But on the second week I bungled a command that did some rebase and stuff and I paniced that I destroyed the work but alas it was all there.

I just hit jj undo and it all was there it snapshotted it all. I fell in love and just kept using it unafraid of making mistakes.

I think the worst part about git is being afraid to bungle up your work, jj makes obvious things obvious, asking someone in git land how to copy a commit as in duplicate it. And they might actually start crying even, if they have been using it for a while.

Generally you just create two branches but what if you want to copy and try to check which branch it would be easier to rebase on to, if you had to rebase.

This is not that hard if you have done it a couple times, but jj makes it so much easier, in big teams I feel like JJ is the only sane choice when git commands can get insane.

Also jj doesn't snapshot large files by default, aka, perfect no one mistakenly commiting a binary again. That is just unhinged when newbies do that and you need to flush that out of your git history.

Just the sane defaults + undo + obvious and simple commands was a huge sell for me and I grit my teeth and learnt it. And ofc in a jj git init repo you can always use git, best part often you can do all the simple operations in jj and then the hard one in git after creating a duplicate of the commit. To ensure nothing gets destroyed beyond recovery, further you can use jj state to recovery bad git state as well. It's all so nice that you will accept doing the harder stuff with git as well.

And less social media shit, maybe adding better LFS alternative similar to huggingface and stuff.

Git isn't the popular choice in game dev because of this assets in tree hosting nonsense, why haven't we fixed it yet.

Similarly many edge cases, also finally they built stacked prs but man does it feel a under baked, and what it's like 2+ years late.

Please just improve Github, make me feel like I will be missing out if I am not on Github because of the features not because I have to be because of work.

We are definitely living in the worst timeline. Lmao

I want to build xyz,

```

jj desc -m "feat: x y & z"

```

do the work.

```

jj split

```

Split up the parts and files that you want to be separate and name them.

This will also allow you to rename stuff.

```

jj bookmark create worklabel-1 -r rev1

jj bookmark create worklabel-2 -r rev2

# Push both commits

# since we just split them they are likely not inter-dependent

# so you can rebase them both to base

# assuming rev1 is already on top of base

jj rebase -s rev2 -d base

jj git push

```

That is it.

Passive investments are good, but if taken too far as they clearly have been in the last decade they become a scam. Everyone is SIPing into it, and there is infinite liquidity. Until one big whale finally decides they are booking it, then all hell will break loose on the same damn day.

Further browsers have a much larger surface area for even minor fuzzing bugs. Curl's much smaller surface area is already well fuzzed and tested.

Chrome has better fuzzing and tests too. Firefox has had fewer resources compared to Google ofc, so understable.

Ofc not saying it wasn't good. But given the LLM costs I find it hard believe it was worth it, compared to just better and more innovative fuzzing which would possibly scale better.

If you don't code ofc you won't be as good at coding, that's a practical fact. Sure, beyond a certain skill level your decline may not be noticeable early because of the years of built-in practice and knowledge.

But considering every year there is so much more interesting technology if you don't keep improving in both hands-on learning and slow down to take stock, you won't be capable of anything more than delusional thinking about how awesome your skill level is.

You can't click on the budweiser logo when watching super bowl ad. But if you sit in your chatgpt window all day then it's probably worth it for advertisers to expect to build familiarity with brands they advertise.

Solves 90% of my problems so haven't felt like I needed any additional tooling on top of jj.

But I am curious is there some edge case on jj that I missed. That you folks are working on improving tooling for?

Just really curious about this new world with some better solutions to git.

I liked pijul a bunch too but lack of compat with git meant I can't use it for work... Haha real sad moment right there.

I don't understand if it's fun for people(in the software development trade) to see everyone complaining about Software...

I as a software developer honestly feel ashamed in the quality of software we provide out there.

I think LLMs should instead be used to automate grunt work to make software better for edge cases, or where you can use it to get more time to improve software quality.

I don't understand this.

I have seen bad takes but this one takes the cake. Brilliant start to 2026...

They mentioned the training data is much higher for an LLM, LLM's recall not being uniform was never in question.

No one expects compression to be without loss when you scale below knowledge entropy that exists in your training set.

I am not saying LLMs do simple compression but just pointing a mathematical certainity.

(And I think you don't need to be an expert in creating LLMs to understand them, albeit I think a lot of people here have experience with it aswell so I find the additional emphasis on it moot).

Comments URL: https://news.ycombinator.com/item?id=46230021

Points: 1

# Comments: 0

Any higher and its not really a disruption