I for one wholeheartedly disapprove of an app like this. We don't need things enforcing old, arbitrary modes of prescriptivist grammar. If you spent time writing this, and you didn't realize along the way that natural language is far too complex and nuanced to support such reductivist, Cartesian attitudes, then you're entirely missing the point of how natural language works.
You're sending the wrong message and masquerading as a language expert, when really you're just codifying meaningless, pedantic complaints about how other speaker groups utilize a roughly identical system using shared constructions. Those attitudes lead to unnecessary prejudice against others at best and bigotry at worst.
You want to know the best way to determine whether or not a construction is grammatical? Whether someone said it, and another person understood it. Because well-formedness is ultimately determined on a gradient scale, and not a categorical one with a series of by-laws like some kind of rulebook for a secret club.
Someone posts a really neat language tool, and the top-voted post is not talking about the technical merits or interest of the tool, but is instead a pretentious rant about how real language is what ordinary people speak and understand. You know, phrases like "Cartesian attitudes" and "identical system using shared constructions". This is the descriptivist-approved language for the masses. Using 'whereas' instead of 'while' is an elitist habit that must be stopped, whereas name-dropping dead philosophers out of context is the mark of a humanist.
Anyway, I'm primarily interested in hacking, not in seeing some angry nerd on the Internet make excuses to take down someone else's technically challenging work.
> well-formedness is ultimately determined on a gradient
> scale
So you're willing to admit there's a scale. That's great. Use stylistic rules as a guide for where you are on that scale; they're a brilliant heuristic. After all, it's not all relative.
Once you've accumulated sufficient experience, you'll know when to ignore a rule.
The bigger problem imo isn't so much that language changes, but that many of the rules enforced here (and in some older style guides) never existed to begin with, in the sense that not even elite or literary usage followed them, let alone common usage. There was an unfortunate period in English grammar, around the late 19th and early/mid 20th centuries, where grammarians tried to "purify" the language with invented rules, often modeled on Latin grammar. Strunk & White's Elements of Style is one of the offenders from that era that sadly lingers on in reprintings.
A tool based on rules that bear some closer resemblance to actual English grammar would be interesting to me. Maybe something that checked against the recommendations in a recent edition of the Chicago Manual of Style. The technical aspects of this tool don't seem hugely tied to the specific constructs it currently flags, so perhaps that could be done (though the source is not available, so perhaps not).
I tried this tool on a few texts and it recommended some fixes on ambiguous and redundant words. It didn't change the style as much as it improved the clarity and comprehensibility of the text.
Whether someone said it, and another person understood it.
The problem for me is that when I write something I may not know that what someone understood is really what I intended. No matter what I write there's no magic way to make every reader understand exactly what I meant, but if I adhere to some rules of grammar that I expect are reasonably comprehended than I have a better chance.
So, for example, if I write, "The result was less bugs", some people may take that to mean the total number of bugs went down (i.e, "fewer bugs") while others may think that the number of bugs stayed the same but that the bugs were not as serious ("lesser bugs").
The framing of the site as overly picky is brilliant. I disagreed with maybe 90% of the suggestions it made on some example text. But no matter how silly the errors and suggested fixes are, the criticism can be defused by noting that the site stated up-front it was going to be nitpicking.
(That said, I can't imagine actually using this except for the humor value of seeing suggestions like replacing "half" in "half a year" with "one-half" or "moiety".)
Nice, but why tell me to 'Avoid using British English.'? Maybe a one-off misuse could be caught, but if you see multiple instances of British English in a text block, maybe you can assume the writer intends it to be that way?
> Maybe a one-off misuse could be caught, but if you see multiple instances of British English in a text block, maybe you can assume the writer intends it to be that way?
Ah. No. This is clearly in Full-On Prescriptivist Peeving Mode, which means The Author Is WRONG!!!! (Yes, all of those exclamation points are required.) It's fundamentally more important to hew to the dictates of someone who thinks English must needs be used according to the rules of Latin than it is to express yourself in a clear and concise fashion. Anyone who disagrees would likely use the non-words 'Electrocution' or 'Television'.
It would probably be better for usability to just have the tool on the front page. The first two steps of the tutorial - "Enter your text." and "Submit to Nitpicker." - will already be clear to anyone who has used a few web forms. You could then include a hint on the Analyze step that one can click on the issues to pop up the menu.
This would lower the barrier to using the tool, and reduce the amount of text and exposition new users have to slog through.
Actually, all the text across the site needs a thorough edit (by a human). Here are just a few examples of what the tool likely passed but that would have been flagged by a competent human editor:
"Nitpicker was written by Niels Lohmann during the final stages of this thesis." > ... his thesis
"At that time, every second page looked like the one on the left when it came back from his supervisor. " > ... on the right.
I pasted in an article I wrote for a publication with a reasonably competent editor a while ago and the number of false positives was, I thought, surprisingly high. Many of the 'errors' it flagged simply wouldn't be considered as such in the context I was writing, and in fact the proposed changes would worsen the reading experience for the intended audience. I realise it's overtly supposed to be nit picking, rather than providing a realistic overview of what an intelligent reader may actually consider errors, but there's no substitute for simply having another human read what you've written when trying to eliminate mistakes.
This is what I want/need, alas it failed my to/two/2 and their/there/they're as well as long sentence punctiontion and multiple and's.
But early days and can only get better. But hey if you chuck shakespere at any modern day spelling check it will barf more than what I have put the english language thru, so it's no easy challenge.
If anything there are more exception in the english language to the rules than there are rules and there be a lot of rules.
I would like to know the privacy aspects of this - does the text gets stored in the server? What about the "report false positive" button- is it entirely algorithmic or actually goes to a human? If the latter, how much of the text is visible to him/her?
Of course, I am not going to use it for anything confidential, but knowing how much privacy I can expect would be nice.
Be careful, they aren't kidding when they say "Nitpicker". Many of the mistakes covered (for example while/whereas) are extremely common and are acceptable in all but the most stringent circumstances.
Now we can all fight the Grammar/Style Nazis around!
On the serious side - should be useful, I'd certainly like to have a local offline version to add it right after aspell in my editing pipeline. Just to see if I agree with the results.
I'm a bit skeptical of a style checker that introduces itself like this:
> Nitpicker is a language style checker which compares your text with a large database of style warnings, common errors, or warnings indicating that you may have once more used the wrong word or preposition. Nitpicker’s goal is to implement the old proverb “shame on you if you fool me once, shame on me if you fool me twice”: the moment your supervisor, colleague, reviewer, or local native-speaker finds a mistake in your text, any repetition is just embarrassing.
Let's nitpick:
> Nitpicker is a language style checker which compares your text with a large database...
I'd probably write "that" instead of "which" - it's easier to say - and maybe "writing" instead of "text"? Not sure if "compares" is the ideal word either, but these are just nitpicks. Rather than tweak these I'd probably rewrite the whole thing.
> ...of style warnings, common errors, or warnings indicating...
Major bug: "or"? Does it check for one of these problems or another, but not necessarily all of them? It should be "and".
Repeating "warnings" twice is confusing too. This second group of warnings: are these style warnings too, or not?
> ...that you may have once more used the wrong word or preposition.
I've read this at least a dozen times now and my brain still slows way down around the "may have once more used" part. I have a few guesses about what it means, but I'm not sure which if any are right.
Also, "word or preposition" doesn't make any sense. A preposition is a word. It's like saying "a parrot or a bird".
> Nitpicker’s goal is to implement the old proverb “shame on you if you fool me once, shame on me if you fool me twice”
Ugh. Way to mangle a beautiful and poetic saying. This version has no power: it's like the nursery rhyme translation, and it's confusing too.
Here's how it really goes:
"Fool me once, shame on you. Fool me twice, shame on me."
Now that is writing.
But what does fooling people and and who should be ashamed have to do with a style checker anyway? And how would Nitpicker "implement" the proverb? Make people feel ashamed when someone has fooled them twice?
> the moment your supervisor, colleague, reviewer, or local native-speaker finds a mistake in your text, any repetition is just embarrassing.
Is it embarrassing to you or to them? And are we saying that the first mistake is not embarrassing, only when you repeat it?
I'm trying to connect this with the proverb, which is all about who should be ashamed, not when the shame happens. It is all very confusing.
Enough nitpicking. How would I write it? Maybe something like:
> Nitpicker helps you find and fix common mistakes in your writing. It checks for thousands of style errors and questionable wording and offers suggestions to improve them. Let Nitpicker find your mistakes before your friends and colleagues do!
[+] [-] languagehacker|13 years ago|reply
You're sending the wrong message and masquerading as a language expert, when really you're just codifying meaningless, pedantic complaints about how other speaker groups utilize a roughly identical system using shared constructions. Those attitudes lead to unnecessary prejudice against others at best and bigotry at worst.
You want to know the best way to determine whether or not a construction is grammatical? Whether someone said it, and another person understood it. Because well-formedness is ultimately determined on a gradient scale, and not a categorical one with a series of by-laws like some kind of rulebook for a secret club.
[+] [-] cynicalkane|13 years ago|reply
Anyway, I'm primarily interested in hacking, not in seeing some angry nerd on the Internet make excuses to take down someone else's technically challenging work.
[+] [-] apl|13 years ago|reply
Once you've accumulated sufficient experience, you'll know when to ignore a rule.
[+] [-] _delirium|13 years ago|reply
A tool based on rules that bear some closer resemblance to actual English grammar would be interesting to me. Maybe something that checked against the recommendations in a recent edition of the Chicago Manual of Style. The technical aspects of this tool don't seem hugely tied to the specific constructs it currently flags, so perhaps that could be done (though the source is not available, so perhaps not).
[+] [-] fredsted|13 years ago|reply
I think you're overreacting.
[+] [-] jamesbritt|13 years ago|reply
The problem for me is that when I write something I may not know that what someone understood is really what I intended. No matter what I write there's no magic way to make every reader understand exactly what I meant, but if I adhere to some rules of grammar that I expect are reasonably comprehended than I have a better chance.
So, for example, if I write, "The result was less bugs", some people may take that to mean the total number of bugs went down (i.e, "fewer bugs") while others may think that the number of bugs stayed the same but that the bugs were not as serious ("lesser bugs").
Pedantry can be helpful here.
[+] [-] jsnell|13 years ago|reply
(That said, I can't imagine actually using this except for the humor value of seeing suggestions like replacing "half" in "half a year" with "one-half" or "moiety".)
[+] [-] tom_usher|13 years ago|reply
[+] [-] derleth|13 years ago|reply
Ah. No. This is clearly in Full-On Prescriptivist Peeving Mode, which means The Author Is WRONG!!!! (Yes, all of those exclamation points are required.) It's fundamentally more important to hew to the dictates of someone who thinks English must needs be used according to the rules of Latin than it is to express yourself in a clear and concise fashion. Anyone who disagrees would likely use the non-words 'Electrocution' or 'Television'.
[+] [-] jrajav|13 years ago|reply
This would lower the barrier to using the tool, and reduce the amount of text and exposition new users have to slog through.
[+] [-] FuzzyDunlop|13 years ago|reply
[+] [-] GiraffeNecktie|13 years ago|reply
"Nitpicker was written by Niels Lohmann during the final stages of this thesis." > ... his thesis
"At that time, every second page looked like the one on the left when it came back from his supervisor. " > ... on the right.
[+] [-] apaprocki|13 years ago|reply
"Make sure you the negation of 'can' applies to the activity rather than to the ability to carry out the activity."
[+] [-] cheald|13 years ago|reply
It'd reuslt in the kline heard 'round the world.
[+] [-] d0m|13 years ago|reply
[+] [-] mindprince|13 years ago|reply
[+] [-] thisishugo|13 years ago|reply
[+] [-] Zenst|13 years ago|reply
But early days and can only get better. But hey if you chuck shakespere at any modern day spelling check it will barf more than what I have put the english language thru, so it's no easy challenge.
If anything there are more exception in the english language to the rules than there are rules and there be a lot of rules.
[+] [-] xtreme|13 years ago|reply
Of course, I am not going to use it for anything confidential, but knowing how much privacy I can expect would be nice.
[+] [-] neumann_alfred|13 years ago|reply
[+] [-] advisedwang|13 years ago|reply
[+] [-] NinetyNine|13 years ago|reply
[+] [-] drbig|13 years ago|reply
On the serious side - should be useful, I'd certainly like to have a local offline version to add it right after aspell in my editing pipeline. Just to see if I agree with the results.
[+] [-] Stratoscope|13 years ago|reply
> Nitpicker is a language style checker which compares your text with a large database of style warnings, common errors, or warnings indicating that you may have once more used the wrong word or preposition. Nitpicker’s goal is to implement the old proverb “shame on you if you fool me once, shame on me if you fool me twice”: the moment your supervisor, colleague, reviewer, or local native-speaker finds a mistake in your text, any repetition is just embarrassing.
Let's nitpick:
> Nitpicker is a language style checker which compares your text with a large database...
I'd probably write "that" instead of "which" - it's easier to say - and maybe "writing" instead of "text"? Not sure if "compares" is the ideal word either, but these are just nitpicks. Rather than tweak these I'd probably rewrite the whole thing.
> ...of style warnings, common errors, or warnings indicating...
Major bug: "or"? Does it check for one of these problems or another, but not necessarily all of them? It should be "and".
Repeating "warnings" twice is confusing too. This second group of warnings: are these style warnings too, or not?
> ...that you may have once more used the wrong word or preposition.
I've read this at least a dozen times now and my brain still slows way down around the "may have once more used" part. I have a few guesses about what it means, but I'm not sure which if any are right.
Also, "word or preposition" doesn't make any sense. A preposition is a word. It's like saying "a parrot or a bird".
> Nitpicker’s goal is to implement the old proverb “shame on you if you fool me once, shame on me if you fool me twice”
Ugh. Way to mangle a beautiful and poetic saying. This version has no power: it's like the nursery rhyme translation, and it's confusing too.
Here's how it really goes:
"Fool me once, shame on you. Fool me twice, shame on me."
Now that is writing.
But what does fooling people and and who should be ashamed have to do with a style checker anyway? And how would Nitpicker "implement" the proverb? Make people feel ashamed when someone has fooled them twice?
> the moment your supervisor, colleague, reviewer, or local native-speaker finds a mistake in your text, any repetition is just embarrassing.
Is it embarrassing to you or to them? And are we saying that the first mistake is not embarrassing, only when you repeat it?
I'm trying to connect this with the proverb, which is all about who should be ashamed, not when the shame happens. It is all very confusing.
Enough nitpicking. How would I write it? Maybe something like:
> Nitpicker helps you find and fix common mistakes in your writing. It checks for thousands of style errors and questionable wording and offers suggestions to improve them. Let Nitpicker find your mistakes before your friends and colleagues do!
[+] [-] gbog|13 years ago|reply
[+] [-] pseingatl|13 years ago|reply
[+] [-] advisedwang|13 years ago|reply
[+] [-] StavrosK|13 years ago|reply
[+] [-] ansman|13 years ago|reply
[+] [-] dbaupp|13 years ago|reply
[+] [-] jvdh|13 years ago|reply
[+] [-] niklaslogren|13 years ago|reply
[+] [-] tremblanc|13 years ago|reply
These things are always mired with troubled.
These things are always mired with trouble.
These things are always mired with troubles.
[+] [-] kepano|13 years ago|reply