A man can dream.
Apple sells huge amounts of ads, I believe it's the fastest growing part of the business right now. Estimates are $5 billion in advertising revenue in 2021, with one projection of $20 billion annually within 3 years. In addition to this, Google pays them $15 billion to be the default search engine and sell ads.
https://www.ft.com/content/074b881f-a931-4986-888e-2ac53e286...
That was just the best answer he had to questions about third-party apps until they had the infra in place for it.
Apple will never tell you ahead of time about changes like that. So, while I wouldn't exactly call it lying, you'll get answers like the above.
Most of the money Apple gets from the App Store is from pay to win games - north of 80%. It came out in the Epic Trial.
Apple doesn’t really care if your banking app is a website or an app. They don’t make money either way.
Most apps on the App Store that could be a web app don’t charge users.
Not only that, the data for those apps are on a remote server that is accessible via any other platform. Even when an app uses the standard file picker, you can choose any installed cloud storage device to save and load files - ie iCloud, Dropbox, Google Drive etc.
But the money they make on ads in the App Store has to pale in comparison to how much they get from Google
Now there must be more but $99 * 20 million is not nothing in annual revenue which is mostly profit.
If you have a company with 100 iOS developers, you only need to pay for one account that has permission to submit an app.
If you are exclusively a Mac developer, you don’t have to pay anything unless you are submitting to the Mac App Store.
A “registered developer” is not necessarily a “paying developer”, you can register as a developer to get documentation etc.
If you are working for a company creating an app, not all developers need to pay to publish an app. Also Mac developers don’t pay to distribute an app.
What the heck is a nothingburger?
It's not a paid app, just part of the walled garden experience.
Based upon their actions I don't believe that Apple cares about the web beyond the bare minimum they have to to provide a tolerable browser experience for their users until everything happens via Apps.
That's even more true for publicly traded companies
They're being paid handsomely, by Google:
> In 2020, The New York Times reported that Apple receives an estimated $8-12 billion per year in exchange for making Google the default search on its devices. According to one analyst, Google's payment to Apple in 2021 to maintain this status quo may have reached up to $15 billion.
1. https://www.macrumors.com/2022/01/05/google-pays-apple-stay-...
Apple sells ads on search indirectly via Google. (Google pays Apple to be the default search engine.)
Google’s payment to Apple is bigger than the revenue Apple makes from the Apple Watch.
There is sadly not a version or equivalent on macOS, but Christian has confirmed it is in-development.
[1] https://apps.apple.com/us/app/amplosion-redirect-amp-links/i...
P.S. Pairs well with PiPifier to bypass YouTube not allowing picture-in-picture or playing in the background without a subscription, and a good ad blocker.
I never understood why AMP gets so much hatred from end-users. I understand why publisher hate it (it takes away control), and I understand the monopoly concerns, but for me, as an end-user, 99% of the times, the AMP version is better: faster, with less of the annoyances you described. As for privacy, these 99% are loaded with Google ads and analytics anyways, so not much of a win there. I don't know what your configuration is, what kind of ad-blockers you are using, but I never met the horrors people make AMP to be.
And sure enough, there are sites that are better than any AMP sites, but these almost never have an AMP version. So for me, AMP makes terrible sites a little less terrible, and good sites unchanged.
But AMP pages are often worse - things like getting cut-down pages with the normal navigation missing or stuff like that. And it breaks URL sharing - if I want to share a link, I just want the regular link, not one with AMP crap in it, but often cutting out the AMP part to try and reconstruct the original URL just doesn't work...
So I've never found it better but only worse, so understandably it just annoys me. If it is just that my tracking and ad blocking is making all the difference, then AMP clearly isn't actually solving the actual issue...
If it did not do that I could care less.
And here lies the problem. Eventually, these two 'annoyances' will become much bigger than that. The biggest threat, imo, is that these will lead to news organization, journalists, and credible publishers to eventually close up shop as they receive less of the share of revenues.
I do agree with your point about AMP sites generally being a much smoother experience. But at what cost?
- The affiliate link "hijacking" was - if I remember correctly - to sites of crypto companies that partner with them. I'd prefer if this didn't happen, but most seem to be fine when other browsers (Safari, Firefox, etc) add something like "?client=safari" when searching or when their search engine (eg: DDG) use affiliate links to sites like Amazon or Ebay. It's not a new thing.
- The money collection (brave rewards)... if one doesn't understand how the system works, it looks like they are stealing money... but the money is returned to the sender after a while if the website/creator doesn't claim it. Is this that bad?
And then there's them not blocking some trackers (Google, Facebook, etc) by default, but if they did, they would break logins on many websites.
Maybe all this is bad, but I'm not sure if there's any browser out there without a history of shady behaviour. Even Mozilla has messed up a few times.
I don't think anyone ever accused them of stealing that money, but yes, hijacking people's personal brands to collect money without their explicit knowledge is a bad thing.
Imagine if I saw the icon and gave "them" money via Brave instead of joining their Patreon or some other official channel that they explicitly set up. If they don't collect, then yes, I might get my money back - nothing was "stolen". But that was money I wanted to send to the content creator in that particular moment, and that creator will probably never see it. The creator got screwed out of money that otherwise would have gone to them.
Imagine I presented you with $10 (from my own pocket), and asked you where it should be spent. You told me "Doc's Pub, over on 9th." So I walked over to Doc's Pub, but found them to be closed. So I waited outside for a few hours, just incase they opened up. I later went home and wrote down "try to spend $10 at Doc's tomorrow."
Brave staked users with BAT (from our token sale). Users could direct that BAT to the sites/properties of their choosing. The BAT then went into an omnibus settlement wallet (note: the BAT originated in one Brave wallet, and was sent to another Brave wallet).
There was no hijacking of brands, or anything of that nature. I would encourage you again to please visit the aforementioned link. In it I mention our blog post on the topic, which includes screenshots and more. I hope this helps!
Brave gave users BAT to tip content creators. They tipped it, if it wasn't claimed in 90d, Brave returned the BAT to the pool. There was no collecting money. It was Brave's promotional BAT and it never actually left Brave's possession unless claimed.
The issue was that it wasn't clear if the creator had or hadn't signed up. Which was fixed within 48h.
At least for websites I have to manually click the Brave Rewards icon (wasn't prompted to do it on a new profile) and it shows if the site is verified or not:
- My personal website (verified): https://i.imgur.com/WZykI2U.png
- Google[.]com (unverified): https://i.imgur.com/89XvzIz.png
And if we hover over the "unverified creator" text, this is displayed: https://i.imgur.com/IfKQUME.png
I guess the right way to do this is to only allow tips/donations for websites already verified... still, if you're going to use Brave Rewards, you probably have an idea of how it works.
Maybe things are different for creators on platforms like YouTube? I don't know how it works. I couldn't find a way to make a direct contribution with Brave Rewards.
Then what should they do? It sounds like they're between a rock and a hard place
This is only a "problem" for some because they think Brave is actually siding with those services, when in reality there's a good reason to not block them by default.
Grassroots users should support it like the godsend it is.
Some of the ire is earned but I have yet to see how Mozilla is supposed to fund Firefox development without that google search bar.
Developing a browser isn't easy and requires a few teams of developers, but in 2020 it spent 242 million dollars in software development costs, 137 million dollars in administrative costs and 37 million dollars in marketing and branding costs. I don't live in a lala land where I think you can develop a browser for free, but I think we can all agree that you don't need to spend this amount of money on it either. Are 100 developers enough? 200? How much does that cost? Do they all need San Francisco salaries to develop a good browser?
In terms of funding, they got 440 million dollars from royalties (what they get from setting default search engines on their browser) and 25 million dollars in subscription revenue (Pocket and VPN subscriptions type things - products they actually sell).
Now, can you develop a browser on 25 million dollars per year? Maybe it's cutting it short, but for sure there could be a strategy to invest more on this side of the equation to phase out the need to be Google's bitch in a more intentional way.
Source: https://assets.mozilla.net/annualreport/2020/mozilla-fdn-202...
100-200 developers probably roughly costs on the order of 25-50 million a year (assuming a made up number of 250k fully loaded cost with benefits and taxes, which might be lower or higher than their average, idk. This number seems almost too conservative to my gut). From this back of the envelope math, I don't think the royalties business is enough to support the development cost alone.
> Do they all need San Francisco salaries to develop a good browser?
If you don't want them to leave to work on safari or chrome, probably? This experiment was more or less tried by opera, right?
Brave gets away with a lot lower overhead by bascially piggybacking off chrome. Opera similarly gave up selling their own browser engine and cut their development teams while switching to another wrapper on blink. Firefox could become yet another skinning of blink and chrome code, but it's not clear to me how that's helping them with being "Google's bitch in a more intentional way".
I think there are problems with Mozilla's side project expenditures, there is definitely some bloat, and they have had some expensive failures like Firefox phone. However I don't think Mozilla could have survived without Google's funding and people vastly underestimate how expensive quality software is.
Chromium get contributions from a few big companies, why couldn't Mozilla create a "contributor community group" something like Java has? Oracle keeps control over Java but there is a democratic process on what gets into Java/JDK. At some point Samsung was contributing to servo when it still was experimental.
Mozilla regularly accepts code from unpaid community members and has since its inception.
Yes, so it Java and Google and Microsoft... you're still missing the point I made there. Make Servo a cross-company product serving more than just Firefox. More companies contribute to it, lower costs of development.
I don't doubt there are developers being paid peanuts with more talent than I will ever see, but I haven't seen them and I know a lot of people spending a lot of money looking, which suggests to me they are rare and hard to build a team out of.
And if they found them, and they fired all the senior bay area folks and rehired these 100k developers tomorrow, are those developers experienced writing fault tolerant c++ in a multi-decade codebase? And if so, how many will relocate or take remote jobs for way higher salaries after a couple years? Google, Amazon and Microsoft would be happy to have them and pay 300k+ TC for those skills, as well as countless banks and fintech firms.
Here in Spain (which is a first-world country :) a good dev would make about 50k or so..
I think the Silicon Valley area has really skewed expectations of salary (and of course living expenses!). Familiarity with codebase is a problem indeed, but they could start setting up a hub somewhere in Europe.
I think part of the problem is that when they incorporated Mozilla they started thinking as a corporation. Multi-million-dollar-paid CEOs, top-heavy with marketing etc. They should have stayed nonprofit IMO.
Did you mean subscription business? Like I mentioned, the royalties were 440 million which is way larger than 25-50. If you put 200 developers on this project, with fully loaded costs of 400k / year each, that's 80 million dollars per year. You add 20 million dollars for administrative and other expenses and we come to 100 million dollars per year burn rate.
In 2020 they made ~25M on subscriptions, if 50 of the 200 developers focus on improving the subscription business, at the current rate of growth they had from 2019-2020, in a few years they could totally phase out of needing royalties at all to cover their $100M / year expenses, with 200 developers earning competitive salaries.
There's a lot that can be added here:
Mozilla promised to opensource pocket server and never did.
They promised to hire someone full-time for Thunderbird, but never did. afair there is a German company that has full-time developer working on Thuderbird.
They promised a VPN... yes, delivered something.
They promised anonymous email, I know they were giving access by invites, but nothing more about it.
They promised to unfork Tor browser and integrate Tor into Firefox, they were even running a few Tor nodes.
Remember how hyped everyone here was for Servo in Firefox and electron competitor?
MDN could have integration with GH or GL and educational content for web development, they literally had resources, brand and ability to join an online university and give degrees or at least serious bootcamps. Mozilla was a meaningful brand to do it.
They had a lot of opportunities to sell privacy, we literally demanded it from them, but they weren't interested in listening. Instead they delivered 6 rebrands each breaking my muscle memory.
Do you remember how they advised EU to regulate monopoly on the webbrowser market? There was time when they had all ability, but 0 will to keep it this way https://en.wikipedia.org/wiki/BrowserChoice.eu
If we’re talking about the Mozilla foundation, they should seek donations and grants and focus on being the best user web tooling.
If we’re talking about the corp, they could’ve kept rust under the umbrella and pioneered the WASI runtime and built an alternative to k8s that runs webassemblies and built out a paid cloud infra.
It doesn’t make sense to have a foundation that is user aligned and a corp that is user hostile. There should be aligned incentives.
1. Stick to building a more private browser
2. Become a non-profit or Benefit Corp
3. Fund #1 through all of us donating and encouraging our less tech savvy connections to use it and donate
Yes, that is part of the reason why they should be a non-profit - sot that they can work for the public good and not for the good of their corporate sponsor.
> 3. Currently, the donation figure is about $20mil.
Donation figure for what? Currently you CANNOT donate towards Firefox development.
To Mozilla foundation. Most people I heard donating to them don't realize their money aren't and couldn't be put into Firefox development.
And Brave has things that not everyone likes like the bat token stuff. And chromium.
They could also try to seek government funding. EU and other non-US countries should have an interest in there being a browser that is not beholden to a US for-profit corporation.
And something like apple's iCloud relay. I would pay for that.
I won't pay for a Mullvad subscription that's less useful than actual Mullvad because it just works in the browser and it doesn't fix all the issues iCloud relay solves.
If they'd sell actual privacy features I'd pay for it totally. Sync too.
Your raising a discussion about mozilla in a brave thread plays right into mozilla's purpose of muddying the waters for actual competition against google. Please don't derail brave discussions with off topic stuff.
Sure it's probably not going to make it but they're still not things they can sensibly afford to do.
Firefox literally sends your keystrokes to Google, right out of the box. Brave, however, was found to be the most private popular browser by reputable researchers: https://www.scss.tcd.ie/Doug.Leith/pubs/browser_privacy.pdf.
Brave [never] hijacked links either. Affiliate Links were offered among suggested sites for relevant search input. So if you searched "Binance," the browser would offer (among other suggestions), an affiliate link for the site. Users could then choose to browse to the property with the affiliate link, and in so doing support the development of Brave. No impact to privacy or security at all.
The mistake here was with input handling. Built to handle search input, this feature also mistakenly handled fully-qualified domains. While we intended the app to offer affiliate links (when relevant) to something like "what is binance?", it was also offering them for "binance.us". The latter case was corrected quickly (and the feature itself was disabled out of the box).
More about that on our blog: https://brave.com/referral-codes-in-suggested-sites/.
To your second point, about setting up crypto wallets and soliciting donations on behalf of non-participating publishers, you're mistaken there as well.
To prime the support-system in Brave (called Brave Payments at the time), we staked Brave users with tokens, inviting them to direct those tokens to creators they would like to support. More clearly, Brave gave Brave users say over where Brave ought to direct its own tokens.
Unfortunately, our UI/UX wasn't very clear about which creators were verified, and which were not (we followed the Twitter approach, marking verified creators with a checkmark, but doing nothing for others). This resulted in some confusion at the end of 2018, where users were directing Brave's tokens to non-participating creators (most notably Tom Scott).
We received considerably helpful feedback about how the system could be improved (both from a UI/UX side, and operationally). Frankly, I don't think I've ever seen our team work so hard, and churn out such a monumental update in so little time. We had made massive changes within 48 hours IIRC. Creators were explicitly marked as verified or unverified in all cases, the BAT that Brave stakes with users would remain in the local wallet until it could be received by a verified creator. And BAT that sat pending for 90 days would be unlocked again for the user to direct elsewhere.
Tom Scott was kind enough to review our changes, and explicitly gave us his approval soon-thereafter. What is now 'Brave Rewards' wouldn't be doing so well today were it not for Tom and so many other incredible users helping us find the best path forward.
More about that on our blog: https://brave.com/rewards-update/
This is a touch of FUD. Firefox sends what you type into the search/address bar to google for search suggestions by default, yes. Your perhaps unintended implication here is that all keystrokes are being logged and sent to google.
> Brave, however, was found to be the most private popular browser by reputable researchers: https://www.scss.tcd.ie/Doug.Leith/pubs/browser_privacy.pdf.
This is a bit of a misstatement of the research; Rather they found that Brave doesn't tag users' installs with persistent IDs by default in startup telemetry or update requests, while most of the rest of the competition do. This means Brave has a superior initial privacy state but says nothing about the overall usage posture. Is it better? Possibly but that's not the conclusion the researchers actually came to.
Here's an example of why the first line of FUD is dangerous: "Brave's BAT is basically just a way for users to be tracked as they go about navigating the web."
This is an easy throw-away statement with little substance and likely little to no truth behind it.
A more concrete example might be that the so-called "Privacy Preserving Product Analytics API" that collects data on Brave users' installs gets enough data to pretty easily fingerprint a user if you wanted to. You don't need many data points to identify an individual from a large group.
It doesn't look like this is the case, fwiw. (Unless your HN profile lists an outdated handle?)
Keystrokes in the URL bar.
It would have been better to say "by default Firefox literally sends every keystroke you type in the URL bar to Google".
In my opinion it is a user-hostile "feature", and should be pointed out, but not in way that could be so wildly misinterpreted.
I agree. I don't like a lot of what Mozilla does but I don't like Brave at all, so I'll gladly defend Mozilla against hyperbole coming from Brave. Brave isn't even a browser, so I just ignore it most of the time.
Describing the cost without explaining the why is really putting a spin on things.
https://www.theverge.com/2017/12/16/16784628/mozilla-mr-robo...
Multi-Account Containers
Temporary Containers
Sidebery
It is a cliche but you are not Google’s customer. You are their product. If you don’t find AMP compelling they don’t want to serve you ads anyway.
So yes, it is a loss for Google.
Not all products are profitable.
Like that time Brave replaced links to a site with their own affiliate links? Yeah, I no longer trust them.
The issue was about address bar input autocomplete for two domains, binance.us and binance.com, along with keywords which all browsers offer several possible autocompletions for, we autocompleted by default not just via dropdown suggestion, with referral code attribute identifying us (not the user) to Binance at end of domain name. We fixed this right away and made nothing off of it. But it was a blunder for sure.
Mozillas effectiveness at diverting/absorbing/stopping other community initiatives that could actually impact google - worth the dollars from google's point of view.
DDG sounds like a promising alternative.
Regardless, it's already in their tool box because they make the browser.
They could be re-writing urls and not showing you, if they wanted to. And as long as no one notices, you can keep doing it.
Unfortunately, switching to a different search engine does not prevent you from ever stepping onto AMP again. For example all links in the mobile version of Twitter used to be AMP until recently: https://www.theverge.com/2021/11/19/22791002/twitter-amp-ios...
I only recently became aware that the app existed, so I don't know how/if it deals with all things AMP-based.
I rarely end up in AMP pages on my mobile, but when it happens I immediately feel like I stepped on a turd, and promptly backtrack / close the tab before it hijacks my back button, half the screen, standard controls (including doing something weird to scrolling) and other unpleasantries like banners whose "x" somehow overlaps my browser's bars, and are therefore out of reach (and said browser bars somehow do NOT autohide when scrolling, unlike on normal pages)
Getting AMP results from Google search has been one of the drivers leading me to switch to DDG, so congrats Google, one less customer.