Hacker News new | ask | show | jobs
by teamskyfi 1250 days ago
SkyFi team here. We did fight hard to make the minimum size of the image lower than current industry standards. Many use cases don't need large swaths and it helps bring down the minimum price – making it more accessible. We also have one individual EULA for all of our data providers which is not currently standard for the industry. We are working on leading the Earth observation industry towards transparent pricing. It makes it a lot easier for the customer, which is our primary focus.
3 comments

Just wanted to congratulate you on this. I know businesses like this are hard in that it's a lot of slow, hard work to bridge (entirely reasonable) user expectations with the (also reasonable) grubby details of technical realities and business practices.

I know my first reaction was to scoff because it didn't do the obscure thing I wanted in the way I wanted. But my second reaction is to appreciate how much of a step forward this is. Best of luck on your hopefully numerous future steps.

Thank you. Remember this is version 1.0. If you recall version 1.0 of your favorite apps they are significantly inferior to their current version. We have much more work to do and hopefully we have all the features you desire. Many more sensors and features to integrate in the next 6 months.
It is pretty cool and I was looking for a picture of my grandparents place as a gift they wouldn't buy since they don't know it can be made, but the mininum 5k area neglects that purpose. They aren't technical at all, but the idea that something so advanced as a satellite could take a picture of their house would blow their minds.

I completely understand if my request is impossible, but at least one other commenter mentioned this idea in the thread, and I think it would be a pretty common thing.

One other point. Would it be possible to subscribe to an area and get notified when photos become available?

Actually a final point: on the website it mentions the technical resolution of the images. Could you have one example of each size photo that I can see? 500cm doesn't really mean anything to me, nor does multispectural.

Why is the 5 sqkm min area a problem? If you're getting 1 px per meter, it doesn't matter how much irrelevant area gets captured around your area of interest (and existing imagery prices are low enough that it shouldn't be a problem).

That said, I suspect Google Maps and other public mapping services likely already have higher resolution pictures. Like you said, I also can't really imagine much under the "100 cm" description, but zooming in on a random place of middle-of-nowhere, Alaska, I can clearly make out the triangular shape of tree shadows that measure around 6 meters length-wise, so I assume the resolution is better than "100 cm". Middle-of-nowhere Siberia was worse, but in a random 360 people village I could clearly distinguish left and right tire tracks.

I only see very few benefits a service selling historical pictures would provide for curiosity/novelty/hobbyist use cases - specific times (including newer imagery) that aren't available in the Google Earth history, getting the picture officially and without watermarks rather than having to screenshot or otherwise extract it, and maybe some edge cases in terms of areas covered.

Being able to request a new picture is much more interesting, but I suspect at the resolutions available, it won't be too useful either (edit: again - for curiosity/novelty/hobbyist use cases, for which pricing will also be a big hurdle).

An existing image that is recent may cost 20 to 30 bucks at 5sqkm. Perhaps that would work. Existing images might only be a week old.
The min area for archive is 5sqkm or about 20 to 30 bucks. The feature of "notify me when X is avail for archive" has been discussed and is on the product roadmap.

We will have blogs and customer education on Hyperspectral. It's quite amazing.

Thank you for working to unfuck this industry. I often have to buy building or block level imagery and it is an absolute nightmare.