I wish the press release had a bit more detail about what this model actually does and whether it's actually useful for the suggested use cases.
However, make no mistake: this is for the scientific community and will not help geospatial data to be commercialized. No one cares about your geospatial crop model or that you can identify energy infrastructure or that there's some activity around that copper mine. Well, at least no one cares that will actually pay you.
(FWIW, I cofounded a geospatial analytics company)
Satellite data is extremely idiosyncratic. It's coarse (~10m at best), infrequent (every few days at best), and oh you have to deal with the fact that the planet is covered in 50% clouds at any moment. Satellite data works best on things that don't move, that are fairly large, and change infrequently. If you find a use case that satisfies those conditions and want to make money, then you need to find a problem that terrestrial sensors haven't solved. And if you find that problem, the cost of building, training, and running your model (plus the cost of the data!) has to be less than the marginal value of your model. Good luck finding those use cases.
The US Government is special. We don't know what's going on in North Korea or Ukraine or the South China Sea so we buy high resolution imagery over those areas (30cm) at great cost. Large ag companies and oil companies know what's going on within their own facilities; and price gives them information about the rest of the supply chain.
In other words, this might be an interesting announcement for scientists, but it won't change the geospatial market at all.
And it took Amazon nearly a decade to get back to their internet bubble highs. PL got SPACed onto the stock market during the SPAC bubble.
That being said, while I do believe that Planet has one of the best business models in the industry, I do sometimes worry that they are a bit early, and their customers aren't ready for them yet.
As someone who has to work with their data (among others), Planet has some of the best APIs in the industry (it's a low bar though).
Not really seen as an AI play. The bull market has been mostly driven by FAANG + Nvidia.
I'm mostly just playing devils advocate here, and the point I'm trying to make is share price alone isn't everything. Planet is growing its revenue YoY, and they might even be profitable soon (lol).
Also, my original point was that the company I work for buys a lot of data from Planet, which is something that's just increasing as we grow.
>Satellite data is extremely idiosyncratic. It's coarse (~10m at best)
>The best commercially available spatial resolution for optical imagery is 25 cm, which means that one pixel represents a 25-by-25-cm area on the ground—roughly the size of your laptop.
I'm an ML engineer working for a geospatial company and I can assure you, we are looking into this.
> Satellite data is extremely idiosyncratic. It's coarse (~10m at best)
10m is the best free imagery, in the commercial domain it goes down to 30cm.
> Well, at least no one cares that will actually pay you.
There are plenty of things people will pay you for. But you gotta find those niches.
> In other words, this might be an interesting announcement for scientists, but it won't change the geospatial market at all.
Maybe, we'll definitely check if it can be fine-tuned on higher res data.
We do sometimes use Sentinel-2 (not a lot though), but can help with those cases.
I think it is very important for people to understand that terrestrial sensors are orders of magnitude cheaper for most applications, and are typically far more accurate too. There's a reason why most remote sensing companies go out of business fairly quickly.
> Satellite data is extremely idiosyncratic. It's coarse (~10m at best), infrequent (every few days at best), and oh you have to deal with the fact that the planet is covered in 50% clouds at any moment.
Are the coarseness and cloud aspects going to become less of a factor now that there are commercial high-resolution synthetic aperture radar imagery providers? I'm just a hobbyist, but the imagery I've seen is <i>sharp</i>, and it even caught the NRO's attention.[1]
InfSar (InSar, SAR, whatever we're calling it these days) isn't a drop in replacement for anything. Its really neither here nor there when it comes to the utility of other dataset. Infsar is amazing, dgmw, but its stands on its own and has its own advantages/ disadvantages.
The ocs point stands. Satellite data is tough because there is a shit ton of atmosphere between you and the target. That issue doesn't go away with infsar and especially not if it isnt coincidentally collected with higher resolution spectral data. I've been in the industry for around 15 years. Things have gotten better, but really, its important to understand the context and limitations of specific platforms. Afaik, there is no panacea.
Super interesting. Hadn’t heard of SAR before. Quickly reading about it, it seems like it works like lidar, what’s the difference between the two techniques? Is SAR like “lidar for space”?
If I'm correct and philosophygeek is Mark Johnson, he cofounded Descartes Labs. It was a pretty cool company with some quite impressive technology. He (they) did a lot.
I'm not far from bashing the VC scene and the adjacent startup culture, but your overly cynical comment was too much even for me. More intellectual humility and less cheap soundbites would benefit society a lot.
Hey, I'm the founding CEO who wrote the article. I hold some of the blame here too for trying to read the market and be something that we weren't. When you hear lots of people raising at 20-30x their ARR, it pushes you as a founder to try to paint a picture of your company through that lens. However, I currently don't see the possibility of anyone in geospatial being widely-used software.
Has anyone tried to write an algorithm that translates any piece of text into "basic English." That seems like a tractable problem, since it's just mapping all the words in English and grammatical structures into Ogden's 850 words and rules.
A language translation target would be kind of fun. I wonder how much of modern idiomatic english could be successfully machine translated . . . or even translated using meatware?
A scientist at Los Alamos National Laboratory trained a GAN to test automated change detection algorithms in satellite imagery, with a link to the associated SPIE paper.
We've had 10 years of frothiness in venture markets. Many employees under 30 don't even know what a downturn looks like. They're used to high pay, constant calls from recruiters, and venture capital flowing into their companies. Even the companies who flop, no big deal, just go down the street to the next company. I worry that that generation is going to have the most difficulty with these changes. They'll wonder why things are contracting, why difficult decisions need to be made, can't we just wait this one out...?
I've never seen anything like COVID-19. I lived through the dot-com bubble, watching friends who want from an IPO party to crying and packing up their car back to Iowa within 6-months. Tech contracted and survived. Then there was the 2008 financial crisis. In that case, the damage was spread out beyond tech centers. Capital markets dried up and life seemed very uncertain for awhile.
With COVID-19, we went from extreme prosperity to an unprecedented shutdown of global economies with massive uncertainty as to how long this will last. Every business will be affected, the vast majority of them negatively.
I really wonder whether everyone is prepared for what it means to live in scarcity instead of abundance.
“We've had 10 years of frothiness in venture markets. Many employees under 30 don't even know what a downturn looks like. They're used to high pay, constant calls from recruiters, and venture capital flowing into their companies. Even the companies who flop, no big deal, just go down the street to the next company. I worry that that generation is going to have the most difficulty with these changes. They'll wonder why things are contracting, why difficult decisions need to be made, can't we just wait this one out...?“
I went through this in 2002. It took me a long time to accept that the good life in the years before wasn’t because I was so smart but because of a favorable market. It was quite a shock to actually having trouble finding something new and also to accept that my pay was significantly lower. Somebody told me at that time “now you see how job search always has been for most of us”.
We've been trying to hire since last year and have really struggled because rates devs have been looking for on the market are well above what we could see them returning value-wise for their experience level (that's just the nature of the market, I get it). When you have people getting £130k for doing React at a fintech company... well, I just hope people are prepared for the market changes that are underway.
Honestly, I'd rather this was all over and things went back to normal, but I do think that there's going to be a fairly hard reset of salaries for people that previously had a really sweet ride.
I don't know about the national level, but within major cities where tech salaries and Airbnb are both prominent, I'm curious to see what happens to the local real estate markets. I know many people in Austin and SF who can afford their home because they A. have a large salary working in tech and B. offset their mortgage by Airbnbing.
I'm not on £130k but I'm in London and also doing some development aa part of my job. Most of the knowledge I gained over the last few years will be rendered useless in a few months from now because there will be hordes of 10 times more experienced desperate for jobs. So be it. I'll pull out my plumbing tools, I haven't used for years, and start providing different services. If that won't work,will have to figure out something else. This will be a massive readjustment for everybody..
OP did say ‘rates devs’ - guessing they mean contractors who are typically £550-600 per day which works out to around £130k p/a (before NI, taxes, etc).
It's an artificial scarcity, though. We have the most productive technologies ever - we are just not very good to put a better part of them flow towards common good.
The amount of senseless applications and waste is staggering.
Good points. Plus the demand for tech is higher than ever due to the productivity advantage. There are pockets of that common-good advantage being demonstrated here and there, so maybe that can be patternized and better demonstrated to the whole. I'm sure there are immediately applicable ways to redirect a lot of the waste.
For a lot of people who hang out here, this amounts to baseless FUD.
Technology is not going to contract any time soon. Quite the opposite. It's also really hard to build and distribute technology products well. Good software engineers and tech sector employees will continue to be in high demand for quite a while.
There might be an interim period of depressed income but that will also likely correlate with reductions in cost of living, as a consequence of the same macro factors. The key is that a good technology employee making wise career choices will continue to be able to live somewhere towards the front of the standard-of-living curve.
What might be lost for a while are some of the more extreme tail outcomes (in the many millions of $s) that had become a bit too easily realized in this latest iteration of frothy tech bubbles. Not a big deal. Most of us won't notice any difference.
You don’t need to worry about people under 30. Every older generation thinks every younger generation isn’t going to handle things well. And every single time, they are wrong.
I left the Bay Area almost 5 years ago now to start a company that I spun out of Los Alamos National Laboratory. Everyone - investors, my friends & family, our founders, even me - thought we would move to San Francisco within a year.
Four years later, Descartes Laba is almost 100 people strong, $38M investes, and thriving in the Land of Enchantment.
The reality is that if you have a strong mission, a viable company, and a great place to live, you can build a company. Sure, we can't hire anyone from Silicon Valley, but that's ok. Our recruiting team looks in all of the places FANG isn't looking - and we've found some incredible talent.
I hope that it's not just major cities that see this Renaissance of tech, but cities across America and the world. This is how we transform society with technology - by bringing a wider audience into the conversation.
I'm the CEO of Descartes Labs. It's indeed true that we've built a data refinery that ingests lots and lots of satellite data. The data refinery can be seen as a two-sided marketplace. On one side, we form partnerships with satellite and other geospatial data companies (in addition to open source data from NASA, ESA,and others) and pull in all of that data. On the other side, scientists can run computations over huge amounts of data from multiple datasets. For now, most of our business has been done on the scientist side. In principle, we could provide our infrastructure to satellite companies so they don't have to build out the software on their own. Most hardware companies suck at being software companies.
Amazon's offering is geared more towards ground stations, but they might move up the stack and start providing data refinery-type services on top of the ground station work.
Oh, our entire stack is built on Google Cloud Platform.
Can you give an example of "refined" data versus what one might get through the AWS product? ... in order to demonstrate the sort of tech skills and effort that differentiate the two. I'm basically hoping for some symbol grounding for "data refinery".
Also, what's the TAM in your specific market vs the AWS product's market? How has the TAM changed in the past 5 years?
Lastly, I heard your head of engineering brews better beer than any of your competitors. Is that true? Can be provide samples?
Refined data probably means getting insights from the vast amount of aerial imagery (and related like SAR or elevation model) data that has become available in recent years. Having access to the data is one thing (e.g. Landsat and Sentinel provided by Google, Amazon and other parties), but processing it efficiently is still non-trivial.
Examples include Land-Cover-Mapping (mapping pixels to classes like forests, urban areas, water, etc.) which can then further be used to do crop monitoring or land-use monitoring.
I guess this is different than the product AWS is offering here, which is more about getting the data from/to the satellite, but not about processing (at least for now).
Yeah these are classic examples of 'providing business value'. Want to be an agricultural tech company? Ingest some satellite data and calculate NDVI, boom you now know machine learning, data science, and have created a great business product that helps save the world by making farmers better.
Satellite data imo sucks, especially aerial imagery. Too many damn clouds to get anything useful in real time haha!
What a fantastic read. As a CEO who has grown my company from 6 to almost 100, this is one of the hardest discussions to have with someone. It's not that they did anything wrong, they're just no "right" for the position in its current state.
He suggests that it's impossible to retain employees through such changes. I haven't found this to be the case. One technique I've used is to keep the company title-light, so that changes in title aren't as stark. I've also had a few people who have gone from IC->manager->IC, which shows the company that I value people no matter what their role is. Retaining people after hiring above them requires a lot of careful, honest conversations not just with the affected employee, but with messaging to the broader team.
I'd love to know how you approach the messaging to the broader team about hiring above someone (perhaps if they've always been an IC, or if they tried management and weren't performing). In my view it seems quite difficult to maintain face in such a scenario, and it's not obvious how one can alleviate that.
Descartes Labs is building models of complex systems on the planet. Though our first product is focused on agriculture, our aim is to tackle tough, meaningful, global problems with science. To do this, we've amassed over 3PB of geospatial data, growing at over 10TB per day, and can spin up tens of thousand of cores to perform calculations. We're looking for engineers who want to evolve our platform and for scientists who want to build solutions on top of that platform. Check us out!
(I'm the CEO - feel free to contact me directly at mark@[company name].com or apply through our site)
However, make no mistake: this is for the scientific community and will not help geospatial data to be commercialized. No one cares about your geospatial crop model or that you can identify energy infrastructure or that there's some activity around that copper mine. Well, at least no one cares that will actually pay you.
(FWIW, I cofounded a geospatial analytics company)
Satellite data is extremely idiosyncratic. It's coarse (~10m at best), infrequent (every few days at best), and oh you have to deal with the fact that the planet is covered in 50% clouds at any moment. Satellite data works best on things that don't move, that are fairly large, and change infrequently. If you find a use case that satisfies those conditions and want to make money, then you need to find a problem that terrestrial sensors haven't solved. And if you find that problem, the cost of building, training, and running your model (plus the cost of the data!) has to be less than the marginal value of your model. Good luck finding those use cases.
The US Government is special. We don't know what's going on in North Korea or Ukraine or the South China Sea so we buy high resolution imagery over those areas (30cm) at great cost. Large ag companies and oil companies know what's going on within their own facilities; and price gives them information about the rest of the supply chain.
In other words, this might be an interesting announcement for scientists, but it won't change the geospatial market at all.