[suchow]

Thanks for the feedback. These are issues we've thought about, and we came to different conclusions:

re 2, you'll see at http://proselint.com/approach/ that one of the guiding principles of Proselint is that we defer to experts. In practice, that's meant almost all the advice comes from Bryan Garner's usage guide, Garner's Modern American Usage. He is a careful compiler of advice and you'll find that he is almost never "totally wrong", and when his advice is debated, he knows it, notes it, and provides a thoughtful discussion.

re 1, we think of Proselint as eventually being useful as a training tool, a way to learn the conventions. Note that natural languages are large, with so many low-frequency terms that nobody can learn the whole language. Why err if an automated tool can help? Consider for example demonyms, what you call people from a certain place. How many people know, for example, that people from Manchester are Mancunians, not Manchesterians? Rather than call someone by the wrong name, with Proselint the voice of an expert gently corrects you, and you learn a cool new word.

We aren't a mob of programmers, we are three people who love language, respect it, and think we're 2% of the way to making a great tool, one that The New Yorker could run over its stories to flag issues that its own editors would flag anyways. (In fact, we've done this, running Proselint over a corpus of highly vetted text, and have found numerous issues.)

[jonstokes]

Calling someone from Manchester a "Manchesterian" instead of "Mancunian" is not wrong, or even necessarily bad. Rather, it communicates something to the reader. Depending on the context, it could mean this person doesn't know that the correct term is "Mancunian", and did not look it up or even know that it should be looked up, all of which gives me useful info and context about the writer and their education level and the amount of effort they put into the piece and the amount of editing it underwent and so on. At the very least I can surmise that the writer is not a Mancunian. Or, it could mean that the writer is attempting to be clever.

Widespread use of proselint to correct this type of thing wouldn't improve writing. Rather, it would just add another interpretive option to the above range of scenarios, i.e. "ah, I can tell that this writer did or did not run that proselint tool before submission, because their text is or is not littered with boilerplate proselintisms."

The way to improve genuinely bad writing is not with rules and tools -- it's with lots of reading, a little mentorship, and lots and lots and lots of practice.

[amperser]

> Calling someone from Manchester a "Manchesterian" instead of "Mancunian" is not wrong, or even necessarily bad. Rather, it communicates something to the reader. Depending on the context, it could mean this person doesn't know that the correct term is "Mancunian", and did not look it up or even know that it should be looked up, all of which gives me useful info and context about the writer and their education level and the amount of effort they put into the piece and the amount of editing it underwent and so on. At the very least I can surmise that the writer is not a Mancunian. Or, it could mean that the writer is attempting to be clever.

If the only goal of writing were to allow accurate assessment of the writer, then I would agree. But there are other reasons for writing — informing, persuading, clarifying, &c. — where writing clear, consistent, and idiomatic prose can help. Yours is a condemnation at all attempts to improve writing beyond the first-draft capabilities of the author.

> The way to improve genuinely bad writing is not with rules and tools -- it's with lots of reading, a little mentorship, and lots and lots and lots of practice.

Agreed, Proselint is not the right tool to improve genuinely bad writing. Reading great authors and sweating through drafts is what we'd recommend to get better at the craft, too.

[kbenson]

> all of which gives me useful info and context about the writer and their education level and the amount of effort they put into the piece and the amount of editing it underwent and so on.

From a reader centrist point of view, I can understand lamenting the loss of this information channel. From the author's stance, I can imagine wanting to tighten up alternate channels of information and present a clearer message. The author always has this ability, through natural circumstance, effort or research, so this tool would do nothing but make it easier. As a reader, it may change the assessment to whether they ran a proselint-like tool or not, but in the end those are just assumptions. The writer could be making specific choices to disregard the linting tool on purpose. In the end, reading is still an interpretive experience, this just allows authors more options.

> The way to improve genuinely bad writing is not with rules and tools -- it's with lots of reading, a little mentorship, and lots and lots and lots of practice.

Generally good advice for any thing, but I think it's worth noting that different people learn in different ways, and providing more methods for learning is generally an improvement, and opens the field to more people. Tools that look to circumvent historical methods for achieving skill often face an uphill battle from those that used those historical methods. It's easy to see why, as it looks like it has devalued much of the hard work they put into their skills. This may be true to an extent, but the gains often far outweigh this, as making a skill accessible to more people has wide ranging benefits for society in general.

In more concrete terms, I see no reason why a tool like this can't be a multiplier for mentorship and practice. At the very least it enables exposure to ideas that might not have been encountered before.

[ianstormtaylor]

Felt like someone should say this in this thread, but calling someone a "Manchesterian" offers no insight into anyone's education level, and I honestly don't even think it's something that we should be focusing corrections on. If anything, it would probably be nice if everyone started using "Manchesterian" instead of "Mancunian" because that seems a hell of a lot more clear to me ;)

To the library authors, Proselint looks very cool!

[zimpenfish]

Do you have any linguists consulting / on staff?

Bryan Garner might be a careful compiler but doesn't seem to be a linguist and seems to be a traditionalist who makes simple errors.

e.g. http://itre.cis.upenn.edu/~myl/languagelog/archives/001869.h...

"His chapter is unfortunately full of repetitions of stupidities of the past tradition in English grammar — more of them than you could shake a stick at."

http://languagelog.ldc.upenn.edu/nll/?p=5630

"So why did Bryan Garner, a highly intelligent and insightful person, make this elementary error?"

http://www.arrantpedantry.com/2007/01/02/editing-chicago/

"A good editor should know that Bryan Garner’s take on the subject is misleading and incorrect. It’s become apparent to me that many of the self-appointed guardians of the language don’t even know what it is they’re guarding."

etc.etc.

[mturmon]

You're implying that there is some kind of well-accepted notion of Bryan Garner being a poor guide to usage, but you link to some articles that are just nitpicking small terminology differences.

The second link in particular is tendentious. It claims Garner gives "a savage indictment of the behavior and character of those who use Stage 1 words [new usages]" in his book MAU.

But if you follow to the linked page from MAU, you read that Garner is, in an appendix, giving a series of wry analogies for the process of acceptance of new terms -- not a savage indictment at all. In other words, Garner is not himself saying all new usages have "a grade of F", etc., he's saying that is how some new usages will be perceived, in a very gross and qualitative sense, by a strict static conception of the language.

Since Garner comes right out and explicitly says all of the above, the link you cite comes off as picking a fight. There's nothing there.

Having read MAU (back in its first edition), I have to say that Garner strikes me as a very good guide to usage. I still enjoy perusing the book.

Taken as a whole, do you really have significant issues with MAU as a usage guide?

[zimpenfish]

> some kind of well-accepted notion of Bryan Garner being a poor guide to usage

Wasn't my intention - merely pointing out that he's not a linguist and making simple errors should give anyone using him as an "authority" considerable pause.

> do you really have significant issues with MAU as a usage guide?

I am neither an American nor a linguist - which makes me doubly unqualified to comment. That I leave to experts.

[jonstokes]

^^ This right here, is exactly what I'm talking about.

Again, the idea of prose linting is not terrible, and in fact I do a hacked up version of it with a set of standard "find/replace" operations for specific writers who have specific issues. But a giant, general-purpose ball of rules of dubious provenance applied to a generic abstraction called "prose", is what I take issue with.

[capnrefsmmat]

Garner's focus is on usage, not grammar, so for a usage linter, this doesn't seem like a big problem.

Is there an accessible, comprehensive, easy-to-read guide like Garner's Modern American Usage that's considered more accurate? There don't seem to be many options.

(I have a copy of GMAU and enjoy it, but mostly for discussion of usage, not the details of grammar)