Goggles: Democracy dies in darkness, and so does the Web [pdf] | Hacker News Reader

Goggles: Democracy dies in darkness, and so does the Web [pdf]

brave.com

130 pointsInvaderFizz137 comments

Wouldn't these "goggles" have exactly this effect? Instead of being silently lulled into the bubble by "the algorithm", users explicitly dig their own hole from which they can be certain never to cast eye upon an opposing idea.

As they describe in the paper, "at least the choice will be a conscious one". If someone wants a wide variety of sources and opposing ideas, there will be a goggle for that. If someone wants only anti-vax articles, there will probably be one for that too.

This is probably much better than what currently happens, where search engines become personalized but users don't realize it. That anti-vax person might think they are searching the whole web when google has quietly flagged them as a user who enjoys anti-vax articles and continues suggesting those sites, taking them further down the rabbit hole.

> This is probably much better than what currently happens, where search engines become personalized but users don't realize it. That anti-vax person might think they are searching the whole web when google has quietly flagged them as a user who enjoys anti-vax articles and continues suggesting those sites, taking them further down the rabbit hole.

...what the what.

Did you just casually declare that Google is feeding people search results tailored to their political worldview? Because if so, you need to actually come up with something to support that sort of claim.

You're one of today's lucky 10000! [1]

Look up Google personalized search [2]. I guess only Google insiders know for sure what's the extent to which your search results get personalized, but it's common knowledge that this happens. It'd be also interesting to find out if they somehow try to reduce the personalization when it comes to 'political' interests. I'd wager that the answer is no.

And of course it's not just Google. Facebook and Twitter too. And probably every site with some kind of a feed trying to keep you engaged.

1. https://xkcd.com/1053/

2. https://en.m.wikipedia.org/wiki/Google_Personalized_Search

No, they suggested that Google tailors results to the things related to the hits that you follow. His example was political but it was just that, an example.

How could you possibly differentiate an extremely personalized profile, and a political worldview for that matter?

So there's two things to consider:

1) Everything is political, and so are search results and suggestions

2) Personalized results/suggestions, moreover, tend to fit the perceived political views of the user (from the algorithm's perspective)

Google and Youtube (like many others) are well-known to have pushed neo-fascist and conspirationist propaganda to random bystanders (Alex Jones or Trump in the USA, Dieudonné or Soral in France) because their "controversial" viewpoints generate more user engagement. Likewise, Google search often promotes white/christian-supremacist "news" websites like FDeSouche in the top5 results for random queries.

Then, there's personalized recommendations. Once you've clicked one of these results, Google's algos will deem you interested in that kind of content, and suddenly it's impossible to avoid that bubble. It will keep recommending the same sites/channels everywhere you go, for months. That's why clicking "critically" on a link because you're curious what an "opposite side" has to say is a dangerous enterprise with recommendation algos: you'll be subtly yet persistently exposed to their propaganda ecosystem as a result.

I would personally welcome "Goggles" as a positive change. I often know precisely what i'm looking for (an article i saw a few years back), but stupid ranking algorithms make it impossible to find in the stream of information. There's not a one-size-fits-all algo that's going to please everyone: decentralizing this power is the only suitable way, in my humble opinion.

Right up until they ban them entirely.

I wonder what the intended tone is here. I would argue that you're exactly right. Google wants to have their cake and eat it too - they want engagement through rage and unreality, but they don't want to be seen as enabling that kind of thing, so they do performative and largely useless measures like banning people who have large following elsewhere and putting Wikipedia links under Flat Earth articles, but don't actually address the core problem: that they make more money when people get angry for no reason, and thus they have to incentivize that kind of art and "journalism" on their platform.

We see this from the other side every year come Pride; companies put rainbow shit all over their sites while maintaining LGBTQ+ exclusive practices because they don't want to lose their bigoted customers.

I wondered this too, though the sibling comments make good counterpoints.

I tend to look to nature for inspiration in thorny problems. For this one - essentially confirmation bias - genetic mutation seems an attractive metaphor. That is: there's always a bit of randomness thrown in. That could be dialled up/tuned to always include some contrarian answers.

I don't expect we'll all return to the great debating chambers of the past, but it'd be an interesting experiment at least: what would happen if every anti-vax search result included something on the benefits?

Why not let people have their own worldviews? Don't you believe the benefits of self-organized communities far outweigh the downsides?

I'd much rather have 10 000 different communities who don't agree on anything, than have a strong hegemonic culture stepping on every disagreeing voice, as we currently have in most countries.

Would you rather dig your own hole or find, at some point in time, you've actually been living in one?

I'd rather not fall for the false dichotomy at all, and instead have the algorithm ensure that I won't end up in any sort of hole whatsoever, by choice or otherwise.

Well, what about some kind of (private) program that runs on top of your set of goggles, and would tell you that your set of goggles that you frequent happens to slice and dice only tiny of the entire dataset and thus will hint to you that you're peering into holes.

I think this paper is great in delivering a first solution to this space. Sure it's not perfect but atleast miles better than an opaque box.

We humans tend to live in our "holes" quite comfortably. You may not realize it, but there's quite a lot of "invisible" social codes and expectations involved in human interactions with your surroundings, and these codes are far from universal.

Whether you believe in God(s), whether you trust the police, whether you're for community autonomy and collective ownership or wage slavery and private property, for jealousy or free love...

There's so many ways to live and for every individual, most of them will seem alien. When the "hole" is big enough and is crushing other holes with its cultural hegemony (eg. industrial capitalism) then it's harder to realize it's here in the first place, like in the movie Matrix ;)

Is that somehow at odds with what I want? If I search for "can you trust the police," I want both sides of the story. I want information and opinions from varied sources in order to get a comprehensive picture of the topic, upon which I will base my own opinion. I do not want ready-made opinions to copy without thinking.

Any form of curation, whether human or automated, will have biases and opinions. For example, historical encyclopedias were strongly in favor of the status quo. Even Wikipedia, which does a much better job, still has quirks and biases:

- the source reputability criteria prevents much of popular knowledge/culture to be referenced (police abuse example: first-hand witnesses are not considered reliable sources unless quoted by a big newspaper)

- the topic reputability criteria prevents many interesting figures from being referenced (see previous HN discussions about why women scientists are mostly absent from Wikipedia)

Neutrality is a lie. Everything is subjective. We can try to provide broader interpretation and critique, but no single entity can cover every aspect and worldview. In my view, that's a feature and not a bug.

"both sides of the story" is already a hole. Why are there just two? Are they equally credible and valid on every issue? Do we want "both sides of the story" when it comes to the Holocaust or slavery, as we see playing out in Texas today? Do we want only two sides of the story when talking about complex policy measures or historical events?

As other commenters have pointed out, you will be in a filter bubble no matter what, because you cannot speak to every person, read every paper, or understand every point of view. Bias cannot be eliminated, only understood and corrected for.

As a militant antifascist, i still find it interesting to have some reactionary perspectives to balance the dominant narrative. Some historical examples:

- abolition of slavery in the USA was not perceived by many (former-)slaves as social progress, but rather as legitimization of a new form of "wage slavery" where your master doesn't own you, but you owe them for food and housing... moreover, prison slavery is still a thing in most countries which do not recognize workers rights for incarcerated people (France, USA, etc)

- Hitler apologists point out that US army atrocities (during and after WWII) have been mostly erased from history books; on the other hand, Hitler's atrocities have been reduced to the eradication of Jews, whereas queers, handicapped folks and tsigans (among others who have also been eradicated) are barely mentioned if at all. Hitler is also "caricatured" as a genocidal freak, failing to represent his earlier figure which is a lot closer to many of our "presidents" across the globe than is comfortable to admit. After all, Hitler was presented by the media and industry (his biggest supports) as the candidate of reason and property against unions and marxists/anarchists and their rule of the mob

- Daech and other reactionary muslim militant groups insist on the atrocities of Western "democracies" to justify their own; these imperialist atrocities are never acknowledged by eg. France/USA who pretend they're on armed humanitarian missions against barbarism and patriarchy (an argument for colonization which dates back over a century!). This invisible spot in western worldview ("let's fix human rights with more bombs") leaves plenty of room for Daech to recruit young disillusioned folks who know their government is lying and killing innocent civilians abroad

Overall, i think any propaganda slips through your brain and leaves traces, just like any form of advertisement. So i can recommend as a general rule of thumb to never read neo-nazi/patriarchal/capitalist propaganda. However, when you're studying a precise topic, i believe getting points from all sides (not just two, as you pointed out) is an important practice that can help you develop a better understanding of the issue to overcome your own conditioning.

Especially in Global North republics who pretend to be democracies (but don't have any democratic features [0]), it can be hard and painful to acknowledge that our governments don't recognize or respect any of the fundamental rights they so eagerly accuse other nations of infringing on, that all political parties and corporations running the show are delusional maniacs, and that no "human rights" NGO will save us. Sure we all have human rights in theory, but practicing them would be considerably better!

[0] No, electing overlords every few years is not a democracy. Those electoral systems were precisely (historically) designed to prevent democracy which was seen as a danger to the 18th century elites.

I'd rather not have to chose between two broken systems. I don't believe choice is good in this case. I don't think it's good that we can whitelist or blacklist parts of the world from ever entering our minds. Sure, there are lots of places that I know of, that I don't visit because I find them offensive. But sometimes they show up anyway, and while I most often won't agree with their views, I appreciate being reminded that they exist.

Better start believing in hole stories.

‘Democracy dies in darkness’ is a funny trope coming from the mouths of the media, which by its very nature has to focus on long tail statistically less likely events and stories (hence the’news’ aspect).

This has the unintended consequence of decoupling readers from reality, because the long trail events start seeming like the norm.

We see it in survey after survey that the more ‘news’ media you consume the less you are in line with understanding what is actually happening.

"Democracy dies in darkness" is fundamentally true: it's a regulatory system, a basic feedback loop where people elect representatives, observe what happens, then possibly make changes if they don't like what they are seeing.

Take out any element, and the circuit is incomplete. The media is what provides the feedback. Without it, it breaks in the same way your home's temperature will get out of control if the thermometer is broken.

The effect that just seeing some crime stories on TV makes people believe crime is everywhere is probably overrated. As is the believe that "the media" is oh so terrible today and was so much better in some unspecified golden age.

> We see it in survey after survey that the more ‘news’ media you consume the less you are in line with understanding what is actually happening.

Yeah, that doesn't seem plausible. I can't even say what "understanding what is actually happening" is supposed to mean? I guess you have some grand theory of what's really happening that they don't want you to know.

(ps: I know "long trail events" sound so much smarter than "rare". But it's "long tail")

The auto correct of tail to trail is a good one! Thanks for catching.

‘What’s actually happening’ is alluding to boots on the ground surveys and government statistics.

E.g. society calls the police on children walking alone to the park far more today than in the past. The amount of fear has decoupled from the risk.

> We see it in survey after survey that the more ‘news’ media you consume the less you are in line with understanding what is actually happening.

[citation needed]?

https://www.frontiersin.org/articles/10.3389/fpsyg.2020.5608... shows a nuanced result:

"First, we find that social media use is associated with several negative consequences, including the endorsement of misinformation related to COVID-19, lower knowledge about how the disease is treated, and greater prejudice toward Asian Americans...

...individuals using print and radio sources were less common in our study, but these sources were also associated with lower knowledge of COVID-19 and greater stigmatization of Asian Americans.

Finally, and in support that some media formats can have positive influences during public health crises, individuals using web news sources were less likely to endorse misinformation. These findings overlap with recent Pew Research Center findings that individuals utilizing national news on news websites are likely to closely follow COVID-19 updates which could help account for why the endorsement of misinformation was lower."

This seems (unsurprisingly) to depend on the source:

"Further, misinformation scores were significantly lower among participants who frequently use CNN (t = 2.054, p < 0.05) compared to those who do not as well as among participants who frequently use the New York Times (t = 4.397, p < 0.001) as compared to those who do not. Prejudice was similarly lower among participants who frequently used the New York Times (t = 3.070, p < 0.01) as compared to those who did not. Finally, participants who frequently used NPR scored significantly lower on misinformation (t = 5.256, p < 0.001) and prejudice (t = 2.082, p < 0.05), and significantly higher on treatment knowledge (t = −2.242, p < 0.05).

People who used Fox News scored significantly higher on misinformation (t = −4.117, p < 0.001) and prejudice (t = −2.392, p < 0.05), and significantly lower on treatment knowledge (t = 2.766, p < 0.01) than people who did not use Fox News. Participants who frequently used Facebook similarly scored significantly higher on misinformation (t = −3.360, p = 0.001) and prejudice (t = −3.574, p < 0.001), and significantly lower on knowledge of COVID-19 treatment (t = 3.900, p < 0.001) and symptoms (t = 2.387, p < 0.05). Participants who frequently used Twitter scored significantly lower on knowledge of treatment (t = 2.621, p < 0.01) and knowledge of symptoms (t = 2.538, p < 0.05)."

So those who consume "legit" (my words) news sources like the NYT or NPR score well on knowledge and low on misinformation. Those who use primarily social media or Fox News score high on misinformation and low on knowledge.

Sadly, Hackernews readers are not broken out among the sample.

Also viewing habits just emphasize different types of misinformation: https://news.gallup.com/opinion/gallup/354938/adults-estimat...

E.g. democrats in the survey were more misinformed on risks of Covid than republicans. That explains why the misinformed fear is causing Californians to visit outdoor pumpkin patches while wearing cloth masks (this is analogous to wearing helmets while driving, which is statistically safer but not seen very often.)

Indeed, but my point was that the reality is complex. It depends on what news you view; a blanket statement that "the more news you consume the less understanding" is apparently false.

Every general statement is false in some way. The average trends are important. E.g. people that consume more fear-driven media tend to have a lower grasp on perceived risk.

That’s a significantly different statement than the one you started with.

I find it weird to post such piece in PDF, which is meant for fixed format like A4 paper. Web is consumed in different clients - PC, phones, smart TVs. PDF is antithesis of web.

Most PDF-creation programs I've tried are quite inflexible. In particular, I've found few that support hyperlinks. For whatever reason. That's definitely the antithesis of the web.

I think it's to have that "research paper" effect.

The original intention of the web was to make hypertext documents available to people - PDF is just as valid of a format for this purpose as HTML is. Perhaps more inaccessible but certainly not antithetical to the web.

Hard agree.

The single biggest difference is in the text flow algorithm. In PDF it's fixed.

If PDFs feel slow or clunky, that's just a matter of software rendering and payload construction.

The web could benefit from additional text flow algorithms. We leaned into the "application" centric web, but there's so much more we could do for documents and reading.

Super semantic time!

PDF isn't (hyper)text. HTML (HyperTextMarkupLanguage) is. PDF is blob.

A PDF containing text is a text document, just not a plain UTF-8 or ASCII document. If it has hyperlinks, it is hypertext.

A PDF is, at best, an image (of a document).

    application/pdf

There's a reason that isn't

    text/pdf.

I've come to call it the "printable document format." Great for dead trees, sucks for screens.

It saddens me greatly that we live in a time where text documents are the "antithesis of the web"

pdf is still much better than 5 MB webpage full of JS, that is the real antithesis of web.

5 mb of javascript is decried here but in practice it doesn't hinder the reading experience for people the majority of time while PDFs do, even more so when reading on a phone.

As much as I disagree with publishing this as PDF, a page with fat JS which locks up my browser and sometimes my whole machine, as I just experienced earlier today, is much worse.

Anyone developing Web sites should occasionally use a 5-year-old low-spec machine for a day or two.

It does happen but it's comparatively rare to actually be hindered from reading an article because of too much js while almost every PDF is a pain to read on a phone, and to a lesser extent on a monitor. You've probably read thousands of other articles on js-heavy pages without having as much convenience/time wasted as navigating a single PDF would.

I understand how you would assume that, but that is not the case for me at all. I'm hindered by JS-heavy pages on a daily basis, to the point where I rarely click any links on HN anymore.

For example, twitter.com takes 40-50 seconds of lockup to even begin displaying content, after which it is still unusable.

PDF inconvenient, and I didn't read this one, but at least I could close it as soon as I decided not to read.

You are in the minority. For most people twitter does not take anywhere near that long to load and I'd wager there's more people using phones to read and even as their primary/only device than there are with a connection/machine like yours.

I'm all for lighter sites but I'd still bet that an article published as a PDF would cause more inconvenience overall than publishing it on a typical js-heavy site. Further, there are more workarounds to js-heavy sites (blocking some js, mobile versions, rss etc.) than there are to pdfs.

> You are in the minority.

This is true, perhaps.

> I'm all for lighter sites but I'd still bet that an article published as a PDF would cause more inconvenience overall than publishing it on a typical js-heavy site.

This feels to me like you're telling me my experience is not valid. Is it because I'm in the minority? Or because you have a different experience?

Do you know anyone who uses a 5-year-old device day-to-day, or have you ever tried?

It feels like you are denying that this experience even exists.

I'm not telling you your experience isn't valid. We are comparing two approaches and I'm saying that while both cause problems one affects more people. That doesn't take away the issues with heavy js, simply makes it a lesser evil in this comparison.

>Do you know anyone who uses a 5-year-old device day-to-day, or have you ever tried?

Sure, I occasionally use my old 8 year old laptop or 6 year old phone and with either of those the article that will inconvenience me the most on the front page today is this one.

I feel like we are mostly in agreement. I appreciate you sharing your experience, and keeping your older devices working and useful.

Aside from usability, PDF also has many other drawbacks, such as being binary and not human readable or editable.

Thank you for this chat!

I read PDFs on my phone just fine. Yes I have to turn my phone sideways and zoom content to fit page width but that's okay. Much better than having fat JS and Css and browser rendering engines in place just so the page can be "responsive". I don't understand why we don't feel positively dirty about the responsiveness monster we've created to solve a non existent problem.

At least Brave lives up to its name to rid of Google as the default search engine, unlike Mozilla who 14 years ago [0] wanted to move away from them and here we are they still cannot live without Google's money and support 'privacy'; which to them is a joke.

Feature-wise, everyone knows Firefox is hopeless and far behind and still catching up with the rest of the browsers.

Maybe it's time to move to Brave then.

[0] https://web.archive.org/web/20120105090543/https://www.compu...

Okay, but...Brave's history includes:

* DNS leaks in VPN mode that were reported to them and fixed in nightly builds but not pushed to a public build until the issue attracted attention in social media

* inserting referral codes into numerous URLs (most of them cryptocurrency related). They've shifted to silently inserting headers when you hit partner websites that identify you as a Brave browser user

* removing ads from sites to put its own ads in place of them

* collecting "donations" on content creator's pages, without notice or consent...and pocketing the money if the creator doesn't "claim" the donations

* whitelisting Facebook, Twitter, and other tracking

* telemetry that cannot be disabled, that includes "non-personally-identifying" usage data

* collecting Brave Rewards program crypto-tokens requires providing extensive personal information to their crypto-wallet partners and "may" require fucking biometric verification and authorizing sharing that data with an facial-recognition-based ID verification service.

Just use one of the modified builds of Firefox, like Librefox, if you want greater privacy than what Firefox provides out of the box.

I currently do not use Brave due to other reasons but still want say:

Dns leaks are bad but Brave never was a serious anonymity browser. Just as Firefox never was and Firefox had many real ip behind proxy leaks as well. Firefox also never intendedly prevented dns leaks because it just takes system settings or specific doh settings. Its a dishonest comparison. Still their tor mode is a joke.

The referral tags aren't a big thing. They are fixed strings that are the same for every user. They do not identify anyone other than as Brave user, which is trivial anyway due to user agent and fingerprinting. Do you remember that Firefox back in the day did the same when it was underdog? If it makes them money,thats a very privacy and intrusion inexpensive way of getting that.

Ad replacement is opt-in only and an actually interesting concept to try fix the spying third party to everything issue. If its not on ads are just blocked.

Pocketing money from the Ad-view rewards if not claimed. I can't tell if they are actually pocket or just wait for registration of $creator. How else would they do it? They can't magically find out their bank account.

You say they grab donations? Like they actually replace Donation buttons in sites by their own? I haven't heard of that.I only know that you can pay into a Brave Wallet and distribute it to signed up creators. But you have to explicitly pay into Brave and its likely to have read how it works on the way to that. Have a source?

Whitelisting Facebook/Twitter tracking? They had a setting for that in main settings. Is it gone? Else give a source or decrypted pcap please.

Last time I checked telemetry code (early 2020) there really wasn't anything interesting to correlate in there.

KYC for using Brave Wallet breaks the whole vision of the ad replacement thingy. Ack. I hate it.

Most modified Firefox builds are extra buggy or make an easy to match browser fingerprint change that makes you unique. Icecat is severely broken. Waterfox can't keep up with releases. More... I use my distros build with custom settings and disabled backchannels.

I dont use Brave because their release management is garbage and their team is kinda opaque.

> Pocketing money from the Ad-view rewards if not claimed. I can't tell if they are actually pocket or just wait for registration of $creator. How else would they do it? They can't magically find out their bank account.

They don't. There was some confusion about things initially, but no BAT leaves a user's machine unless the creator the BAT is assigned to signs up.

Which is a way weaker statement than what the parent implied.

I didn't "imply" anything. I outright stated that they were running donation scams. The person who replied to me didn't understand what I was talking about.

https://web.archive.org/web/20181221180137if_/https://twitte...

Brave put a "donate to this person" button on creator content, took donations of cryptocurrency on that person's behalf, and then pocketed the crypto if the creator didn't register for Brave's program.

Because that is not obvious and your post looked like uncharitable interpretation of uncertain things.

Are there others than Tom Scott claiming this?

Are there underhanded motivations propable on Tom Scott's side against BraveBrowser or the other way around?

My motivation? Wanting a browser that is actually pro-user.

If BraveBrowser (wont ever use brave as adjective there) used their platform to do underhanded tactics against people that didn't attack first than this would be interesting and a reason to toss any thought about using it. I don't see a trend there yet. More sources? I crave.

I'd love if Mozilla got their shit together but I am not seeing it coming.

> Are there underhanded motivations propable on Tom Scott's side against BraveBrowser or the other way around?

Before making a claim like this, perhaps you could google Tom Scott, a well-known YouTuber who is frequently linked on HN.

The incident in question is covered here: https://news.ycombinator.com/item?id=18734999

I've been using Firefox since, I don't know, 2006? Maybe 2007? Still use it for everything (note that my web dev ability and interest don't really reach beyond HTML and ASP.Net). My main reasons for using Firefox are it works well and uBlock Origin.

What features do other browsers have that Firefox doesn't? What's so compelling about these features? Does Brave still do that ad replacement thing?

Another Firefox user here (although I'm waiting for a fork of the latest one since we cannot get rid of Mozilla any other way it seems).

I also cannot see what I should get technically that will be worth giving up tree style tabs and also inching closer to full Chrome monoculture.

Stock Chromium (which Brave mostly is, UI wise) is snappier, and has things like tab groups which are amazing. I don't remember the last time Firefox introduced a new UI thing that made the browser nicer to use. Containers, probably? The browsing experience on a mostly stock Chromium doesn't differ much from Firefox apart from containers vs. separate profiles, tab groups and sheer speed.

Edge and Vivaldi have a ton of actually novel UI innovations built in. Microsoft's doing a lot of menus as floating panels that can be pinned, which turns the menu into a sidebar. Very normie-friendly but powerful. They also have vertical tabs and can hide the titlebar if they are in use. If they weren't a privacy disaster, they'd probably be just the best browser on the market.

Vivaldi goes more the tweaker way, their tab strip is an inhouse custom implementation and they have settings for just about everything. They have three different tab group setups alone, can slap tabs on any edge of the window you want or do away with them entirely, etc.

I think mobile is where the bigger differences are: Mobile Chromium browsers have tab groups, and tab groups are just as awesome on mobile. They're also clearly snappier than mobile Firefox IME. Stock Chrome falls behind because it doesn't have any adblocking ability on mobile and no extension support. Most forks do, with varying degrees of power.

> Does Brave still do that ad replacement thing?

Replacement is a bit of a wrong word, in that it implies putting their own ads in place of the normal, usually evil ones. Brave has, separately, a traditional ad blocker (built into the browser instead of being an extension, so it doesn't get nerfed by Manifest v3, and can do CNAME uncloaking like uBO on Firefox can) and an ad delivery system that shows ads from a local catalogue as the background in new tab pages and as toaster popups. Both can be toggled off (and the popups are off by default) and their frequency adjusted by the user.

If you turn the ad stuff off, it's close to being a stock Chromium with a bunch of tinfoil and a standalone sync infrastructure.

> What features do other browsers have that Firefox

this is not a meaningful question. I don't know any major browser that still misses features. If people don't use firefox, it likely that they're not happy with factors such as snappiness, its UI, power usage or synchronization of extensions.

I agree with you. GP stated "feature-wise".

For me, brave comes with built in features (ad blocking, ipfs, tor) so I don't have to install 3rd party extensions... In the end I have zero extensions installed.

Also they are investing a lot into security and privacy (check out the Brave Blog for examples), which is a glimmer of hope for me.

What's wrong with Ublock Origin on Chrome? Or is there other reasons (read:Google) you don't use Chrome?

It is indeed that I will not use anything from Google unless I absolutely have to (doesn't happen often but image search on DDG sometimes misses results Google shows).

I'm fortunate in that I have never bought into Gmail, maps or any other product of theirs. I did however sell my London flat without qualms to a Google employee who offered way more money than the asking price.

https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-b...

Brave's operation is all the more impressive when you consider that Mozilla gets $500M every year from Google. I would be surprised if Brave has invested half of that in total over the last several years.

Brave’s total funding has been $42mm. Mozilla burning $500m/yr to accomplish practically nothing is what you get when you push out the technical CEO who is an HN regular, and instead have a lawyer milking it as a resume checkbox.

Firefox is losing market share and has no plan B from Google’s generosity. It’s looking inevitable that Google will turn off funding once FF declines a few more years market share. Then it’s Google, Microsoft, Apple as main browsers. SEC has already given the ok for three major players in a market a la T Mobile / Sprint merger.

You're comparing apples to an apple orchards.

Mozilla produces browser, email client, and calendaring apps (among other things.)

Mozilla's nonprofit arm engages in tech advocacy and betterment work, including sponsoring fellowships (looks to be about thirty this year) and innovation awards. The work they fund covers a mind-bogglingly vast range of topics in the humans-and-tech area: https://foundation.mozilla.org/en/what-we-do/

Brave is a browser run by a guy who has done just about every slimy thing he can think of to make money, including donation scamming and affiliate-code insertion.

Never heard of any substantial thing there. Most "reports" state obvious things. Never seen anything but vaporware beyond Firefox. Thunderbird does their own thing now. Seamonkey is dead.

Both of your accusations of "slimy things done to make money" fail to pass Hanlon's razor. They can be easily attributed to incompetence and even if malicious they would not be that profitable to begin with.

Re: Mozilla's "work". What is the point of being so protective of the "apple orchard" and having justification for all the money and resources that it spends, if the results are completely worthless?

What is the point of spending HALF A BILLION DOLLARS PER YEAR, and get a web that is increasingly dominated by Big Tech and surveillance capitalism? Who cares about "innovation awards" if all the products they have still depend on the company that exploits consumers and consumer data?

I don't want to look at Mozilla and think "oh, but look at all the nice things they are doing". "Niceness" is overrated. We need something that can actually solve the problem.

Mozilla is not doing ANY of that, and its CEO is still taking $3M/year home. That is not just slimy, it is diabolical.

Mozilla is more of a progressive activist organization and fat paycheck for Mitchell Baker than it is a software development organization these days. Do not expect Mozilla to prioritize maintaining Firefox or helping its user base get off Google.

An activist organization that is so intransparent that I don't even get what they are activists for.

I have only ever recognized blog posts about interpersonal matters by them which are rather cringy and don't add anything to the preexisting matter. All about first world problems.

Do you happen to know what concrete and half effective activism they actually do?

I don't know if any of it is "half-effective" but it's the standard social-justicey things like "women and BIPOC in tech", as well as more recently, alerting people to the dangers of the open internet and pressuring platforms to take more aggressive censorship actions: https://blog.mozilla.org/en/mozilla/we-need-more-than-deplat...

I think Mitchell Baker would rather chair something like the SPLC than an organization merely devoted to the development and maintenance of open source software, and that's what she's turning Mozilla into.

I think they were one of the main drivers behind let's encrypt and maybe some other stuff so all is not wasted.

My main criticism is that they "milk the cow dry" by extracting so much money from the browser (which is their single largest source of income) that it starves.

I wonder how conscious the choice of “goggle” was because of its visual similarity to google. They could have called these “lens” or “vistas” or “scenes” or “filters” or so many other terms, but they chose the one that was ironically most like google (one of the main sources of problems as addressed in the paper).

Do the authors wistfully hope for a future in 20 years where we refer to “goggling” for things and “goggle it”? And “remember when” we used to google things? So MySpace.

The title is unfortunate, but I found this from a comment Brendan Eich made in another thread.

The TL;DR is that it's explicit filtering and ranking that you can apply to your search results as subscription feeds.

Like, say, never show Pinterest.

< Like, say, never show Pinterest.

Oh dear lord, I live in hope. Pinterest results have completely ruined image search.

Add: '-site:pinterest.com' to your Google image search to avoid pinterest results.

> Add: '-site:pinterest.com' to your Google image search to avoid pinterest results.

Even though I usually remember to include this manual filtering for images in my searches, it's cumbersome and inconvenient to need to type it, and I definitely don't want to have an extension that handles it for me. And even worse, is the enormous and growing proliferation of sites that just scrape StackOverflow and spam the top 10 places of code-queries with their robot-written word-salad of stolen results.

I've even thought about writing a terminal script that includes all of these sites, and which I could use as an intermediary step for opening searches in my browser. But honestly - a simple exclusion list should be a part of the built-in search.

edit: btw the exclusion needs to be '-site:pinterest.*' to totally exclude this pestilential image-stealing 'service' from results

Good point. I do indeed do this in my personal searches. But much of my searching is in front of students. As an art/design teacher, image search has become an invaluable part of my teaching. Much of this is ‘performative’… searching whilst teaching. Having to include Booleans adds an extra layer of complexity to what should be a simple operation. I don’t want to lecture on Google-foo when I had planned on lecturing on Post-Impressionism.

Would super-love the ability to ‘own’ my search criteria. Ideally, via defined profiles.

Due to ideas like this, I learned that Google limits you to "32 terms" in your query. Bing is a little more generous, but also breaks horribly with queries that are long for various reasons (like site restrictions or exclusions).

https://github.com/pistom/hohser

This looks interesting. Will check it out.

Title mocks WaPo, don't take it too seriously.

I agree with the mocking and immediately understood the reference. I was saying it's unfortunate because as a headline on HN, it lacks what I would call critical info to get people interested in clicking and reading.

I think it could get real traction on HN if there were a blog post on Brave summarizing the paper in a few paragraphs with a title something closer to:

Goggles: User Generated Search Rankings and Filters that you can subscribe to.

Why does it mock WaPo?

“Democracy dies in darkness” is their main slogan.

Mostly to drive subscriptions.

no.

No isn't good enough here. Their whole 'Democracy dies in darkness' thing happened in sync with a huge push to get people to subscribe. Now that may be coincidence but I highly doubt it was.

That doesn’t explain how this is mocking them by just reusing it.

That's a good thing in my book.

what degree of seriousness would be excessive for "democracy dies in darkness"?

also kinda lists that search only these sites eg if you programming you probably want only search stackoverflow, github and official documentation not random content farms.

Eich is rather conservative, politically.

This feels less "hide pinterest" and more "further political divisiveness. by letting people hide news sources that don't match their worldview."

Brave is starting to set off crypto-fascist caution lights for me.

Oh no, people get to explicitly define their list instead of some overseers tweaking the algorithm to show them what they need to see.

At least one is explicit

By that matter.

@BraveBrowser people

Add some things as user controlled public key pinning or even web resource version pinning in an advanced menu.

or honestly @Any Browser Dev people

Resource pinning isn't gonna work unless the app isn't actively being developed. And if it isn't, pinning is redundant.

Public key pinning would disregard the PKI RFCs, so not a good idea either.

If people want to whitelist only some certs they already can by providing the browser only with the specific one's they want to trust.

That would be disingenious. Ever had a thing that wanted to encrypt text in a browser? Those things could finally make sense. Or that one offline wallet generator. It would be sweet if software become actually finished.

A brave claim from someone who adopted Google's browser engine, which goes hand in hand with their monopoly they complain about.

You mean the one Google forked from Apple's WebKit, forked from KHTML? Stop the ritual impurity / genetic fallacy. The issue is not code provenance, it's web compatibility and the incremental fork costs.

Hoist Google on their own petard. It'd serve them right if the open source they contributed to helped speed a correction of the state of the web.

The issue is to not create a tower of Babel that can only be rendered by one specific piece of convoluted software.

I can see that being an issue if the web is centralising around one proprietary engine, but why does that apply to Blink? If Google turns completely evil that won't take away from all the webcompat, optimisations, and security in Blink, and any bad feature (ManifestV3) can be patched out easily. Mozilla could switch to Blink (and I think they should, and Apple too, they clearly don't have the manpower to maintain a whole engine and keep it modern and secure, see my other comment in this thread; there are many more 0days with Safari and Firefox than Chrome) and it wouldn't harm the web in any way, because they can just swap out bad components with their own, and still reduce their dev time requirements by orders of magnitude compared to today. You have Intel, Samsung, tons of companies contributing to Blink, to a much, much greater extent than Gecko or WebKit.

If every browser switched to Blink, that would actually benefit consumers, because right now Chrome basically has the enterprise market on lock, since nobody cares to test internal webpages on other browsers, and Chrome has better enterprise features (though I don't know how much of that is to do with the engine); and many consumers also just switch to Chrome as the first thing they do, because they've run into compatibility issues with Firefox or Safari and they just want to access the web without issues. If every browser used Blink, you'd have a lot more people using Safari and Firefox, just because most people don't want to keep two browsers around (Firefox + Chrome or Safari + Chrome) when websites break (which is not often, but often enough, trust me). I really hope we see this.

Serious question:

What specifically made you switch from gecko to chromium in the Brave browser? I know compatibly with modern web apps is frequently cited as the reason but using FF as my daily driver I never run into any issues.

Was it easier to develop BAT integration and the custom ad/tracking blockers?

I'm glad they switched to Blink, and I hope TorBrowser switches too. Blink is more secure and modern than Gecko. Brave still wouldn't have site isolation, an important security feature, if they stuck with Gecko. Whether due to bad priorities or lack of funding, Firefox is just really falling behind, but in many cases, in ways that aren't easily apparent. Sure, websites load, but Firefox didn't have sandboxing until (2017? I think), which Chrome had since 2008. It's not really defensible for a web browser with such a big attack surface. And Firefox today is still pretty far behind in security, to the point where even if I valued Mozilla and thought Firefox was the best browser ever, I still wouldn't use it and wouldn't recommend anyone else use it.

Bashing Gecko security in such ways is dishonest. Did you ever actually look at their architecture or do you just parrot a lazy blog post?

The real problem here is browser ever increasing scope. I never cared for VR or gaming in browsers. SPAs are user unfriendly garbage that I seldom see used for anything else than preventing user agency.

I don't think it's dishonest. It's easy to find areas where Chrome is ahead, and I haven't found one area (in security) where Firefox is ahead.

https://madaidans-insecurities.github.io/firefox-chromium.ht...

If you ask most browser exploit researchers, this would probably be pretty much a consensus view. Very few of them that I know use Firefox.

> The real problem here is browser ever increasing scope. I never cared for VR or gaming in browsers. SPAs are user unfriendly garbage that I seldom see used for anything else than preventing user agency.

I don't care what Mozilla or the Safari team think I should be allowed to do in a web browser, I want to have the option of doing practically anything. I'm glad I can play Cyberpunk 2077 in a browser, because otherwise I wouldn't be able to, period (I don't have Windows). People complain that the web's prominence means every app uses Electron nowadays, but I'm glad I can run anything on Linux or Mac, and the "development costs" argument that kept apps Windows-only no longer applies. The web saved the desktop, it didn't ruin it. I view Google's Chrome web app push in the same vein as Steam's Proton. Decreased reliance on just 2 big, somewhat-dysfunctional vendors (Apple & Microsoft), to not be locked into a declining ecosystem.

There were a ton of issues using Gecko, starting with (at the time) no CDM (HTML5 DRM module) so no HD video content from the major studios, Netflix, Amazon, etc. -- Firefox had an Adobe deal but it was not transferable or transferred to any other browser that used Gecko -- and running the gamut of paper-cuts to major web incompatibilities especially on mobile, vs. WebKit-lineage engines such as Chromium/Blink.

> who adopted Google's browser engine

I'd love to see a "from scratch" Chrome competitor, but man, that would be expensive to put together.

Relevant tangent, as Goggles initiates a great conversation and it's a great opportunity for us to expose our (related) vision to the HN community for the first time. I hope I'm not too OT and that you don't mind.

At Monitoro[0], part of our long term vision is to transform the web into a push model instead of pull as it is currently.

Right now, you go to a search engine and ask it what you want (based on which it builds a profile, trying to guess what you're interested in, hit or miss). This is the pull part.

Our vision is to turn this upside down:

- Specify what you're interested in at a super fine grained level. Can be a color or a price range if it's a product (I want to know about headphones between 70 and 150 in navy blue, that support Bluetooth 5.1). Could be a location (I'm interested about gigs in any of these 5 cities, bands from this list or this set of genres ...). A keyword maybe (news items about company X mentioning Y in region Z, or new law covering chemical X ...) Any information attribute that exists can be used to filter your interests.

- Receive matching information as it happens (in real time), or at your own agenda (in a digest for example), or any mix and match between the two.

- Most importantly, it would be data first. Contrary to the HTML/JS/CSS heavy internet we live in, only structured data would be pushed to you, and the presentation would be adaptable. Good defaults, ecosystem of presentation layers, or make your own are all possibilities.

Why? You might ask.

- We live in a JIT world. Everything is changing quickly, the impact of these changes is increasing, and chaotic systems/snow balls became more common recently. Our existing information infrastructure is not optimized for that. You cannot expect people to guess when to look something up and even then it's horribly inefficient.

- You explicitly say what you're interested in. No guess work (profiling) is necessary which leads to a better model for privacy on the internet, and the possibility to adopt profiles (groups of interests) that are separate from your identity and from each other as well.

- Tying data to presentation (websites) is what led to this situation in the first place, where websites are silos that you cannot easily interact programmatically with. Many websites have no issue policy-wise with this, yet the very nature of the technology makes it hardly tractable.

- Another pattern is the explosion of endpoints and devices, and most importantly contexts. Voice, mobile, wearables (so many subcategories in it), digital signage, but also video streams, radio channels ... Each of these media not only has different presentation requirements, but separate interaction contexts. How do you expect content publishers and information providers to adapt to this?

It's a classical MxN problem which can be partially addressed by reducing information from websites to data, and then providing different presentation layers adapted for the medium and the context.

You might want to ask Alexa is a new shoe matches your interests, while you want the newest NFT collections to arrive as a daily digest. You also want to archive news about your school's alumni in a spreadsheet somewhere to check whenever you want to.

tl;dr: Instant information that is relevant, in the medium and context where it is relevant, no tracking involved. (Efficient markets hypothesis here we come!)

Thank you for reading and again, sorry for the tangential hijack. I will expand on our vision in a separate place if there is interest.

PS: what about subscriptions, follows, newsletters? They're great validations of this model! But they're too fragmented ultimately and suffer from the information silo issue. The new model would comprise these as use cases and liberate this information as well.

[0]: https://monitoro.co

Reply on news.ycombinator.com