Can you train it better than anything already available on model zoos for free though? If not then beside learning I dont see the point.

should update torrent client to say "LLM training agent" instead, legally safe /s

Well then you are already doing what I suggested.

I'm not saying adopt Chrome as your main browser, I'm saying if Chrome alone has unique feature you think you might need, try it there. For example that allowed me to discover https://github.com/ArcaneNibble/awawausb

Consequently if you are curious about vertical tabs, do try it, even in a container, e.g. Webtop, if you want to be extra safe. I'm not suggesting to switch.

FWIW I do think you should give it a go.

Firefox is main browser, on desktop and mobile, and I'm not changing that.

Still, I do have Chromium installed for very specific usages, e.g. WebXR tests, WebBT and WebUSB devices, e.g. flashing keyboard firmware, controlling Lego bricks, etc. I'd rather only have Firefox to be clear but some things are not supported, marked as not planned, and I believe using Chromium on specific Websites, not for browsing, is perfectly fine when no alternative is available.

No idea how it works where you are but in Belgium and in the EU we have https://en.wikipedia.org/wiki/Ombudsman which according to https://en.wikipedia.org/wiki/Ombudsman_services_by_country is available in most places. Anyway I can tell you when companies receive an email from them, they don't mess around. I've been waiting for a bike part for a year. I contacted the ombudsman, no cost, just 2 emails, suddenly my carbon belt was shipping the very next day.

• their repo (checked the commit graphs and basically they did most of the work, 2nd dev agree with them, covers 90%+) their choice of governance
• their repo, their choice of tooling
• I genuinely believe they think are doing "good enough" code and they are probably right about it in their context
• they do have fair points on the economical power dynamics, namely that yes Anthropic is slightly less worst than Meta, Google, OpenAI, Microsoft, etc (... but IMHO honestly that's a damn low bar)

but also

• obfuscation rather than discussion (closed the issue and limited to maintainers only) so clearly the signal is precisely "my repo, my choice"
• no mention of the copyright or license washing
• no mention of ecological impact

so I would personally consider instead Bottles, GOG (have different problems), Steam (obviously not open source and basically monopolistic position), etc.

Overall I think preventing discussion is unhealthy (even though sadly sometimes needed, here I lack context, maybe the issue poster did this numerous time on other platforms, title definitely was provocative) but removing provenance is NEVER a good choice. They want to use Claude on their repo? Absolutely fine (even though not to me) but hiding it makes it instantly untrustworthy to me. In fact I even argued in the past that even though I personally do not use GenAI/LLMs (for coding or otherwise) except for testing it should always be disclosed precisely so that others can make THEIR choice in consequence, including using or contributing, cf https://fabien.benetou.fr/Analysis/AgainstPoorArtificialIntelligencePractices

(pasting a Mastodon post I wrote few days ago on StackOverflow but IMHO applies to Wikipedia too)

"AI, as in the current LLM hype, is not just pointless but rather harmful epistemologically speaking.

It's a big word so let me unpack the idea with 1 example :

• StackOverflow, or SO for shot.

So SO is cratering in popularity. Maybe it's related to LLM craze, maybe not but in practice, less and less people is using SO.

SO is basically a software developer social network that goes like this :

• hey I have this problem, I tried this and it didn't work, what can I do?
• well (sometimes condescendingly) it works like this so that worked for me and here is why

then people discuss via comments, answers, vote, etc until, hopefully the most appropriate (which does not mean "correct") answer rises to the top.

The next person with the same, or similar enough, problem gets to try right away what might work.

SO is very efficient in that sense but sometimes the tone itself can be negative, even toxic.

Sometimes the person asking did not bother search much, sometimes they clearly have no grasp of the problem, so replies can be terse, if not worst.

Yet the content itself is often correct in the sense that it does solve the problem.

So SO in a way is the pinnacle of "technically right" yet being an ass about it.

Meanwhile what if you could get roughly the same mapping between a problem and its solution but in a nice, even sycophantic, matter?

Of course the switch will happen.

That's nice, right?.. right?!

It is. For a bit.

It's actually REALLY nice.

Until the "thing" you "discuss" with maybe KPI is keeping you engaged (as its owner get paid per interaction) regardless of how usable (let's not even say true or correct) its answer is.

That's a deep problem because that thing does not learn.

It has no learning capability. It's not just "a bit slow" or "dumb" but rather it does not learn, at all.

It gets updated with a new dataset, fine tuned, etc... but there is no action that leads to invalidation of a hypothesis generated a novel one that then ... setup a safe environment to test within (that's basically what learning is).

So... you sit there until the LLM gets updated but... with that? Now that less and less people bother updating your source (namely SO) how is your "thing" going to lean, sorry to get updated, without new contributions?

Now if we step back not at the individual level but at the collective level we can see how short-termist the whole endeavor is.

Yes, it might help some, even a lot, of people to "vile code" sorry I mean "vibe code", their way out of a problem, but if :

• they, the individual
• it, the model
• we, society, do not contribute back to the dataset to upgrade from...

well I guess we are going faster right now, for some, but overall we will inexorably slow down.

So yes epistemologically we are slowing down, if not worst.

Anyway, I'm back on SO, trying to actually understand a problem. Trying to actually learn from my "bad" situation and rather than randomly try the statistically most likely solution, genuinely understand WHY I got there in the first place.

I'll share my answer back on SO hoping to help other.

Don't just "use" a tool, think, genuinely, it's not just fun, it's also liberating.

Literally.

Don't give away your autonomy for a quick fix, you'll get stuck."

originally on https://mastodon.pirateparty.be/@utopiah/115315866570543792

It's always cool to see such projects, and I was discussing about the topic just yesterday.

... but please, don't spend $150 on that. This is limited has it has :

• no controllers (and gamepads aren't spatial),
• no hand tracking (which would be spatial),
• no 6DoF, so you can't move your head or body around, just turn your head around
• a very limited software stack, e.g. no OpenXR (AFAICT),
• a cable to be plugged to a computer, i.e. it's not standalone

so it's basically a more open but not standalone version of the Oculus Go. For context the Go is from 2018 and back then was $200 while being standalone. Note also that the Go is rootable, cf https://developers.meta.com/horizon/blog/unlocking-oculus-go/

There are also other DIY VR HMDs, e.g. https://github.com/relativty/Relativty from 2020 which similar limitations. In fact this made so much buzz back then the founder managed to ride the hype and make (sadly) a VC funded startup. I say sadly because the initial project was all open VR and openness but once the money was locked-in... well I let you check.

So... again this is VERY cool to build but please do not consider this anything but a way to learn. If you do want to play with VR with a limited budget consider instead an accountless (meaning no Meta involved) second hand Meta 2 then Alvr or WiVRn (cf https://lvra.gitlab.io/ for more).

Google was "cool" 2 decades ago... or before they combined both monetization (through ad) with power (through monopoly) they inexorably transformed EXACTLY to Microsoft the same way Facebook/Meta did the Google playbook.

Those huge tech corporations are pulled by the market to follow the same, sadly successful, strategy playbook and keep no uniqueness.

Google has been the new Microsoft for years already but through careful marketing consumers somehow believe they aren't.

For years... well pretty much since I had a PC, I had a Windows partition. Why? Well because I (sadly) paid for the damn thing (damn OEM deals). Plus, I admit, sometimes they were things that only ran on Windows.

For few years now though, everything, literally, from the latest tech gadget to playing games to VR, works on Linux.

Few weeks ago I deleted the Windows partition. I didn't have to. I didn't boot on it for months. It didn't affect me.

Still, I now feel ... safer, more relaxed, coherent.

When I see shit like that, I feel even better!

So... they are a non-profit (as they initially were) or a public research lab then. That would perfectly fine to say the path that they chose and so happen to make them unbelievably rich, is not viable.

They don't have a business if they can't legally make profit, it's not that hard. I'm sure people who are pursing superhuman intelligence can figure out that much, if not they can ask their "AI" some help to understand.

What a joke.

Convenience is one of the most addictive drugs.