While we are just one industry, it would definitely be interesting to see several different industries overlaid on a common data fabric.

In my town of 100,000 people, there are probably at least 5 places that make plausible indoor maps for my usage. (In areas like DC, that number skyrockets — per-capita-indoor-area counts are huge relative to population, but.) Even if we say that in Cambridge, there’s only two — say, the Galleria and the Harvard Museum of Natural History — that would mean that there is one map per 50,000 people. In total, that would mean more than 100,000 maps worldwide — and so far, in 4, or 2, or 1 year of operation, Micello has gotten 15,000.

(Granted, last week apparently your website listed 6000 maps, based on https://twitter.com/eichin/status/326333064512937986, so I’d accept that there has been a lot of growth recently — but none of the growth affects any of the things I mentioned caring about in my post, other than the Postal Museum, which wasn’t even interesting enough for me to go to.)

In reality, I think the number is closer to one map per 10,000 people — MIT, Harvard, Harvard Museum of Natural History, the Galleria, Mt. Auburn Hospital, both Shaws/Star Markets, Whole Foods. (That’s 8, in a city of 100,000 people.) That would put the number of maps at 500,000 for the world. Your coverage report still says “Meet the Mapper” — maybe you have many, it’s impossible to tell from your website, but it seems like at the very least it’s internal, not crowdsourced. (Which is supported by the the 14:1 bias towards North America.) Even with better software, you might get 10 times better at doing this — which still leaves you a decade away from covering the things I would consider important.

I don’t think this problem can be solved with a small team of people. (I’m not *sure* it can be solved with a large team of people/crowdsourcing either, but I don’t think it’s being tried.) I think it requires involvement of a community.

Frankly, OpenStreetMap has a much better chance of solving this problem than Micello *or* Nokia. The question is just when they’ll get there — it’s only a matter of time.

I know 15,000 is ‘just another number’ but just like the outdoors, it does take some time. Before we know it though, in the next year or two all the interesting places in the world will be mapped out, from the inside!

Ping me at ankit dot agarwal at micello.com if you’d like to chat offline.

There is a lot of money changing hands under the status quo. Changing the status quo won’t really make more money appear, not tangibly.

It’s harder to get rid of a cost like that, because you’re moving the cost from being a smaller per-use cost to an overall infrastructure investment. It’s not a lot different from the open source argument — yes, a successful open source project may benefit everyone involved, but in the end, *someone* has to step up and do the initial investment — and without a promise of a reward (if the project fails, etc.). I don’t think we’re likely to see a change in this status quo any time soon.

I’m back in the property industry at the moment, and slowly mapping *every single house* in Australia. My workplace has valuers, who visit and measure, with laser distance measurement tools, a house every day. There are many who are in the same market, competing with us, doing the same measurement process.

This seems silly. The majority of our worth is to determine what a property is valued at - so why can’t we spend 5 minutes less measuring, and more time assessing, based on shared, open map data of particular spaces?

OpenCV offers a way to also reduce the cost of map maintenance - recognising shapes from aerials can surely be automated, much as google has already done.

Why can’t we link the building industries, architects, and all of the property industry with the transport, delivery and anyone who routes from A to B for commerical reasons in a common data fabric?