I'm amazed that someone who has been this badly burned by a corporation controlling their API access would even think about writing another app that uses third-party APIs, to be honest.

If you had read the article you would find that this app doesn't really need YouTube API access. It's just an iframe. It's 1998 tech.

In that case, isn’t it a race to the bottom with just someone else doing something similar? What makes this special?

Good product is first about understanding the user and the problem statement very well foremost. Most product moats are just that, everything else is a function of that.

Designing a great UX to interact with the system is the other key ingredient, that requires step 1 and also a great deal of creativity.

Anyone can copy same the features after someone as good as Christian Selig has made an app, Few can do similar or better starting on their own, especially indie developers, so he can always be ahead if he wants to.

Christian also chooses apps to work which are third party platform controlled for a reason I think. He can operate in markets like this as a extremely talented indie developer that very few competent teams with capital funding would attempt with platform risk. Beeper is the most recent example on Apple, Christian himself got burned in Reddit[1][2].

Finally he prices at a point so low that people are just paying for the brand - for a well designed reliable software which won't crash on them.

He likely will not lose all that much sales if a lower priced/free product comes out Safari browser based Youtube.com is already there .

---

[1] He can afford to in the sense his monthly cash burn is very low compared to any normal company and he doesn't have 100's of employees to worry about if he gets kicked out.

[2] Even then he has carefully choose an API that Google will have a hard time just blocking him ( and not every other use of embedded playback), and he also is careful not to use APIs to render the UI he has just skinned the main website with light CSS.

Technology doesn't need to be "special" to be useful.

Nothing makes this special, maybe a well-known indie developer.

The fact that he wrote the apollo app? I don’t know. It’s not exactly revolutionary is it

It's the first one so it's currently the best one. If someone else makes a worse app, why would you use it? I don't get what's confusing here.

"There are three ways to make a living in this business: be first, be smarter, or cheat."

If you have a link to the other similar ones that would be useful.

I thought the article did a good job addressing it. There were a lot of nice touches which make this app work well with the vision pro.

The iframe embed API is API access, and YouTube can remove, paywall, or rate limit it any time they want. How old the underlying technology is is completely irrelevant to that.

Unlike Reddit, Apollo was stopping users from seeing ads and Reddit gaining ad revenue from them, so they went to charge the Reddit app devs for this loss in revenue.

The YouTube embed API supports ads, and works perfectly with Premium so Google are not losing any potential revenue with this app existing.

Sounds like Christian learned his lesson with his experience with Reddit: "don't get in the way of the company's ad revenue".

Your statement still stands though, you are ultimately correct.

That's an interesting way to frame the reddit debacle. Reddit could have mandated ads to be displayed as a part of their TOS of API usage, but just decided not to - for the clear reason of centralizing users to their app. It wasn't /just/ about the loss of revenue - it was also about the metrics they can collect on their platform which they could not do on others'.

YouTube disabling embedding would be fucking insane.

I'm 100% expecting YouTube to cease&desist him

> https://support.google.com/youtube/answer/171780?hl=en

The YouTube API Terms of Service and Developer Policies apply to all access and use of the YouTube embedded player.

> https://developers.google.com/youtube/terms/developer-polici...

You and your API Clients must not, and must not encourage, enable, or require others to:

use YouTube API Services to create, offer, or act as a substitute for, or substantially similar service to, any YouTube Applications. API Clients must not mimic or replicate YouTube's core user experiences by recreating features or process flows unless they add significant independent value or functionality that improves users' interactions with YouTube. For example, an API Client must not recreate the browse experience from any YouTube Application without adding significant independent value to that flow.

I think it's a reasonable argument that because YouTube have deigned to not support this device with a first party app that they are indeed adding "significant independent value or functionality that improves users' interactions with YouTube" with this program?

Specifically this appears to offer a better experience than what YouTube choose to offer.

He's not selling it as a subscription so there's very little downside. Some people will send him $5 and if/when YouTube cuts it off (can they really?) he doesn't really owe anyone anything.

>if/when YouTube cuts it off (can they really?)

It's called "Juno for YouTube" so they could definitely send a cease and desist for the YouTube trademark.

It uses the magic word: “for”

They’ll probably still send a C&D but it will be defensible.

But.. aren't most of the c&d defensible? It's like a mafia scare tactic to dominate something - sometimes just the "or else" destroys ppl lives

It seems that he found a (momentary?) gap in the Vision Pro app market and promptly seized the opportunity.

Absolutely. He's an indie product hacker who we could all learn from.

"oh, he's entirely reliant on the platforms!"

Meanwhile, his cap table is himself, he does quite well, and doesn't owe anything to anybody.

https://www.cbc.ca/news/canada/nova-scotia/largest-spca-fund...

Nah this approach is solid. It’s web views not api. It’s basically a web browser. I imagine other iOS browsers like brave will come to vision and have YouTube etc video playback and demonstrate it’s tolerated even with Adblock probably. But maybe easier to get by if more clearly a multipurpose web browser.

You don't need third-party APIs to make a YouTube viewer. There's a bunch of 3rd-party YouTube viewers like SmartTube, ReVanced, etc. that bypass ads and don't use the official API, plus of course the yt-dlp downloader.

I'd say the lesson here is NOT to rely on official APIs.

However, upon reading this blog post, it does seem he's using the official API so I guess he thinks he'll be fine as long as he doesn't block ads. Time will tell.

I wanted my bot to "see" a youtube video and sum it up - so me being the naïve 1998 kid me, spent like 2 hours setting up the API access to get the transcript of a video only to realize I can use this API only on videos I uploaded which is a complete bullshit because as soon as that happened I ditched it and wrote a script using puppeteer scraping the the transcript of ANY video, which ironically took less time than setting up the API.

So yeah I learned my lesson I should not resort to piracy, but start with it

> However, upon reading this blog post, it does seem he's using the official API so I guess he thinks he'll be fine as long as he doesn't block ads.

The idea of strapping something to my face that's going to project ads into my eyeballs that I cannot look away from--well, let's say it's pretty clear technology took a wrong term some time ago.

Apple should do the right thing and enforce a strong "no ads" policy for this product. Keep it premium for people who shell out thousands of dollars for it.

You can get rid of the ads on YouTube by paying 0.4% of the Vision Pro base model price per month.

That does nothing for all the "sponsor segments" embedded in the videos.

Try sponsorblock, or stn which includes it.

AFAIK, sponsorblock isn't going to work with a 3rd-party YouTube viewer app like this.

Yeah, I guess STN (a viewer app in which sponsorblock works) is of the variety that TFA says makes Google "grumpy."

You can look away from it. The OS doesn’t give apps access to the eye tracking info.

A lot of the internet would break if YouTube removed/tweaked their embedded video player so I doubt he has to worry.

Keep in mind that this didn't prevent Facebook and Twitter doing the same, Google is just behind with these patterns, just like with everything else

Seems like a lot of the motivation was just recouping lost revenue from his existing YouTube player integration code.

I have a feeling is going to be a speedrun of the reddit saga. Google obviously doesn't want a smooth youtube experience on vision pro.

Why is that? Obviously they don’t want a third party app to be that experience, but they make native apps for everything else.

Also, Christian can’t help himself but attach his apps to large companies that can cut him off overnight. Haha.

Short answer - usually individual app developers, even of Google’s size, need the platform (iOS) more than the platform needs them. This means Apple has historically driven hard bargains with even the most popular apps. Now Apple is launching a new platform (visionOS) the 3 most popular in their categories - YouTube, Netflix and Spotify decided that visionOS needs them more than they need visionOS. For now.

It’s possible they might use this leverage to negotiate better terms on iOS. For example, Netflix would like to offer in app subscriptions and to keep more the revenue without sharing with Apple.

If Apple sells millions of visionOS devices then that gives Apple more leverage and these 3 might come crawling back.

Long answer - The Apple Vision Pro’s Missing Apps by Stratechery (https://stratechery.com/2024/the-apple-vision-pros-missing-a...)

That's just apple being apple though, shipping half baked stuff 3 years before they're ready

> For example, Netflix would like to offer in app subscriptions and to keep more the revenue without sharing with Apple.

Netflix already has that option[0] at a 15% commission rate, but they snubbed their nose. Allegedly because they didn’t want to play nice with the TV app, like other streamers do.

Personally I think it’s that (and the potential loss of data) + them just wanting to pay $0.

0: https://developer.apple.com/programs/video-partner/

Tons of respect for Ben Thompson, but reason for not shipping visionOS apps (or allowing your iPad app to work on them) for these big co's is literally a matter of "Bang for your buck":

> Building a new app from scratch makes zero sense for the size of the install base. Only reason to ever do that would be to get some love back from Apple in the form of features and attention (which, for YouTube / Netflix / Spotify are hardly necessary) > Allowing YOUR iPad app to function on visionOS means that your customers will hold YOU responsible for its functioning. At the size of customer base of these companies, that's a bunch of risk for no reward > When your users use a browser that promises 'regular access to all websites' (built by Apple) to access your service, the responsibility for that experience lies with the browser builder, not you

There's 100% no negotiation over fees happening with individual developers, regardless of how big they are, regardless of what type of support for a platform they promise Apple, as that's exactly what has gotten Apple and Google in hot water with regulators worldwide.

Ben worded it well in the article:

"It’s certainly possible that I’m reading too much into these absences" < Yes

A example would be Windows phone where they just not only did not make a YouTube app they denied access to the YouTube app made by Microsoft.

That app literally broke YouTube's agreements with the music labels though (by allowing free downloads etc), which this one doesn't.

> but they make native apps for everything else.

They do. But it's difficult to call anything they make "smooth." Google does some decent backend stuff but their frontend experience is not.

the android youtube / music app is much better than any other audio / video app i have including spotify, netflix

Come on.

The YouTube Music app blocks you from navigating to a different song at the same time as playing a track if it decides that the track is primarily aimed at under 18s (such as the theme from a retro cartoon).

It's UI might charitably be described as a total catastrophe.

That is a COPPA compliance thing and is true on the regular YouTube app as well:

https://www.androidpolice.com/2020/01/30/why-youtube-videos-...

Well compare that to Spotify which can only show videos full screen in one orientation and needs to be force stopped constantly because podcasts block a lot.

Also, I never encountered the problem you complain about.

Everything Google does in their own ecosystem is smooth.

Not as smooth as Apple.

I don’t know if that’s still true. The Music app is still sluggish, Notes has poor UX, almost every built-in app has a better third-party replacement. It has been downhill since iOS 7.

What the replacements for contacts and clock would be?

Contacts+ and Cardhop are two.

To be fair the clock app is a lot better since they introduced the sleep schedule, but I used to have a separate alarm app, and still use Sleepytime to calculate wake up times.

What's so poor about Apple Notes? It's my most used app. Just curious.

And in all 3 cases, the Google equivalents are worse. History shows they’ll probably just be replaced with a different product to solve the same problem which will be worse in its own ways.

Subjective, but I use both Android and iOS daily. Interesting byproduct of Android being the favorite of those obsessed with customization is that the stock apps are almost universally bad because everyone just replaces them with different niche alternatives.

Has it ever been the case that the built-in apps are the best in their category? Should it be the case? Apple's strategy seems to be to make a simple offering that appeals to most people, and to leave the advanced/special/power features to third-party developers. I think that's a pretty healthy arrangement, though I bet many devs would prefer Apple not offer defaults in some categories at all.

until the product gets killed

The YouTube app would have just worked on the vison pro if they hadn’t explicitly opted out.

>Why is that?

Operating system fatigue, supporting three native apps as well as apples own browser engine is a lot of engineering time.

End of the day Vision Pro needs YouTube more than YouTube needs it.

On the other hand, I have to imagine that to some extent YouTube is making maintaining their apps across multiple platforms harder than it has to be.

The app is almost entirely made up of tableviews/collection views/recycler views, save for the video player… really not rocket science. If YouTube’s public API were more capable I’m positive that third party devs would have no issue maintaining their YouTube apps across N platforms simply because they wouldn’t be overcomplicating them like Google is theirs.

Why do you say that?

I don't think the Youtube product managers really care enough about Vision Pro to prioritise making an app for it. That doesn't mean they strategically disgree with the product and actively wish to hamper it.

Indepedently of Vision Pro, I think they just might not be that enthusastic about third party youtube apps.

They explicitly opted the YouTube iPad app out of working on vision pro

Maybe it was a UX bug that hampers the experience to the point that where the website is a better experience, and they felt that a bad app would hurt the brand more than no app.

Or, maybe there wasn't but they didn't care enough to test so they disabled it 'just to be on the safe side'.