I learned in a recent discussion on Fecebook that the semantic web was discussed during the Telluride Tech Fest back in 2002.
I’m not a member of the professional classes and don’t attend conferences and the like, but I certainly remember “semantic web” being quite the buzzword for a season or so. I looked it up a few times but never figured out for sure what it refers to. Just now looked it up on Wikipedia. It seems that distribution of machine-readable data is a large part of it. It seems to me that to the extent that machine-readability itself is monetizable (and it seems to me to be VERY monetizable) it will not be freely distributed. To quote James Alexander Levy, “For information to be free, the coordinates of the information must be free.” You can have all the speedy and public trials in the public record (as required by the US Constitution) but if machine readability of the public record is proprietary, there will be a business model for Intelius-like malignancies to offer $35-a-pop peeks at “the public record.”
I once downloaded and played with a MediaWiki plugin called Semantic MediaWiki. I found its markup schema too finicky and too labor intensive to be useful. Perhaps if I had a data entry staff… And there you have it. Workers deserve to be paid, so work product deserves intellectual property protections. But with data entry as a line of work degraded all the way to Mechanical Turk level piecework pay, you’d think machine readability would be too cheap to meter.
Perhaps it is, in terms of production costs, but the strategic advantages of information asymmetry (including asymmetric levels of machine readability) far outweigh the monetization opportunities of selling access to machine readability. To offer semantic web functionality as a product would be to leave money on the table, so instead of semantic web we have “big data.”