4 ms·
Quora has been completely ruined by dogmatic Silicon Valley thinking and poor monetization attempts. Content cloaking, removal of the credit system, deliberate
by joelthelion 10y ago
Quora has been completely ruined by dogmatic Silicon Valley thinking and poor monetization attempts.
Content cloaking, removal of the credit system, deliberate choice to promote crappy clickbaity answers as opposed to quality ones, aggressive question length limits...
I think there is definitely space for someone to recreate the original Quora without the delusions of grandeur. A place for people to exchange questions and answers on any topic, without misguided attempts to remove "bad" questions, whatever that means.
- KyleBrandt 10y agoSounds like you are describing reddit to me :P
- acdha 10y agoI added quora.com to my search result spam blacklist for content cloaking back before Google made them tone it down so you'd at least see the text you saw in the search results. It'd take a top-level management change to consider going back.
- michalu 10y agoStackExchange? Btw. same here, I used it before it reached it's first million users, even went to few Quora meetups but I blocked the site on notebook two years ago. I feel it became a place for chronic procrastinators and self-promoters + there's very little unique knowledge... as a content researcher, 99 out of 100 I find better information elsewhere.
- littleweep 10y agoJust curious, where do you go the other 99 times?
- dredmorbius 10y agoNot OP, but I've increasingly been searching within published papers and books for information. Rather than rely on Internet re-tellings (sometimes reliable, sometimes not) of what's in such references, got to the source. The availability of technical articles and books through (extralegal) services such as SciHub and LibGen means that the quality of information online has increased greatly. Not that articles and books are always valid, but they tend to have a lower bullshit level. If you're into older sources (1924 and prior, some after), The Internet Archive and Project Gutenberg, as well as other archival sites, have a wealth of information. Google's Ngram Viewer is an amazing way to trace the emergence of specific terms (and occasionally names). I'll often start my search from Ngram Viewer, find an early period of pre-emergence of a key phrase, and look for first instances of mention. For technical papers, further drilling down to references can turn up first instances. I've used this successfully numerous times to track down idea origins (or near-origins), examples being the concepts of money as based on energy (Frederick Soddy -> H.G. Wells -> A.C. Clarke -> Kim Stanley Robinson), or the modern misapprehensions of Adam Smith's "invisible hand" metaphor (Jacob Viner -> Paul Samuelson -> FEE/Leonard Read -> Regnery Press -> von Mises/Rothbardian Libertarianism -> General popular use ~mid 1960s). Wikipedia can be useful, though you want to rely on its references section for verification. Reddit's "Ask" subs (Science & Historians) can be useful. For programming / inforamtion technology technical info, StackExchange is pretty good.
- littleweep 10y agoInteresting, thanks
- notacoward 10y agoI was a Top Writer on Quora for two years in a row. I ended up quitting it entirely for a whole laundry list of reasons. Chief among them was the uneven enforcement of rules and manipulation of answer rankings so that people who promote themselves and/or generally post catchy-but-crappy answers always end up at the top for every question and every topic. Quora is more like those "one weird trick" listicle sites than a place to find actual information or insight, and it's quite intentional on their part.