now. It *does* offer a number of useful search operators, see <a href="https://blog.mojeek.com/2023/08/mojeek-operators-a-guide.html>A Guide to Mojeek Operators</a>
-
-
-
Friday 4th August 2023 10:22 GMT Bebu
Just wondering...
Some years ago when the main stream search engines (ok I mean google) started spewing crap in response to simple purely technical queries (and the decrepit altavista was returning better results) I wondered whether a search algorithm that discarded any page that directly or indirectly tried to sell you something, discarded pages that had any social media links, anything that had links to any other ratbaggery and skulduggery could be tuned to return reasonable answers to reasonable questions.
eg
Q. Whats the meaning of life, the universe and everything?
A. 42 and bugger off. Your card is marked son.
Q. Whats the part number for the cmos ram used in the Applied Technology Microbee.
A. 6116 you will find a datasheet on the Renesas web site. From where did they disinter you grandpa?
-
Friday 4th August 2023 16:09 GMT Rich 2
Re: Just wondering...
Great idea except it seems every single website these days has links to anti-social media, adverts, bla bla. Even websites that are otherwise “useful” (yea, I know - I use the term “useful” very loosely here; a truly useful website is as rare as rocking horse shit these days)
-
Friday 4th August 2023 19:14 GMT Anonymous Coward
Which would stop in a hurry if having them dropped instead of boosted your SEO rankings
My site auto-hides the effin things unless the user clicks on them first. So it's not like this will break the web by itself. Still need to pay the piper though. Sites gotta run on servers somewhere. I'd just like for that to not be Facebook, Twitter, or Google.
-
-
-
Friday 4th August 2023 15:34 GMT Anonymous Coward
Untrustworthy results
Google & Bing will give you the results they want you to see. They will not give you the results you want to see.
As an academic exercise, a few years ago I did some searching for a vaccination card template. Nowhere to be found on Bing or Google. I did, however, find tons of pages with the template when searching via Startpage & Duckduckgo. These were not random pages hosted on shady websites either. These were government websites (mostly individual state agencies) re-posting official CDC documents which contained the template.
I learned something very important. Big Tech cannot be relied on to be impartial in their search results. These were results for official government websites that they were intentionally leaving out of their search results. This was not Google/Bing prioritizing one product or service over another because they got paid advertising dollars. This was blatant suppression of official government documents publicly released on official government websites.
Don't get wrapped up around the covaids topic, that is not my point here (an academic exercise, okay?). During a time of crisis they manipulated their search results to conceal official government information pertaining to the crisis. That is an immense amount of power & control that they have shown a willingness to use. This should scare everyone one of us.
-
Saturday 5th August 2023 14:19 GMT Anonymous Coward
Re: Untrustworthy results
yup, AC- as well,TPTB dictate the SE script[sic] by Always omitting the downsides of the official line/pablum fed via msm...synthesis vs antithesis be dammned-it's our way,or the highway...fer instance,enter le clotshot- ddg gives same crapp as goo, but use yandex&get gobsmacked by the reality of what's behind the curtain/swept under the rug at our mortal expense- viz.,
despair.com/cdn/shop/products/sacrificeposter.jpeg?v=1433971332
=Text in image=
SACRIFICE
All we ask here is that you give us your heart.
"Sometimes, emplovees are asked to make sacrifices for the company.
Other times, they must be sacrificed for it."
-----
youtube.com/watch?v=E5VMZqgVzRo SIXTEEN TONS - Tennesse Ernie Ford w/ Lyrics
-
-
Friday 4th August 2023 16:01 GMT Anonymous Coward
I don't know much of anything about how search engines work, so how necessary are indexes today? They were designed at a time when even business lines were slower than residential lines today. Are index just the solution we already know or is there a reason there hasn't been a more streamlined solution for smaller or decentralised engines?
-
Friday 4th August 2023 19:24 GMT Anonymous Coward
In short yes, there are reasons
While there are other possible architectures, without some kind of index you would need to load a copy of every page of every webserver in the world that could possibly contain the thing you are looking for.
This is a terrible and impractical idea, as then probably 90% of the world internet traffic would be inefficient searches, the internet would literally break if you tried it. A shared index lowers the server load for everyone, and is many orders of magnitude faster, more efficient etc.
The trick is that crawling the entire web requires a huge amount of resources, and so the companies that have a big, current, and searchable index don't just share it for free, or let you use it directly.
There are other viable solutions to online search than the Bing/Google duopoly, but not likely anyone is talking about them outside of their own companies or projects, or at least a stiff NDA. Any good idea that a VC won't pay good money for will probably still get snapped up by one of the majors to protect the search monopolies.
-
Friday 4th August 2023 20:40 GMT Anonymous Coward
Re: In short yes, there are reasons
Surely most searches are to a relative handful of resources (I'm sure I read some unassailable statistic somewhere). Wouldn't it be more efficient to model where a search is most likely to be resolved, and only index the ones that aren't reliably resolved by the model?
-
Sunday 6th August 2023 17:08 GMT Jonathan Richards 1
Re: In short yes, there are reasons
I'm not sure that I'm following your logic here. Is the strategy that you're suggesting along the lines of pointing searchers towards specialized repositories, a bit like a librarian pointing you towards the History of Medicine shelves if you're interested in penicillin? Because although you could then browse the [books | resources] there, you wouldn't be that much closer to whatever particular aspect of penicillin you were curious about. A really good librarian will help you find the right index term in the Universal Decimal Classification, or Dewey (depending on the library, of course).
What I hope you're NOT suggesting is that a significant majority of what any particular search is about is to be found in a handful of web sites. That cannot be the case, and goes against every concept of distributed knowledge and the World Wide Web, which even in its entirety is not the Sum of Human Knowledge.
If you're looking only to index [1] a sub-set of the Web, how do you identify that sub-set? Seems to me that you'd have to index the whole thing to find out.
[1] Indexing used to be a skilled job, where someone with subject matter expertise would assign index terms from a controlled vocabulary to a document, so that it could be retrieved with precision later on. Sometimes the terms were filed on index cards, and then we started using computers. I know, I know, I'm old. Nowadays, Google et. al. don't have an index in that sense, at all. They have a searchable concordance, and the precision of the search results depends directly on whether the set of words in the document tell you what the document is about. Generally, it does not, and then there is all the SEO and "algorithm" finagling on top of that.
-
-
-