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

They’re in age order and age 15 starts at about a quarter of the way down.

Shame on the flaggers.


That’s what it looks like to me, but I haven’t yet seen a good explanation of their motive. Why would the development of `rv` be such a threat to them?

I know specific individuals hate Andre and have had beef with him for years, but it’s hard to see what might have motivated Shopify and specifically Ufuk Kayserilioglu to carry this out.


> Why would the development of `rv` be such a threat to them?

Well, package managers and language bundlers/runtimes are the hottest new luxury item for big tech - maybe they're worried rv gets bought in the same way that Anthropic bought bun, and OpenAI bought uv (Astral). Though at the time, none of that had happened yet.


sad but true


Some estimates are about two million but I think that’s an extremely loose definition of Ruby Developer.

I run rubyschema.org which maintains the rubocop JSON schema that’s pulled via schema store. I can see there are about 21k unique downloads each month, which I think is a pretty reasonable lower bound.

Most text editors will pull this schema when opening a project with Rubocop.


> Ruby Central’s actions during this period were taken in response to a breakdown in a working relationship with an individual who had significant access to infrastructure and code.

This is the first time they’ve actually admitted that this was all about Andre.

> At the time, we believed a serious risk had been introduced to RubyGems and related services.

This doesn’t add up. Access was revoked and then temporarily restored. Nothing about this was mentioned in the meeting that took place before the access was removed again. See https://archive.org/details/gmt-20250917-160422-recording-64...

And what’s more, they didn’t even try to remove Andre’s access to AWS until he told them to.

> As stewards of services relied upon by millions of developers, we took that risk seriously and made the decision to act quickly to protect that infrastructure.

That’s not what Freedom said. Freedom said they needed to act quickly or lose funding.

https://apiguy.substack.com/p/a-board-members-perspective-of...

> A full, independent security audit has now been completed. The review was ultimately inconclusive because key logs required for a complete analysis were no longer available. We recognize that this creates continued uncertainty.

This makes it sound like there was some big security incident that they had to respond to. What actually happened is they forgot to remove Andre’s access to AWS and he told them and then they removed it. That’s it.

> Our intent was to stabilize a situation that was quickly escalating to work toward an amicable resolution.

If you watch the meeting (linked above) it’s clear that’s not what they were doing. This is a new spin they’ve come up with to justify it.

> Ruby Central did not initiate litigation and has consistently sought a path that would allow the community to move forward without prolonged conflict.

That is not what I’ve heard, but I’ll wait for others to post details of what’s happening in this space.

> At the same time, we recognize that aspects of how this situation was handled and communicated did not meet the expectations of the community.

They keep trying to admit fault in communication as if communication was the problem in an attempt to distract us from the fact they literally stole open source projects in a hostile GitHub takeover and used their privileges as administrators of RubyGems.org to take over the `bundler` package.


It can be a mistake though to assume that the DOM hasn’t changed since it was rendered. Browser extensions, ad blockers and other JavaScript can modify the DOM.

I know it’s more expensive, but it’s like 1ms to render a document on the server and 3ms to morph it in the client. If you keep an SSE connection open, Brotli compression is very effective when you send almost the same HTML again and again.


I don’t hate SPAs, I just think some apps are better off being MPAs. I wouldn’t build a todo list app as an MPA. But many apps really are just CRUD forms and tables.


Thank you so much. Please ping me if you have any questions about these techniques. I’m `joeldrapper` on Discord and GitHub.


I enjoy writing mostly SSR apps with just a few specific Svelte components mounted as custom elements. It works really well.


My specific use case was building a form where each change to an input would fetch a new copy of the form from the server and morph it in place.

It means the server-side code can be really simple. You can make parts of the form depend on the values of other parts. For example you can show/hide a section based on a checkbox or fill a select with options based on a previous selection.

Because it was a form, it was really important to maintain object identity and state perfectly so the user would not be interrupted.


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

Search: