Hacker News Reader

Comment by CommieBobDole | Hacker News Reader

Google often seems to have the attitude "if we build it, they will come, and if they don't, the idea was never feasible in the first place".

I think Google has to some extent been crippled by their search success - they built something and everybody came and uses it with little or no promotion or support, and its fantastic profitability has colored their view of everything else they try - if it's not an immediate massive success (and at Google, something that makes a hundred million in profit every year is generally not even worth the trouble), they lose interest and the people internally working on it disperse to other projects with better prospects.

They have the best hammer in the world, and it's cost them the ability to see anything that's not a nail.

View on HN

> Google often seems to have the attitude "if we build it, they will come, and if they don't, the idea was never feasible in the first place".

This comment echoes Paul Graham's observation about the wrong attitude founders often have about initial growth [1]: "A lot of would-be founders believe that startups either take off or don't. You build something, make it available, and if you've made a better mousetrap, people beat a path to your door as promised. Or they don't, in which case the market must not exist."

[1] http://paulgraham.com/ds.html

I wouldn't say that directly applies to what Google is trying to do.

I think what Google is doing is just re-investing back into x amount of companies/ideas where even if 1-2 of those companies bounces back and becomes profitable they make up their investments + they make profit.

I think they do realise how much money there is to be made in the gaming world and such. Its just time and money against profit for them.

Throw eggs in the wall and see what sticks or not.

Personally I don't like that model and thats why I use google services less and less. Even gmail is a huge no-go for me due to their nonexistent support.

It's funny though, I remember the old days before Google and they built something that was demonstrably superior for their users than what was out there at the time. They did win with a superior product.

Stadia feels like something that got designed by asking "how can we make something that's worse for our users than the other alternatives that are out there now?" and then going and building that.

Now had they come out with virtual machines better than what anyone could get at home, maybe?

Don't get me wrong though; I'm sure a lot of this is death by a thousand compromises.

As a side note: I was watching the Product Keynote from Flutter Interact '19 and when the person from the Stadia team was speaking about their app, it reminded me that despite all the negativity Stadia is getting, we should remember that there are real people building it that are enthusiastic about what they're doing and trying to make something good.

It was actually the same with Gmail as it was with Search.

Gmail was just a much better service than Hotmail or Yahoo! mail, arguably their biggest competitors at the time. They also offered a ridiculous amount more space.

I think those two successes alone led a lot of us into a place of just using Google products by default for a while.

It’s an utter shame what’s happened with the company, especially with its privacy issues.

> Stadia feels like something that got designed by asking "how can we make something that's worse for our users than the other alternatives that are out there now?" and then going and building that.

My friend has Stadia and it's actually a great product. The latency is negligible. I would be the exact target audience. I usually play 8-10 games a year. I don't really want to invest in a console or gaming PC just for that. The ~100 upfront cost is great. But Google's reputation kills it. I won't pay full price for AA titles when it can be taken away in any minutes. Till that fixed, I just stick with buying ~5 years old games from Steam that runs on my MacBook

Forgot about http://paulgraham.com/ds.html. Great post and relevant as ever today.

I think/suspect/speculate that since Google has search/ads billions raining from the sky, nothing else it does really matters to them.

So all side projects are done kinda like a hobby.

Makes sense to me. And I think another factor is that with ads, users are the product, not the customer.

If customer service is the foundation of a business, it drives a certain attention on the individual experience of every user, and there's a real incentive to be a customer's #1 choice by the customer's own criteria.

But Google's relationship with the userbase is statistical, not individual. As along as enough people look at their ads, that's good enough. They were never intending to please everybody, and so don't really have to pay attention to the details.

So it's entirely unsurprising to me that they're failing at gaming, which is very much about delivering amazing individual experiences to one's chosen niche. Ditto their failure in social networking. In both cases they built something that sounded good on paper and could be adequate for some people, but never succeeded in being truly great for enough people to capture an audience.

Google built a pretty successful social network, which even grew faster than Facebook for the first half decade. And like with their messaging services, they then decided it was better to just create a new one from scratch rather than trying to work with what already successful.

Orkut?

Yeah.

Orkut? I'm not sure we can really count that as something Google created. As I recall, it was a single engineer's side project based on work he did previous to Google.

Sure, Google did sort of adopt it. But as you say, it got pushed aside.

Reader was kind of a social network too, and would have equally been a great foundation to build on instead of shutting it down for google+

It's weird though, because GSuite -- esp google docs, google sheets, and google calendar -- are really good and pretty well supported. So they _can_ do it. Now to be fair, we're rapidly approaching $15k/year in spend with them.

I'm desperately wishing google would build a company-aimed wiki because confluence is atrocious.

And if anyone is going to deliver cloud gaming, my bet is it's google. It plays to their strengths: their ability to orchestrate and deliver scaled computing at the highest possible speeds. Further, it addresses one of the few markets at a large enough scale to make an impact on their revenues.

I believe docs and sheets were from their acquisition of Writely in the mid 2000s, not something that's homegrown.

I wouldn't call docs and sheets good by any means; they're passable at most, with tons of missing features.

No doubt they're an ok word processor and a mediocre spreadsheet.

That's not what they are though. They're really good group collaboration on docs, particularly content, and collaborative lists with owner assignment. We do a huge amount of editing / signoff / planning / workbacks / etc in them.

> I wouldn't call docs and sheets good by any means; they're passable at most, with tons of missing features.

And yet they entirely replace Microsoft Office for millions of people.

Docs and Sheets were entirely rewritten in house many years after the acquisition. They don't work at all like the original products.

I'm guessing you're getting down-voted for that last bit.

For one, it's missing the point about what everyone said is important about gaming, and hits the point that everyone said google doesn't understand about customers. why do you think 'deliver scaled computing at the highest possible speeds' = good at gaming? its kind of typical google, 'we offered the best X statistic, therefore they must like it.'

for "if anyone is going to deliver cloud gaming, my bet is it's google.", No. If I were to bet money, I would say it would be someone who is good at cloud, or good at gaming. Google is good at neither.

I don't think it's sufficient, but it is necessary. I'm not sure who else has an edge network that can deliver comparable compute and comparable latency? A very short list, if anyone.

And Google isn't good at cloud? Maybe not gcp, but their internal application delivery capability is second to none.

What's so bad about GCP/Firebase?

It’s not the technology, Google is horrible at Enterprise sales and support. Firebase also isn’t going to win them large enterprises.

+1. They're an ad company and ultimately they feel clunky and condescending to deal with in a consumer context.

Azure Solutions Architects: You want to move your SQL Server 2003 database running on Windows 2000 on Azure? Ok we can do that and come up with a project plan to modernize your stack and extend your support.

GCP SA: We are “smart people” (tm) you should move your entire workload to $cool_technology before we migrate and do things the modern way.

We are not game developers, our opinion doesn't count.

edit: I just remembered: I've actually begged people not to make projects dependent on knol or wave. Its unsurprising how those projects ended after they ignored me. What did surprise me were the people who identified the use of those platforms as a failure in judgement and left immediately.

For cloud gaming, the hard and highly relevant infrastructure work is about how many edge locations you have.

On that metric, Google is far ahead of everybody else in that particular field.

Now if Cloudflare or Akamai were to enter the cloud gaming business...

There is a lot of competition in the cloud gaming space, and there is a low technical barrier to entry, especially since the streaming server + client are likely to commoditize (they’re all based on similar codecs, and require hardware acceleration that’s provided by the GPU).

The number of edge locations you have doesn’t matter; latency does. It looks like startups in this space can achieve “good enough” latency with a small number of edge locations.

For instance, Shadow covers the vast majority of the US population with four data centers (they have six, total, but two are in Europe):

https://status.shadow.tech/

Coverage map:

https://shadow.tech/usen

> their ability to orchestrate and deliver scaled computing at the highest possible speeds.

The gaming industry is already good at that. Like, really really good. Do you think companies that operate world-spanning low-latency gaming experiences are going to be quaking in their boots because Google?

What Google is really, really bad at is humans. And the gaming industry is all about understanding that there are a raft of interests ranging from casual to shooter to arty to... whatever. There's still a lot of room for taste-driven sleeper hits (think Untitled Goose Game) as well as research & marketing driven extravaganzas like GTA Online or FIFA whatever-the-year-is. And wrangling social media, forums, engaging with fan wikis and chat and stuff, which is part of the fabric of gaming, is the sort of thing Google are absolute shit at. Gaming chat is famously terrible, but even gamers look down their nose at YouTube comments.

> I'm desperately wishing google would build a company-aimed wiki because confluence is atrocious.

have you tried Google sites?

Ironically I am yet to find a Wiki as good as confluence.

I'm not sure I can call that a real exception. Calendar is something they needed themselves, and also use as a "soak up everybody's data to make things more sticky" tool. It's good for what it is, but if it were a serious product offering, I'd expect it to be more flexible, more feature-rich.

Docs and Sheets strike me as probably in the same bucket. They're perfectly cromulent, but there's nothing amazing about them. They were acquisitions, not an in-house creation. All they really do is take the basic features of any number of competitor products going back 20 years and put them on the web. If you like what they deliver, great. But if not, I wouldn't wait around for Google to add features aimed at specialists. It seems more like a basic utility than a consumer product.

Have you worked collaboratively with Google Docs?

Sure. That's what I mean by "on the web". They have decent basic real-time collaboration features. Otherwise, they're basically the same as Word and Excel were 25 years ago.

But even for collaboration I think there's a long way to go. It's not like they significantly rethought the product in terms of true collaboration at a global scale. As far as I've noticed they've added non-conflicting, real-time edits and showing cursor location to the previous Word/Excel collaboration features like comments and proposed changes.

Just what you want from your service provider.

Actually most of the Google projects exist to defend Ads/Search combo and to napalm potential threats.

My thoughts are my own but I think Google is very serious about diversifying their sources revenue away from mostly advertising.

Sure, but it's a "nice to have" serious.

Not the "life or death" serious companies are for their main business.

I don’t doubt they are trying, they just haven’t shown they know how to compared to other large tech companies.

Google is the world's most expensive post-college hiring and training program, subsidized by the core search/ads cash cow.

But it's a really bad one. They killed Reader because it would let them abandon some legacy infrastructure. All other businesses don't have that luxury. They deal with annoying challenges head on. Google's unlimited money basically lets them abandon products when maintaining them stops being fun, or when extra work is required to power through the last 10% and make that product viable for take-off.

I think it's more that Google's management/culture currently doesn't work.

Everyone I've ever met there subscribes to the ideology you can soft launch things and build over time. "It worked for Gmail" they say. But ultimately Gmail didn't ask people to shell out real money. And it was a long time ago. Consumer culture has changed, and a launching half finished messy products with no real support or direction really alienates people now.

That approach also only really applies if you can give a product away for nothing for a long time to build a userbase. Ruth Porat simply isn't going to let that happen, because Google's new fiscal discipline means she kills products if they don't have a clear path to profitability after three years.

And that's the real contradiction. There's nothing wrong with Porat's approach, but Google structurally is not set up to deliver products like that at all. And Google's culture did used to be successful, but can't work with what the Stock Market has made it very clear that Porat is to do (it's extremely clear stockholders like the three year requirement).

Ultimately you've got to look at the top of the company for the rot, because senior leadership do not seem to be pulling in the same direction, and as a result you end up with now... when was the last time Google launched a successful consumer product? The Nest Home Mini is probably the closest one, and I suspect that still hasn't made any actual profit with how many they give away.

Google's easy to sunset policy makes them an easy target in any industry that isn't ripe for change or that provide any setbacks. If the going gets tough Google exits.

Google's size gives them a big advantage that they never figure out how to use it.

> Google's size gives them a big advantage that they never figure out how to use it.

Good for us consumers, isn't it?

Like any tool it depends on how they use it. Leaving the abusive fiefdoms of others in place doesn't help consumers. It is a matter of nuance essentially. Google and all competitors may do things good, bad, or both for consumers and their size doesn't have much to do with it.

iirc Google Maps was initially not a success. MapQuest had a majority market share of online maps up until the release of the iPhone. So they floundered for about a year or two.

https://www.businessinsider.com/mapquest-a-symbol-of-everyth...

So you're saying Google Maps wasn't a huge success until Android?

... but totally in a non-monopolistic way, of course.

I'm saying that Google does sometimes stick with products that aren't doing well initially if they see the promise in it.

Reply on news.ycombinator.com