And it works for generating text that would be tedious to do manually.
But for searching information, I can’t bring myself to trust it.
And it works for generating text that would be tedious to do manually.
But for searching information, I can’t bring myself to trust it.
Don't ignore rate of change, recognise it can only improve with time, you're looking at very early system that will quickly be orders of magnitude better.
https://en.wikipedia.org/wiki/Wikipedia:Academic_use
For day to day use it's fine though and the chances of your queries intersecting with a page that has errors on it and/or has been vandalized are relatively remote. But to be skeptical about what you read on WP isn't a bad thing per-se.
@mirekrusin: >> Your comment reminds me of a comment made almost 2 decades ago by professor regarding wikipedia - it's just random people writing text, it'll never compete with encyclopedia, I will never trust it.
Don't ignore rate of change, recognise it can only improve with time, you're looking at very early system that will quickly be orders of magnitude better.
@jacquesm: >>> Your professor was and is correct in one respect: Wikipedia should not be cited whereas an encyclopedia in print can be cited. […]
===
The reason why @mikerusin was invoking the Wikipedia analogy was to point out that at a certain moment in time in the heady .com bubble days Wikipedia was an acorn and it was hard to imagine how it could ever grow to compete with the likes of Britannica. I remember the arguments at the time. Some people said "no way" and other said "huh, wouldn't be so sure, just you wait and see". Turns out the latter group were not only correct they were very correct. Wikipedia has entirely supplanted Brittanica and its ilk. I wouldn't even like to guess how more used Wikipedia is than its print rivals.
(And for the purposes for which Wikipedia is used people are aware of its limitations. It's not "At best [] a tertiary source", it's a secondary source for when it comes to citations but if one needs information in a hurry people the information they retrieve from it directly and do not go through the hassle of looking up the primary sources unless they have to. If you use it any other way I'd be very surprised. But this is by-the-by.)
As a response to @hagbarth's pessimism this perfectly echoes the debates we had around the time of the birth of Wikipedia. I wouldn't be so sure. Don't bet against it. Etc. Wikipedia scaled quickly because of crowd-sourcing but a ChatGPT turbo-charged Bing (or whatever) may not need a democratised version of ChatGPT, it may just need to harness the relentless pace of change in the hardware/software sector.
Long story short: I believe @mikerusin is correct and the analogy is a good one, I believe @hagbarth needs to try harder to see the potential here, and I reckon you're responding to a argument that was never put forward! (and I have no idea why I spent 20 minutes going through all this rather than taking a shower and starting my day :/ but such is life …)
Sorry, but that's incorrect. Wikipedia mostly uses secondary sources, rarely primary ones that makes it for the most part a tertiary source.
https://en.wikipedia.org/wiki/Wikipedia:Primary_Secondary_an...
> "For example, Wikipedia itself is a tertiary source."
As for the rest of the comment: it's a free world. And if you want to save some time there are 'vote' buttons which allow you to express the same sentiment in a less nuanced way (though I appreciate the effort) which would allow you to start your day on time ;)
I'd like to see empirical analysis substantiating this claim. Not saying you're wrong, just saying I'd like some hard data.
Hmm, having said that, look at what I found: https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_is_a_terti...
“
Wikipedia is not a primary source
Wikipedia avoids describing topics that never have been described before — doing otherwise qualifies as performing original research. Unsourced eyewitness accounts or other unsourced information obtained from personal experience should not be added to articles, as this would cause Wikipedia to become a primary source for the added information (see Wikipedia:Verifiability).
Wikipedia is not a secondary source
Wikipedia does not offer interpretations or analyses that deviate from previously published interpretations and analyses — doing otherwise qualifies as performing original research.
Wikipedia is a tertiary source
Wikipedia summarizes descriptions, interpretations and analyses that are found in secondary sources, or bases such summaries on tertiary sources. Wikipedia illustrates such summaries and descriptions with material that is as close as possible to the primary source(s) on the described topic.
”
> As for the rest of the comment: it's a free world. And if you want to save some time there are 'vote' buttons which allow you to express the same sentiment in a less nuanced way (though I appreciate the effort) which would allow you to start your day on time ;)fair point :)
Still, it's awesome how Wiki evolves and moves so fast. And most of the articles are pretty great. My wish is that the foundation would spend money on academics and researchers to enhance contributions, rather than a lot of the junk they spend money on today.
With chatGPT I wrote a few sentences about our warehouse and asked it how to optimize some part of our process. Then it spat out 5 suggestions that were quite tailored. The same as I found when googling, but instantly in the correct context without me having to skim loads of google results and try various queries to avoid spam.
Sure there's some prompt optimisation on the GPT side but it's nowhere as complex as navigating through Google and the results are just there.
"How can I code a mock for a boto3 dynamodb client, using unittest.mock, using a patching context manager"
And it will give me a highly relevant example, often right or close enough.
I will occasionally ask to come up with a function based on requirements but I don't find that nearly as useful as using it as super advanced search engine.