3 ms·
Google Scholar is still quite useful for general information, but it's also generally pretty dense and hard to parse so it will take a bit of work. The days of
by naravara 7y ago
Google Scholar is still quite useful for general information, but it's also generally pretty dense and hard to parse so it will take a bit of work. The days of just having a question and being able to find an easy layman-level answer about it are gone (outside of Wikipedia).
There was an economics paper published recently making an argument that we have more to fear from shitty AI/robots than good ones and I think it's spot on. (In other words, they're good enough to justify firing the person who used to do the job, but they're objectively worse at the job. What ends up happening is aggregate wealth goes down because not only are the jobs gone, but the job itself is being done less efficiently than before.)
The point where Google (and Yelp and a bunch of other services) started to turn was when they stopped just being a thing you go to when you need them to being things that get in your face to anticipate your needs. They're bad at anticipating our needs in any way that's actually useful for us a lot of the time, so it ends up putting a lot of cognitive load on us to make sure the apps/services are working as intended. Instead of just using technology as a tool as we need it, we end up needing to babysit technology that's being fussy, opaque, and unpredictable. And babysitting is the best case scenario, in other cases it's being outright hostile.
- pasttense01 7y agoA problem with Google Scholar is that 75% or so of the results are behind paywalls.
- squiggleblaz 7y agoAnd the quality of results there seems to have gone downhill now too. I can turn to Scholar to find a lost article, only to have to go through pages. Again, I think it's also geared towards the common and the popular at the expense of the detailed and the idiosyncratic - it's just using a different definition of common and popular. Ten years ago, I gave up subject specific databases for Google Scholar. For most of that time, I was happy. Nowadays I wish I had access to a subject specific db. As for paywalls: I sometimes find better results by adding filetype:pdf to the title of an article in DDG than I do in scholar. They're both completely hit and miss. I think they're trying to respect copyright more nowadays than before. Unfortunately, of all the unreasonable copyrights in the world, journals and academic books are the most unreasonable, since they're trying to get libraries to buy them.