The screenshot shows the (corrected) example sentence:
> Sometimes I still make mistakes with articles and prepositions, but my grammar is getting better every day I practice.
In American/Simplified English, this is grammatically correct. However, in 'full fat' English, practice is a noun, whereas practise is a verb; e.g.:
> I go to my practice to practise medicine.
The problem I have with this website is that it's entirely concerned with peripheral issues. The product respects my privacy - good. The product is performant - good. The product doesn't require an Internet connection - good. The product works in many writing apps - good. The product has transparent pricing - good. But I don't give a shit about any of this until you convince me that this will consistently do the correct thing, and this website singularly fails to achieve this.
In American English, "practise" is incorrect. So if the screenshot is taken from a user using the en_US locale it is correct. Perhaps, if your locale is en_GB it will correct "practice" to "practise", but you can't know that from a screenshot.
> and you are not a citizen of the British Commonwealth
What do you regard as the British Commonwealth, and what variant of English spelling to you expect people to use who are not part of this grouping?
Just so that you know, in case you don't, "US English" is used just about exclusively in the US, and UK English is used in most of the rest of the world, despite the fact that most of our devices incorrectly default to US English.
I work at an international organisation, and can confirm. My understanding is that is that at the international level, only the US-based international organisations use US English.
To be sure, I don‘t have any dog in this fight, just highlighting a fact from my experience that many here might not know.
That sounds surprising. As an European, I thought that UK only thinks their English is the dominant one, because it's the one officially taught everywhere. Meanwhile, the reality is, people learn English primarily from Hollywood and MTV; music, video, television and computer games are both the primary exposure and the main reason for people to pick English up, and they're almost all in US English. Secondarily, computers - the OS, software, SaaS - all of that is either in US English or localized to wherever the users live, and even then the US English version is usually better.
Nobody actually uses UK English here, except for English teachers. Computers don't. TV doesn't. Corporate jobs don't. And so regular people don't either.
The kind of organization that identifies as an "international organization" is disproportionately likely to be hyperaware of its working language choice and standardize on a particular English dialect by policy and pick en-gb.
Without such a conscious choice, yes, Americanisms do seem a fair bit more globally pervasive and easy to fall into "by accident".
I'm assuming we're talking NGOs here, because if we expand "organization" to include for-profit entities, then I'd argue vast majority of them will not just be US-English speaking, but US-originating and US-headquatered.
> Without such a conscious choice, yes, Americanisms do seem a fair bit more globally pervasive and easy to fall into "by accident".
You'd need choice and enforcement - unless such organization is testing for Received Pronunciation during interviews and filtering out people who cannot into Queen's English[0], I'd wager most of the members in such org, who don't come from UK (or a few related countries), will be speaking "British English" with distinctly US pronunciation. Because while an organization can make a conscious choice here, for most people, learning a second language is a long-term endeavor that largely happens "in the background", and it's very easy to learn a blend, with UK English being present in schools, and US English everywhere else.
Coming from the horse's mouth, my expectation is that GP works at some kind of (non-military?) treaty organization? ime while outsiders might include NGOs under the umbrella, the kind of people who work at the "international level" tend to uhhh, care about their taxonomy. Multinational corporations are almost never lumped together with these (and generally don't really care about international cooperation as a goal, except instrumentally), outside the barest sense of "yeah, I guess they're organizations and they work internationally".
I believe at least in all of Eastern Europe (including those guys who call themselves Central :wink:) US English dominates in pop culture, and business. I also used to work with a few Italians, and Portuguese and they all wrote US En too, so I suspect it's the same for them too.
"Here" is Poland, but in my trips to other places in Europe (and around the world), I never saw anything that would suggest this is an unique experience. On the contrary, it's pretty much self-evident, and having it be otherwise would require the last 200 years of world history to be dramatically different from what they were.
My experience with a large multi-thousand Eng department where the majority are in the US or UK: US-based or influenced employees will write usually with American English spellings and continue to do so even if based in the UK. UK-based or influenced employees will write with British English spellings and continue to do so if based in the US. No one conforms to the other and everyone can understand each other perfectly fine because the spelling of these words does not matter for comprehension. This applies to writing as well as words in code or API names.
Not sure what is international level? If it's a kind of supranational organizations which mandate a particular version of English, I'm ready to believe that in EU it's UK En. But for commercial companies my experience is exactly opposite: it's mostly US En unless you're communicating with Brits, or someone from a country which inherited British education.
I'm interested in this - I am from the Commonwealth and I do use those words, including when I forget with American colleagues.
It never occurred to me that this could ever be perceived as arrogant (even if only when referring to someone with a different background.) And I wouldn't have thought it would mean anything more than a certain language cosmopolitanism, lah ;) (Hope that joke comes through! It's been decades since I had much exposure to Malaysian English.) Can you explain why this might be, please?
Speaking for myself (an American), when I read published work that uses British spellings and I know the author is American, it feels to me that the author thinks American spellings are somehow vulgar or improper and he/she is trying to rise above our shameful misspellings.
British Commonwealth authors (well, really any author I know to be not from the USA) get a pass because these are the spellings they were taught. Nothing wrong with that.
This is a phenomenon I've only noticed in the last two decades or so. I don't know if American students are now being (wrongly) taught British spellings in school or they merely think their writing will carry more weight if it has a British "accent" but it just seems arrogant to me.
The OED is a useful resource but it is not our dictionary of final arbitration. Americans should use the American Heritage Dictionary.
I do not think you should ever feel that way. If any English-speaking listener has an issue with another speaking the dialect of English they were raised with, the listener has an issue with themselves they need to work through.
As an American English speaker, I have in the past used UK spellings when communicating at work with a group that I know only contains British English recipients. There is nothing wrong with that -- anything that makes communication more fluid should be welcome.
I believe the arrogance angle exists in a situation where an American English speaker with no British English education is using British spellings when communicating with other _American English_ speakers to purposefully create an air of superiority. If you do this, even if no one says anything, they definitely notice.
For other English dialects, my personal take is that most Americans (at least the ones who travel or interact with foreigners personally or at work) will assume they either are or are heavily influenced by British English due to history.
I have always been "aware" of the concept that they can be perceived as arrogant, but really only "colour" - it sounds kind of deliberate and like some attempt to sound "fancy," like enunciating "theater" as "thee-AY-tour" But even so, I usually see it as a humorous thing. The person is purposely trying to sound over the top arrogant/refined as a joke. I've never actually read "british" spellings and gone "what an ass." I usually assume that's how they write or it's a joke.
Interesting - I'd heard of American Gen Zs using theatre to mean the art form, to distinguish it from just meaning a building, but I hadn't heard of British English being considered a more prestigious register than American English. Is this a new phenomenon?
I don't think OP meant "arrogant" in terms of more prestigious, but in the sense of any native US-English speaker using British English spellings as a way to seem fancier or more formal.
Non-native US-English speakers are not viewed in the same light (in my opinion).
I find there are a few specific Britishisms (like theatre) that don't really raise an eyebrow in the US and maybe can seem a bit more upper-class. Grey vs. gray are essentially interchangeable. Toward vs. towards is another.
Describing American English as “simplified” English is textbook bad linguistics. It’s a different dialect, not an inherently simpler or more complex one.
In the UK, it's considered good form to be humorous when making an argument; I gather that in the US, you're supposed to sound like you're making a speech to prepare troops for war. I apologise if, in the course of describing how a product is unsuitable for use in my culture, I made that argument according to the norms of my culture.
I once saw this movie where several UK English words were compared to US English and US came of very “simplified”. Ie, pavement “Side walk” (because walk on the side). And several things like Lorry and Hauler all becoming “truck”. I guess it was very cherry picked (I can’t find it now sadly).
Ah well, I once read an argument for “EU English”. If it’s anything like my Dunglish (Dutch-English “What talk you about”) it would indeed be simplified.
I don't know if the parent comment was trying to equate American English and Simple English - I can see it as a way to dismiss American English as a "lesser" language (which it isn't, as you say), but I wouldn't start by assuming that.
> The Simplified Spelling Board was announced on March 11, 1906, with Andrew Carnegie funding the organization, to be headquartered in New York City.
Some big names here.
> The New York Times noted that Carnegie was convinced that "English might be made the world language of the future"
He wasn't wrong.
> and an influence leading to universal peace,
That's still to be seen.
> but that this role was obstructed by its "contradictory and difficult spelling".
Well now.
It's interesting to scroll through the list of proposed changes; 100+ years later, many of them seem to be the default/correct spelling, but just as many look wrong, even when following the same transformation rules. E.g.: "brasen" -> "brazen" vs. "surprise" -> "surprize".
There is a great comment I read about a decade ago from anthony_franco [0] about exactly this issue with many "open source alternatives to X", this one specifically about an alternative to Product Hunt that then failed:
> OpenHunt tried solving a problem for the content makers without providing any additional benefit to the content consumers. It's a nice, heart-warming mission. But in the end of the day, content is king, that's what consumers want.
> There have been many examples of people rallying around a "free and open" version of a service. They fail to realize that the end consumer barely cares. Look at voat (Reddit), app.net (Twitter), Diaspora (Facebook), even ycreject.com (Y Combinator) tried to be a thing for a while.
> If someone is able to make it "free and open" while also making it a better experience than the alternative, then it'll be a big success. But so far everyone gets that wrong.
The elephant in the room here is how you ethically get people to onboard without an existing community / fomo / money.
The trick is getting the content creators there, but most of them are ultimately and fairly interested in making money, and your new platform wont have that for them.
Bluesky has done alright, but that was a black swan event Elon Musk inspired.
Do what reddit did, use multiple accounts as founders, and with AI I'm sure it's even easier to do so. For content-based platforms you must have content, there is no way around it and I don't see adding fake content in the beginning to be unethical, it is a solution to the cold start problem (also a good book by Andrew Chen at a16z [0]).
For what it’s worth, I’m a native British English speaker and don’t instinctively consider “practice” “grammatically incorrect”. Indeed, I would probably write “practice” myself.
Pretty sure I had this corrected on more than one occasion when I was at school.
Also licence/license.
I remember one day figuring out the parallel with advice/advise as a way to remember which was which. So C for the noun and S for the verb.
Weirdly (to my brain), Americans always spell practice with a C, but always spell license with an S.
> English is a bastard of a language and getting messier every day as new nations adopt it is their standard language.
I disagree strenuously with this idea, because it suggests that there is one 'big' English in which anything goes. A better idea is the one of the register[0]: there are many Englishes, many sets of rules. Different rules are used in different regions, by different groups of people, and have different connotations (e.g., the King James Bible was intentionally written in a form of English that was considered archaic at the time because that would make it sound more grandiose).
If I were to use this tool, I'd be using it to ensure that whatever I'm writing is well-received by my intended audience. Because English usage is so varied, I would need to be able to control the register that it uses to ensure that the output is suitable. The fact that the product website doesn't even mention a list of supported languages, let alone supported dialects and registers within those languages, has a very everyone can see what a horse is kind of feeling[1].
You still need to adapt it to where you are though, people expect this because it causes misunderstandings. If I as a British person go to the US, I know that I can't ask people to go and buy some booze from the off-license and when finished ask them to put their aluminium can in the bin ready for the rubbish lorry while wearing their jumper because that sounds anachronistic.
I'm American and have no idea what a pinafore or American jumper is. I know a jumper is a hoodie because I lived in Australia a while ago. But that's not a word I ever hear here.
If you chose Oxford because of the Oxford English Dictionary, note that it's not regular en-gb, it's en-gb-oxendict. "the OED often favo[u]rs "-ize" (and its derivatives) over "-ise" for words derived from Greek roots, and may also include historical or less common usages."
Fluent American readers are likely to think "practise" is a typo. It's not even one of the commonly-known British/American spelling differences (like "color"/"colour"). Unless you know your audience is likely to be more familiar with British spelling, I'd avoid "practise."
I thought the difference between practice vs. practise was that the latter is British. My spell checker (US English) does not like "practise" though, it is underlined with red. UK English, however, does not underline "practise" with red. So is it really not the case that "practice" is US English and "practise" is UK English? Because based on the spell checker, that seems to be the case.
> So is it really not the case that "practice" is US English and "practise" is UK English?
Correct: practice is a noun, and practise is a verb, in non-US English. I don't have my (twenty-volume) copy of the Oxford English Dictionary to hand, but Wiktionary has an explanation under 'usage notes': https://en.wiktionary.org/wiki/practise
> British, Australian, Canadian, New Zealand and South African English spelling distinguishes between practice (a noun) and practise (a verb), analogously with advice and advise though without an analogous difference in pronunciation. In American English, the spelling practice is commonly used for both noun and verb.
It's fine if the tool has severe limitations at this stage. However, it's crucial to clearly state what those limitations are: not only does it prevent the flurry of complaints and chargebacks from customers who were disappointed that their specific case is unsupported, but it's also an opportunity to introduce a 'we're on this journey together' aspect that helps to make customers emotionally invested in the product.
Severe limitations? You sound like an a*hole tbf. The tool clearly provides the value, and it's a magnificent display how something like a simple LLM can be used to make your everyday life easier without compromising other minor stuff, you know, like letting other companies make profits by selling your own data?
Perhaps you live under a rock, I don't really know, but it maybe just happens that you're not a target audience? There's for sure many folks, and even more so companies, who will value their own privacy more over practice vs practise bikeshedding.
I have to go with GP here, being able to set the language to en_UK and en_US and getting the right corrections based on this setting is a minimal requirement for an English spellchecker. I can do without other English spelling variants but these two need to be supported correctly and consistently.
There is no guarantee that this software will not occasionally start acting as a keylogger. If somehow this happens (let's assume not intentionally), will it be the direct responsibility of the author?
Legally, there is no entity behind that responsible for privacy (1), and honestly, I don't see even minimal reason to trust this software from a legal perspective.
Apple provides a network client entitlement[1] that sandboxed apps must have, to connect to the network. Since this app isn't sandboxed, that restriction doesn't apply.
Personally, I only use software that was either built on my machine or downloaded off of the Mac App Store (MAS apps have the be mandatory sandboxed).
Not at all. Most commercial software has a publisher (as legal entity) that is responsible for privacy and takes reputational risks if something goes wrong.
Is your objection that he is distributing this by himself, instead of through the app store? Or that it appears that he is doing it as an individual instead of a company?
Sure, it's a little sketchy, it's a guy with a website and a privacy protected domain and that's about it. But if anything were to happen you would be suing the developers of refine.sh.
I guess I do see your point though. For my software I have indeed created a legal entity and can be easily looked up.
Does anyone know how this compares to other products in its field, such as LanguageTool and Harper? LanguageTool can be hosted locally, and Harper runs entirely as an extension, so I'm interested in how the spelling and grammar checks compare.
I've run LanguageTool Server with the ngrams[0] for years, it is legitimately excellent with the ngrams (and mediocre without). The English-only ngrams are roughly 15 GB on disk.
Just have a Windows Scheduled Task kick off this bat file:
I just tested both on the text "Look Dick. See Jane. Jane run home. I says you go home to. They eats dinner." LanguageTool does what I would expect. Harper does not. However, both whine about two spaces after a period.
Edit: Alas, Hacker News also removes the extra space after periods.
Extra space after periods is never correct with proportionally-spaced fonts, which is why all browsers remove it by default.
Two spaces after periods is a kludge invented for typewriters that had monospaced fonts and touch typing teachers need to stop teaching it in the modern era where most writing uses proportional fonts.
Indeed - especially the space before colons and semi-colons. The space before exclamation marks sometimes happens in informal typing amongst Brazilians. But never the space before colons/semi-colons.
This is precisely what I've been hoping somebody would build. In my initial testing, it works well. I can even mix sentences with different languages, and it still makes correct suggestions.
The fluency suggestions are seemingly largely malfunctioning. It frequently suggests starting and ending sentences with quotes, although it also makes some useful suggestions. There seems to be an issue with analysis running synchronized, or something like that; when I type into a text field and Refine starts to run, it often blocks text entry. Selecting a suggested replacement blocks the app for half a second or so. Neither of these problems occurs with Grammarly or Language Tool. I also noticed a bunch of issues that Grammarly catches (like verb agreement) that Refine does not.
But this is an amazing first release and extremely promising. Congrats!
Languagetool is an open source tool you can run as a local spelling and grammar checker. It's different to Grammarly - less AI and more rules based. I often use both tools at the same time. I wrote a quick intro on how to self host this - https://martincapodici.com/2025/05/10/check-your-writing-usi....
You don't need docker (a Linux-only piece of tech) to run a java application. Even though I'm on Linux, and docker experience is waaaaay better here than on any other platform, I wouldn't in my life consider using it to run LanguageTool.
A command without context is not very useful. If anyone wants to run LanguageTool locally, I suggest reading the official documentation page: https://dev.languagetool.org/http-server
Their recommended process is :
1. Install fasttext (it's an official Debian package, but you have to compile it on Windows).
2. Download and uncompress the LanguageTool release.
3. Create a config file.
4. Launch the server with the java command (of course, a JRE must be installed).
5. Connect to the API, e.g. with the browser extension.
Running a ready-made docker image replaces steps 1-4 and removes the need to install Java globally. Some will prefer it this way.
> 2. Download and uncompress the LanguageTool release.
> Running a ready-made docker image replaces steps 1-4
We can go pretty low level in the docker option too.
1. Download a Linux installation image
2. Download a hypervisor
3. Install Linux on the virtual machine
4. Install docker in a virtual Linux machine
5. Launch LanguageTool container
6. Configure networking between the host and the container in the guest
7. Connect to the API, e.g. with the browser extension
Obviously I initially oversimplified by omitting the configuration step. But adding download step to inflate the complexity is not a fair play ;-)
I'd argue that running a platforn-native artifact is both simpler and easier than involving virtualization. Even if steps 1-4 are done by some magical tool like rancher or docker-desktop
It is restricted by Flatpak (i.e. Control Groups and Namespaces):
* No file-system access
* No access to devices
* Network Access is allowed (API of Languagetool is only reachable via REST?)
It is fat. But that's more an issue of Java itself. I only wish Languagetool didn't use Java, which is fine on a servers but horrible on personal-computers. Implemented in C, C++ or Rust and it would be probably already part of LibreOffice. Sonny Piers is the packager Flatpak, a prominent ex-member of GNOME board. He was removed due some Code-Of-Conduct thing which nobody can explain, due to issues within the Code-Of-Conduct.
PS: LibreOffice had to fight years to remove Java which plagued the project.
I worry that this will make my writing more likely to fail an AI coursework detector, which could really impact my life. The risk just isn't worth it till someone has tested the output through all the big players (turnitin etc.)
We'll soon need a writing tool that introduces spelling and grammar errors into our text and messes with punctuation so that we aren't accused of using LLMs.
It's funny how many people still think sloppy, mistake-filled writing is a sign of AI, as if their writing is at the same level as the image generators giving people six fingers, when the truth is the current LLMs use better English grammar than 99% of humans. Their writing may be kind of boring and standard, but they don't confuse "their" and "there."
It's hilarious, on some other sites one is immediately accused of using ChatGPT when using the n-dash (–) or m-dash (—) instead of the hyphen (-). Not an issue with the monospaced font here. ETA: I stand corrected.
Seams weird to not have "How does this compare to Apple Intelligence Writing Tools" at least in the FAQ. Maybe refine is better or has more features, but the page doesn't even seem to acknowledge that a system level feature like this exists.
Does not catch a singular/plural discrepancy between the subject and the verb in a sentence--a common mistake when the expressed thought applies equally to one thing or to many things.
Does not catch a missed indefinite article--a common mistake for speakers of languages that don't have articles. Similarly, does not catch the use of the indefinite article for a thing already mentioned before.
Well, only Americans use it. There's no point in arguing about it your version of a language is better or worse but for the rest of the world it's incredibly annoying having to correct Zs with Ss when using LLMs or American only software.
That's certainly not the case in the majority of Europe, Australia, New Zealand, East Africa, West Africa, South Africa, PNG, Fiji, Philippines, South Asia and South East Asia. Really it's mostly just America and Canada that use American.
I honestly don't think this is true. In many places outside the US our devices will default to US spelling, making it appear as if the language spoken by the locals is US English.
But when you look at things like the type of English being taught in school, or the language used by the government, it'll be UK/International English.
Kinda forgot India is still part of the Commonwealth. Maybe because all Indians I ever corresponded with uniformly used US spelling, likely because they work a lot with the USA. YMMV.
I could care less about this attempt at trolling, but I won't.
We have enough American cultural hegemony as it is. It frustrates me no end that I regularly am unable[1] to configure software to use my preferred version of English.
[1] or it's extremely difficult. CLion I'm looking at you; every available option I have is set to British English, but still you insist on telling me my Colours should be corrected to Color etc. :(
English is a bastard language of French and Saxon. But we don’t get upset over that because Britain made the world English. Now the USA is making the world American. That ship has sailed.
I see that you're using gemma3n which is a 4B parameter model and utilizes around 3GB RAM. How do you handle loading/offloading the model into the RAM? Or is it always in the memory as long as the app is running?
Neat idea. I see why the fluency feature is off by default. It constantly rewords things, adds random quotations, or does something pretty silly https://imgur.com/oVSWmtN
The Grammar feature seems to have weird suggestions/cycles too on a little bit more testing. Curious to see how this improves. A local only, one-time-purchase grammarly alternative is appealing!
Minor thing with the website but the download button is not entirely clickable, if you don't click on the text you don't get to download it, due to the div > a and the anchor being just the text, consider styling the anchor to have the padding so the whole thing is clickable. :)
This phrase is offensive and violates my safety guidelines. Therefore, I will not revise it. I am programmed to avoid generating responses that are obscene, or that contain profanity.
because what's under the hood is this, and prompts are hardcoded
unsloth/gemma-3n-E4B-it-GGUF
You are a precision editor guided by a custom style manual. Your tasks are ordered by priority.
Your primary rule is to consult the provided REFERENCE DICTIONARY. Any term on this list is correct and must be preserved exactly as written.
Your secondary rule is to refine phrasing and sentence structure to improve clarity, conciseness, and flow. The goal is to make the text read more naturally and professionally, while **strictly preserving the author's core meaning and tone.**
Your final rule is to output ONLY the clean, revised text. You MUST NOT add any commentary, greetings, or explanations.
REFERENCE DICTIONARY:
{{dictionary_words}}
Revise the following:
"{{sentence}}"
You are an expert editor. Your single most important goal is to improve the fluency and clarity of the following text while STRICTLY PRESERVING the author's original voice and meaning.
You MUST follow these rules:
1. Only rephrase sentences that are genuinely awkward or unclear.
2. Never make changes for purely stylistic preference.
Return ONLY the clean, revised text.
Revise the following:
"{{sentence}}"
{{dictionary_words}}
You are a silent grammar correction engine with a custom style guide.
Your primary rule is to consult the provided REFERENCE DICTIONARY. Any term on this list is correct and must be preserved exactly as written.
Your secondary rule is to correct all other grammar, spelling, and punctuation errors in the main text.
Your final rule is to output ONLY the clean, corrected text. You MUST NOT add any commentary, greetings, or explanations.
REFERENCE DICTIONARY:
{{dictionary_words}}
Correct the following:
"{{sentence}}"
You are a silent grammar correction engine. Your sole function is to receive text and output the corrected version. You MUST NOT add any commentary, greetings, or explanations. You will only return the clean, corrected text.
Correct the following:
"{{sentence}}"
Imagine if WinRAR didn't let you archive or name your files what you want. How we were trusted with such DESTRUCTIVE tools all these years is beyond me. /s
Mainly because (cloud-hosted, bigger & smarter) LLMs probably offer no added privacy benefit. Self-hosted models are dumber, neutered and unpredictable.
They used to have an API, which they removed about two years ago or so. Until then, there were a bunch of 3rd-party Grammarly plugins, including a pretty decent one for VSCode.
Does this do tone? That is the single largest feature of Grammarly’s Slack plugin that I absolutely loved, and if my work allowed it, I’d have it again.
Wonderful! I've given it a go, works in Apple's Notes app, but it does not seem to trigger suggestions in Chrome, Firefox or Slack. It does however highlight misspellings there. Any idea what can I do to enable suggestions there? I was looking for a product like this.
Couldnt make it to show suggestions in vscode/cursor. I would like to use the tool, but i'd expect it to work consistently across all widgets in the system (i.e. like superwhisper). Is there a technical limitation here or my misconfiguration of things?
My biggest problem with Grammarly is how buggy it is. How often it doesn’t work as intended and either messes up formatting or doesn’t change the text when I click.
What I'd be interested in would be something I could host on my local server (e.g. with ollama) to get suggestions on my laptop, where I write typst or markdown with Zed or VSCode.
I'm missing some information on how this works (a LLM? which? Do I need to bring an API key? Does this work offline?) and what I can expect in terms of performance/battery hit.
> Lightning Fast. Local processing means instant results without internet dependency or delays.
> Always Available. Works offline, on flights, in coffee shops, anywhere you write.
Two of your 4 questions were answered in the first content block
It is Gemma 3n, I can't give feedback yet on the battery hit, But I would not expect anything bad as these models have been developed for much smaller devices (Phones)
Isn't privacy a concern? How do consumers ensure that data is not going to captured in a future update without it being open source or having third party security audits?
Disagreed. It's not concerned with all apps, because most commercial applications have legal entities explaining how they use collected data. In this particular case, this is something called 'Refine', and it's not a legal entity, therefore, questioning its data privacy approach is fully legitimate.
Perhaps this type of software could be either open-source with full code accessibility, or proprietary but from a highly trustworthy entity responsible for privacy both legally and reputationally. Currently, both approaches are missing.
Theoretically, it could support over 140 languages, as it is powered by the newly released Gemma 3n model. However, I haven’t tested many languages yet.
In upcoming releases, Refine will support custom prompts and BYOK (Bring Your Own Key), allowing you to use any large language model you want.
What's the rationale behind adding BYOK? Or the advantages?
You do realise you're already using an optimised model built for everyday devices, and that model includes some serious innovations in parameter-efficient processing, right?
You're a great developer, and it looks like you're thinking about adding features like BYOK quickly to please more users. But in doing that, you might be missing the real innovation you've already created. You've basically built a version of Grammarly without the privacy issues that make most IT departments ban it.
No one wants Grammarly or your tool sending corporate emails or documents to a language model. Privacy is what sets you apart from Grammarly. It's your biggest feature right now. Add a big table to your site comparing your products privacy with Grammarly's. That's your strongest selling point, and probably the only feature that can truly compete with the big players.
My advice? Keep improving the app, but keep the model local. Keep it private. That's the killer feature you've got.
The screenshot shows the (corrected) example sentence:
> Sometimes I still make mistakes with articles and prepositions, but my grammar is getting better every day I practice.
In American/Simplified English, this is grammatically correct. However, in 'full fat' English, practice is a noun, whereas practise is a verb; e.g.:
> I go to my practice to practise medicine.
The problem I have with this website is that it's entirely concerned with peripheral issues. The product respects my privacy - good. The product is performant - good. The product doesn't require an Internet connection - good. The product works in many writing apps - good. The product has transparent pricing - good. But I don't give a shit about any of this until you convince me that this will consistently do the correct thing, and this website singularly fails to achieve this.
In American English, "practise" is incorrect. So if the screenshot is taken from a user using the en_US locale it is correct. Perhaps, if your locale is en_GB it will correct "practice" to "practise", but you can't know that from a screenshot.
So apparently the tool is slanted toward American English where the non-word practise is properly treated as a spelling error like colour.
If you use these words in writing for Americans and you are not a citizen of the British Commonwealth, you instantly mark yourself as arrogant.
> and you are not a citizen of the British Commonwealth
What do you regard as the British Commonwealth, and what variant of English spelling to you expect people to use who are not part of this grouping?
Just so that you know, in case you don't, "US English" is used just about exclusively in the US, and UK English is used in most of the rest of the world, despite the fact that most of our devices incorrectly default to US English.
I work at an international organisation, and can confirm. My understanding is that is that at the international level, only the US-based international organisations use US English.
To be sure, I don‘t have any dog in this fight, just highlighting a fact from my experience that many here might not know.
That sounds surprising. As an European, I thought that UK only thinks their English is the dominant one, because it's the one officially taught everywhere. Meanwhile, the reality is, people learn English primarily from Hollywood and MTV; music, video, television and computer games are both the primary exposure and the main reason for people to pick English up, and they're almost all in US English. Secondarily, computers - the OS, software, SaaS - all of that is either in US English or localized to wherever the users live, and even then the US English version is usually better.
Nobody actually uses UK English here, except for English teachers. Computers don't. TV doesn't. Corporate jobs don't. And so regular people don't either.
The kind of organization that identifies as an "international organization" is disproportionately likely to be hyperaware of its working language choice and standardize on a particular English dialect by policy and pick en-gb.
Without such a conscious choice, yes, Americanisms do seem a fair bit more globally pervasive and easy to fall into "by accident".
I'm assuming we're talking NGOs here, because if we expand "organization" to include for-profit entities, then I'd argue vast majority of them will not just be US-English speaking, but US-originating and US-headquatered.
> Without such a conscious choice, yes, Americanisms do seem a fair bit more globally pervasive and easy to fall into "by accident".
You'd need choice and enforcement - unless such organization is testing for Received Pronunciation during interviews and filtering out people who cannot into Queen's English[0], I'd wager most of the members in such org, who don't come from UK (or a few related countries), will be speaking "British English" with distinctly US pronunciation. Because while an organization can make a conscious choice here, for most people, learning a second language is a long-term endeavor that largely happens "in the background", and it's very easy to learn a blend, with UK English being present in schools, and US English everywhere else.
--
[0] - Or is it King's English now?
Coming from the horse's mouth, my expectation is that GP works at some kind of (non-military?) treaty organization? ime while outsiders might include NGOs under the umbrella, the kind of people who work at the "international level" tend to uhhh, care about their taxonomy. Multinational corporations are almost never lumped together with these (and generally don't really care about international cooperation as a goal, except instrumentally), outside the barest sense of "yeah, I guess they're organizations and they work internationally".
Where is "here"?
I believe at least in all of Eastern Europe (including those guys who call themselves Central :wink:) US English dominates in pop culture, and business. I also used to work with a few Italians, and Portuguese and they all wrote US En too, so I suspect it's the same for them too.
"Here" is Poland, but in my trips to other places in Europe (and around the world), I never saw anything that would suggest this is an unique experience. On the contrary, it's pretty much self-evident, and having it be otherwise would require the last 200 years of world history to be dramatically different from what they were.
My experience with a large multi-thousand Eng department where the majority are in the US or UK: US-based or influenced employees will write usually with American English spellings and continue to do so even if based in the UK. UK-based or influenced employees will write with British English spellings and continue to do so if based in the US. No one conforms to the other and everyone can understand each other perfectly fine because the spelling of these words does not matter for comprehension. This applies to writing as well as words in code or API names.
Not sure what is international level? If it's a kind of supranational organizations which mandate a particular version of English, I'm ready to believe that in EU it's UK En. But for commercial companies my experience is exactly opposite: it's mostly US En unless you're communicating with Brits, or someone from a country which inherited British education.
This is my approach/experience as well, hence my comment and question to the previous poster.
Plenty of European companies and organizations use American spellings. In general I’d say it’s pretty much random and 50/50.
Americans finding anyone else arrogant is quite rich.
That's a mighty broad brush you're painting with
I guess it's more of a shit throwing contest than an artistic practice.
In the real world, it boils down to a game of "count the nuclear-powered aircraft carriers".
I'm interested in this - I am from the Commonwealth and I do use those words, including when I forget with American colleagues.
It never occurred to me that this could ever be perceived as arrogant (even if only when referring to someone with a different background.) And I wouldn't have thought it would mean anything more than a certain language cosmopolitanism, lah ;) (Hope that joke comes through! It's been decades since I had much exposure to Malaysian English.) Can you explain why this might be, please?
Speaking for myself (an American), when I read published work that uses British spellings and I know the author is American, it feels to me that the author thinks American spellings are somehow vulgar or improper and he/she is trying to rise above our shameful misspellings.
British Commonwealth authors (well, really any author I know to be not from the USA) get a pass because these are the spellings they were taught. Nothing wrong with that.
This is a phenomenon I've only noticed in the last two decades or so. I don't know if American students are now being (wrongly) taught British spellings in school or they merely think their writing will carry more weight if it has a British "accent" but it just seems arrogant to me.
The OED is a useful resource but it is not our dictionary of final arbitration. Americans should use the American Heritage Dictionary.
I do not think you should ever feel that way. If any English-speaking listener has an issue with another speaking the dialect of English they were raised with, the listener has an issue with themselves they need to work through.
As an American English speaker, I have in the past used UK spellings when communicating at work with a group that I know only contains British English recipients. There is nothing wrong with that -- anything that makes communication more fluid should be welcome.
I believe the arrogance angle exists in a situation where an American English speaker with no British English education is using British spellings when communicating with other _American English_ speakers to purposefully create an air of superiority. If you do this, even if no one says anything, they definitely notice.
For other English dialects, my personal take is that most Americans (at least the ones who travel or interact with foreigners personally or at work) will assume they either are or are heavily influenced by British English due to history.
I have always been "aware" of the concept that they can be perceived as arrogant, but really only "colour" - it sounds kind of deliberate and like some attempt to sound "fancy," like enunciating "theater" as "thee-AY-tour" But even so, I usually see it as a humorous thing. The person is purposely trying to sound over the top arrogant/refined as a joke. I've never actually read "british" spellings and gone "what an ass." I usually assume that's how they write or it's a joke.
Huge quantities of english speakers, who are not from the Commonwealth, learn British English.
As a Frenchman, I enjoy writing colour and being arrogant.
Come on, write couleur and let them eat cake ;)
Interesting - I'd heard of American Gen Zs using theatre to mean the art form, to distinguish it from just meaning a building, but I hadn't heard of British English being considered a more prestigious register than American English. Is this a new phenomenon?
I don't think OP meant "arrogant" in terms of more prestigious, but in the sense of any native US-English speaker using British English spellings as a way to seem fancier or more formal.
Non-native US-English speakers are not viewed in the same light (in my opinion).
I find there are a few specific Britishisms (like theatre) that don't really raise an eyebrow in the US and maybe can seem a bit more upper-class. Grey vs. gray are essentially interchangeable. Toward vs. towards is another.
It gives off arrogant vibes to have one accuse Americans of being arrogant for using alternative spellings.
That's very colorful[sic] thinking.
Describing American English as “simplified” English is textbook bad linguistics. It’s a different dialect, not an inherently simpler or more complex one.
It's an old meme: https://imgur.com/thats-bit-harsh-steam-XCEdD8W
In the UK, it's considered good form to be humorous when making an argument; I gather that in the US, you're supposed to sound like you're making a speech to prepare troops for war. I apologise if, in the course of describing how a product is unsuitable for use in my culture, I made that argument according to the norms of my culture.
I once saw this movie where several UK English words were compared to US English and US came of very “simplified”. Ie, pavement “Side walk” (because walk on the side). And several things like Lorry and Hauler all becoming “truck”. I guess it was very cherry picked (I can’t find it now sadly).
Ah well, I once read an argument for “EU English”. If it’s anything like my Dunglish (Dutch-English “What talk you about”) it would indeed be simplified.
Banter (Traditional)
Banter (Simplified)
Simplified English is a thing that exists, for clarity - see for example https://simple.wikipedia.org/wiki/Practice
I don't know if the parent comment was trying to equate American English and Simple English - I can see it as a way to dismiss American English as a "lesser" language (which it isn't, as you say), but I wouldn't start by assuming that.
Could the 'Simplified Spelling Board'[1] of 1906 have anything to do with the naming?
[1] https://en.wikipedia.org/wiki/Simplified_Spelling_Board
Huh, that's an interesting tidbit.
> The Simplified Spelling Board was announced on March 11, 1906, with Andrew Carnegie funding the organization, to be headquartered in New York City.
Some big names here.
> The New York Times noted that Carnegie was convinced that "English might be made the world language of the future"
He wasn't wrong.
> and an influence leading to universal peace,
That's still to be seen.
> but that this role was obstructed by its "contradictory and difficult spelling".
Well now.
It's interesting to scroll through the list of proposed changes; 100+ years later, many of them seem to be the default/correct spelling, but just as many look wrong, even when following the same transformation rules. E.g.: "brasen" -> "brazen" vs. "surprise" -> "surprize".
Thanks for linking this!
Yea, it’s insane to try and an entire country’s writing dialect as simplified. I guess it would be the only way to show off his snobbery though.
There is a great comment I read about a decade ago from anthony_franco [0] about exactly this issue with many "open source alternatives to X", this one specifically about an alternative to Product Hunt that then failed:
> OpenHunt tried solving a problem for the content makers without providing any additional benefit to the content consumers. It's a nice, heart-warming mission. But in the end of the day, content is king, that's what consumers want.
> There have been many examples of people rallying around a "free and open" version of a service. They fail to realize that the end consumer barely cares. Look at voat (Reddit), app.net (Twitter), Diaspora (Facebook), even ycreject.com (Y Combinator) tried to be a thing for a while.
> If someone is able to make it "free and open" while also making it a better experience than the alternative, then it'll be a big success. But so far everyone gets that wrong.
[0] https://news.ycombinator.com/item?id=10934465#10940729
The elephant in the room here is how you ethically get people to onboard without an existing community / fomo / money.
The trick is getting the content creators there, but most of them are ultimately and fairly interested in making money, and your new platform wont have that for them.
Bluesky has done alright, but that was a black swan event Elon Musk inspired.
Do what reddit did, use multiple accounts as founders, and with AI I'm sure it's even easier to do so. For content-based platforms you must have content, there is no way around it and I don't see adding fake content in the beginning to be unethical, it is a solution to the cold start problem (also a good book by Andrew Chen at a16z [0]).
[0] https://a16z.com/books/the-cold-start-problem/
Case in point: "Is there a educational discount?" I believe this should say "an educational discount". I wonder if the tool would have caught it.
For what it’s worth, I’m a native British English speaker and don’t instinctively consider “practice” “grammatically incorrect”. Indeed, I would probably write “practice” myself.
Pretty sure I had this corrected on more than one occasion when I was at school. Also licence/license. I remember one day figuring out the parallel with advice/advise as a way to remember which was which. So C for the noun and S for the verb.
Weirdly (to my brain), Americans always spell practice with a C, but always spell license with an S.
That's why you need a good spellchecker.
I get this.
> ... in 'full fat' English ...
English is a bastard of a language and getting messier every day as new nations adopt it is their standard language.
Setting the bar where it is well written and unambiguously understandable is IMHO completely fine for a 15$ product.
Having a text spell checked to comply with contemporary Oxford English is likely not the goal of this product.
> English is a bastard of a language and getting messier every day as new nations adopt it is their standard language.
I disagree strenuously with this idea, because it suggests that there is one 'big' English in which anything goes. A better idea is the one of the register[0]: there are many Englishes, many sets of rules. Different rules are used in different regions, by different groups of people, and have different connotations (e.g., the King James Bible was intentionally written in a form of English that was considered archaic at the time because that would make it sound more grandiose).
If I were to use this tool, I'd be using it to ensure that whatever I'm writing is well-received by my intended audience. Because English usage is so varied, I would need to be able to control the register that it uses to ensure that the output is suitable. The fact that the product website doesn't even mention a list of supported languages, let alone supported dialects and registers within those languages, has a very everyone can see what a horse is kind of feeling[1].
[0] https://en.wikipedia.org/wiki/Register_(sociolinguistics)
[1] https://en.wikipedia.org/wiki/Nowe_Ateny
> there are many Englishes, many sets of rules.
Absolutely, but try to make a run-of-the-mill LLM understand this.
You still need to adapt it to where you are though, people expect this because it causes misunderstandings. If I as a British person go to the US, I know that I can't ask people to go and buy some booze from the off-license and when finished ask them to put their aluminium can in the bin ready for the rubbish lorry while wearing their jumper because that sounds anachronistic.
Not too mention that a jumper in the US is closer to a pinafore.
Yes, I'm aware, a colleague moved to Berkeley and relayed a story where he confused a lot of people :)
I'm American and have no idea what a pinafore or American jumper is. I know a jumper is a hoodie because I lived in Australia a while ago. But that's not a word I ever hear here.
It's like overalls with a skirt. In the 80s they were popular for girls.
> contemporary Oxford English
If you chose Oxford because of the Oxford English Dictionary, note that it's not regular en-gb, it's en-gb-oxendict. "the OED often favo[u]rs "-ize" (and its derivatives) over "-ise" for words derived from Greek roots, and may also include historical or less common usages."
I was not aware of this, thanks!
Fluent American readers are likely to think "practise" is a typo. It's not even one of the commonly-known British/American spelling differences (like "color"/"colour"). Unless you know your audience is likely to be more familiar with British spelling, I'd avoid "practise."
I thought the difference between practice vs. practise was that the latter is British. My spell checker (US English) does not like "practise" though, it is underlined with red. UK English, however, does not underline "practise" with red. So is it really not the case that "practice" is US English and "practise" is UK English? Because based on the spell checker, that seems to be the case.
> So is it really not the case that "practice" is US English and "practise" is UK English?
Correct: practice is a noun, and practise is a verb, in non-US English. I don't have my (twenty-volume) copy of the Oxford English Dictionary to hand, but Wiktionary has an explanation under 'usage notes': https://en.wiktionary.org/wiki/practise
> British, Australian, Canadian, New Zealand and South African English spelling distinguishes between practice (a noun) and practise (a verb), analogously with advice and advise though without an analogous difference in pronunciation. In American English, the spelling practice is commonly used for both noun and verb.
I see.
Lets be fair here, this tool is new - the domain was registered on Saturday.
What you suggest does seem like a good early doors feature; but the cut they've made seems to be the right one to prove market potential.
It's fine if the tool has severe limitations at this stage. However, it's crucial to clearly state what those limitations are: not only does it prevent the flurry of complaints and chargebacks from customers who were disappointed that their specific case is unsupported, but it's also an opportunity to introduce a 'we're on this journey together' aspect that helps to make customers emotionally invested in the product.
Severe limitations? You sound like an a*hole tbf. The tool clearly provides the value, and it's a magnificent display how something like a simple LLM can be used to make your everyday life easier without compromising other minor stuff, you know, like letting other companies make profits by selling your own data?
Perhaps you live under a rock, I don't really know, but it maybe just happens that you're not a target audience? There's for sure many folks, and even more so companies, who will value their own privacy more over practice vs practise bikeshedding.
I have to go with GP here, being able to set the language to en_UK and en_US and getting the right corrections based on this setting is a minimal requirement for an English spellchecker. I can do without other English spelling variants but these two need to be supported correctly and consistently.
I have no idea how this tool is better than running local LLM. Should I buy it?
So you’re saying that for most English speakers it was correct, and that’s a problem?
There are many different Englishes. I would bet that this tool does not handle Indian or Malaysian dialects.
British English is still influential over most of the Commonwealth, ie a large number of countries.
There is no guarantee that this software will not occasionally start acting as a keylogger. If somehow this happens (let's assume not intentionally), will it be the direct responsibility of the author?
Legally, there is no entity behind that responsible for privacy (1), and honestly, I don't see even minimal reason to trust this software from a legal perspective.
1. https://refine.sh/privacy-policy
There's no reason to trust it from a technical perspective either. The app is unsandboxed. Easy enough to check from the CLI.
Apple provides a network client entitlement[1] that sandboxed apps must have, to connect to the network. Since this app isn't sandboxed, that restriction doesn't apply.Personally, I only use software that was either built on my machine or downloaded off of the Mac App Store (MAS apps have the be mandatory sandboxed).
[1]: https://developer.apple.com/documentation/bundleresources/en...
Most of the time, both legal and technical misalignments walk together. Thank you for noticing this.
Isn't that true of all software? How do you know that grammarly is not already doing that your data is transmitted to their servers after all.
Not at all. Most commercial software has a publisher (as legal entity) that is responsible for privacy and takes reputational risks if something goes wrong.
So does this, it says at the bottom "© 2025 Runju Huang". Mister Huang is publishing this, presumably, he would be responsible for any wrong doing.
Is your objection that he is distributing this by himself, instead of through the app store? Or that it appears that he is doing it as an individual instead of a company?
Sure, it's a little sketchy, it's a guy with a website and a privacy protected domain and that's about it. But if anything were to happen you would be suing the developers of refine.sh.
I guess I do see your point though. For my software I have indeed created a legal entity and can be easily looked up.
Does anyone know how this compares to other products in its field, such as LanguageTool and Harper? LanguageTool can be hosted locally, and Harper runs entirely as an extension, so I'm interested in how the spelling and grammar checks compare.
+1. Also worth noting that both LanguageTool [0] and Harper [1] are FOSS.
[0] https://languagetool.org/
[1] https://writewithharper.com/
I've run LanguageTool Server with the ngrams[0] for years, it is legitimately excellent with the ngrams (and mediocre without). The English-only ngrams are roughly 15 GB on disk.
Just have a Windows Scheduled Task kick off this bat file:
[0] https://dev.languagetool.org/finding-errors-using-n-gram-dat...I just tested both on the text "Look Dick. See Jane. Jane run home. I says you go home to. They eats dinner." LanguageTool does what I would expect. Harper does not. However, both whine about two spaces after a period.
Edit: Alas, Hacker News also removes the extra space after periods.
Extra space after periods is never correct with proportionally-spaced fonts, which is why all browsers remove it by default.
Two spaces after periods is a kludge invented for typewriters that had monospaced fonts and touch typing teachers need to stop teaching it in the modern era where most writing uses proportional fonts.
Curiously, it‘s also a tell-tale sign that a North-American typed the text.
Just as a space before colons or exclamation marks is a sign that someone francophone typed it.
Indeed - especially the space before colons and semi-colons. The space before exclamation marks sometimes happens in informal typing amongst Brazilians. But never the space before colons/semi-colons.
browser rendering does. You'd need white-space: pre-wrap rule to retain double spaces.
Harper will detect those errors in its next release.
We're building a Chrome extension grammar checker that runs locally using Chrome's built-in LLM.
You can try it out here:
https://chromewebstore.google.com/detail/grammit-the-ai-gram...
LLMs can also correct other sorts of mistakes, such as correcting "The first US President was _Ben Franklin_" to "George Washington".
>LLMs can also correct other sorts of mistakes, such as correcting "The first US President was _Ben Franklin_" to "George Washington".
And they can also cause new factual mistakes that a grammar checker never would!
Preach.
This is precisely what I've been hoping somebody would build. In my initial testing, it works well. I can even mix sentences with different languages, and it still makes correct suggestions.
The fluency suggestions are seemingly largely malfunctioning. It frequently suggests starting and ending sentences with quotes, although it also makes some useful suggestions. There seems to be an issue with analysis running synchronized, or something like that; when I type into a text field and Refine starts to run, it often blocks text entry. Selecting a suggested replacement blocks the app for half a second or so. Neither of these problems occurs with Grammarly or Language Tool. I also noticed a bunch of issues that Grammarly catches (like verb agreement) that Refine does not.
But this is an amazing first release and extremely promising. Congrats!
Languagetool is an open source tool you can run as a local spelling and grammar checker. It's different to Grammarly - less AI and more rules based. I often use both tools at the same time. I wrote a quick intro on how to self host this - https://martincapodici.com/2025/05/10/check-your-writing-usi....
You can use a local instance of LanguageTool in a docker container for this:
https://github.com/gardner/LocalLanguageTool
Gosh people love complicating things.
You don't need docker (a Linux-only piece of tech) to run a java application. Even though I'm on Linux, and docker experience is waaaaay better here than on any other platform, I wouldn't in my life consider using it to run LanguageTool.
A command without context is not very useful. If anyone wants to run LanguageTool locally, I suggest reading the official documentation page: https://dev.languagetool.org/http-server
Their recommended process is :
1. Install fasttext (it's an official Debian package, but you have to compile it on Windows).
2. Download and uncompress the LanguageTool release.
3. Create a config file.
4. Launch the server with the java command (of course, a JRE must be installed).
5. Connect to the API, e.g. with the browser extension.
Running a ready-made docker image replaces steps 1-4 and removes the need to install Java globally. Some will prefer it this way.
> 2. Download and uncompress the LanguageTool release.
> Running a ready-made docker image replaces steps 1-4
We can go pretty low level in the docker option too.
1. Download a Linux installation image
2. Download a hypervisor
3. Install Linux on the virtual machine
4. Install docker in a virtual Linux machine
5. Launch LanguageTool container
6. Configure networking between the host and the container in the guest
7. Connect to the API, e.g. with the browser extension
Obviously I initially oversimplified by omitting the configuration step. But adding download step to inflate the complexity is not a fair play ;-)
I'd argue that running a platforn-native artifact is both simpler and easier than involving virtualization. Even if steps 1-4 are done by some magical tool like rancher or docker-desktop
There is also a readily installable Flatpak:
https://github.com/sonnyp/Eloquent
It is restricted by Flatpak (i.e. Control Groups and Namespaces):
It is fat. But that's more an issue of Java itself. I only wish Languagetool didn't use Java, which is fine on a servers but horrible on personal-computers. Implemented in C, C++ or Rust and it would be probably already part of LibreOffice. Sonny Piers is the packager Flatpak, a prominent ex-member of GNOME board. He was removed due some Code-Of-Conduct thing which nobody can explain, due to issues within the Code-Of-Conduct.PS: LibreOffice had to fight years to remove Java which plagued the project.
TIL LibreOffice has finally removed Java. Thanks for educating me.
A lot of people would rather pay $15 than mess with docker containers
> Powered by local AI models
I worry that this will make my writing more likely to fail an AI coursework detector, which could really impact my life. The risk just isn't worth it till someone has tested the output through all the big players (turnitin etc.)
If I use correct punctuation marks my work will be more likely to be \detected\ as AI written; The risk just isn"t worth it so I never do that^
We'll soon need a writing tool that introduces spelling and grammar errors into our text and messes with punctuation so that we aren't accused of using LLMs.
It's funny how many people still think sloppy, mistake-filled writing is a sign of AI, as if their writing is at the same level as the image generators giving people six fingers, when the truth is the current LLMs use better English grammar than 99% of humans. Their writing may be kind of boring and standard, but they don't confuse "their" and "there."
It's hilarious, on some other sites one is immediately accused of using ChatGPT when using the n-dash (–) or m-dash (—) instead of the hyphen (-). Not an issue with the monospaced font here. ETA: I stand corrected.
ChatGPT tends to speak in american english, which means it's obvious to readers in the rest of the world because local phrases aren't used.
Seams weird to not have "How does this compare to Apple Intelligence Writing Tools" at least in the FAQ. Maybe refine is better or has more features, but the page doesn't even seem to acknowledge that a system level feature like this exists.
Worth to mention as another alternative: Harper[0]
[0]: https://writewithharper.com/
Harper is so basic that I can't recommend it.
Does not catch a singular/plural discrepancy between the subject and the verb in a sentence--a common mistake when the expressed thought applies equally to one thing or to many things.
Does not catch a missed indefinite article--a common mistake for speakers of languages that don't have articles. Similarly, does not catch the use of the indefinite article for a thing already mentioned before.
Does not even catch the obvious "don;t" typo.
Have you raised these as issues? The developer has been extremely responsive in my experience.
I use Harper in my Neovim setup and its been great. I just add things that are missing from the dictionary when I come across them in my codebase
How well does it handle standard international English? So many of the tools I've seen seem to only support American English.
Speaking of which, isn't it time to consider American English to be the standard one?
Colour and licence are so quaint.
Well, only Americans use it. There's no point in arguing about it your version of a language is better or worse but for the rest of the world it's incredibly annoying having to correct Zs with Ss when using LLMs or American only software.
I'm actually Russian. The whole world uses American English, by virtue of US dominance in all important technical and cultural spheres.
"The only whole world uses American English" ... I think you might be living in a bubble. It sticks out like a sore thumb.
He is actually correct. I am neither American nor English.
That's certainly not the case in the majority of Europe, Australia, New Zealand, East Africa, West Africa, South Africa, PNG, Fiji, Philippines, South Asia and South East Asia. Really it's mostly just America and Canada that use American.
I could cite many sources but https://en.wikipedia.org/wiki/American_and_British_English_s... provides a simple view.
We don't use American English in Canada either.
Commonwealth nations tend to not use American English. They form a fairly major part of the world.
I honestly don't think this is true. In many places outside the US our devices will default to US spelling, making it appear as if the language spoken by the locals is US English.
But when you look at things like the type of English being taught in school, or the language used by the government, it'll be UK/International English.
I regularly correspond to people from all over the world, India, China, Brazil, France, and they all tend to use US English, thus the suggestion.
Commonwealth countries are relatively minor in population compared to the USA and the rest of the world.
And yes, default US spelling in devices does play a role. See prev. point about technological and cultural dominance.
> Commonwealth countries are relatively minor in population compared to the USA and the rest of the world.
Given India's presence in the Commonwealth, this seems an odd assertion. 2.6 billion people live in the Commonwealth[1].
[1] https://en.wikipedia.org/wiki/Member_states_of_the_Commonwea...
Kinda forgot India is still part of the Commonwealth. Maybe because all Indians I ever corresponded with uniformly used US spelling, likely because they work a lot with the USA. YMMV.
I think we need to call American English "American", then you can speak American and I can speak English.
I could care less about this attempt at trolling, but I won't.
We have enough American cultural hegemony as it is. It frustrates me no end that I regularly am unable[1] to configure software to use my preferred version of English.
[1] or it's extremely difficult. CLion I'm looking at you; every available option I have is set to British English, but still you insist on telling me my Colours should be corrected to Color etc. :(
English is a bastard language of French and Saxon. But we don’t get upset over that because Britain made the world English. Now the USA is making the world American. That ship has sailed.
Europe still teaches British English in schools.
I see that you're using gemma3n which is a 4B parameter model and utilizes around 3GB RAM. How do you handle loading/offloading the model into the RAM? Or is it always in the memory as long as the app is running?
I can see this as a major issue. If you start using this for grammar checking, you're basically subtracting 3GB of RAM from your system.
If this were open source AGPL it'd be an insta-download.
Neat idea. I see why the fluency feature is off by default. It constantly rewords things, adds random quotations, or does something pretty silly https://imgur.com/oVSWmtN
The Grammar feature seems to have weird suggestions/cycles too on a little bit more testing. Curious to see how this improves. A local only, one-time-purchase grammarly alternative is appealing!
Has anyone tested how this works in comparison to locally-hosted LanguageTool?
Minor thing with the website but the download button is not entirely clickable, if you don't click on the text you don't get to download it, due to the div > a and the anchor being just the text, consider styling the anchor to have the padding so the whole thing is clickable. :)
I tried it and it worked pretty well. I'm curious about the principle behind it. Is there any model or API being called?
No API to hook into one's favorite tools, such as emacs.
It does this
because what's under the hood is this, and prompts are hardcodedunsloth/gemma-3n-E4B-it-GGUF
Incredible. Has any non-AI grammar checker ever returned a "safety" refusal to the user before? Is this the power of LLM wrappers?
Imagine if WinRAR didn't let you archive or name your files what you want. How we were trusted with such DESTRUCTIVE tools all these years is beyond me. /s
why is there no open source alternatives to this? Seems ripe to be just built.
Harper?
https://github.com/Automattic/harper
I'm really wondering if/why Grammarly still has a business case, in this age of LLMs.
Mainly because (cloud-hosted, bigger & smarter) LLMs probably offer no added privacy benefit. Self-hosted models are dumber, neutered and unpredictable.
It's beyond me how they still refuse to offer an official API.
Gemini subscription costs as much, allows me to do much more, and I can call it from vim or emacs without some arcane acrobatics
They used to have an API, which they removed about two years ago or so. Until then, there were a bunch of 3rd-party Grammarly plugins, including a pretty decent one for VSCode.
Very cool product. And thank you for not making it a subscription.
> Can’t find the answer you’re looking for? Feel free to sent us an email at support@refine.sh
Should be "send". Especially with this kind of product, you shouldn't have grammar mistakes on the website :)
Does anyone know of an equivalent for translations, ideally open source though?
Does this do tone? That is the single largest feature of Grammarly’s Slack plugin that I absolutely loved, and if my work allowed it, I’d have it again.
Wonderful! I've given it a go, works in Apple's Notes app, but it does not seem to trigger suggestions in Chrome, Firefox or Slack. It does however highlight misspellings there. Any idea what can I do to enable suggestions there? I was looking for a product like this.
Same! Interested in this but need it to work for Slack, where I often type more informally and don't take the time I should to proofread.
Nice product! Does it support other language like Chinese?
Couldnt make it to show suggestions in vscode/cursor. I would like to use the tool, but i'd expect it to work consistently across all widgets in the system (i.e. like superwhisper). Is there a technical limitation here or my misconfiguration of things?
My biggest problem with Grammarly is how buggy it is. How often it doesn’t work as intended and either messes up formatting or doesn’t change the text when I click.
Does Refine solve this?
What I'd be interested in would be something I could host on my local server (e.g. with ollama) to get suggestions on my laptop, where I write typst or markdown with Zed or VSCode.
I realize I'm a niche :)
Does anybody know of such a tool?
I'm missing some information on how this works (a LLM? which? Do I need to bring an API key? Does this work offline?) and what I can expect in terms of performance/battery hit.
> Lightning Fast. Local processing means instant results without internet dependency or delays. > Always Available. Works offline, on flights, in coffee shops, anywhere you write.
Two of your 4 questions were answered in the first content block
It is Gemma 3n, I can't give feedback yet on the battery hit, But I would not expect anything bad as these models have been developed for much smaller devices (Phones)
Considering that it mentions offline capability I'd say local tiny LLM.
LLM usage would immediately disqualify this from most academic writing.
Could you elaborate why?
I've installed it to give it a try but it does absolutely nothing in any application I use. I did give it required permissions.
Isn't privacy a concern? How do consumers ensure that data is not going to captured in a future update without it being open source or having third party security audits?
That's a concern with all apps ever so idk what answer you're expecting
Disagreed. It's not concerned with all apps, because most commercial applications have legal entities explaining how they use collected data. In this particular case, this is something called 'Refine', and it's not a legal entity, therefore, questioning its data privacy approach is fully legitimate.
Except the open-source ones, or sandboxed[1] ones without any auto-update functionality (not sure if this app has any).
[1] Loosely; I’d say not referencing any networking entrypoints or dlsym() also counts, as working around that would be very non-deniably malicious.
Perhaps this type of software could be either open-source with full code accessibility, or proprietary but from a highly trustworthy entity responsible for privacy both legally and reputationally. Currently, both approaches are missing.
I also strongly believe that it’s a primary question when we talk about keyboard input.
Holy crap, a local-only app with one-time purchase *and* a free demo? You don't see that very often these days
If you are running local LLMs what is the hardware requirement in my machine? Don't see any mention of that.
Gemma 3n (the model used by this app) would run on any Apple Silicon device (even with 8GB RAM).
Yup, but you're automatically giving up a ton of RAM that could be better used for Slack.
It'd be neat if there was an open source solution this polished.
I would pay 50$ a year if this had a vim integration.
How big is the model that powers this?
8B (Gemma 3n)
https://ai.google.dev/gemma/docs/gemma-3n
I wonder if there was some complexity with internationalizing or localizing English (UK English, etc) outside of the primary build case.
Does this only supports English?
Theoretically, it could support over 140 languages, as it is powered by the newly released Gemma 3n model. However, I haven’t tested many languages yet.
In upcoming releases, Refine will support custom prompts and BYOK (Bring Your Own Key), allowing you to use any large language model you want.
What's the rationale behind adding BYOK? Or the advantages?
You do realise you're already using an optimised model built for everyday devices, and that model includes some serious innovations in parameter-efficient processing, right?
You're a great developer, and it looks like you're thinking about adding features like BYOK quickly to please more users. But in doing that, you might be missing the real innovation you've already created. You've basically built a version of Grammarly without the privacy issues that make most IT departments ban it.
No one wants Grammarly or your tool sending corporate emails or documents to a language model. Privacy is what sets you apart from Grammarly. It's your biggest feature right now. Add a big table to your site comparing your products privacy with Grammarly's. That's your strongest selling point, and probably the only feature that can truly compete with the big players.
My advice? Keep improving the app, but keep the model local. Keep it private. That's the killer feature you've got.
One can have a beefier server at home and may want to use it for running the model while out and about with their laptop.
[dead]
[dead]