Live data from Hacker News

Show HN: TabNine, an autocompleter for all languages

tabnine.com

141–150 of 193 posts

Re: Show HN: TabNine, an autocompleter for all languages

#142
post #131

Earlier quoted context omitted.

That can lead to ("Work", "Worry", "Word") so you'd then have to type the 'k'. Now you could have ("Work", "Worker", "Worked") and still are missing the variant you want. It'd be nice to long press "Work" at step one, get that completed without a space being inserted, then tap 'i' to get ("Working", "Workings", "Workingmen")

It turns out that iOS prediction will make a provisional guess based on what you typed and will go back and adjust its autocorrections as you type subsequent words. You can see this more clearly if you use dictation, but allegedly it won't do as well if you use the word corrections every time.

I’ve noticed that. All my work-related jargon is suggested pretty damn well when it’s appropriate.

What I fear is that the autocomplete can reveal my NSFW jargon. It’s the same keyboard even when Safari is in Private Mode, right?

Re: Show HN: TabNine, an autocompleter for all languages

#143
post #67

I've prototyped something like this in the past using n-grams and it was surprisingly effective. When I think it gets really interesting is if you marry ML/NLP tactics with traditional static code analysis. So you can imagine the ML engine generating the suggestions with the static analyzer ranking the suggestions intelligently. It's kind of similar to the original AlphaGo where you have the model generate the potent…

I believe the Visual Studio team is working on something like this to provide better autocomplete for specifically C#.

Re: Show HN: TabNine, an autocompleter for all languages

#144
post #115

Played with free for a bit, 200KB is quite a bit low, didn't get any completions. Purchased the premium licence. Gotta say stripe integration is very smooth. Overall after a couple of hours of playing with this. My mind is quite blown away. This is absolutely amazing. Hopefully Microsoft or someone acquires this technology for a fat sum and open sources it. I've thought about code completion smarts for a long time. Y…

Did you see much improvement in the suggestions after purchasing and using the larger index?

I was using it on a large project so 15 MB got me no completions on the files I cared.

$30 is a pretty cheap price for a pattern based completion engine.

It’s the first time I’ve seen it work well. It was completing fairly long statements and I was pleasantly surprised how close the first few results were to what I wanted.

The whole configurationless, all language completion using pattern analysis and fast index lookups in a very easy to install delivery is great execution.

This are the kinds of little small things that make me think “why didn’t I do this?”

I wish the author gave, 30 day free premium trial. A lot people would be willing to spend on the license IMHO.

Re: Show HN: TabNine, an autocompleter for all languages

#146

Earlier quoted context omitted.

To allow the person(s) who implemented it to get paid for their efforts and hard work?

It's also a very competitive price. $29 is excellent for a piece of software that helps my day to day. It's really a sweet spot between very reasonable, and a bit pricey. I'm so happy about this project, and hope it works well (I'll be trying it tonight after work)

No craftsman of any kind would whine about a (good) tool that costs $29 and makes him more productive. I do not see why a progrommer should do that.

Re: Show HN: TabNine, an autocompleter for all languages

#147
post #122

Earlier quoted context omitted.

...as risky as installing a proprietary editor plugin which updates automatically, yes. Also, AFAIK most understandings of MIT, BSD, and Apache 2.0 licenses require you to acknowledge the copyright holders of the source code you compile into your binary, even if the licenses permit binary distribution. I can't find your "Copyright (c) 2018 Tokio Contributors" or "Copyright (c) 2014 The Rust Project Developers" that I…

MIT only requires source attribution. It's the BSD licenses that require attribution for binary forms of redistribution. Still, it is good manners and good cover-your-arse practice to attribute whatever free software work they used (Google does this with their giant "open source licenses" page).

MIT has no special language regarding source or binary distribution, it simply states:

> The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

It's up to a court to decide what "copies or substantial portions of the Software" means.

Personally, I've always very much interpreted that to mean both binary and source.

Post reply on HN