Hacker Newsnew | past | comments | ask | show | jobs | submit | archagon's commentslogin

Because when it eventually and inevitably corrupts your data, you won't know what to do or have any recourse?

Surely any sane person vibe coding a note taking app just has it save all the notes as markdown files to disk? At that point making a backup is trivial and they're unlikely to get corrupted.

So why vibe code a version of a thing that already exists in a dozen different permutations, and with actual eyes on the codebase?

In a typical open source project only one person has had a look at a particular piece of code. Only in the larger and more mature projects do people actually spend time reviewing code. Also, if you don't pay for the free code, there is often no serious recourse to recover your data either.

As stated in my first comment, Obsidian does not support Emacs keybindings properly, nor is it open source. Writing an extension to add Emacs keybindings is not at all trivial, because you have to work around a lot of existing and undocumented functionality.

There are other reasons for not vibe coding your own alternative, but as LLMs keep progressing, these reasons may become less relevant.


For some reason, AI boosters can't help but condescend. I've never seen this with the rollout of any other technology. It's like this stuff immediately becomes a core part of their personality.

I agree that this new thing is polarizing, but as with the rollout of any groundbreaking technology, the ones looking backwards is just going to be left behind.

As I see it, the only reason to make such a bold claim (as opposed to just doing what you do and seeing how things shake out) is insecurity. Especially if you’re condescending about it.

This discussion is not new. Books was criticized for being addictive and antisocial. TV for rotting our brains. And even if there are some truths to these claims, I do appreciate both.

If you want to be passive aggressive without AI, the more tokens there is for the rest of us ;)


I largely agree with you but I still can't stand the "you're gonna be left behind!!" framing that is really common with people who are enthusiastic about AI. What does it even mean to "be left behind" in this instance, it's just a vague emotional expression.

These AI tools are not hard to learn, in fact they're super easy when you have some experience programming, so the only people who are going to be left behind are the ones who simply refuse to use the tools out of principle. And why would they care about being "left behind"? They're making a conscious choice to not use the tools. They want to be left behind!

And not everyone who is skeptical are that out of principle, some just don't see the value yet or are slowly and cautiously adopting it into their workflow. If AI powered coding ends up being even half as good as promised, so good that denying the evidence is impossible, they can just start using it and catch right up. So who exactly is "being left behind" here? It's complete nonsense while simultaneously being extremely condescending and I get triggered every time I read the phrase.

I don't mean anything against you personally with my ranting, it's more a general observation. Perhaps you and some others do mean it as a genuine bit of advice, like "hey, you should learn these tools or else you might struggle to find work in the future", but the sense I get most of the time are people who are gleeful that the non-believers are soon to be homeless or whatever.


Why would I use a social network run by someone with opinions straight from the pages of Stormfront? I don't need white supremacy mixed in with my light recreational reading, thanks.

[flagged]


If you truly believe "those days" are over, you're in for a depressing few decades as the world moves on without you. Most reasonable people — even white people like myself — are disgusted by the sort of rhetoric that Musk retweets and endorses on a daily basis[1]. It is, without any exaggeration, literal Stormfront shit as I remember from a decade ago. We will fight this putrid garbage and we will win.

[1]: https://www.theguardian.com/technology/2026/feb/12/elon-musk...


If a social network stays comparatively small but still active, I see that as a huge win. Half the people I follow are happily on Mastodon. I don't see that changing anytime soon.

I am on a bunch of socials but as time goes by I like my cohort on Mastodon better and better.

Many a Discord server would agree

Yeah. You sure learn a whole lot of horrible shit about the world in gender-integrated social spaces. (Like how many women start getting unwanted adult male attention when they turn 12.)

If they use any social media like Instagram or TikTok, just ask them to show their DMs as well. I was a douchebag when I was younger, and that was the first step into realizing that I was not aware of the women experience as I thought as I was at the time.

Same as any other conspiratorial thinking: they hold themselves in too high a regard and want to think they’re privy to some secret knowledge that the rubes have missed.

Somehow I doubt the precedent will apply to common folk. There will be some obscene carevout like “piracy fair use for corporations training frontier models” and we’ll be even worse off than we were before.

It’s weird that you think a social network needs to have some sort of technological drive or entrepreneurial spirit to be relevant. If anything, this will lead to its inevitable enshittification, as you can plainly see in Twitter and Facebook.

I agree that the sprawl of objective, trying to cover more and more, is quasi horrific in the ways we have seen it: seen through market captures and monopolization and acquisitions of Instagram, etc.

But for computing, not having a network fabric that is adaptable to purpose has kept people from having any starting point to play with. Every time you want to build a social system, you start with your own website (or go even lower level to reinvent), with your own bespoke back end, your own unique internal and or external API, rather than having set protocols you can adapt.

This isn't strictly true. There have been many attempts to build more good protocols, with endurance and reuse and adaptability. RemoteStorage, inventor of the web Tim Berners-Lee's Solid, HyperCore/Swarm, or Dat (which atproto cto @pfrazee built a browser for, Beaker). (Mastodon was never one of these & is not now.) This is not the place for writing a Speaking for the Dead for each of these, (and none of them are dead beyond grasp), but so far this question of what gets the most use, what has seen the most apps built upon it, the most end user adoption: it has a clear answer and it's atproto.

Atproto layers social networking right. You are self sovereign to a data store that you control (absolute authority, can-move/credible exit), that has your records in it (https://overreacted.io/a-social-filesystem/). It makes syndication cheap & easy. It makes relays that aggregated PDS's while presenting a like API surface to the individual user, to facilitate syndication en masse. It layers indexers (appviews) that provide aggregation etc. This mirrors a functional social network system, goes beyond protocols to be a complete system, that is imminently individually maintainable, at scale.

It provides a platform that anyone can use to be creative, in a connected fashion. Your phrasing hints at the whiff of terror that these large companies bring, the reign they have over our lives. The drive they have is to saturate, to take our attention. I see the role of atproto & it's drive as a network technology layer to allow new social technology layers to be created at very low cost. It allows individuals to build amazing services, very lightweight and client only / isolated web apps that speak the protocols and which can put or retrieve records or appviews, for any kind of social network technology.

That to me lacks the scariness you speak of. There is not one overarching drive or ambition. It is not centralized or directed intent, working cohesively to capture society. Both are expansive & space filling, but their characteristics are as far apart as I can imagine.

@pfrazee distilled it well in Atmospheric Computing (https://www.pfrazee.com/blog/atmospheric-computing), a lovely post. That to me resonated strongly everywhere but especially the cold start problem. That it used to be hard for people to make connected things. And now it's not. That's the essence of the cold start problem. And rather than control the things we create, the role of the creator is different: they facilitate creating records, tools to help people build Abramov's Social Filesystem (previous link). They don't nor should they want to own the network, the protocols, the data, the users, the accounts. Those are atmospheric ambient systems that exist.

> It’s weird that you think a social network needs to have some sort of technological drive or entrepreneurial spirit

I perceived a noted lack of techno-social substrait upon which individuals or entities could collaboratively build or explore. The systems humanity has built have been dead end systems, built for narrow set purposes. There has been a missing General Systems Research component to computing for decades. There has never been a successful General Network Research effort, beyond the web itself (what a hit), and the vision TBL had for the web as a bidirectional writable author able medium never materialized. BlueSky corporation has me at least convinced that nothing can stop credible exit, that this is not their network, and independent actors like BlackSky have somewhat proven that out already in astoundly sophisticated end to end fashion, with incredible thought to their own moderation systems etc.

Tl:dr: actual humans have lacked the means to explore and build the techno-social, and it is absolutely nuclear hot fire that I have all this from little apps a bunch of creative friendly awesome neat people built that lets me capture this stuff (https://pdsls.dev/at://did:plc:zjbq26wybii5ojoypkso2mso/). This is proof that we have a general social networking system right here today available for use, experiment, & play. Not proof, but imho the roots are strong & good. This is how we escape the capture-entrepreneurialism. We have lacked fertile soil & good foundation for new things to get started on. I'd characterize the distinction as alive, adaptable, usable systems, versus stagnant dead systems, and imo, stagnant fixed systems aren't serving us well now, and over time decay & en-badification to serve us worse. Hence my advocacy of aliveness & possibility. (And my disdain for Mastodon as barren anti-ecosystem.)


“and many do”

Yeah, that’s definitely citation needed.


If many people did not want to read other people’s controversial opinions, we wouldn"t be having this whole debate.

It seems like most of the people complaining about this aren't actually on Mastodon.

Well, obviously you’re not going to be on a platform you dislike. I have signed up for Mastodon several times, but always ended up finding the instance dependency off-putting. I wish for a protocol where the instance mostly doesn’t matter and you can trivially switch to a different one, like with e-mail.

How is ActivityPub instance migration worse than switching email providers?

E-mail providers allow you to use a custom domain, so if the one you’re using suddenly goes away, you can just point your DNS records at the new one and that’s it. If the ActivityPub instance you’re using suddenly goes away, you can say goodbye to your account.

Unless I'm missing something, that's exactly the same.

If you're using someone else's email domain (like, eg. gmail.com), you cannot migrate to another provider. If you use your own domain, you can change who is hosting it.

Like email, there are plenty of Mastodon hosting companies. You can use your own domain, and migrate between them at will. If you use eg. mastodon.social, you can even migrate in most circumstances (but not if the instance vanishes).


The difference is that you cannot use someone else’s instance with your own domain. You have to get your own instance, which is much more expensive and creates a lot of duplication.

Uh, you mean like a spam filter? I would not use an e-mail service without one.

No. A spam filter still allows you to view the e-mails and reply to them if you want to.

Also, I would certainly not want an e-mail service where the spam filter is based on the admin’s political opinions.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: