Re:Ask HN: why isn't RDF more popular on HN?
by acdha on Saturday March 29, 2014 @09:48PM
(#7494784)
There are two separate thoughts here:
1. I'm not saying the sem web concept is useless but that poor quality and unnecessarily high costs make it a worse deal. More practical focus on specs and solid software engineering would avoid making that trade-off worse unnecessarily.
2. Many topics which are of interest to academics either aren't generally useful and too many academics are dismissive of mundane concerns like software engineering. My first point falls solidly into that latter case. The former case isn't proven – it's entirely possible that we'll eventually see something interesting develop with more time, although as another commenter pointed out, it's likely that we'll see something different which uses the good parts but doesn't give up as much for theoretical purity. Of course, it's also possible that they'll spend more time having academic debates over HTTP range-14 and what will actually be adopted will come from outside the current semweb community entirely – after all, the “Web” in SemanticWeb refers to a technology which saw rapid growth due to ease of adoption, even though the academic hypertext community wasn't impressed because it was too simple.
In every case, more attention to building useful tools would massively increase the number of people working with semweb technology and lower the cost for everyone doing research. Everyone who wants that class of technologies to succeed should be focused on lowering that barrier to entry.
1. I'm not saying the sem web concept is useless but that poor quality and unnecessarily high costs make it a worse deal. More practical focus on specs and solid software engineering would avoid making that trade-off worse unnecessarily.
2. Many topics which are of interest to academics either aren't generally useful and too many academics are dismissive of mundane concerns like software engineering. My first point falls solidly into that latter case. The former case isn't proven – it's entirely possible that we'll eventually see something interesting develop with more time, although as another commenter pointed out, it's likely that we'll see something different which uses the good parts but doesn't give up as much for theoretical purity. Of course, it's also possible that they'll spend more time having academic debates over HTTP range-14 and what will actually be adopted will come from outside the current semweb community entirely – after all, the “Web” in SemanticWeb refers to a technology which saw rapid growth due to ease of adoption, even though the academic hypertext community wasn't impressed because it was too simple.
In every case, more attention to building useful tools would massively increase the number of people working with semweb technology and lower the cost for everyone doing research. Everyone who wants that class of technologies to succeed should be focused on lowering that barrier to entry.