A Critique of Dictionary Websites and Apps

Like the author, I used to be an avid user of dictionaries. In fact, my interest in them led to me freelance part-time as a lexicographer for a number of years. And, like the author, I find online dictionaries and apps to be a mixed bag.

But the biggest problem with conventional dictionaries, whether paper or digital, is that they cannot tell you what a word means in the specific context in which you encountered it. If you come across the word canonical, to use the OP’s example, and you look it up in a dictionary, the dictionary won’t tell you whether, in the text you’re reading, it means “conforming to a general rule or acceptable procedure,” “of or relating to a member of the clergy,” “of, relating to, or forming a canon,” or something else.

Take the following instance of canonical, from a recent Ezra Klein podcast:

“One of the things I always think when I hear this argument about loneliness is I don’t think we’re online because we’re lonely — I think we’re lonely because we’re online. ... And the loneliness is partially a product there. Sometimes you’re lonely being online with people you know — the canonical kids texting their friends instead of hanging out in person. But I also think that, even for people who are not lonely online, there is something really disastrous about the politics it produces.” [1]

None of the definitions of canonical shown in the OP's screenshots, or in the other dictionaries I checked, matches that usage.

LLMs do much better. Here is what Gemini gave me:

https://g.co/gemini/share/156820176dba

And Claude:

https://claude.ai/share/7fb2aabd-fb29-439c-925a-c2d4b167b35e

[1] https://www.nytimes.com/2025/09/19/opinion/ezra-klein-podcas...

	▲	treetalker 9 hours ago \| parent [-]
		That dovetails with the standard prescriptive vs. descriptive issue, don't you think? The speaker in your podcast example seems to be (mis- ?) using the word to refer to a stereotype, likening an Internet trope to reality, and (thus) implying that the Internet now has the status of a canon. So, it's an inaccurate use, prescriptively; but it's sufficiently related to the prescriptive use that it gives us some insight into the speaker — probable age, reading habits, opinion about the validity of what the speaker reads online, etc. — provided we have some context. It's a pleasing instance of the game of language being played. Later edit: I guess one point I'm making is that real dictionaries are still of great use and need in the age of LLMs.

▲

treetalker 10 hours ago | parent | prev | next [-]

The online version of Webster's 1828 is pretty good (third-party project because the work is in the public domain):

https://webstersdictionary1828.com/

To go the better route, here's a HN post about adding Webster's 1913 to the macOS dictionary app (the dictionary is very good):

https://news.ycombinator.com/item?id=29733648

If anyone is aware of a .dict file to add Webster's New International Dictionary, Second Edition (1942) to the MacOS Dictionary app, please post! That would be the holy grail for me.

▲

latexr 10 hours ago | parent | prev [-]

Real post title is “Dictionaries”, but that’s so devoid of information that a bit of editorialisation to make it more descriptive seemed appropriate.