Street View turns 15 | Hacker News Reader

Street View turns 15

blog.google

I think street view was one of the most audacious features I remember ever seeing in a product.

I wish I knew the strategic and business analysis that was behind thinking up the feature, green lighting and funding it.

I definitely did not have the foresight to see how street view could ever be worth the investment 15 years ago.

Today I am more inclined to see the tremendous value of google maps as a whole.

Street view has saved me quite a few times. Searching for a shop or office and being able to see what the front looks like when there’s no signage.

Street view is an amazing product on its own. I've seen it used in all manner of instances from facility managers counting windows in order to write bids to sightseeing by people going on vacation to people checking where they need to go.

Come to think of it, i've seen more professionals use it than anything else. The ability to go somewhere, without actually going there and actually see it is truly a asset.

It's astounding that surveillance advertising is so profitable that it can fund enterprises like this.

“Surveillance” isn’t very profitable; Google ad targeting isn’t even good and when it is good it doesn’t need your data. (For instance, if you search for “buy ps5” they don’t need a secret profile to show you ads for a PS5.)

The reason their ads make infinite money is they’re the best at serving them, they’re vertically integrated, and their ad auction systems are set up to cheat in their favor.

If that were true then they wouldn't collect it.

It doesn't mean it's effective, but it's still clearly profitable.

I said “very profitable”.

“Surveillance capitalism” earns nothing compared to how much cash Apple makes, but there isn’t a field of studying “cell phone capitalism”.

There’s plenty of study of the practice of one entity selling a good or service to another entity. You might call it “economics” or “business.”

“Surveillance capitalism” is interesting/unique because not all of the relevant parties in the exchange (users, platforms, advertisers) actually understand their role in the exchange.

Google started out collecting it with no idea what to do with it. User data was a waste product like gasoline was before someone invented a mass market car. Google figured out they could buy the then-leader in targeted advertising and apply their data to it when the tech market turned and investors suddenly started asking about business models.

Then why do I have to sign off on 65 pages of “we collect every scrap of info about you short of sequencing your genome, and we would do that too if it weren’t a logistical nightmare” every time I install a piece of software or buy a device with wifi capability? Someone ought to relay this insight to tech companies - it would save them a fortune.

Because their having your data is a liability for them, not an asset, which is why they want to offload some liability on you.

If it's a liability, where are the company ending fines and lawsuits?

There aren’t any because you signed the TOS. Many people have tried class actions about these things, but they never get anywhere meaningful. Europe is trying with GDPR to tip the balance back to the users, but the countries are stuck with massive backlogs that prevent progress.

I just cashed a $400 check from Facebook for illegally using my data. I imagine they'll be more careful next time.

> For instance, if you search for “buy ps5” they don’t need a secret profile to show you ads for a PS5.

Sorta. The real money is from competitors. How much would Microsoft spend to convince someone interested in a ps5 to to spend $500 on an xbox instead? Sony wouldn’t spend as much as Microsoft here except to outbid them out of that ad.

And you’ve just shown your hand at having $500 to dispose, maybe a local casino ad would fit well here?

While the geographies of gaming consoles overlaps well, it doesn’t for automobiles, restaurants, service providers (mechanics or dentists or whatever).

>> The real money is from competitors.

Yep:

https://www.google.com/search?q=free+cad+software https://www.google.com/search?q=free+office+suite

Asking for open source or GPL will usually get rid of the ads at the top. Companies can clearly pay to get results at the top that don't even fit your search terms.

Ads must be relevant to be shown. The casino ad would probably not be shown at all, but even if it were they would have to pay more than Sony.

>> For instance, if you search for “buy ps5” they don’t need a secret profile to show you ads for a PS5.

>Sorta. The real money is from competitors. How much would Microsoft spend to convince someone interested in a ps5 to to spend $500 on an xbox instead? Sony wouldn’t spend as much as Microsoft here except to outbid them out of that ad.

But they still don't need a secret profile to show you ads for an Xbox when you search for PS5...

That’s why it’s not a great example. But if I’m searching for a fiat, it wouldn’t make sense to target a Renault ad against me because they don’t sell those here.

Targeting is only one of the reasons for the large amount of tracking that Google does. The second reason is conversion tracking, and that one's way more important to Google's customers. They do want to know that if you see a specific ad for a specific thing, whether you now buy it or not.

An attempt to answer the statement "I waste half my money advertising, problem is I don't know which half"

Is that still the case? Do we know how much of Google's ad revenue is directly from their own search product? Obviously the genius of putting ads on a search engine is that it's a perfect fit: the users are literally typing in what they're interested in. But surely Google is also making big bucks now from ad products where this isn't the case (or is less so the case), like third-party web ads (AdSense) and mobile app ads (AdMob), as well as embedded ads in other Google products (like Gmail), and presumably those do rely heavily on tracking user behavior.

Given that one of the early team members was a world-famous wardriving expert and that the cars were initially “accidentally” scraping all open data transmissions they could find, the initial strategic/business analysis probably was some form of: what if we drove cars around that just physically scraped data from unsuspecting neighborhoods Wi-Fi networks for our advertising algorithms, and in exchange we’ll make photos of those neighborhoods publicly available.

Note that when the FCC investigated this practice they found that the data collection was definitely a deliberate design decision, which obviously makes sense given they moved Milner (creator of Netstumblr) over from YouTube to work on it.

So yeah, it’s just a fancy hacking project on wheels against unencrypted networks and with a corporate wrap on the car to support the ad network.

I’ve heard this theory a few times but it doesn’t sound very plausible: could you speak more to why you think the goal was always to mine wifi network information?

We know that Google was doing that, but I thought it was an after thought associated with street view, not the motivation for street view.

Because why else would you do it?

I don’t think they cared about unencrypted networks, but more about the ssid and macs of devices. Today they use Wi-Fi networks near you to determine your position when GPS is not available.

1) What would be the rationale for such a project otherwise, per GP?

2) What especially would be the rationale to put a wardriving expert on the project?

“Data-hungry advertising network grows sensor array with cars” is far more believable to me than “data-hungry advertising network maps the world for free.”

Getting email address, CC info, passwords, names, Wi-Fi SSIDs, MAC addresses, etc all lined up with precise physical locations? That’s of value to the company. Having pictures of houses? Not sure what value that has to the company, and if it had any it’ll be similarly just an input into the same advertising algos.

1) The Google that built Street View is not the soul-less, profit-and-shareholder-value-optimizing enterprise it is today. There is no chance they would greenlight that project in 2022 if it didn't exist yet.

They just happened to have a wardriver on a project that looked an awful lot like wardriving?

When you are doing an expensive operation such as putting a driver in the field traversing thousands of road miles, it makes sense to collect as much data as humanly possible from that platform instead of realizing later that there are certain things you wanted but can't repeat the driving easily. Or, you think of things to do with the data you never imagined before.

The SSID collection only turned out controversial later, and they adjusted their data collection. I still don't really see a problem with it, TBH. You radiate into public space, you bear the consequences.

Yes this does make sense if you are, at bottom, an advertising company... in need of data... In which case your "putting a driver in the field traversing thousands of road miles" is called "wardriving."

Walmart's fleet, for example, drives 700 million miles. How much data do you think they "accidentally" sniffed and recorded? I'll bet precisely zero bytes, because they are moving goods, and when that's a company's MO it makes no sense to have wardriving equipment or expertise involved.

> How much data do you think they "accidentally" sniffed and recorded? I'll bet precisely zero bytes, because they are moving goods

I'll bet dollars to donuts they track exactly where their vehicles go, and penalize drivers that exceed a certain deviation from the expected route/timing.

Either way seem like perfectly legal things to do.

Uhhh, which is completely unrelated to what Google was doing?

And hey I never claimed it was illegal. I just claimed it was a likely rationale, that’s what OP was asking about.

Google’s famous motto: “Don’t get us fined too much.”

Right, also back to that original point: I could totally believe that a 2009 Google would run that project for other reasons than advertising, such as prestige or engineer-driven because they thought it was cool. I mean there are posts here from people who were part of the original project. Today a project like that would be beancounter-driven and only allowed if it had a clear short-term ROI.

Did Street View have a positive ROI over time? I don't know. It certainly helped put Google Maps front and center. Maybe you can think of it as a very successful advertising campaign - instead of blowing millions on SuperBowl ads you blow millions on cars taking pictures of every public road mile.

This is a pretty insane conspiracy theory considering people happily give Google their CC (Google pay), emails (Gmail), physical address (maps), password (chrome), wifi info (android).

It was just a mistake in a setting. They were accidentally capturing random packets, and of course you collect enough random packets and they will contain every conceivable type of data. It wasn't used for anything. It's not a conspiracy. They did street view to do street view, which is why they kept doing it for 12 years (and counting) after the packet collection was fixed.

The “insane conspiracy theory” is not that an ad network did everything it could to scrape more data for ads, lol. The insane conspiracy theory is that the ad network decided to photograph people’s property because it’s a nice thing to do.

There’s obviously all sorts of data that wardriving can pick up that none of those services could (even if they existed at the time, which most didn’t). And obviously you or I have no clue what the data was used for. My guess is that, being a business, they used it for their core business, and your guess is that they spent exorbitant amounts of money to accumulate their pot of gold for… well no gosh darn reason.

Now that Android exists, Google has pretty good reason to continue investing in mapping. But let’s keep in mind that Android also only exists to power the revenue-generating part of the business (ads). This isn’t a critique of Google, this is how businesses work.

I would have gone with "they mined wifi network information to use in Android's location system", but while the data was almost certainly used for that I'm not sure the timeline lines up for that to factor into greenlighting the project.

I doubt it was the only reason, a large part of the project IMO being a show of technical prowess, a “come and compete with us if you think you're hard enough” message, but I similarly doubt it was at all an afterthought.

Location information is valuable for advertising and other uses, and if you combine many users turning GPS off to conserve battery¹ when they didn't specifically need it, with cell-based location information not being particularly fine-grained[2][3], being able to track location by wireless AP proximity is a pretty valuable extra option. I expect there would have been a fair few people tasked with coming up with ideas for how to collect enough data for it to be a practical option⁴.

[1] IIRC this was before easy controls to stop apps having access to location data if the hardware was turned on

[2] in a crowded city you might be able to track someone to within less than half a mile, but elsewhere this potential error is much higher

[3] or available at all due to reception issues

[4] StreetView came before Android saw public release, so siphoning off location data from users of that OS would not have been an option that early, and even if it was would still rely on GPS use to make the data usefully accurate

Also, many devices didn't have built-in GPS (the original iPhone, for one), so how do you run location services?

There were, at the time, several crowd-sourced sites of Wifi AP locations, which devices could use to estimate location from Wifi RSSI. So making a private, comprehensive database of the same is a pretty obvious task for a company that is either trying to catalog the world's information, or that wants to do something with location-based services on mobile devices in the future.

Beyond that, getting an approximate location first is useful, even if you do plan to use GPS. It takes up to 30 seconds to download the ephemeris for a GPS satellite over the air (assuming no uncorrectable errors), after first locking on to the frequency and PRN of each satellite. Cold-start times of 5 minutes under realistic urban conditions were not unusual. Do you want to wait 5 minutes for the map to load on your phone?

However, knowing your approximate location (from nearby APs) narrows the PRN and frequency search space (and you also know which satellites are above the horizon). With ephemeris data downloaded over the internet, every cold start can be as fast as a warm start: Just a couple seconds.

There are plenty of reasons to think that Google intended to war drive from the beginning.

GPS doesn't use battery; a better description of users is they do completely random stuff because they think it might increase battery life or they dreamed someone told them to do it once years ago.

Killing apps after they're done with them is the other big one there.

> GPS doesn't use battery

Tell that to my phone. And my sports watch. GPS being available doesn't use battery on its own, but if apps actually ask for precise location information so it gets used, that doesn't happen for free power-wise.

I will not be able to prove it to you, but your theory is incorrect. As stated elsewhere, the project was kickstarted at the height of the mid-2000s map-tech explosion, and it was a loss-leader much like Google Earth and Google Maps (which were barely monetized, even to this day).

If nothing else, remember that in mid-2000s, Google was overflowing with cash and was growing rapidly. It was just beginning its transition from "search and only search" into apps (gmail, docs, etc) and mobile. Web ads were pouring in money at a rate hard to comprehend, and the "value" of SSIDs or whatever else you suggest were being farmed, was not even a rounding error.

> Given that one of the early team members was a world-famous wardriving expert

TIL why NetStumbler stopped getting updated in 2004

"Our company mission is to organize the world's information and make it universally accessible and useful. "

-- seems very in line with mission of google -- take photos of everything so you could later run some (at the time OCR, now "ML") analysis over it & presents the information usefully --

Street view may be tricky to justify from a financial perspective, but it's pretty easy to justify from a mission perspective, which for Google is to "organize the world's information and make it universally accessible and useful". The mark of a good mission statement is that many seemingly unprofitable things that you do to pursue the mission end up paying dividends in the long term.

I'm also not sure how expensive early street view actually was, does anyone know how many cars/drivers it takes?

We had 2 initial prototypes to index the bay area.

I think there were maybe 10s of cars in the first big sweep over the US.

I have no idea how many cars they're running these days.

> which for Google is to "organize the world's information and make it universally accessible and useful"

Those missions statements always make me laugh. What if someone at Google found a great and cheap way to do that (organize the world's information blah blah blah) in a way that's unfortunately completely unmonetizable? Of course, Google would not implement it - and would patent the idea if possible, so that no one else can.

The only true mission is to make as much money as possible.

But Street View itself is not monetized (not directly, and Maps itself is barely monetized). Street View proves your thesis wrong.

Maps is very well monetized. They make money from maps ads, as well as maps API[1]. Their minimum $ for maps was very expensive last I checked too.

[1]: https://www.investopedia.com/articles/investing/061115/how-d...

How can something be completely unmonetizable? If users want to use it, then you can sell them access or show them ads. It'd be strange to have a product that's logically only usable by people without money.

I mean, it's possible if the content you're indexing is offensive or pornographic of course.

Street views was actually started as a Stanford project called CityBlock.

LOLOL, business analysis? The idea was entirely an engineering nerd project. We wanted to index the physical world, just like the book scanning project and a bunch of similar ideas.

Source: I helped build one of the original street views vans as my 20% project.

Prototyping and building this must have been a fun experience. It must be amazing to see a moonshot like this to become a real product that is used millions of people every day. I find street view to be tremendously helpful in many situations. It's a shame that my fellow German citizens are so scared of new tech, that Google had to stop indexing German streets.

I was always surprised how they were afraid of credit cards tracking everything they do.

OTOH in France we have people afraid of (someone, not sure whom) tracking their life through the computerized power meters...

It was an interesting mix of "let's do a cool thing" and "imagine what can be done with this data!" Early on someone mentioned that NYC doesn't actually know for sure where all of the fire hydrants are _actually_ placed, and it may be possible to automatically extract the locations from the collected data.

More obviously, looking for the address signs of addresses that are expected along a street can dramatically improve driving directions.

Of course, the biggest business value was being able to generate the actual, underlying street maps without having to purchase that from companies that had already digitized and driven the streets.

who was the berkeley guy at the corner of Kittredge with all the SSDs then?

Digression, but this is my current complaint at my job at a Big Tech Company: we're now so big that we have product managers for everything that dictate what we can and can't work on and they hold a monopoly on customer data that's used to justify new product decisions. Engineers who have been around for years and started the product from the ground up, worked with customers to understand their needs, and have a good idea of what features may be valuable to them going forward are being ignored because now everything needs a business analyst to approve the idea.

Some of the best products in the world came out of experimenting and just building something cool - I could never imagine many products that now support business analysts and product managers salaries getting the green light from them if they were pitched from scratch.

I wonder how to avoid that kind of thing from happening?

It's almost like big business edict is a predatory entity, like a grossly gargantuan gelatinous cube absorbing whatever it can control into it.

It's absolutely possible to be both rigorous and risk-tolerant at the same time. It's mostly about ever-expanding investments. Here's a good example from a seed stage investor: https://www.youtube.com/watch?v=Wi3PiZsIfBU

The very first StreetView was a relatively slow-paced one-two hour tape with Larry, Marissa, Schwim (I think) and a fourth (?) person driving around town in 2000 or 2001. Larry then took the tape to Stanford and a while later the project got funded by Google. Source: I saw the footage years ago. If the project hadn't became what it became, the VHS video would have been a truly unremarkable one like millions of others before and after it.

There was a Virtual Paris for Mac which was the same as Street View.

Also, Encarta and such had 3D views too.

Great projects that work come from tinkering, not strategic or business analysis. With business analysis you get stuff like Meta.

I doubt one would say Facebook or Oculus came out of "business analysis", right?

It's surprising it took Google so long. I remember "street view" projects like this when the Web was in its infancy back in 1994. You couldn't spin freely around, but you could go forwards, back, turn left/right etc. I was planning one with my home town, but it burns up a lot of 35mm film and you had to then find a good color scanner (usually handheld then) or use Kodak Photo CD to get the images into the computer.

Google were standing on the shoulders of hundreds of prior projects. They just did it on a scale no-one else could muster.

One of the neatest yet subtlest features in my experience is going back through time to see the streetview for a given position over time. It’s especially interesting in Boston’s Seaport or other areas that have seen a lot of development in the last ten years, where you can see the changes over time.

I’m not sure where the option is on mobile (or if it’s even available), but it’s a great feature on desktop.

The blog post details that viewing historical images is now available on mobile

Ah sorry I meant mobile web, where I had checked since I usually use desktop web. Thanks for calling that out!

It's been such an incredible discovery seeing what my neighborhood looked like in 2008. You can quite literally go back in time with this technology and it's nothing but astonishing.

Historical Street View is the main reason I hope Google is still around in a hundred years. It is the single most comprehensive record of the built environment ever assembled, and its simple, broad accessibility belies the depth of its value as a historical tool.

In Chicago, there's no more profound Street View experience than virtually walking the streets of the infamous Cabrini-Green housing project in its twilight of 2007, just as the last tenants had been moved out and the towers were beginning to come down. Even through the grainy, first-generation Street View camera, the haunting specter of neglect and despair comes through clear as day. Jump forward a few years, and the towers disappear, leaving greenfields. A few more jumps, and construction fences appear, quickly followed by condos. Less than a decade later, the area is completely unrecognizable.

And now – can you imagine what it would be like to look back fifty years in Street View? A hundred?

Completely agree. Reminds me of a longer-term https://1940s.nyc/:

Between 1939 and 1941, the Works Progress Administration collaborated with the New York City Tax Department to collect photographs of most buildings in the five boroughs of New York City. In 2018, the NYC Municipal Archives completed the digitization and tagging of these photos. This website places them on a map.

If there's one product that showcases Google's strengths and Apple's weaknesses it's Street View.

Google has mapped/streetviewed the world twice over, producing useful photo's that help in navigation. And now is building a new camera so they can map the Amazon as well.

Meanwhile Apple is juuuuust about ready polishing and color grading their 20th run of 4K "Look Around" footage of 5th Avenue in New York, fingers crossed this time it will be production ready! You can expect footage of your medium-to-large city any decade now!

I live in a small town and look around on Apple Maps is everywhere. What’s wrong with it in NY?

Bold of you to claim that Gmaps is actually updated just because they say it is.

Brisbane (the Australia one) has a 3D satellite map from 2010 despite the 2022 copyright date on it. And if you walk along the street view images, it tends to tell you you’re currently viewing the underground tunnel you’re over rather than the surface street you’re actually looking at, which is a funny bug.

There must be something going on with the tunnels in Brisbane in google maps. I recently looked up directions to a restaurant in the valley and they were leading me to the clem 7 directly beneath it.

Just for the record, there are two dates: the copyright one (useless BTW) and data recording one.

You don't think the fact that Google Maps had a 10 year lead makes a difference?

The initial iPhone was feature-poor, yet everybody seemed to love it (don't ask me why though).

Look Around is available on 20% of earth (area, not population) - most extreme example I can find is the middle of Australia: https://i.imgur.com/ZHYbh58.png

In the US, it's actually more limited than in other areas. It's certainly true that Google is ahead, sure - but Look Around isn't just 5th avenue of new york city.

One area lookaround does best is spatial navigation. You can click on bridges, lanes, etc. to jump to them and the animation is MUCH smoother than Google Maps.

Surprised to see no mention of Geoguessr yet - awesome game and community built on top of Street View - https://www.geoguessr.com

I love Geoguessr but it has a distinct randomisation problem (certainly in the competitive pool) that it hasn't solved.

If you play regularly you'll begin to notice that around 1 in 10 cities are in fact Vienna.

I'm not sure how it chooses, I suspect it first picks a country, then picks a city.

Without weighting for population or other factors, you end up with a significant number of Taiwan, Singapore and Vienna, far more than their populations or areas should suggest.

Which game mode are you playing where you have that issue? I've played countless hours of Geoguessr and haven't noticed that

The competitive mode in which you try to guess before your opponent, it's the standard mode now. (They've rebadged the old mode "geoguessr classic".)

Man, since when did you have to make an account to use this site? I've played it before and this definitely is a new addition. I just simply refuse to make accounts for every little game I spend 5 min playing on the internet.

It's been there for quite some time. I believe the street view API is really not cheap for them to use, and the accounts are there to enforce the time limitation.

Ah, that makes sense. Thanks for the explanation. Guess it's been longer than I realized since I last played.

Gee, it sure is lucky that no one could just create another account to get around the time limitation!

Discussed a long time ago:

GeoGuessr: Guess the location from Google street view - https://news.ycombinator.com/item?id=5748923 - May 2013 (97 comments)

Geoguessr - https://news.ycombinator.com/item?id=5682367 - May 2013 (2 comments)

GeoGuessr: Let's explore the world - https://news.ycombinator.com/item?id=5572324 - April 2013 (2 comments)

For those interested in the history of such software, I can recommend "The Billion Dollar code" on Netflix.

It's the story about the team behind TerraVision, a predecessor of Google Maps.

Though the story is not depicted completely accurate, it gives a great hacker vibe from the early days of computing. Can recommend.

...meanwhile, StreetView imagery for Germany has turned 14 - i.e. in much of Germany (major cities only), StreetView still shows the images from the original drive-through from 2008. After that (and a series of legal challenges which led to a ruling saying that buildings have to be blurred if one inhabitant objects to it being visible on StreetView), Google has apparently lost interest. The "3D buildings" feature, which is not affected by this ruling for some reason, gives a much better overview anyway...

In Slovakia, the StreetView is quite up to date but curiously the satellite/airplane photos are about 10 years old and as a result, some neighbourhoods are hard to recognize.

Last summer in berlin I saw several ‘streetview’ like rigged cars, amongst them a Apple branded car with dslrs(!) mounted all around on a space frame... prolly autonomous driving tests though?

it could be just cars checking paid parking, quite common in Prague, it looks like this here

https://cdn.forbesmedia.cz/images/eyJ1IjoiXC91cGxvYWRzXC8yMD...

I don't think they've given up - I saw a streetview car driving through my village a couple weeks ago.

Zoom out to an overview of Europe. Drag the street view-guy. Be amazed.

My city's street view is gone, for one.

They did give up. They are driving cars there for years but only to update street information, they are destroying the pictures taken.

I like Street View and I use it regularly, but it's an interesting case of privacy vs speech/expression. I can see why the Germans might not want Google wandering around and aggregating pictures of all their homes.

> The "3D buildings" feature, which is not affected by this ruling for some reason, gives a much better overview anyway...

Depends what you use it for.

For example I often use Street View (in Europe) to find a street with enough free parking before I drive somewhere. You need photos for that.

It is interesting that there is a large number of consultants (I.e: low wage workers) working behind the scenes to support G Map [0]. Cognizant is their employer which is another surprise for me as I always thought it is an Indian company mainly employs people in India for outsourcing projects. The gas price is so ridiculously high in WA now that they are asking to postpone the RTO full time plan in June. I have switched to Apple Map since 2018 and still don’t miss GMap.

[0] https://www.nytimes.com/2022/05/23/technology/google-maps-wo...

I'm curious if you've seen anything about Apple's equivalent. My impressions is basically every mass data set has an army of low paid workers somewhere, but maybe not and specifics are always interesting.

| high cost of housing in Bothell

I've been away from the PNW for a long time, apparently.

Reading further, one person spoke of a 70 mile commute from Olympia; another of 50 miles from Puyallup, and how they'd have to quit because they can't afford the commute.

For those not from the area, they're commuting from one less-densely populated area, ACROSS a densely populated area, to another less-densely-populated area.

Of course, we all have reasons to live where we live (family connections, other employed members of the household close to work, etc). But in this particular example, the commute has basically nothing to do with housing costs. If it was about costs, they could live more cheaply much closer to Bothell. It's all about commute choices.

I'd love if they improved navigation while in Street View.

It works okay when the geometry is simple - like on a long road/street with no junctions. But in more geometrically complex areas - for example, in urban settings around squares or on wide boulevards which have multiple paths - it's frustrating. Often I find the only way to move to where I want is by going back to the map view and clicking somewhere on a nearby blue line.

Same in areas there are no roads/paths but have lots of user submitted panoramic photos - common in historical areas - for example around Roman/Greek/Egyption ruins - you have to go back to the map.

I'd love for them to integrate NeRF with street view to allow free fly

https://www.youtube.com/watch?v=6lGMCAzBzOQ

Love streetview through a VR headset. So many empty museums to see, so many places to visit around the world, except Germany

Street view is almost required when booking Airbnbs now. It's incredible how often the Airbnb photos look great and then you pull up street view and realize why the place was 20% cheaper than similar stays...

Airbnb's don't have exact addresses and usually no exterior photos. How do you find them on street view? I'd be interested because it would help me as well.

Well they typically give a rough location. I'm mostly looking to see if the neighborhood seems okay or not.

This tech is amazing as far as the mapping effort in a random poor country like islands.

I wonder if somewhere out there like a supercomputer they have a 1:1 mesh/depth map of the world ha as they've mapped it.

On a side note I also applaud/miss their 360 photosphere app that tied in with the g cardboard.

Curious that no one else has mentioned what I find most addictive -- the unfiltered view of anthropology at street level spanning most of the world.

Local fashion and architectural styles, history preserved or not, public safety infrastructure, private wealth versus public wealth, rule of law or local strong men, trust in law enforcement ... Clues everywhere, if you take the time to look.

I had a project collecting photos from street view.

https://instagram.com/noplacetosit

I stopped using insta because the format was wrong for this project. One day I'll make a dedicated website for it... one day.

I remember in 2009 being part of an online community where I preferred to be anonymous and while reading a thread where people were sharing photos of sunsets, had the thought that eventually, there would be enough information out there that my precise location could automatically be determined based on me taking a picture looking out at the city from my balcony. That's a pretty creepy thought, and it seems that google is approaching it, given that they can have you point your camera "across the street" to determine your location from their glorgabytes of street view imagery.

Geoguessing has been going on for a while, and some people are very good at it.

One of the situations where it’s deployed for the good of all is real estate: it’s very common for brokers to hide the exact location for various reasons, but there are forums and individuals who from just a few pictures pointing outdoors and a rough location (e..g closest city) can very quickly find the exact location the picture was taken from.

Super useful so you don’t have to waste time visiting in order to find out that the house which looked good over the internet is sandwiched between a 6-lane highway and a pig farm, or is in a gated community, or can only be accessed through an easement.

Yes, I intentionally used the word "automatically" to my description, since it's very obvious to me that it's already often doable if a human is willing to engage their brain and do research. Whenever I use airbnb, I make a game out of finding the actual addresses of the places I'm considering based on clues I can see in the photos and approximate maps on the site.

I also find locating images an interesting skill to develop. I originally only wanted to mention https://old.reddit.com/r/picturegame,* which I find more "organic" than Geoguessr: players submit single images, which are identified using whatever means are available. It has a quirk in that often people submit images found online, which must be obfuscated to prevent trivial reverse image searching. A randomly transformed translucent mask is recommended, but those are still relatively easily reversed! I haven't tried, though--maybe there's greater difficult than at first apparent.

* It turns out, after an HN search, that this is a more common organized activity than I'd thought; this thread lists several other communities:

https://news.ycombinator.com/item?id=17692172

I've been doing something like this privately for my own picture library – because my backlog of pictures-to-be-geotagged is about 15 years, I don't necessarily remember all the details where exactly a picture was taken. Of course I could just drop the location marker in roughly the right location and call it a day, but pinpointing the location as accurately as possible is usually too tempting.

> […] there would be enough information out there that my precise location could automatically be determined based on me taking a picture looking out at the city from my balcony.

In Japan someone doxed a singer by zooming in on the reflections that were in her eyes from a selfie she took:

> A Japanese man accused of stalking and sexually assaulting a young pop star told police he located her through the reflection in her eyes in a picture, according to local media reports.

* https://www.bbc.com/news/world-asia-50000234

* https://petapixel.com/2019/10/14/attacker-used-eye-reflectio...

I'm curious to know what became of the case.

IIRC he was lying and it was actually in the EXIF data.

Cool. I guess something that 'mundane' didn't make as much of a splash as the initial report.

People have triangulated down to the exact apartment balcony from a picture taken from it, multiple times. Google maps helps in that it allows someone not from the area to do it, but it's always been possible if there's enough information available in the picture (building skyline is the main one).

The key thing I was talking about was the difference between it being possible to deduce the location with enough effort and it simply being automatic. Google has automated it in one specific scenario, which is that you are looking at something which is substantially covered by street view imagery taken from a similar vantage point.

There's a fascinating technolibertarian metaverse dynamic to Minecraft anarchy servers, where gameplay is driven by an arms race of normalized cheating. Players are careful about what screenshots they share online, as this concern is the reality there. There's kind of a similarity to a future real reality where the entire Earth has been autonomously scanned: the terrain of the entire map of a Minecraft server is known to all, since the seed for procedural world generation can be reversed from nearby terrain, and reverse engineering of the world generation has resulted in several methods of determining in in-game coordinates of a player taking a screenshot (terms to search: trees, texture rotation, bedrock).

Related are the impressive endeavours that determined the seed of the screenshot used in the game's splash screen and the seed of a 128px (tiny) screenshot used in the UI. The top comment of the first link explains it better than I could here:

Minecraft's “Pack.png” Seed Reversal Methodology https://news.ycombinator.com/item?id=24621303 – 2020-09 (97 comments)

Minecraft@Home https://news.ycombinator.com/item?id=23895789 – 2020-07 (75 comments)

There've also been few server exploits for locating other logged-on players:

Nocom – 2b2t Minecraft server exploit using Monte-Carlo localization https://news.ycombinator.com/item?id=29615428 – 2021-12 (80 comments)

It's interesting how the anarchy server hacking scene is sort of like science, except with the god-like abilities of reverse-engineering the game's binaries.

edit: fix cut off sentence

> technolibertarian metaverse dynamic

What?

Inspired by 1) an HN title claiming that Minecraft servers are the/a metaverse[0] and 2) the old Net philosophy of minimal regulation of technological advancement. Treating a server as a society, political and economic concepts apply quite well, including the advancement of the "technology" of gameplay: strategies, in-game inventions, client-side modding, etc. This applies to every game, but Minecraft in particular is known for its extensive unintentional emergent gameplay.

And server owners--the government--typically draw some line before major glitches and definitely before blatant the cheating of server-side exploits, banning players who do not obey the rules. But the owner of an anarchy server explicitly does not do this, and instead treats any advancements in even cheating technology as legitimate in-game technology. Hence, a technolibertarian metaverse.

[0] https://news.ycombinator.com/item?id=29083271

>...there would be enough information out there that my precise location could automatically be determined based on me taking a picture looking out at the city from my balcony.

Not yet automatic, but the folk at Bellingcat are impressive at figuring out location from pictures and videos (manually, I assume).

AFAIK Google Maps for Android now supports calibration by using phone camera.

Like Google search and maps, Street View quality seems focused on large towns, cities and areas of high population. I only live a few miles into a large rural area from one of UKs largest conurbations and our village is 13 years out of date for street view, and 9/10 years old for Satellite imagery. Google maps is missing most of the nearby small single track unpaved roads and about 75% of footpaths (all marked on definitive maps are public rights of way), its so bad there is a 2 mile long, 1 mile wide woods on the edge of our village that is completely missing (including the road and footpaths through it), there is just a big blank, yet Street View clearly shows the entrances and the woods behind.

Doesn't it make sense to you that such data collections would be population-weighted?

It make sense for Google to steer their products, not for generalised mapping.

Our small town gets a "pass" every once in awhile (amusingly enough you can be on one side of the street and see the older view as it's clearly summer, the other side of the street is newer and you can see the new gas station in winter).

I suspect that for smaller towns they have some sort of trigger based on things like new gas stations opening, etc, that causes a car to go by eventually.

Last year I was living in a small village in the UK. Interestingly, street view for my street there was updated about 8 months ago so perhaps it will get to you!

So does this mean they are going back and updating old imagery? My neighborhood (in a small city) hasn't been updated since 2011, and quite a few areas have street views that don't match reality anymore

Looks extremely simple for anyone to do this for google now, so I'd guess yes

I think that's what they are doing with the live view street directions. They can easily recreate the new imagery based on the users photos.

I very much doubt it. The entire point of that is to gauge the location of the user based on the existing street view imagery. Perhaps they'll add something eventually where users can contribute imagery from their phones, but I think that feature won't be the way to do it.

They already have user contributed imagery. In parks, Google usually maps the paths but you can see dots off the paths for user added points.

Looks like streetview of center of Helsinki was last updated 2009. Granted it's not very big either but capital of a european country still.

That has been happening all along, but I would expect not everywhere.

Does anyone actually use street view for anything but novel purposes? I've used it maybe 5 times when I couldn't find the place I was looking for using the standard map view; However in nearly all cases, resorting to StreetView meant the place I was looking for was gone anyway.

Yes, almost daily. When I'm going somewhere new, what does the front of the building look like? Where is there parking? How do I access the parking? What are the parking fees and restrictions? How visible is a landmark from the street? Which street is the entrance on? And that's just when I'm planning a trip, one of many use cases.

It was helpful in my truck driving days when going to customers I'd never been to before, and it was absolutely a daily tool in telecom construction/maintenance.

Example - random sheriff's office calls in a phone line ripped down when I've got one foot out the door for the day. I pulled up the address on Streetview while she was explaining the scene to me, and it sure looked like coax on that side of the street, our lines were on the other side. Sure enough, 45 minute drive later I called the sheriff and told them to call the cable company.

When you cover an area hundreds of square miles you can't go look at every job before sending a crew out. Pull it up on Streetview, even if the images are 10 years old it's probably still the same.

Yes, all the time. Before going to a place to see parking, how it looks, etc

Not too long ago I had a leak that I suspected the previous owner had repaired. I was able to confirm with street view pictures that it's been repaired and patched over at least three separate times since 2008!

I recently used it and its 12 years of history to demonstrate to a commission of the City of Berkeley that the same handful of cars were perpetually parked in free, no-time-limit street parking spaces which local merchants incorrectly believed were useful to their customers. This is good data that can change outcomes for city planning and regulation.

I often use it with family members who lack the confidence to find a location that they need to get to, and the skill to read conventional maps. "Walking off" the route in StreetView beforehand is nearly 100% successful.

I use it all the time in civil engineering. There's always some question I have for which we didn't collect data or photos about and before I send someone out to look, I check streetview.

I also use it to preview a route when navigating from one place to another in unfamiliar locations. That way, I typically don't need any further instructions, the guidance ends up being review anyway and i don't need to worry soich about poorly timed or ambiguous instructions.

Work in a company that handles delivery - Customer service often uses it to see properties for deliverability concerns.

Sometimes I just load up google maps, drop the streetview pin in an arbitrary spot, and click around looking for something weird or interesting. I always wish they'd add more countries, especially in Asia and Africa but I'm sure the business case for "guy likes to idly click around and wants more countries" isn't very strong.

I would kill for a driving simulator with streets and texture data from Street View!

Microsoft Flight Simulator 2025: The One With The Ground Planes.

Google had a very productive 5 year period where every year they were releasing massive groundbreaking products that were 10x better than competitors: Gmail, Maps, Street View, G Suite, Google Earth, Translate. Since then I'd struggle to count even 5 products that stand out: Android, Chrome and what else?

I’m continually jealous of the street view dataset and disappointed there’s no public/free alternative.

They provide API access but it’s very hamstrung and extremely expensive, making it essentially useless for anything except pulling up a single image of a location. It’s useless for any real mapping/localization/analysis project.

There’s Mapillary (acquired by meta but imagery is free to use in most cases) and KartaView (Grab).

Both are crowdsourced platforms.

https://www.mapillary.com/ https://kartaview.org/landing

I didn't know about these!

Still, since they're crowd-sourced from phone data, they only have a single view (no 360 in most places), inaccurate (or at least not reliably accurate) camera calibration, no accurate localization, and possibly most importantly, no LiDAR data.

That's a lot of things that make it much harder to deal with and limit the potential compared to street view.

Incredible stuff. The various Google properties have performed an incredible objective of recording our history. One could imagine the ad supported empire as this personal consumption tax that funds human information recording. Glorious.

I remember where I was when I first used streetview (and I have a generally bad memory for things that long ago). Was such an amazing feeling and a real living in the future kind if thing. More so than using an iPhone for the first time

I wonder how much Streetview has changed the shape of buying real estate remote. Used to be a time you couldn't just see the history of a neighborhood and condition of an area over time.

It definitely helps get a "feel" for the neighborhood but it can often be out of date.

I've noticed some real estate listings now include drone shots, perhaps to compensate.

Dang, I've actually never thought about this but it is such a powerful tool for checking how the neighborhood looked like before.

Does anyone remember when Amazon (briefly) had a Street View competitor? Having a hard time finding any trace of it on the web.

Will they now allow mobile users to see older street view images? At the moment it's only possible to see this on desktop.

From the article:

> Starting today on Android and iOS globally, it’s now easier than ever to travel back in time right from your phone.

So yes, this is possible now.

Now if only they could add that to the Desktop version of Google Earth, too…

(Weirdly enough, historic aerial imagery is only available in the desktop client. You can also look at current street view images there, but for historic pictures you then always need to pull up the same location in the browser…)

Great, now can we get it to scale with processing power. It was somehow more usable 10yr ago on most hardware...

Thinking about driving without google maps sounds so stressful.

I strictly use OsmAnd~ on Android and it is a pain to use, specially in states where openstreetmap.org don't map street addresses like in Georgia. I often have to convert addresses to latitude/longitude first...

And still no way to un-block street view from my address. It's the only house on the block that is blurred, which makes it _less_ private. (The previous owner had it blocked.)

Can you explain why it's less private?

Is Street View still illegal in Germany? Are there places in the world that haven’t been co-opted by surveillance capitalists?

Most major cities (where most of the population lives) have been mapped out.

I don't think that's a fully correct characterization. Those scenes that exist are still from 2008. They seem to have given up after the initial storm. And I'd not deem it completely out that another protest wave would start should they try again.

It's old data. But the 3D view from the plane cams is a good enough replacement at this point (where it is available), if you want a better view of an area than a satellite gives.

Apart from just being nosey, i find street view helpful when driving to an address you don't know, for picking out landmarks etc. The 3D view just about makes up for the lack of street view in that regard.

I find street view useful, but I also think it's a massive invasion of privacy and if it were banned I would not be upset.

It has never been illegal. Public opinion against it was stronger than Google and they have obviously not dared to try again.

if the Asian couple was always happy in their house why did they move out?

Reply on news.ycombinator.com