We won't discuss the industry.
Recruiters seek out the people we place, and we find them anywhere they are. It is what we do - especially good recruiters - we leverage information in order to reach out to the folks who are doing what is in demand. We find them where they are in order to put them where we need them to be.
That is how it has always been and that is, I suspect, how it will always be until there is no longer a need for us, and oh how I hope I don't see that day.
Currently there is a lot of information out there for the parsing. One might say there is more information on all industries readily accessible to our grabby mitts than there ever has been - thanks to social tendencies, cloudy skies and a growing emphasis on analytics and data. A savvy recruiter in most any industry can find prospective candidates out there on the internets doing what prospective candidates do - which is working the job they currently have.
Technical recruiters can peruse places like github and stackoverflow to their hearts content, and other industries and specializations also have their special places. While it is incredibly helpful, there is a dark side to all of this readily available information. Sometimes recruiters can forget that all of these social profiles or online work examples are not replacement resumes.
A resume, in whatever format, is a carefully constructed argument for consideration. Even without a cover letter it exists in order to be examined in relation to a need. This is different even than an online portfolio or presence which is more a location for activity as it is an argument for consideration.
I've seen a lot of discussion online from developers regarding how employers and recruiters use of github and stackoverflow in hiring, much of it is pretty unpleasant, and deservedly so.
Not only are many recruiters pretty hamfisted in how they use information on these sites, when we pretend that places like this provide accurate representations of individuals we are missing something important - their own ability to curate and position. It isn't just about being able to read code, we miss the context for the code (or discussions) that we see, and if we think that isn't relevant information we are wrong.
I've seen managers become less interested in candidates who didn't have a github profile or whose profile was pretty empty, even though there may be perfectly good explanations for there being a lack of code they can share publicly or semi-publicly, or even though a lack of code on a public site doesn't mean they won't be a good employee. Sometimes that disinterest continued even after code samples were provided in other ways.
I've seen people pass on candidates because they didn't like the kind of code that was contributed, or didn't see enough of something they wanted to see, or too much of something they didn't like. The conclusions we can jump to when looking at prospective candidates doing what they do online are actually pretty amazing - and many times can be terribly off. We don't always have the context for content we find online. We can't be sure that the content we find has been positioned for eyes like ours. In a way we are interlopers. We are going out to the places where folks do what they do and seeing them in the wild, so to speak, and some of us, sometimes, think that read in the wild can replace a resume or cover letter. I think we need to be careful. If it is possible for us to misconstrue carefully constructed information on a resume/cover letter, and it is, how much more careful must we be when assessing a candidiate based on information we collect on our own?
We need to remember the difference between information we collect online on our own, and information a candidate provides us in relation to a job search, or interest in an opportunity. The context of the content can be the most important thing we don't have.