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

GitHub Copilot Workspace: Technical Preview

github.blog

Post reply on HN
Print view View original

321–330 of 333 posts

Re: GitHub Copilot Workspace: Technical Preview

#321

Post by screye »

packetlost wrote: post #10
studenthrow3831 wrote: post #3

Student here: I legitimately cannot understand how senior developers can dismiss these LLM tools when they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years. I'm not a singulatarian, but this seems like a brutal S-curve we're heading into. I also have a hard time believing that there is enough software need to make such an extreme prod…

The hardest part of software development is not writing code, full stop. It never has been and it never will be. The hard part is designing, understanding, verifying, and repairing complex systems. LLMs do not do this, even a little bit.

LLMs remove the most annoying bits, making it possible for 1 person to do 3 people's work. LLMs are also good at fixing minor bugs. So version upgrades, minor maintenence and finding the right API handshakes will soon be doable for big LLMs without user supervision. Lastly, LLMs are accelerating existing tailwinds towards software commodotization. If an LLM can create a good-enough website on a low code platform, how likely are you to hire a front end engineer for the last 10% of excellence ?

Think about how many jobs are 'build a website', 'build an app' or 'manage this integration' style roles. They are all at risk of being replaced.

> hardest part of software development is not writing code

I agree, but you have to write a lot of code before you become good enough to think that clearly. If juniors don't get the opportunity to work their way up to a senior, then they might just never pick up the right skills. What's more like is is that CS education will undergo drastic changes, and masters/specialization might become a more degree requirement. But, those already on the market are in for a big shock.

Re: GitHub Copilot Workspace: Technical Preview

#323

Post by Sammi »

WorldMaker wrote: post #317
Sammi wrote: post #299

Earlier quoted context omitted.

I can regularly get it to autocomplete big chunks of code that are good. But specifically only when it's completely mind numbingly boring, repetitive and derivative code. Good for starting out a new view or controller that is very similar to something that already exists in the codebase. Anything remotely novel and it's useless.

I have strange documentation habits and sometimes when you document everything in code comments up front, Copilot does seem to synthesize from your documentation most of the "bones" that you need. It often needs a thorough code review, but it's not unlike sending a requirements document to a very Junior developer that sometimes surprises you and getting back something that almost works in a PR you need a fine tooth c…

Yes I use this trick regularly too.

Re: GitHub Copilot Workspace: Technical Preview

#324

Post by Roonerelli »

PoignardAzur wrote: post #51

I think Microsoft is going the wrong direction with Copilot (though it's a reasonable direction given their incentives). Right now Copilot is terrible at large changes to complex codebases; the larger and more complex, the worse. But it's great at suggesting very short snippets that guess exactly what you were in the middle of writing and write it for you. I wish Copilot focused more on the user experience at the sma…

Jetbrains are doing something like this (in the .net space) https://blog.jetbrains.com/dotnet/2024/04/30/jet-brains-ai-a...

Re: GitHub Copilot Workspace: Technical Preview

#325

Post by zackmorris »

zackmorris wrote: post #217
studenthrow3831 wrote: post #3

Student here: I legitimately cannot understand how senior developers can dismiss these LLM tools when they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years. I'm not a singulatarian, but this seems like a brutal S-curve we're heading into. I also have a hard time believing that there is enough software need to make such an extreme prod…

After programming for 35 years since I was 12 and learning everything from the transistor level up through highly abstracted functional programming, I'm a total doomer. I put the odds of programming being solved by AI by 2030 at > 50%, and by 2040 at > 95%. It's over. Programming (automating labor) is the hardest job there is IMHO, kinda by definition. Just like AGI is the last problem is computer science. You notice…

Hey I was feeling particularly dour when I wrote this, but not everything is doom and gloom. AI/AGI will be able to solve any/all problems sometime between 2030 and 2040. I take the alarmist position because I've been hit with bad news basically every day since the Dot Bomb and 9/11, and don't feel that I'm living my best life. But we can manifest a brighter tomorrow if we choose to:

'ChatGPT for CRISPR' creates new gene-editing tools:

https://www.nature.com/articles/d41586-024-01243-w

https://news.ycombinator.com/item?id=40205961

The intersection of AI with biology will enable us to free ourselves of the struggles of the human condition. Some (like me) are concerned about that, but others will run with it and potentially deliver heaven on Earth.

The way I see it all going is that a vanishingly small number of people, roughly 1 in 10,000 (the number of hackers/makers in society) will work in obscurity to solve the hard problems and get real work done on a shoestring budget. But we'll only hear about the thought leaders and billionaires who do little more than decide where the resources flow.

So the most effective place to apply our motivation, passion and expertise will be in severing the hold that capital has over innovation. Loosely that looks like UBI and the resource economy I mentioned, which I just learned has the name Universal Basic Services (UBS), and intentionally avoids complications from the money side being manipulated by moneyed interests:

https://en.wikipedia.org/wiki/Universal_basic_services

The idea is that by providing room and board akin to an academic setting, people will be free to apply their talents to their calling and work at an exponentially faster rate to get us to a tech utopia like Star Trek, instead of being stuck in the path we're on now towards a neofeudalist tech dystopia.

Sorry if I discouraged anyone. I truly believe that there is still hope!

Re: GitHub Copilot Workspace: Technical Preview

#326

Post by BoorishBears »

Zambyte wrote: post #305
BoorishBears wrote: post #257

Earlier quoted context omitted.

Llama 3 just came out so they couldn't have used it, and Claude Haiku is the smallest cheapest closed source model out there from what I've seen. Github is likely using a GPT-4 class model which is two (massive) steps up in capabilities in Anthropic's offerings alone

Yeah I just mentioned Llama to point out that the open weight models have been really catching up. Microsoft is almost certainly using GPT-4 given their relationship with ClosedAI, but I would definitely not put GPT-4 (nor Turbo) "two massive steps up" from Claude 3 Opus. I have access to both through Kagi, and I have found myself favoring the responses of Claude to the point where I almost never use GPT(TM) anymore.

You're misreading in multiple ways, maybe in a rush to dunk on "Closed AI".

Github Copilot is not the same as Copilot Chat which uses GPT-4, there still some uncertainty on if Copilot completions use GPT-4 as outsiders know it (and iirc they've specifically said it doesn't at some point)

I also said Haiku is two massive steps behind Anthropic's offerings... which are Sonnet and Opus.

Anthropic isn't any more open than OpenAI, and I personally don't attribute any sort of virtue to any major corporation, so I'll take what works best

Re: GitHub Copilot Workspace: Technical Preview

#327

Post by Zambyte »

BoorishBears wrote: post #326
Zambyte wrote: post #305

Earlier quoted context omitted.

Yeah I just mentioned Llama to point out that the open weight models have been really catching up. Microsoft is almost certainly using GPT-4 given their relationship with ClosedAI, but I would definitely not put GPT-4 (nor Turbo) "two massive steps up" from Claude 3 Opus. I have access to both through Kagi, and I have found myself favoring the responses of Claude to the point where I almost never use GPT(TM) anymore.

You're misreading in multiple ways, maybe in a rush to dunk on "Closed AI". Github Copilot is not the same as Copilot Chat which uses GPT-4, there still some uncertainty on if Copilot completions use GPT-4 as outsiders know it (and iirc they've specifically said it doesn't at some point) I also said Haiku is two massive steps behind Anthropic's offerings... which are Sonnet and Opus. Anthropic isn't any more open tha…

I... don't think I misread you? Maybe you didn't mean what you wrote, but what you said was:

> Github is likely using a GPT-4 class model which is two (massive) steps up in capabilities in Anthropic's offerings alone

Comparing GPT-4 to Anthropics offerings, which, as you say, includes Sonnet and Opus.

> Anthropic isn't any more open than OpenAI, [...] so I'll take what works best

I understand that, and same here. I don't prefer Claude for any reason other than the quality of its output. I just think OpenAIs name is goofy with how they actually behave, so I prefer the more accurate derivative of their name :)

Regarding what model Copilot Completions is using - point taken, I have no comment on that. My original comment in this thread was only meant to point out that open weight models are getting a lot better. Not saying they're using them.

Re: GitHub Copilot Workspace: Technical Preview

#328

Post by ianbutler »

A little late to the party here, but we (https://www.bismuthos.com) offer something already to scratch this itch. We provide a workspace to build Python backends. Chat on the left, code and visual editors on the right. However, we also handle deployments, data storage (we have a blob store), serving (we built a home grown function runtime) and logging. The experience is tightly integrated with the copilot and the idea is to get ideas off the ground as quickly as possible. We're early days and you can try it for free right now.

Re: GitHub Copilot Workspace: Technical Preview

#329

Post by sussmannbaka »

imiric wrote: post #286
sussmannbaka wrote: post #284

Earlier quoted context omitted.

No. The progress is not being ignored. Normal people just have a hard time getting excited for something that is not useful yet. What you are doing here is the equivalent of popular science articles about exciting new battery tech - as long as it doesn’t improve my battery life, I don’t care. I will care once it hits the shelves and is useful to me, I do not care about your list of acronyms.

I was arguing against the claim that progress has flatlined, and when I gave concrete examples of recent developments that millions of people are using today, you've now shifted the goalpost to "normal" people being excited about it. But sure, please tell me more about how AI is a fad.

You are seeing enemies where there are none. I am merely commenting on AI evangelists insisting I have to be psyched about every paper and poc long before it ever turns into a useable product that impacts my life. I don’t care about the internals of your field, nobody does. Achieve results and I will gladly use them.

Re: GitHub Copilot Workspace: Technical Preview

#330

Post by BoorishBears »

Zambyte wrote: post #327
BoorishBears wrote: post #326

Earlier quoted context omitted.

You're misreading in multiple ways, maybe in a rush to dunk on "Closed AI". Github Copilot is not the same as Copilot Chat which uses GPT-4, there still some uncertainty on if Copilot completions use GPT-4 as outsiders know it (and iirc they've specifically said it doesn't at some point) I also said Haiku is two massive steps behind Anthropic's offerings... which are Sonnet and Opus. Anthropic isn't any more open tha…

I... don't think I misread you? Maybe you didn't mean what you wrote, but what you said was: > Github is likely using a GPT-4 class model which is two (massive) steps up in capabilities in Anthropic's offerings alone Comparing GPT-4 to Anthropics offerings, which, as you say, includes Sonnet and Opus. > Anthropic isn't any more open than OpenAI, [...] so I'll take what works best I understand that, and same here. I d…

I used "in Anthropic's capabilities" intentionally: it's two steps up in what they can do from Claude Haiku
Post reply on HN