Hacker News Reader

>> a) These areas are restricted airspace, and most of the aerial imagery Google et al buy is taken from airplanes.

Just replying to point out that this is true. I worked 3 summers for "Aerometric", a company that does exactly this. We flew flight lines from 10-20 thousand feet with a huge camera rig. Those photos were transmitted back to corporate, where they are stitched back together to form large high-resolution aerial maps. Our customers were usually timber companies, gas/oil companies, and occasionally FEMA.

Until that job, I naively assumed everything was a satellite image with some Bond-like zoom capabilities.

What's your career path?

It shoots upwards for about thirty minutes, followed by a feeling of just sort of gliding along for a while. Then it quickly comes back to the ground.

Geospatial anything. Geomatics, GIS, geodesy, surveying, photogrammetry, optics, whatever is related.

Coming from a background in remote sensing, and having been to several conferences where companies desperately try to sell destitute students instrumentation worth hundreds of thousands of dollars...

A lot of remote sensing folks come out of geography degrees into specialist Masters' programs. The use cases are all the various things you'd use high resolution aerial maps for. To be clear most of the scientists are users of this data, rather than the creators. Historically there were more computer vision scientists involved, but we've mostly solved the big problems like image stitching and camera calibration. Then GPS came along and solved the geo-referencing problem as well.

Most academics just buy their data or get it from a space agency, do some analysis in GIS software and get on with answering their research questions. There is still a healthy research field in remote sensing, but fewer people worry about the collection side. There are also plenty of people involved from the space side of things, but again, unless you're on an instrument team you're a consumer not a creator.

So the career path depends. Typically degree in geography with a specialism in remote sensing, then into a career with an appropriate company. Most of the airborne cameras are produced by large orgs like Leica Geosystems. Also look at Hexagon AB who have bought up just about everyone else.

The people that generate this data tend to be small companies who own a plane (or three) kitted out with expensive surveying cameras. They are then contracted to fly over places and they provide calibrated (ortho-rectified, geo-referenced, etc) data to the client. Depending on the budget and requirements, this can include (multispectral) imagery and/or LIDAR mapping.

If you're interested in camera development that spans the engineering gamut from electrical (sensors) to optical. There are also aeronautical engineers and of course pilots. Nowadays if you have some machine learning knowledge, you're probably useful.

>> What's your career path?

Programmer/sysadmin type, the aerial surveying was for fun in my 20's.

> Until that job, I naively assumed everything was a satellite image with some Bond-like zoom capabilities.

Google Maps really messed with people by calling their photo view "Satellite view". Myself, it took me couple of years to realize those couldn't possibly be satellite photos, and to learn that in fact most of them are aerial.

Even if Google and the US military have an agreement not to update images of millitary bases, it's perfectly understandable and doesnt infringe on anyone's rights. I agree this is sensationalism piggybacking on real privacy and government power concerns.

> It's not like they're all buying Android phones and getting turn-by-turn directions in their Humvees.

Err.... you may be surprised.

I've driven on bases before. They're massive and you need directions to get around. Usually the folks on the ground like those letting you on will tell you what turns to make.

Don't forget that satellites in polar orbits don't always cover everything since their orbits can still tend to sync up to the earth's orbit in such a way to leave regular tiny gaps at certain latitudes, hence one of the needs for aerial imagery. If one of the gaps of your favorite imaging satellite happens to line up with a no-fly zone, then you can end up with artifacts such as this

b) You might be surprised. https://www.zdnet.com/article/fitness-app-polar-exposed-loca...

>c) There's also satellite imagery, but no imaging company is imaging the entire globe at a resolution Google and friends are interested in. You need to pay for specific areas, which brings us back to b), what's the commercial interest?

Isn't Planet Labs (https://www.planet.com/) doing exactly that?

Planet is Medium Resolution, not high res like in Google Earth. Plant is roughly 4m resolution where google is sub meter resolution. But yes, planet now has access to 100% earth's landmass daily.

A key component of sensationalism is inaccuracy. What's inaccurate? What's being hyped? They admit their assumption. The questions propose the possibility of deception, "what are they hiding here?" but this isn't inaccurate or hype. It's a question, not a statement, and not misleading supposition or speculation.

If you're going to presuppose the piece is sensationalism, the speculative explanations are themselves sensationalist, because you're making value judgements: the explanations are reasonable, and some assumptions should be off the table. These are misdirection. Instead come in the form of additional questions.

I think it goes without saying there could be reasonable explanations, but why speculate or assume? If they exist, and it's a simple matter, then Google could simply answer. But they haven't. Why not is a completely reasonable question, even if it's a valid opinion to consider it boring and a waste of time.

> This is just sensationalism.

This is just art.

Satellite imagery in the raw form when it is not yet registered to a map has a beauty to it that is lost in everyday mapping products, e.g. satnav and weather maps. Normally satnav and weather maps interpret the image for you. With raw imagery you have to interpret the image for yourself. The imagery may not be in normal 'visible' colours, it can be like HDR on steroids, to reveal patterns and detail only once you adjust the levels/curves to fit the limited range of colours that a computer screen can show.

Perhaps the most fascinating remote sensing imagery is of the areas of the world you don't normally see, e.g. polar latitudes, the seas off coastal areas and the deserts. It is a bit like looking at the stars with a very powerful telescope but in reverse.

The thing is though is that you need to have quite a lot of inner child or actually be aged nine to truly appreciate the wonder and explore the dataset that bit more.

I think that good art has to get people asking questions and thinking a bit. Usually, as per the recent 'Banksy' spectacle (art business) there is a big price tag attached to the 'art'. This 'art' fails to get people beyond the question of 'how much is it worth in dollars' whereas here I think that with this Area 51 inspired 'installation' we have much better questions being asked and answered, as per what you wrote here. It is also good that the star of the show is the remotely sensed image, not the painting of it, no disrespect to the painting which I think is rather special and does the subject justice.