Re:Explaining SQL and NoSQL, to Grandma
by spectramax on Tuesday September 03, 2019 @11:07PM
(#20873660)
Also, the entire article would be much clearer with plain language. Analogies are overrated and ELI5 method actually hurts in grasping complex abstract topics.
[ Parent
| on HN ]
Re:Explaining SQL and NoSQL, to Grandma
by sebscholl on Wednesday September 04, 2019 @08:36AM
(#20876543)
I have a really hard time swallowing the idea of analogies being over-rated. There is so much literature behind the effectiveness of analogies in learning, particularly in sciences: https://pdfs.semanticscholar.org/8518/367cc1ef090ea263c39e91...
Why do you feel this way?
Re:Explaining SQL and NoSQL, to Grandma
by spectramax on Wednesday September 04, 2019 @02:16PM
(#20879929)
I should correct myself and say that Analogies are "Overused". Feynman's teaching style had frequent use of Analogies but as a way to clarify the concept, not manifest the concept using an analogy in a didactic manner to students. He explained things in plain english first.
Analogies definitely have a place in learning but it should be carefully prefaced with a warning that analogies can only go so far. You can't explain Quantum Electrodynamics using analogies or you're misleading the students.
Re:Explaining SQL and NoSQL, to Grandma
by sebscholl on Wednesday September 04, 2019 @08:21PM
(#20882742)
Challenge accepted.
Re:Explaining SQL and NoSQL, to Grandma
by aloknnikhil on Tuesday September 03, 2019 @11:56PM
(#20873867)
I think the whole purpose of ELI5, when done right, is to introduce you to the concept. Not help you master it. Similarly, analogies do help, especially for a non-native speaker.
Re:Explaining SQL and NoSQL, to Grandma
by spectramax on Wednesday September 04, 2019 @12:37AM
(#20874032)
Introduction shouldn't be complicated to explain in plain english. Instead of introducing to the concept, it misleads from the beginning.
Re:Explaining SQL and NoSQL, to Grandma
by aloknnikhil on Wednesday September 04, 2019 @12:24PM
(#20878724)
Fair enough. ELI5s wouldn't be necessary if the original abstract was good enough.