Hacker News Reader

Comment by UncleMeat | Hacker News Reader

The actual big goals appear to be to get GCP out of third place and to do something nebulously magic with AI. Both of these have made progress, but not in a way that actually makes the company money. Other than that I agree with you, there doesn't seem to be any sort of coherent product vision other than "ads in search/youtube" and "keep building the handful of actually popular products we've had for ages."

It is definitely hard as hell to develop another business the size of ads alongside it, but it really doesn't seem like there's a clear idea here.

View on HN

"GCP is not shutting down. Rest assured we're always working on bringing more great servers to the platform."

Unlike Stadia, I do actually think that killing GCP would destroy the company. So I don't expect the same end outcome.

Right, they're trapped in a terrible position where it's sucking the company dry of its profits while also being necessary to continue or they will lose all credibility with enterprise and developers forever. That said, they could shut down Verily and save some $1B/year.

The thing that is bizarre to me is that Google has absolute shitloads of money. 120B+ in the bank. Megabillions in profit each quarter. Like, they could buy Nintendo for less than half of their cash on hand.

I understand that the market demands infinite growth but internally it feels like horseshit to see the ridiculous numbers and also get told that basically nothing is high priority enough to get funded.

Nothing is high priority enough to get funded because they have a firehose of money coming out of ads, compared to which nothing looks like a good enough opportunity.

Exactly. The IRR in investing more into ad platforms for Google is still higher than any other project so they keep investing in it and increasing their ad revenue. Alongside Meta, they have a duopoly (monopoly?) on that market for now so they are just milking it for all its worth.

You don't just build stuff for the sake of it. It has to be a better investment than what you currently have, and Google doesn't have anything better than doubling and tripling down on ads.

Cloud's losses ($1.8B in H1) are about 4% of the operating income of the money-making parts ($46B in the same timeframe). Sure, a billion here and a billion there, pretty soon you're talking real money. But it's hardly "sucking the company dry of its profits".

Much of cloud's real costs are hidden in other parts of the budget, such as global infrastructure expansion.

So you're suggesting that the accounting is fradululent, and nobody figured it out except you? Not the auditors with actual access to the raw data. Not the hedge funds who could have made a pretty penny by shorting the stock and then releasing this information. Not even any whistleblowers who would presumably be eligible for an award.

Or maybe, just maybe, you've just made that up.

291 titles. Not great, not terrible.

Azure became interesting when it got better integrated into the Windows/AD/Office worlds. You're not having to manually integrate so much anymore.

The way for GCP to compete is to give customers (presumably high level customers) some real deep integration into Search/Gmail/YouTube/Ads/Drive/Maps, including data. That would not only be a killer differentiating feature, it would signal to everyone that Google is serious and won't shut this down on you like they did all those other services.

They probably won't do that, but I'm not sure why they really want to run GCP if not. If they're not going to give customers something only Google can give them, I don't see them capturing much more of the market.

Don't forget about shoving Shorts into faces of their paying Youtube Premium users without a toggle to permanently disable it. That will definitely outcompete TikTok any day now!

If they made it possible to cast Shorts like other videos, I might actually use them. I've tapped a Short several times because it looked funny, but it wanted to start playing on my phone instead of the TV. I watch TikTok alone on my phone and YouTube with my wife on my TV, so Shorts just doesn't fit into that routine.

It's mindboggling that you can't cast Google Shorts to a Google Chromecast. My hypothesis is that because Chromecast casting is so buggy and unresponsive (compared to Bluetooth), swiping from one Short to another would be such a frustrating experience that they dare not enable it.

My 2yo smashed my Oh So Smart TV with the remote and I got a spare dumb TV and plugged in Chromecast into it since everything is streams nowadays.

Dear God how annoying it is to use. Buggy, laggy, drops randomly ... still can't play videos from my computer in an easy way.

I forgot how bad Chromecast were since I last used it 5 years ago.

When the original came out ir was practically magic to be able to have the TV on while hanging out with roommates or friends, and all be able to contribute to the entertainment for the night right from your phone...then it didn't improve in any meaningful way, and regressed in others over the following 8+ years.

Really sad with the state of things.

When he broke my TV I was in the store looking for a new one. I had one condition -- there had to be a sane way to write with the remote. None of the TVs had any sane way but used arrows and enter or some cursor marker for the on screen keyboard.

Dunno why none used a T9 type of letter entering system. All remotes but one had number pads. It is really inconvenient to use "Smart" TVs.

So I went with the spare for a while.

Get a Bluetooth keyboard.

I used to maintain the Chromecast app of a decently sized streaming company for years, and oh boy. I damn near lost my sanity working on that thing. The app was broken repeatedly by silent firmware/OS updates pushed from Google, and trying to regression test updates across devices and versions was as close to hell as I've ever been.

On top of that, if you have a Pixel, the only thing it can cast to out of the box is Chromecast. Miracast (which e.g. Roku supports) - nope. Hell, they don't even do video-out over USB-C.

And I remember that it was different back in early Nexus days - they had Miracast, but deliberately ripped it out.

It seems like within the last two weeks they changed how Shorts and Chromecast works, where I'm at least able to cast the video.

I'm still annoyed at the "hover to play" that I have to disable seemingly every few days.

Also they are testing 5 preroll ads now. Idk why as a premium customer I do not have the option to filter out shorts. It is really frustrating. Also I just want youtube with no ads I do not care about youtube music that is bundled with it.

>something nebulously magic with AI

And even then I wouldn't place a bet on Waymo producing anything interesting from a revenue perspective for the next couple decades+. And, while I'm sure Google is doing other interesting things with AI--especially from an internal operations perspective--it's not like Google Home is doing anything earthbreaking or that Google search etc. is--from a consumer functionality point of view--particularly differentiated from other major players. And plenty of AI research is happening outside of Google.

Reply on news.ycombinator.com