3 ms·

I love using WA as a calculator since it's so good at handling units. Silly example: I just tried "average distance to moon/Eiffel Tower height" in Kagi and go

I love using WA as a calculator since it's so good at handling units. Silly example: I just tried "average distance to moon/Eiffel Tower height" in Kagi and got "= about 1.167 million" (powered by WA).

rqtwteye 3y ago
Good to know! This will help in figuring out when reading the news and they describe everything in tennis courts , school buses and football fields. How do you convert from tennis court to school bus?
- ziddoap 3y ago
  You forgot Olympic swimming pool as both a length unit and a volume unit.
  - samatman 3y ago
    I once worked out that an Olympic swimming pool would hold about 500,000 human being's worth of blood. The topic was Communism.
rgovostes 3y ago
It can't handle non-numeric quantities like "average distance to moon" but Google's calculator has handled calculations with units very well for 15 years at least. It definitely puts Apple's Spotlight calculations to shame. I find that W|A works fine until it fails to parse my query and then I have no idea how to fix it, because the syntax is unclear.
- 7thaccount 3y ago
  There should be some doc on the English syntax. I wonder if WA lets you fall back to the proper Mathematica representation. Something like the below but in reverse as WA typically does the free form by default I think. https://wolfram.com/mathematica/new-in-8/free-form-linguistic-input/ https://wolfram.com/mathematica/new-in-8/free-form-linguisti...
  - rgovostes 3y ago
    You can input Mathematica (ahem, Wolfram Language). I suppose yet another use for LLMs: better translation of natural language input into structured Wolfram calculations. Unsurprisingly, they're already advertising "Wolfram GPT" and have added LLMSynthesize[] etc. to Mathematica.
    - 7thaccount 3y ago
      I assume the translation is a bit more rigorous than running through an LLM.
      - rgovostes 3y ago
        Which is the problem: It is using natural language processing techniques that were state of the art in 2009, but have been completely eclipsed in the past couple of years. The "rigor" tends to be at odds with the fluidity of user input: typos, search query-ese, ambiguity, etc. The challenging problem that Wolfram|Alpha tries to solve is conversion of natural language queries to structured ones. Although I doubt Wolfram's parser has been completely static for a decade, the most recent generation of language models are vastly better at translating natural language queries to structured ones. See also: how terrible Siri is at everything.
        pbronez 3y ago
        Ugh I can’t wait for Apple to roll out Siri II, re-baselined with contemporary LLM tech. So much of Siri’s architecture is exactly what I want… it’s just too dumb.
dawnerd 3y ago
I just tried some time math and it got it wrong 3:38 - 0:10 = 8 hours and 32 minutes I'm guessing it was trying to convert it to times of day instead of 3 minutes 38 seconds. Interestingly if I click the quick answer button it gives 3:48 as a result. Google - doesn't even try. WA directly also had to nudge it to use it as a unit instead of a time of day. Still some work to do here it seems!
- jjtheblunt 3y ago
  > got it wrong But your query was ambiguous as written. 8 hours and 32 minutes is how long it takes to go from 3:38 pm to 10 after midnight? Perhaps the am/pm thing is also underspecified, so it chooses the shorter time delta?
  - dawnerd 3y ago
    That makes sense, ideally it should come back saying it's unsure but here's the best guess. I like the way WA explains it directly. The problem with these quick answers is you still can't blindly trust them unless they show their work.
    - jjtheblunt 3y ago
      Exactly!

rqtwteye 3y ago

Good to know! This will help in figuring out when reading the news and they describe everything in tennis courts , school buses and football fields. How do you convert from tennis court to school bus?

ziddoap 3y ago
You forgot Olympic swimming pool as both a length unit and a volume unit.
- samatman 3y ago
  I once worked out that an Olympic swimming pool would hold about 500,000 human being's worth of blood. The topic was Communism.

rgovostes 3y ago

It can't handle non-numeric quantities like "average distance to moon" but Google's calculator has handled calculations with units very well for 15 years at least. It definitely puts Apple's Spotlight calculations to shame. I find that W|A works fine until it fails to parse my query and then I have no idea how to fix it, because the syntax is unclear.

7thaccount 3y ago
There should be some doc on the English syntax. I wonder if WA lets you fall back to the proper Mathematica representation. Something like the below but in reverse as WA typically does the free form by default I think. https://wolfram.com/mathematica/new-in-8/free-form-linguistic-input/ https://wolfram.com/mathematica/new-in-8/free-form-linguisti...
- rgovostes 3y ago
  You can input Mathematica (ahem, Wolfram Language). I suppose yet another use for LLMs: better translation of natural language input into structured Wolfram calculations. Unsurprisingly, they're already advertising "Wolfram GPT" and have added LLMSynthesize[] etc. to Mathematica.
  - 7thaccount 3y ago
    I assume the translation is a bit more rigorous than running through an LLM.
    - rgovostes 3y ago
      Which is the problem: It is using natural language processing techniques that were state of the art in 2009, but have been completely eclipsed in the past couple of years. The "rigor" tends to be at odds with the fluidity of user input: typos, search query-ese, ambiguity, etc. The challenging problem that Wolfram|Alpha tries to solve is conversion of natural language queries to structured ones. Although I doubt Wolfram's parser has been completely static for a decade, the most recent generation of language models are vastly better at translating natural language queries to structured ones. See also: how terrible Siri is at everything.
      - pbronez 3y ago
        Ugh I can’t wait for Apple to roll out Siri II, re-baselined with contemporary LLM tech. So much of Siri’s architecture is exactly what I want… it’s just too dumb.

dawnerd 3y ago

I just tried some time math and it got it wrong 3:38 - 0:10 = 8 hours and 32 minutes I'm guessing it was trying to convert it to times of day instead of 3 minutes 38 seconds. Interestingly if I click the quick answer button it gives 3:48 as a result. Google - doesn't even try. WA directly also had to nudge it to use it as a unit instead of a time of day. Still some work to do here it seems!

jjtheblunt 3y ago
> got it wrong But your query was ambiguous as written. 8 hours and 32 minutes is how long it takes to go from 3:38 pm to 10 after midnight? Perhaps the am/pm thing is also underspecified, so it chooses the shorter time delta?
- dawnerd 3y ago
  That makes sense, ideally it should come back saying it's unsure but here's the best guess. I like the way WA explains it directly. The problem with these quick answers is you still can't blindly trust them unless they show their work.
  - jjtheblunt 3y ago
    Exactly!