Live data from Hacker News

Copilot regurgitating Quake code, including sweary comments

twitter.com

331–340 of 672 posts

Re: Copilot regurgitating Quake code, including sweary comments

#331

Earlier quoted context omitted.

> The python example is using floats for currency. Dumb question, but what is the proper way to handle currency? Custom number objects? Strings for any number of decimal places?

This is a complex topic, mainly for two reasons: 1. it works on two layers (storage and code) 2. there is a context to take care of. [Modern] programming languages have decimal/rational data types, which (within limits) are exact. Where this is not possible, and/or it's undesirable for any reason, just use an int and scale it manually (e.g. 1.05 dollars = int 105). However, point 2 is very problematic and important t…

> How do account 3 items that cost 1/3$ each (e.g. if in a bundle)?

You never account for fractional discrete items, it makes no sense. A bundle is one product, and a split bundle is another. For products sold by weight or volume, it's usually handled with a unit price, and a fractional quantity. That way the continuous values can be rounded but money that is accounted for needs not be.

Re: Copilot regurgitating Quake code, including sweary comments

#333
The uproar over Copilot is kind of hilarious. Maybe it's SWE's realizing that they might not be as irreplaceable as they seem but its an awful lot of salty comments. If anything I think Copilot is a really cool PoC and shows just how close we are getting to automating large portions of the code writing process which we should all welcome as more cycles can be spent on architecture and system design.

Re: Copilot regurgitating Quake code, including sweary comments

#334

I hate to be the one that says this but I think it‘s true: "So you are an SWE and you take a break from work to go to Hackernews to complain that Github's Copilot, which is an AI-based solution meant to help SWEs, is utter shit and completely unusuable. And then you go back to writing AI-based solutions for some other profession. Which is totally not shit or anything.“ Can anybody put this more elegantly?

You mean like the insanely annoying AIs that replaced Google search? The idiotic one that files Javascript books under "Law" in Amazon or the insulting one who runs Ad Sense and thinks my wife isn't good enough and I am stupid enough to leave her for some mail order bride?

Javascript books under "Law" is hilarious

Re: Copilot regurgitating Quake code, including sweary comments

#335
post #149

Based on all the negative comments so far, and based on this website's aptitude at predicting the viability of a product, it really seems like Copilot is bound to be a success.

Right?

Even given all its initial problems, I don't see a world where people completely avoid using it.

Re: Copilot regurgitating Quake code, including sweary comments

#336
post #203

Earlier quoted context omitted.

The problem is also stupid people and companies. My last job they wanted me to invoice them hours worked, which was some number like 7.6. This number plays badly when you run it through GST and other things - you get repeaters. So I looked up common practice here, even tried asking finance who just said "be exact", and eventually settled on that below 1 cent fractions I would round up to the nearest cent in my favour…

Well, they did say to be exact, and you handed them approximations, so...

The “exact” version they wanted was full of approximations too. They just didn’t have enough numerical literacy to understand how to say how much approximation they are ok with.

I guarantee nothing in anyone’s time accounting system is measured to double-precision accuracy. Or at least, I’ve never quite figured out the knack myself for stopping work within a particular 6 picosecond window.

Re: Copilot regurgitating Quake code, including sweary comments

#337

Earlier quoted context omitted.

> This product seems like it needs some more thought. Maybe a way to comment, flag, or otherwise call out bad output? Wait for your colleagues to use it, fix the bad code in the pull request, and wait for copilot to learn from the new training data you just provided!

This is actually a good idea that is missing from nearly every machine learning product. How do you back propagate lessons from user interaction into future training of the model? It can be done, I can't think of a place I've seen it done though.

It's tricky, because once you start accepting user feedback, you need to moderate it, or else someone will poison your model for fun and profit.

Re: Copilot regurgitating Quake code, including sweary comments

#338
post #30

They have 4 hand picked examples on their homepage: https://copilot.github.com/ One has the issue with form encoding: https://news.ycombinator.com/item?id=27697884 The python example is using floats for currency, in an expense tracking context. The golang one uses a word ("value") for a field name that's been a reserved word since SQL-1999. It will work in popular open source SQL databases, but I believe it would bom…

> The ruby one isn't outright terrible, but shows a very Americanized way to do street addresses that would probably become a problem later. As someone who has been coding up address storage and validation for the past week in my current job, that one really made me laugh. Mostly because it tries to simplify all the stuff I have been analyzing and mulling over for a week into a single auto-complete. Spoiler: The Gith…

I spend most of my time thinking about the problem, how to solve it, and considering ramifications of my decisions before ever putting code in the IDE. So Copilot comes around and it autocompletes code for me. But I still have to read what it suggested, making edits to it, and consider if this is solving the problem appropriately.

So, rather than helping people program better, all its done is replace a bunch of the offshore cut-and-paste shops with "AI."

Re: Copilot regurgitating Quake code, including sweary comments

#339
post #30

They have 4 hand picked examples on their homepage: https://copilot.github.com/ One has the issue with form encoding: https://news.ycombinator.com/item?id=27697884 The python example is using floats for currency, in an expense tracking context. The golang one uses a word ("value") for a field name that's been a reserved word since SQL-1999. It will work in popular open source SQL databases, but I believe it would bom…

What are the statistics of Copilot based on a validation set? How often does it get code right? I want to see hard statistics, not 4 hand-picked examples.

Yeah. And like how would you even devise a metric? Like compile it down to assembly and see if it’s similar logic?
Post reply on HN