lytedev

I'm using a nearly default configuration which seems to not have this problem. curl still works and so do downloads.

I guess if your cookie expired at just the right time that could cause this issue, and that might be worth thinking about, but I think "breaks the web" is overstating it a bit, at least for the default configuration.

It includes links with explanations, but the page does kind of "fly by" in many cases. At which point, would you still leave?

I'm guessing folks have seen enough captcha and CloudFlare verification pages to get a sense that they're being "soul" checked and that it's not an issue usability-wise.

I never claimed confusion, but that there is more to "intent" here than what one page of documentation says.

I think it's clear the author "desires" or "wants" folks to keep the images. However, I think the author also "wants" users to use the software without restriction, hence the license.

If I say I intend one thing in one place, but then also say another thing orthogonal to that thing elsewhere that seems to be at odds, what was my intent truly? If my actions do not line up with my words, how do external parties judge what is the socially acceptable approach given my two statements that are at odds?

I simply think the choice of license says a lot more about intent, and is, in fact, the mechanism by which a creator decides how their code may be used. If the author truly intends their software to be used a certain way, the license is _the_ way to have control over that.

I believe this conversation is a bit more nuanced than you are making it out to be and the discussion around "what is open source" is where this discussion begins and ends. I'm not going to try and argue about what the author "wants", which, I agree with you, seems clear, but is not expressed fully, given the chosen unrestricted license.

Yeah I had no idea that some folks would get so passionate about making changes to a piece of FOSS based on a request on a certain footer-esque documentation page.

I think its a great discussion though that gets to the heart of open source and software freedom and how that can seem orthogonal to business needs depending on how you squint.

I'm not aware of any ignorance (feigned or otherwise) on my part or in my comments about my ignoring the author's request. I'm aware that I am doing so and have made that clear and shared with others that also would choose to do so as they deem reasonable.

If you'll allow me to make assumptions, given that the author neither demands -- and is, in fact, explicit about not doing so -- nor licenses the software in such a way as to prevent this use case, I am guessing the author had at least some intent or foreknowledge around some folks wanting to swap the images. I further assumed that such use cases were for instances such as those the author wrote Anubis for to begin with, protecting small git forges with little resources. Now, I admit my server is not small and I have resources, and so am happy to pay for and donate towards open source software, but in this case the only option was to contact the author, which is something I deemed overkill in this case. I would simply wait and see how the author planned to approach the issue and revisit at that time.

Perhaps I've made the wrong move socially or ethically, which I think is at least a worthwhile discussion to have, and if I should decide I feel like I've made an ethically sideways choice, I will eat my words and make things right as best as I can.

However, if we're going to talk about intent, I an guessing there is a bit more nuance to bring to the conversation. Or perhaps the author can chime in or update the documentation to be more clear, because the liberal license says quite a lot about intent to me. I think it's at least a little disingenuous to say that the software license carries no intent behind it (spirit of open source and all that) and is "only" an enumeration of my rights.

I'm surprised to read this from you, somebody I and many others hold in high regard as accepting and knowledgeable, insulting someone's character because they didn't like some specific aspect of your work or opinions or chose to ignore an ask in this particular use case.

I didn't implement this out of fear or some lack of courage. In fact I had the original avatars up for quite a while. I simply wanted my own logo so visitors wouldn't be potentially confused. It seemed to fit the use case and there was no way to achieve what I wanted without reaching out. I didn't feel comfortable bugging you or anybody on account of my tiny little no-traffic git forge even though, yes, that is what you politely asked for (and did not demand).

I think if you do feel this strongly you might consider changing the software's license or the phrasing of the request in the documentation. Or perhaps making it very clear that no matter how small, you want to be reached out to for the whitelabel version.

I think the success story of Anubis has been awesome to read about and follow and seeing how things unfold further will be fun to watch and possibly even contribute to. I'm personally rooting for you and your project!

You are correct in that I ignored a specific request, but you are also ignoring the larger social contract of open source that is also at play. To release software with a certain license has a social component of its own that seems to be unaccounted for here.

Your analogy to me seems imprecise, as analogies tend to be when it comes to digital goods. I'm not taking pennies in any sense here, preventing the next person from making use of some public good.

You can make a similar argument for piracy or open source, and yet... Here we all still are and open source has won for the most part.

> You ignore the authors words and published intent, and go so far as to undo them

I guess I took the MIT license as the author's word and intent. Are you saying their choice of license is not? It clearly outlines that I am free to use the software without restriction which you conveniently leave out of your core argument.

If you want to talk about open source and the social contract, this is the heart of it: freedom, which I have exercised. If I was using it for commercial purposes and doing something more against the "spirit of open source" I think I might be inclined to agree with you. But I'm not.

While I think you have a point buried in there worth discussing, I simply can't equate me wanting slightly different functionality from MIT-licensed software and making it happen equivalent to these kinds of breakdowns in society.

The author also went of their way to indicate this license, for what it's worth.

If you believe using MIT licensed software is doing so, I suppose? I do not hold that viewpoint, however.

I did so, though I would hardly call using MIT FOSS for my personal projects a breach of the social contract of open source. This was easier than forking, building a docker image, etc. I'm guessing it will be much easier for others, too, since the recommended config has you dink around with reverse proxy configuration no matter what.

My "workaround" for this MIT-licensed software that does not allow me a simple and common customization was to have my reverse proxy redirect requests to the images. https://git.lyte.dev/lytedev/nix/pulls/92/files

Hope this is useful to others!

I was doing this last night with open-codex, a fork. https://github.com/ymichael/open-codex

I'm probably on the wrong side of history and generally think LLMs are way overhyped, but Gemini 2.5 Pro is the first one to consistently impress me for pretty much everything.

I self-host Postgres at home and am probably screwing it up! I do at least have daily backups, but tuning is something I have given very little thought to. At home, traffic doesn't cause much load.

I'm curious as to what issues you might be alluding to!

Nix (and I recently adopted deploy-rs to ensure I keep SSH access across upgrades for rolling back or other troubleshooting) makes experimenting really just a breeze! Rolling back to a working environment becomes trivial, which frees you up to just try stuff. Plus things are reproducible so you can try something with a different set of machines before going to "prod" if you want.

Indeed we have it tough out here. I like the excitement, though, I must say! Funny to find the sentiment here on HN, though!

Reading the article, it looks like they are writing go, so will probably be distributing go binaries.

The same reason you might care about the artist or producer of a song: its context generally implies certain characteristics you might associate and appreciate.

For example, most Rust projects use clap which provides really nice command line help menus and also have relatively good performance in general.

It seems to be squaring up directly against the mac studio with its efficient APU and big memory bandwidth use cases with a cheaper price tag. At least that's the loose sense that I got based on their keynote.

Can you elaborate? Why not?

I think they're working on it? MCP: https://www.anthropic.com/news/model-context-protocol

I'm pretty clueless with regards to modern LLM coding tools. What tools make this possible; talking to the computer to change code and whatnot?

I work on tools for other developers!

Not sure but gonna guess power.

Nah the space savings can significantly cut down on bandwidth costs at scale. They'll get (and have been?) pushed by Google and friends for that reason.

Explained it perfectly. Thank you!

I really like Axum. Like... a lot.

However, it feels petty weird sometimes as the extractor thing results in somewhat unusual-looking function signatures. Not a real problem and something I'm sure I would appreciate more if I understood exactly how it worked...

But it makes for really ergonomic definitions of route handlers. Even if it presently feels a little bit like weird or gross dependency injection.

Obviously what keys to include, what they do, and maybe other choices to make when designing a keyboard is almost entirely subjective, but here's my opinion:

Layering, chording, combos, and other methods of producing different keycodes aside from just throwing more buttons and adding more surface area ARE a good solution and arguably superior due to the following benefits:

- Ergonomics are better due to less overall movement and effort needed to produce a certain keycodes by moving two fingers slightly instead of moving your wrist or arm.

- Keyboard is generally cheaper to produce due to fewer parts required, simpler circuitry, smaller footprint, and other benefits. The arguable downside is that you are "hiding" the complexity in software (firmware?).

- It's more fun! This one is more subjective, but thinking about clever and more efficient ways to setup your keyboard can be a good time indeed! It feels cool to use.

I ordered my AMD one late in 2023 and love it.

It's not perfect as battery life compared to the Apple Silicon MacBooks is simply not a competition, but everywhere else it's close enough for me.

Being able to upgrade the screen already was a huge "yes, thank you" as instead of needing to buy an entirely new machines for a particular upgrade I was able to quite simply swap in the new part, which is what I'm used to on the desktop hardware side of things. It's awesome! The community and Linux support is also very cool and I'm excited about a possible RISC-V board to pop in and the overall tinkering possibilities.

That said, if you just need a personal computer, it's hard to recommend in the most general sense I think. But for businesses I can see it being a no-brainer.

I have never thought of this and it's brilliant. I should definitely be able to whip up a relatively simple script for tagging arbitrary files with arbitrary tags. I was wondering how to achieve something like this for backup purposes and this should make it a breeze.