Hacker News Reader

Comment by panarky | Hacker News Reader

"Insanely good ideas"?

With the months of nonstop and over-the-top hype of "Apple Intelligence" these all seem underwhelming.

Even in the idealized world of demoware there's no killer feature, not to mention that none of it is even available to real customers yet.

I was expecting something only Apple could do with their vast ecosystem of IP and partnerships, the must-have cool new thing that would stimulate a leap in demand for iPhone.

It's hard to see how any of this is enough to trigger a new upgrade cycle instead of just waiting until battery life goes to shit or the screen cracks so you have to upgrade.

View on HN

> With the months of nonstop and over-the-top hype of "Apple Intelligence" these all seem underwhelming.

None of that was coming from Apple, though. I don't think they can be held responsible for hype that gets put on them.

IMO these are all solid moves into AI integration without getting carried away by hype and doing things that will have very middling results.

True. All the whining was coming from lazy "industry observers" who never mustered up an example of what, exactly, they expected Apple to be doing with "AI."

Yeah all of that annoying whining was from them. No one other than "lazy" ""industry observers"" were expecting "so-called AI" stuff from Apple.

We just watched an Apple event where they discuss Apple Intelligence.

My understanding is people knew Apple Intelligence existed before this event.

Is it accurate to say Apple wasn't involved at all in touting it? All external parties? Sounds wrong to my ear

It was brought up in their software yearly event - WWDC. It was actually a pretty big part of that keynote, even as it was light on details.

idk if its that light on details tbh:

- https://buttondown.com/ainews/archive/ainews-talaria-apples-...

- https://buttondown.com/ainews/archive/ainews-apple-intellige...

The sources in those articles are user conversation and speculation from Reddit, Twitter, and Discord. There's very little information in that massive article that actually came from Apple.

Apple was notoriously quiet on AI until they recently revealed Apple Intelligence, it made sense to me since iPhones already have access to all of the user's data, and data is the core of all of the best use cases for AI.

Apple called AI “Apple intelligence”. It’s on them to live up to that smug claim. They didn’t

I think you're seriously underestimating the impact of actually-useful and reliable AI features at your fingertips. This is what everyone imagined computers would be, since I don't know... the 80s?

Nice to have sure, enough to consider spending upwards of $1000 on upgrade? don’t think so.

Sure, but this isn’t one of those AI startups like Humane where you’re expected to drop $800 on a brand new thing. Most people aren’t buying the top end models and they probably aren’t paying full price due to promos, trade-ins, etc.

That means Apple has a much simpler problem: they don’t need to convince you that AI as a concept is worth hundreds of dollars, they just need to convince you that the hundreds of dollars you were already going to spend on a smartphone should go to them instead of Samsung or Xiaomi.

Are you sure ? People do not move in significant numbers from Android anymore, a lot of people just do not want to learn a new interface and workflows and happy with their ecosystem.

Anecdotally I tried to get my dad to a iPhone 14 and shift from Android, he basically kept using his Android and never switched,so my mom is using that device now(along with an Android) not too happily just cause no else one is using it and will switch back to Android fully next upgrade. There are plenty of people who simply prefer the Android interface or by now so used to it that don't want to change and vice-versa. I myself use and need both and cannot imagine exclusively committing to one ecosystem.

Inertia is real but that would make it all the more important not to lose existing users or appear behind to new buyers.

Siri (for all the absurd whining about it) met those expectations 10 years ago.

Hmm, my experience with Siri is that it's good for setting timers, okay at setting reminders, and that's about it.

Even reminders is pretty bad:

"Hey Siri, add celery to the shopping list" "Okay John, I've added celery to the shopping list list (sic)."

Shopping list: - Celery

"Hey Siri, remove celery from the shopping list" "Okay John, I've added celery to the shopping list list."

Shopping list: - Celery - Celery

This is not meeting expectations.

Ha hahah, that is indeed not good.

Although somewhere I'm sure we can find an '80s pop-culture vehicle that predicts those shortcomings as well.

Your comment reminded me of the Apple Knowledge Navigator.

https://m.youtube.com/watch?v=umJsITGzXd0

Great opportunity for someone to make a side by side comparison video & go viral.

> photos/album search now includes video understanding, which imo seems very good from the first 2 examples they showed. includes scroll to exact time of the moment you describe.

This is a killer feature. Everyone can understand the usefulness of just describing the video moment you want to find. Digging through years worth of photos and videos looking for that one specific moment can be a huge chore when you want to show it to someone.

> there's no killer feature

You know when you go to a restaurant and the menu is like 3 different cuisines, it's 12 pages, and each page has 20+ items?

The Intelligence section of the presentation felt a bit like that.

Generally, I was surprised how lightly Intelligence featured into the keynote: it wasn't a standalone talking point, nothing on the Apple site talks to it directly. It's conspicuous, given that they've dedicated hardware real estate to it ... and the hype cycle.

Didn’t it have a pretty large thing on the keynote where it was actually announced?

No, Apple Intelligence is worse.

I go to a restaurant and look at the menu/see a commercial on TV, I can usually order the item now (other than films).

Apple Intelligence - certain feature(s) (can't recall now) are US-only in November, rolling out to other English-speaking countries in December. No idea when non English-speaking countries will get that feature.

Who knows what Apple Intelligence features are in the device on day 1?

I thought you were going to say the iPhone would be able to load the QR code for that menu, summarize it, sort it by ingredient or calories and give you a recommendation without you having to read it... Now I'm kinda bummed out.

See, now THERE's a FEATURE.

Where were you when they were coming up with "when a friend texts a song name to you, remind them they can play it in Apple Music"??!

Apple rarely ever innovates, so it's not surprising these new features seem underwhelming. What Apple does best is to perfect (others') existing ideas so that they work more much more reliably than the competition. The number one pain point of technology today is that hardly any of it consistently works correctly more than 95% of the time. But when Apple releases a feature, it works 99% of the time instead. That's their key differentiator.

>But when Apple releases a feature, it works 99% of the time instead.

"You're holding it wrong" - Steve Jobs

The action button being prime real estate I think shouldn't be understated.

It's a button that'll be in the hands and pockets of >1 billion people soon.

Each press allows these billion people to connect with a supplier of Intelligence. Whether it's ChatGPT, Google's search, Apple AI or whatever supplier. Apple controls the hardware.

Just think of Apple's contract with Google that nets them $20 billion a year in profit. For comparison, Walmart, Amazon, NVIDIA, Volkswagen, VISA, all had about 20 billion in net income in 2023.

Or think of the appstore as an analogy. Apple controls the gateway between consumer and supplier, and gains from it. This action button will be no different.

1) i am as ready to shit on apple and siri as the next guy, but imo the HN cynicism is overbearing here. apple isnt building to impress internet neckbeards like us. they are building for the normies. features they understand intuitively and cannot fuck up. maybe wait til it gets in the normies hands to judge. ~none of us here know what its like to work on a hardware/software platform for >1 billion? people.

2) i actually do think dedicated visual intelligence button IS a killer feature. suddenly phone is ai's view into the world. yes probably v1 today will disappoint. but 5 years from now the kids will laugh when we say we had to take photos and upload them to an app back in our day. or manually enter in any information from the real world into our calendars and emails and texts.

Highly doubt normies give a crap about any of this.

What? How could you possibly say that? They can take a picture of something and then auto make a note, or a reminder, or an email, or find out what it is.

Normies struggle with the nuance of devices. They understand what they want, they don't understand steps, like start here then do this, then go here and copy this, then go there. This SKIPS all the steps. This is huge for them, but (probably) just mildly convenient for others.

>find out what it is

Google Googles did this 13 years ago https://en.wikipedia.org/wiki/Google_Goggles

It remains interesting to me how much we have been able to do, already, with technology. But discoverability, walled gardens, and <hand waving>marketing/positioning/hype</hand waving> have obscured awareness/adoption.

This is really really old tech that normies definitely do not care about.

Asking when your moms flight gets in, to pull up the picture of you and your partner on vacation at the beach or to take a picture of a restaurant from afar at 5x zoom and get reviews is insanely nice.

This is the perfect use of AI, at least for me.

> apple isnt building to impress internet neckbeards like us. they are building for the normies. features they understand intuitively and cannot fuck up. maybe wait til it gets in the normies hands to judge.

Apple has lost its way, just look at how the "normies" and the "neckbeards" responded to the Vision Pro. Apple doesn't know how to innovate anymore, just how to market, and mass produce things, which is honestly good enough to keep making money for now.

you’re suggesting apple planned to sell a $3.5k headset to normies? That doesn’t make sense. Sure, many normies had an opinion on it. Most normies never tried it so i don’t care what they think. It’s like asking a normie 20 years ago about what driving an electric car is like.

Vision pro was an over-hyped dev kit that offers a taste of the future. That’s it.

They have classes full of "normies" trying them out every time I pass by my local store. Sure looks like its meant to be something other than a dev kit to me.

> Apple has lost its way, just look at how the "normies" and the "neckbeards" responded to the Vision Pro.

Such people are likely confused. The AVP makes sense as a high end POC and dev kit. It takes a minute for a new modality of computing to make sense to people, even for Apple's own devs to figure out how people can use it, to iterate the OS and default app suite for a few years.

Agreed, and Apple's differentiator has always been seamless integration. With the iPhone, they own the boundary layer between the physical and digital worlds. It's an obvious place to introduce AI for daily tasks without requiring any behavioral changes from the users.

I agree but all the normies I know say Siri is shit or just have it disabled

They aren’t that dumb to think calling it now Apple intelligence powered will make Siri work

As a "tech person" I feel like we've all been told for 20+ years that we can't have UX that makes any sense to us, or lets us do simple things we want to do, because we are power users and software has to cater to some imagined demographic of incredibly dumb people who can't understand basic computing metaphors. It can get frustrating.

Which would be fine if they easily allowed us to write software and scripts for our own devices.

Shortcuts.app exists, you can even tie those scripts to a hardware button now or ask Siri to execute them or place them as app-like icons or as interactive widgets on your home screen or Lock Screen, or tie them to automation events (schedules, but even more fine grained like "when I get a message from this person, send them an email", or "when I get home, log the current weather to a new line in a table on a spreadsheet". A script gallery even exists where scripts can easily be shared and ran by anyone or shared across iMessages.

Not everything is available like on a rooted Android of course, but I can at least create a Shortcut script and give it to any iOS user regardless of their technical expertise and they can utilize it, which is leagues ahead of the situation amongst non-rooted Android and non-technical Android users.

I'm a Growth Engineer. What that actually means is "make the UI so incredibly simple people who are functionally illiterate somehow make their way to the place you want them to be so you can eventually monetize them".

I view users moving through interfaces in a mass abstracted form like electrical or hydrological currents. My job is to widen the pipes.

Well, cynically that fits with my own experience. :) But I don't know if I'd say it's simpler, just that you don't have any choices. For me Apple UX often feels complicated and I can't figure out how to do what I want, or figure out what is going to happen when I accept some dialog they want me to accept. But I guess I end up going through the pipe eventually, out of frustration.

My impression of my non-tech friends is that they aren't too dumb to follow instructions, they just don't enjoy operating a computer like I do, so they are more okay with their choices being restricted if they don't have to waste time doing something they find boring.

Semantic search was always going to be the killer feature of this AI boom. But it’s just a cherry on top of existing software and requires training users to expand what they think is possible.

As far as I can tell, AI is one of several features that have actively made search worse. There are a handful of novel things it does well enough, but I'd rather engage those in their own LLM sandbox than have the facilities formerly available in a searchbox replaced with a guessbox.

What kind of search has it made worse? I’m not talking about Google search. I’m talking about searching your personal collection of emails or photos or text messages.

I would guess is you're probably in the minority with that opinion.

They showed up the full Apple Intelligence featureset at WWDC months ago. This was a shorter recap.

Thanks for noting this. I was surprised how little it featured into today's announcement, so this fills in the picture.

Technically it's an iOS 18 feature rather than a phone feature, except the only previous phone model with the hardware for all of it is the iPhone 15 Pro, so it doesn't really feel like an iOS 18 feature

On device and not shared with advertisers is the Only Apple can do part.

Like what? People have been whining "where's the innovation" for years about phones, totally ignoring the fact that this is a mature product category.

Where's the innovation in word processors?

At this point, we're often talking about regression. Apple removed the audio output from its best-selling music player. It removed TouchID. It removed the SIM slot.

You know what would be "innovative?" Making a goddamned phone that can withstand being used for its primary purpose. The failure of the iPhone's design has been confirmed by millions upon millions of people, who have concluded that they must bury their "thin, elegant" phones in bulky, tacky cases. You have to shake your head at Apple's breathless announcement of NEW COLORS!!!! when the vast majority of them will never be seen.

Then there's the fact that Apple ignores an entire market: men who don't wear cargo pants all day every day. Maybe some of us don't want to look like a schlub with a TV tray jammed into his pocket (again, made even worse with a dumb case).

The moment you think a portable electronic system is a mature product category is the moment you become ‘old guard’.

Edit: and currently, that looks to be you and the entire iPhone engineering team

Oh really? I already provided some suggestions on what they could do. Where are yours?

I mean ‘could do’ is subjective. The ROI that apple receives on R&D is substantial, so on that basis, include a portable projector in phones such that I can replace my tv with my phone - not being able to use my phone while watching tv wouldn’t be nice, so add anti-shake gyroscopes such that as long as my phone is pointing in vaguely the correct direction, the projection stays stable.

Possible with current tech? Hell no. But within the realms of reality for handhelds to do this in the next 15 years? I think so. Probably less valuable over health and other potential device functions, I don’t think about these things and so I wouldn’t know.

Thus, we are not at a point where portable computers should be considered ‘mature’ from an innovation standpoint, even if current innovation is slow.

Reply on news.ycombinator.com