The article appears to be conflating three different things:
1. The decline and eventual collapse of Gigablast
2. The decline of Google/Bing
3. The poor quality of other search engines like Chatnoir.
It's not clear that these three things are related. Gigablast seems as though it were a personal project, and indications are that it was mismanaged. (Posting bizarre political messages -- the sort of thing you'd see on novelty bumper stickers -- on your company's website does not inspire confidence.)
The decline of Google's search engine has been well characterized, e.g. in "Enshittification" and Ed Zitron's work, and it's quite well understood as the reflection of a perverse incentives problem.
Small search engines, e.g. Chatnoir, were never good. Bing, also, was never good -- outside a few narrow niches, like video search, and that very temporarily.
Fair. Now, I don’t understand why some big company doesn’t see that all the search engines are terrible quality, and then make a better product and steal the market. Sure, google does their anti-competitive things, but I think at this point there are enough people who dislike the current situation, and a good search engine could survive anyway.
Yeah, Microsoft tried it by making a product that has (as far as I know) always been smaller and less useful than Google. Do you think it is impossible to make a better product than Google?
They did blind "taste tests" that showed people preferred Bing. Even Microsoft can't overcome Google's inertia with a better product even with the marketing blitzes they've done and all their efforts to push it through Windows.
It's not impossible but it's clearly a hard problem, and all the other big companies seem to prefer doing something else.
Can you provide a link to these “taste tests”? Personally I could see preferring bing over google just for the UI up until maybe 2018, but back then Google was better too. As for the quality of results, I wasn’t paying attention back then.
I am guessing that these “taste tests” were done around that time or before, and I wonder how they would go today.
It was a long time ago. They stripped it of identifying features for the test. Finding information on it is hard on account of the overall decline in search quality. There might be a post on it here on HN but I couldn't find it on a quick search.
I found that by searching with DuckDuckGo (which is Bing) for "prefer bing" "taste test".
If that is what you are remembering, then I would point out that the article makes it sound like Microsoft conducted the test, and if that is the case I could see the test results being less trustworthy. Also the sample size was only 1000 people, which is really small.
Bing it On was what they called the campaign, that’s pretty funny.
I could see this statistic combined with Bing’s market share being a reason why companies would decide not to try making a search engine, so you have a point there. Considering how terrible both search engines are though, I think it’s a dumb decision. We need a better search engine.
Either you are saying that malware in images isn’t very common, or you are saying that it is common and you live in a really scary place where you should never buy candy
One is that I cannot select part of the text, then copy and paste it into the "reference material" I keep in plain-text files on my local computer.
Another is that there is no convenient way to search, i.e., Ctrl+F does not work.
Another is that making the text larger[1] is less convenient: in general I need to right click > "Open image in new tab", then do the zooming rather than just do the zooming, and then the zooming is jumpier somehow (which I can handle, but it slows me down) and text never reflows. Sometimes after I zoom the image to make the text large enough for me to read easily, a column of text has become too wide for my screen so that I need to scroll back and forth horizontally for every line of text. That was not a problem with this specific image file, but IIRC I gave up right before I noticed that that would not be a problem.
Finally, even when the text I want to read is actual text (not an image) sometimes I have to resort to Ctrl+A (select all) then copy and paste the text into a text editor (because every web site is a snowflake, and some of the snowflakes are gnarly); knowing that this last resort is not available with an image file makes me less likely to try to read the image file.
The reasons I give above add up to a situation in which it is slower and more annoying to read from an image file than to read ordinary HTML text or plain text.
Also, I was already annoyed by archive.org even before I figured out that the thing I might want to read is an image file: the way my browser is configured (OS zoom set to 200%, then Chrome's zoom set to 80%, so the effective zoom of the viewport is 160%, but the browser chrome occupies twice as much vertical real estate as it would if OS zoom were 100%) when I land on the web page, more than half of the viewport is occupied by a plea for money at the top plus the two menu bars in the site header. (And the text in the plea for money is larger and sharper than the text of the content.)
Uh, there is a global secret state level private enterprise looting the world and even conventional governments are there pwn?
You know, when China “hacked” the US telco infrastructure, that back door is actually owned by a private intelligence community enterprise that rents it back to the FBI the CIA and NSA.
These purchased all of this and more using post 9/11 “debt cards.”
Who else understands information is the next black gold and has Americas’ complicity to back up their thirst for more everything?
I’m sorry, I’m having a little trouble understanding. Are you saying that you think it is a private company that is paid by governments to sabotage search engines?
I like your username by the way. It sounds like the evil version of sun screen.
Gotcha. If something like this was happening, it would be a huge story if news broke about it. How would you go about finding proof? Make a search engine and see what happens I suppose?
If this is what is happening, why hasn’t anyone brought it to light already, especially with how many people would have been targeted so far?
The enemy of whom I speak is your thought control.
The Power eluded to in the links below are of the nature of those reported in the CIA Stargate programs. Those are cargo cult interfaces for the mainstream governments’ interface with Power. Remote viewing? That’s how they found Bin Laden. Puffs.
Power is real and we are not alone in our own minds.
These other things are just games for Power, for fun and for profit. To be your invisible God profit.
I am completely open to the idea of people trying to coerce search engine operators into shutting down or having a worse product. Essentially it would be the same as what was talked about in the Twitter Files.
Are you essentially saying that people who want to control information would also run a bunch of propaganda about how there isn't a problem with search engines ("thought control" through propaganda and gaslighting), and also threaten people who run search engines so that they don't talk about censorship? If so, I think that would be possible but I wouldn't say that's the case unless I found proof.
Or are you saying that there are people who use actual telepathy to control people? Because if that is the case, I don't think that telepathy is real. Now, it might be possible to convince someone of something, or make them feel like an alternative idea is crazy, and then whenever they try to think critically about the situation they put up a mental block because they don't want to feel crazy. That's just social engineering. This might seem like a voice in their head in certain circumstances, or perhaps if the victim had schizophrenia it actually would be a voice in their head, but never actually telepathy.
And in either of these cases, are you saying that it's hopeless to try to make a search engine and see for yourself if there is real censorship taking place, because people wouldn't believe you if you talked about it? If you are saying that, I disagree.
I would love to hear your theories as to why, or your reasoning for why this is normal.
The article appears to be conflating three different things:
1. The decline and eventual collapse of Gigablast
2. The decline of Google/Bing
3. The poor quality of other search engines like Chatnoir.
It's not clear that these three things are related. Gigablast seems as though it were a personal project, and indications are that it was mismanaged. (Posting bizarre political messages -- the sort of thing you'd see on novelty bumper stickers -- on your company's website does not inspire confidence.)
The decline of Google's search engine has been well characterized, e.g. in "Enshittification" and Ed Zitron's work, and it's quite well understood as the reflection of a perverse incentives problem.
Small search engines, e.g. Chatnoir, were never good. Bing, also, was never good -- outside a few narrow niches, like video search, and that very temporarily.
Fair. Now, I don’t understand why some big company doesn’t see that all the search engines are terrible quality, and then make a better product and steal the market. Sure, google does their anti-competitive things, but I think at this point there are enough people who dislike the current situation, and a good search engine could survive anyway.
Microsoft tried that. Its market share 15+ years on is probably why no one else tried.
Yeah, Microsoft tried it by making a product that has (as far as I know) always been smaller and less useful than Google. Do you think it is impossible to make a better product than Google?
They did blind "taste tests" that showed people preferred Bing. Even Microsoft can't overcome Google's inertia with a better product even with the marketing blitzes they've done and all their efforts to push it through Windows.
It's not impossible but it's clearly a hard problem, and all the other big companies seem to prefer doing something else.
Can you provide a link to these “taste tests”? Personally I could see preferring bing over google just for the UI up until maybe 2018, but back then Google was better too. As for the quality of results, I wasn’t paying attention back then.
I am guessing that these “taste tests” were done around that time or before, and I wonder how they would go today.
It was a long time ago. They stripped it of identifying features for the test. Finding information on it is hard on account of the overall decline in search quality. There might be a post on it here on HN but I couldn't find it on a quick search.
Is this what you are remembering? https://www.searchenginewatch.com/2012/09/06/bing-it-on-bing...
I found that by searching with DuckDuckGo (which is Bing) for "prefer bing" "taste test".
If that is what you are remembering, then I would point out that the article makes it sound like Microsoft conducted the test, and if that is the case I could see the test results being less trustworthy. Also the sample size was only 1000 people, which is really small.
Bing it On was what they called the campaign, that’s pretty funny.
Hey, you might be able to do a test like this today, as an impartial third party. The Cross Origin Policy might make it harder though, see this: https://web.archive.org/web/20230321113801/https://gigablast...
I could see this statistic combined with Bing’s market share being a reason why companies would decide not to try making a search engine, so you have a point there. Considering how terrible both search engines are though, I think it’s a dumb decision. We need a better search engine.
If this had been ordinary text rather than an image with text in it, I would've read it.
Alright. I OCRed it with Tesseract, and then proofread it and fixed things that were broken like links, and then I put it up on github: https://n1xis10t.github.io/search-engine-article/
It is one left aligned column of pre-formatted text in an html file, which contains no style information or javascript.
Is this because of the malware risk inherent in images?
malware in images is as prevalent as razorblades in candy
Either you are saying that malware in images isn’t very common, or you are saying that it is common and you live in a really scary place where you should never buy candy
No, it is several things.
One is that I cannot select part of the text, then copy and paste it into the "reference material" I keep in plain-text files on my local computer.
Another is that there is no convenient way to search, i.e., Ctrl+F does not work.
Another is that making the text larger[1] is less convenient: in general I need to right click > "Open image in new tab", then do the zooming rather than just do the zooming, and then the zooming is jumpier somehow (which I can handle, but it slows me down) and text never reflows. Sometimes after I zoom the image to make the text large enough for me to read easily, a column of text has become too wide for my screen so that I need to scroll back and forth horizontally for every line of text. That was not a problem with this specific image file, but IIRC I gave up right before I noticed that that would not be a problem.
Finally, even when the text I want to read is actual text (not an image) sometimes I have to resort to Ctrl+A (select all) then copy and paste the text into a text editor (because every web site is a snowflake, and some of the snowflakes are gnarly); knowing that this last resort is not available with an image file makes me less likely to try to read the image file.
The reasons I give above add up to a situation in which it is slower and more annoying to read from an image file than to read ordinary HTML text or plain text.
Also, I was already annoyed by archive.org even before I figured out that the thing I might want to read is an image file: the way my browser is configured (OS zoom set to 200%, then Chrome's zoom set to 80%, so the effective zoom of the viewport is 160%, but the browser chrome occupies twice as much vertical real estate as it would if OS zoom were 100%) when I land on the web page, more than half of the viewport is occupied by a plea for money at the top plus the two menu bars in the site header. (And the text in the plea for money is larger and sharper than the text of the content.)
[1] My eyesight is substandard.
I see, thank you for explaining. How well does the version that I OCR’d and then put in an html file work for your purposes?
It works fine. No complaints.
Perfect.
Fascinating.
Uh, there is a global secret state level private enterprise looting the world and even conventional governments are there pwn?
You know, when China “hacked” the US telco infrastructure, that back door is actually owned by a private intelligence community enterprise that rents it back to the FBI the CIA and NSA.
These purchased all of this and more using post 9/11 “debt cards.”
Who else understands information is the next black gold and has Americas’ complicity to back up their thirst for more everything?
I’m sorry, I’m having a little trouble understanding. Are you saying that you think it is a private company that is paid by governments to sabotage search engines?
I like your username by the way. It sounds like the evil version of sun screen.
Private enterprise is the new secret state. The govs are squares with day jobs. Think about it.
This new breed is uninhibited by state safeguards. And those tolerate them blindly for the govs enjoy outsourcing and plausible deniability.
I like to think I’m the good version of starscream.
Edit: oh, and they want to control search engines as they curate the catalog of nearly all internet accessible content.
If you cannot find it, it might as well not exist.
It is one of many information flow control strategies. There are so many.
Gotcha. If something like this was happening, it would be a huge story if news broke about it. How would you go about finding proof? Make a search engine and see what happens I suppose?
If this is what is happening, why hasn’t anyone brought it to light already, especially with how many people would have been targeted so far?
You are many times fooled.
The enemy of whom I speak is your thought control.
The Power eluded to in the links below are of the nature of those reported in the CIA Stargate programs. Those are cargo cult interfaces for the mainstream governments’ interface with Power. Remote viewing? That’s how they found Bin Laden. Puffs.
Power is real and we are not alone in our own minds.
These other things are just games for Power, for fun and for profit. To be your invisible God profit.
https://anonpaste.com/share/american-thought-control-and-tho...
https://anonpaste.com/share/the-secret-war-ca40445143
https://pastebin.com/3cPXbjhA
I am completely open to the idea of people trying to coerce search engine operators into shutting down or having a worse product. Essentially it would be the same as what was talked about in the Twitter Files.
Are you essentially saying that people who want to control information would also run a bunch of propaganda about how there isn't a problem with search engines ("thought control" through propaganda and gaslighting), and also threaten people who run search engines so that they don't talk about censorship? If so, I think that would be possible but I wouldn't say that's the case unless I found proof.
Or are you saying that there are people who use actual telepathy to control people? Because if that is the case, I don't think that telepathy is real. Now, it might be possible to convince someone of something, or make them feel like an alternative idea is crazy, and then whenever they try to think critically about the situation they put up a mental block because they don't want to feel crazy. That's just social engineering. This might seem like a voice in their head in certain circumstances, or perhaps if the victim had schizophrenia it actually would be a voice in their head, but never actually telepathy.
And in either of these cases, are you saying that it's hopeless to try to make a search engine and see for yourself if there is real censorship taking place, because people wouldn't believe you if you talked about it? If you are saying that, I disagree.
I am saying we have been rooted and all of our bases do not belong to us.