Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Building a no-code toxicity classifier by talking to GitHub Copilot

surgehq.ai

Post reply on HN
Print view View original

101–110 of 152 posts

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#101

Post by dgb23 »

caymanjim wrote: post #59
junon wrote: post #53

Earlier quoted context omitted.

If this is what software engineering turns into, put a bullet through my head please.

I haven't tried Copilot, and based on what I've read, I don't think I'd want to use it in its current form, but I'd love for software development to evolve to the point that I never have to write any boilerplate code again. Even with DSLs, code generation, autocompletion, snippets, and countless libraries and frameworks to draw upon, the bulk of what I do as developer is write the same boring code over and over again…

The boilerplate problem has been solved decades ago. You can use macros or code generation. Letting Copilot do that for you seems a bit like an ugly workaround. We sometimes need those, but at some point we should be taking a step back and design a program instead of hacking away.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#103

Post by dgb23 »

ShamelessC wrote: post #66
tomerv wrote: post #34

The first comment asks Copilot to import all the libraries needed for a toxicity classifier, and it imports libraries such as re (regex engine) and nltk (natural language toolkit). But what if I wanted a classifier for toxic chemicals and not toxic speech? That was my first thought when I saw "toxicity" in the title. I'm now imagining a very frustrated junior developer a few years from now trying to argue with Copilo…

> I'm now imagining a very frustrated junior developer a few years from now trying to argue with Copilot to write code for a classifier for chemical compounds, but it just spits out code for classifying text. So just the future version of a junior developer not knowing how to use their tools? Yeah, that scans. Still sounds incredibly useful however. The alternative is, of course, a junior developer fumbling as they t…

That’s a very important thing to experience.

There’s not always a quick fix or easy path. You can’t always patch existing stuff together or just wait until the problem goes away.

And when a tool helps you too much, then is there really a point in what you’re doing? It’s not even a learning experience anymore.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#104

Post by ShamelessC »

IshKebab wrote: post #93
ShamelessC wrote: post #70

Earlier quoted context omitted.

> The negativity here just seems like sour grapes or weird goal posts. Indeed. Every negative comment I have seen here has been a shallow dismissal by someone who clearly hasn't engaged with the tool. I'm not sure why people here are so primed to shit all over anything potentially innovative, seemingly even without background knowledge. Like, is there something inherently offensive to coders about a model that threat…

> I'm not sure why people here are so primed to shit all over anything potentially innovative Maybe jealousy - people often downplay others' achievements to make theirs feel better. Or pride - "I don't need no stinking AI assistant! What are you saying? I couldn't write this myself?". I find the latter is a common reaction to static types too.

> Maybe jealousy

Somehow, I suspect this the most. It's the defensive tone they tend to strike, I think. Particularly apparent in threads about new research. Lots of "I actually had this idea", and "my concern for this is [slippery slope fallacy here]".

I sincerely hope that I never become so egotistical about my own achievements (or lack thereof) that I instinctively despise those who achieve more. Fuck that.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#105

Post by xodjmk »

ShamelessC wrote: post #70
hombre_fatal wrote: post #60

This is absolutely insane. I had no idea Copilot was this good. The negativity here just seems like sour grapes or weird goal posts. Sure, it makes mistakes and needs verification. But know what also makes mistakes and needs verification? All the code I already manually write as I tediously ratchet towards a solution. Removing some cycles from that process is a win. Just stubbing out close-enough boilerplate is a win…

> The negativity here just seems like sour grapes or weird goal posts. Indeed. Every negative comment I have seen here has been a shallow dismissal by someone who clearly hasn't engaged with the tool. I'm not sure why people here are so primed to shit all over anything potentially innovative, seemingly even without background knowledge. Like, is there something inherently offensive to coders about a model that threat…

You can't imagine how some people might have an adverse reaction to a low-barrier of entry arbitrarily defined self-appointed moral policing 'AI' tool generating framework? Not all ideas are good ideas. It doesn't mean the ops are not talented, just misdirected.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#106

Post by Aeolos »

qayxc wrote: post #100
Ozzie_osman wrote: post #9

This is a game-changer, even if it doesn't work 100% of the time. I only infrequently need to use notebooks and dataframes, I'd say once every few months. Frequently enough that I have a vague idea of what I need to do but not frequently enough that I can remember syntax. With this, I don't need to memorize the syntax OR be bottlenecked on looking at documentation or stack overflowing the commands I need.

> With this, I don't need to memorize the syntax OR be bottlenecked on looking at documentation or stack overflowing the commands I need. In other words: you're celebrating the fact that a tool allows you to become more and more incompetent. I don't have much hope for future generations at this point.

People said that during the switch from assembly to C, and then again from C to managed languages. Yet, at each point, there are more and better software engineers, solving more and more challenging problems at each step - the modern web would not exist if all we had at our disposable was 80's era assembly.

Aren't you at least a bit curious what new possibilities this technology could enable? What new discoveries could e.g. an expert doctor or a biologist achieve given access to programming tools without spending decades learning programming?

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#107

Post by Aeolos »

faddypaddy34 wrote: post #39
da39a3ee wrote: post #30

Earlier quoted context omitted.

I think you're missing the point: this is starting to open the door to people who can't code.

I think the point is if you cannot/do not know how to code you cannot confirm what co-pilot is doing. Especially when it comes to complex topics like drawing context from natural online language using machine learning.

That's where testing comes in, no?

Entering the programming field then becomes an iterative loop of you instructing an AI to generate code, generate tests and iterate / re-adjust until it does what you want.

"You" as in the next generations of programmers in a decade or two.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#108

Post by ShamelessC »

xodjmk wrote: post #105
ShamelessC wrote: post #70

Earlier quoted context omitted.

> The negativity here just seems like sour grapes or weird goal posts. Indeed. Every negative comment I have seen here has been a shallow dismissal by someone who clearly hasn't engaged with the tool. I'm not sure why people here are so primed to shit all over anything potentially innovative, seemingly even without background knowledge. Like, is there something inherently offensive to coders about a model that threat…

You can't imagine how some people might have an adverse reaction to a low-barrier of entry arbitrarily defined self-appointed moral policing 'AI' tool generating framework? Not all ideas are good ideas. It doesn't mean the ops are not talented, just misdirected.

> low-barrier of entry arbitrarily defined self-appointed moral policing 'AI'

So, a software developer?

Just kidding. Perhaps it would clear things up to know that "toxicity" classification is simply an introductory topic in natural language and machine learning. It is an interesting "problem" to try to solve precisely because of how ambiguous language gets. As far as I can tell, the article is mostly not concerned with the quality of the the classifier and is meant to be a proof of concept.

In any case, people have been coding and using such classifiers all over the internet since their inception. Believe me - this isn't accelerating that and the damage is likely mostly done. My advice? Self-host a web server.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#109

Post by ShamelessC »

dgb23 wrote: post #103
ShamelessC wrote: post #66

Earlier quoted context omitted.

> I'm now imagining a very frustrated junior developer a few years from now trying to argue with Copilot to write code for a classifier for chemical compounds, but it just spits out code for classifying text. So just the future version of a junior developer not knowing how to use their tools? Yeah, that scans. Still sounds incredibly useful however. The alternative is, of course, a junior developer fumbling as they t…

That’s a very important thing to experience. There’s not always a quick fix or easy path. You can’t always patch existing stuff together or just wait until the problem goes away. And when a tool helps you too much, then is there really a point in what you’re doing? It’s not even a learning experience anymore.

> And when a tool helps you too much, then is there really a point in what you’re doing? It’s not even a learning experience anymore.

Have you seen the documentary about AlphaGo? After watching it, and seeing Lee Sedol just utterly devastated by losing to a computer, I felt like I too would surely feel the same thing in my life. I mean, surely Lee Sedol is a far more skilled Go player than I am a skilled programmer.

Anyway, sorry for the rambling. I agree - it's deeply important to _actually_ learn how things work. That's why I wouldn't recommend copilot to a junior dev. Unfortunately, the way things are going - those junior developers are going to use it anyways and I tend to be more of a realist than an idealist.

Re: Building a no-code toxicity classifier by talking to GitHub Copilot

#110

Post by visarga »

BeefWellington wrote: post #45
abeppu wrote: post #35

We're all focusing on the weaknesses of co-pilot (the comments can be longer than the code produced; you need to understand code to know when to elaborate your comment, etc). But also ... what do you need to know to recognize that the concept of a 'toxicity classifier' is likely broken? We can do _profanity_ detection pretty well, and without a huge amount of data. But with 1000 example comments, can you actually get…

A few years ago I did some work with IBM's Watson Twitter integration. One of the fun things you could do was sentiment analysis. It was reasonably accurate for the extremes but anything in the gray area would be wildly off. A politely worded tweet that was scathing would come across high on the positive sides of the scale, whereas a perfectly reasonable sentence that included profanity as used in a quote would immed…

> I think the most basic english language parser could output something along the lines of what was suggested, given an understanding of what valid Python should look like.

OK, I am waiting for you to propose a basic language parser that can do it. There's a reason we're only now having this debate - it was unconceivable 5 years ago, in the era of basic language parsers.

Post reply on HN