The world of fashion forecasting has become a crowded market of late, in part because of the changing technologies that have become

available in the past decade or so, and in part because it's a lucrative market, especially for

tech-savvy folk who are capable of putting together a solid back end of services for fashion

and fashion-related companies, and for fashionistas who want to get out of the making of clothing side of things,

and into the B-to-B side, selling services, selling panning equipment to the gold miners,

rather than mining for gold themselves, essentially.

Fashion forecasting is a business with three main components.

First, they capture data. second, they crunch that data, and third, they make that data available in useful ways. To capture data,

they employ and hire tens of thousands of people to go out into the world and photograph

fashion and style as it happens in real life.

And in some cases, to interview the people wearing the clothing they photograph.

These forecasters attend conferences and music festivals, fashion runways and red carpets,

they walk the streets of New York City and Paris to see what people are wearing, their photos and any

supplementary information they gather is then added to a database of other photos and

information and tagged with metadata, making it organizable and searchable. In recent years,

this information is also often combined with the same types of photos that

are instead collected from social media, especially networks like Instagram, which allows

forecasters to find and capture a lot of the same information, but from wherever they happen

to be, rather than out on the street or at a festival.

This is where the crunching of that data comes into play.

Now that all these photos from around the world, organized by demographics and geographic

location and type and time period and color, and all kinds of other identifying information

...