jaredsohn

The last thing I saw on it was a video saying that calling Devin an AI software engineer that does Upwork tasks is a lie based in what was shared in an earlier video.

https://www.youtube.com/watch?v=tNmgmwEtoWE

"I broke down the Devin Upwork video frame by frame, and here I show what Devin was supposed to do, what it actually managed to do instead, and how bad a job of that it did.

On the whole that's not surprising given the current state of Generative AI, and I wouldn't be bothering to debunk it, except: 1) The company lied about what Devin could do in the video description, and 2) a lot of people uncritically parroted the lie all over the Internet, and 3) That caused a lot of non-technical people to believe that AI might replace programmers soon. "

My understanding this is about using Authy on Macs rather than on mobile since it talks about the MacOS store.

You should be able to provide more data than that in the input if the output doesn't use the full 4k tokens. So limit is context_size minus expected length of output.

Yes, it is better if for example using it via an API to classify. Deterministic behavior makes it a lot easier to debug the prompt.

"Customize GPT" -> "Custom Instructions" lets you do this; tell it your preferred language assuming you want code examples.

>I would argue the problem is how dependencies in general are added to projects

I haven't done anything with this myself (just brainstormed a bit with chatgpt) but I wonder if the solution is https://docs.npmjs.com/cli/v10/commands/npm-ci

Basically, enforce that all libraries have lock files and when you install a dependency use the exact versions it shipped with.

Edit: Can someone clarify why this doesn't work? Wouldn't it make installing node packages work the same way as it does in python, ruby, and other languages?

or Gamestop

I think you should look into making your extension 'private' at least at first and only grant access to people who pay for it. You need to add each user's gmail account to a list in the Chrome Webstore or associate a Google Group with it and then add peoples' emails to that list.

Maybe set up a server where people can pay and people can auth against later on to automate the payments and updating the users list.

I'd also like to see the results of hand-tracing of variable values done separately by the developer for each copy. Maybe have several developers do it as an extra check.

I think that would ideally be done via a separate linter set up in CI.

It has bugged me when using python interactively but fortunately ipython will ignore it.

I was under the impression that Product Hunt and such sites are good if you want to get eyes in general on your product and maybe increase your popularity within the tech scene but those people might not match the target demographic for customers. Is that wrong? I'm guessing it depends on who your market is.

Wonder if people where you're from run into problems where the manhole covers tip onto their diagonal and fall in (since that's the answer I memorized and I think is known as the predominant reason.)

There were ways to get problems, though, such as prior questions for the ACM programming contest. Those tended to be a little harder, though.

Also, topcoder started in 2001.

This book was about puzzle interview questions from Microsoft in the early 2000s: https://www.amazon.com/How-Would-Move-Mount-Fuji/dp/03167784...

Also, a famous question around that time was 'why are manhole covers round?'. (Thought there was a book from that era with that name but not seeing it; only found a newer book.)

Remember doing interview prep at the time by trying to memorize the answers to a bunch of these puzzles. Leetcode is useful in comparison to that. I had an interview once where the interviewer could tell I knew these and he was asking me to tell him about others I knew about.

Your title is too vague and people likely won't bother reading this post. Is "Ask HN: Has anyone ever asked this before?" in case it gets changed.

Also, the answer is probably 'Yes'.

Think the reality is people don't read everything and use titles as a guide to determine what they're interested in reading. The Internet offers endless choices compared to a newspaper.

I think professional passionate reasons would work well here, too.

If you think of XML (and JSON, YAML) as a standardized CSV that can handle hierarchy I think that's beyond a useless fad. There are lots of edge cases in parsing data and having standardized tools for that makes it a lot easier for systems to communicate with each other.

I think the hype was there because companies needed to update their data to follow standards and update how that data interacts with other companies. The specifics of XML are less relevant - once things are standardized there can be standardized ways of converting to other formats.

Imagine needing to write custom parsers for every API; not needing that enabled SaaS APIs to be much more practical.

LLMs can do this for you (although no guarantee it will be a hit)

50-year-you has lots of experience in the domain of 'you' and knows the future (via your lens) from when you were 20. 50-year-olds when you're 20 will lack those things.

Also, everyone has holes in their knowledge/perspective. Could be the older people in your orbit when younger have thoughts that are too limiting compared to where you end up when making decisions as a 20-year-old.

This isn't an effect name but if I see this in a bug report I'd tag it a 'works for me' / 'unable to repro'

I look forward to the day that I can run a local LLM (for confidentiality reasons) to automatically reorganize commits within my PR. Should be very safe compared to generating code or merging/rebasing other code since it is just changing the grouping of commits and the final code should be unchanged.

This would free up developers to spend more time solving problems for the business instead of tidying up code.

Could also delay doing this to when somebody is trying to understand the code (during review or later) or doing a bisect.

Might depend on how companies use git but I just verified I can see commits for PRs from 2017 in my work codebase; could bisect that if needed (although code likely is no longer relevant.)

I think splitting PRs into multiple commits can make sense when there are only 2-3 commits but you're right that it doesn't when there are more than that.

I prefer to add commentary in the PR on the final changes - what does it do, how does it work, what are some pieces of code I'm less sure of, etc. It is accessible since the squashed commit gets linked to it and that provides a place for people to ask questions.

Also is a lot faster to write and easier to communicate compared to messing around with moving code across commits, ensuring tests pass across them, etc.

Basically I think people care more about what was built and how it works rather than how to split it up step by step (although if the latter is important I can add a comment for that on the PR - although ideally it would have been separate PRs.)

>- are easier to debug (and possible to bisect; it's not possible to bisect a squashed commit).

In practice, if bisecting indicates a commit caused the problem you can narrow it down further by checking out that branch and investigating further within that branch (which ideally isn't too large). Also, if you have a highly structured PR then you may need to be careful to ensure that each individual commit passes CI.

At that point you might as well ship each commit via a separate PR. (While in development you can temporarily set the branch for Part 1 as the base for Part 2.)

I think the argument against rebasing is spending time polishing a PR to have a cleaner history for one feature that will maybe get looked at during review (rather than the final output) vs also making a PR for a second feature during that same time.

And you can eventually have github squash for you anyway.

Maybe betterhelp.com or something else online could get you an appointment sooner. Haven't tried it myself but I'd be surprised if it took three weeks to get an appointment there. Although maybe you want a meeting with your therapist who knows you better; just offering another option.

Sounds like the now defunct https://en.wikipedia.org/wiki/DMOZ

Also, this just helps you find general sites and doesn't help you narrow down on answering specific questions (who's that actor? how do i fix this bug in my code? why won't my iphone power up any more?) although LLMs are probably taking over those areas anyway.

I've had 13 Amazon orders this year and I got them all within a few days of ordering them. Just mentioning since threads like this get dominated by people experiencing problems and want to point out it isn't always like that.

Also not GP.

I create profiles for starting up each part of my work tech stack (backend, frontend, workers, specific ssh connections, etc.) and I can run/switch to them in separate tabs via shift-cmd-o and typing a few characters of the name. Can also start up all of them at once via an 'Open All' command for a group of profiles.

Seniors can be mentored, too, and there are always ways to grow. A few examples: learning new languages/frameworks, applying AI, having more of a leader role, communicating more effectively, getting more involved in product, trunk first development, writing better tests, etc.

People shouldn't get the 'senior' title and just stop growing entirely (although it happens for some people)