play around with local LLMs and image upscaling

FWIW I did that for a bit https://fabien.benetou.fr/Content/SelfHostingArtificialIntelligence and I stopped doing it. I did it mostly from FOMO and that, maybe, truly, it wasn’t just hype. Well I stopped. Sure most of those (then state of the art) models are impressive. Yes there are some radical progresses on all fronts, from software to hardware to mathematics underpinning ALL this… and yet, that is ACTUALLY useful in there? IMHO not much.

Once you did try models and confirm that yes indeed it makes “something” then the usefulness is so rare it make the whole endeavor not worth it for me. I would still do it again in retrospect because it helps to learn but… honestly NOT doing it and leaving others to benchmark, review, etc or “just” spending 10 bucks on a commercial model will save you a LOT of time.

So… do what you want but I’d argue gaming remains by far the best usage of a local GPU.

Sadly not the only one https://www.bbc.com/news/articles/cx23ke7rm7go

Well I get why you stick to a hardware device you like… but honestly that’s 15 years old. You can get something better and cheaper delivered to your door tomorrow.

I personally went down a similar path while discovering https://www.rockbox.org/ was still a thing, looking for old iPod or Archos I could refurbish, checking 2nd hand market, etc. As much as it pains me to say, unless you are a collector it’s not “worth” it. You can get something ridiculously smaller, with more memory, more features, etc for the price of a meal.

IMHO it’s better to get rid of Windows by purchasing new hardware that is genuinely interroperable by supporting standards.

Ideally you’d check something like https://www.hanselman.com/blog/how-to-update-the-firmware-on-your-zune-without-microsoft-dammit but it might be more work than you want to put it. Maybe your local HackerSpace could help though.

My point finally is that freedom is quite important and feeling trapped daily is not worth ~$50.

Oh noes… who will think of the shareholders?! /$

What’s next, he’s also going to ask people not use terms like “slop” like MicroSlop CEO?

Honestly a lot of people, including me, who do complain about the AI hype (not public research in AI) were, or still are, sadly, customers. People who wanted to buy MORE of his products but not fund something that has so many predictable negative consequences with so little upsides except for his bank account.

Nothing you (nor I) know of but that doesn’t mean it’s the case. I can’t evaluate but https://www.openimagedenoise.org/ is publishing by Intel and in 2026 so maybe it’s good.

Right, then I can’t help you.

To clarify for others though as I guess I wasn’t clear based on the downvotes : I’m not suggesting a single piece of software is a viable alternative to Lightroom. Rather I’m saying Lightroom itself is a collection of algorithms dedicated to photo editing wrapped in a UX one is familiar with. On the other hand ImageMagick (just to pick one I know relatively well) is a set of command line tools for image editing. It’s mostly used as a backend with other tools as interface. I imagine there are plenty of alternatives to ImageMagick too, probably some that can include arXiv STOA algorithms for photo editing, maybe some even with a GUI but my point again is to reconsider the workflow to understand how the tools one rely on actual work.

So to hopefully express myself better this time, ImageMagick + Gimp + Krita + some script in a Github repository based on an arXiv publication + I don’t know what + … all together or in part might be better for some people but no I don’t know an all-in-one open source alternative that cover ALL needs without them being expressed first.

Goes against https://duckduckgo.com/privacy which is basically also their main selling point. If they don’t have that, they have nothing. So… FUD until you do have data. You clearly did run an experiment, show us the proof.

Not.

Now to be slightly more helpful (apologies for the provocation) I suggest you consider alternatives to Lightroom. I know that instantly you will receive countless comments on how alternatives are just nowhere near as good as Lightroom… and that’s OK. IMHO it’s OK because I bet YOUR usage of Lightroom isn’t the usage of others. So… I recommend you forget the brand “Adobe” or the product “Lightroom” and instead you list here the actual function of a tool you need.

This way, by listing actual needs rather than a bundle product with branding and specific UX, you go back to the root of your problem, namely WHY do you need such a piece of software in the first place.

Sure, you might end up with an entirely different workflow. Sure it will probably be absolutely alien at first… but so was learning how to use that piece of software in the first place too. Right now you do have the concepts, so replacing one click by a command line tool, or 1 piece of software by 10, is IMHO acceptable. What you will hopefully have in the end if YOUR workflow that is even more adapted compared to what you had first. It will be “weird” and maybe nobody else will get it but for you it will be exactly what you need.

You are right to be wary of that behavior as it is spreading.

That being said I believe it’s good to be mindful of the “paranoid privacy nut” in the sense that… WHO are you concerned might have WHICH information about you?

I don’t know you but… if I were to find your name I could find your address or at least roughly where you live. In fact depending on what you post one could potentially know your neighborhood without even knowing your name. Now I didn’t know you had a pet, now I do. That being said even without a pet… if you live somewhere you are expected to walk around your neighborhood. It might be to go buy milk, help neighbors, drop mail, etc. There is relatively no new information there. Your neighbors might know you are around, their doorbell cameras might have footage of you doing so… and what? How is the confirmation that a perfectly average behavior is indeed coherent? What also NOT having that footage bring? Maybe you are traveling and thus not with your pet but maybe you also are sick.

So… I agree with you that all those cameras with footage are not healthy but also what do they genuinely add or remove? I would argue in principle a lot but in practice not so much. I would even argue to biggest impact is unwarranted stress and concerns, the chilling effect.

Walk freely however you want, it’s your neighborhood, ignore the cameras.

PS: as someone already pointed out, one does not need video cameras to track your movement and patterns, using wireless signal (5G/4G, WiFi, BT) on your phone is enough.

Why would you feel bad, the interview is a 2 way process. They are evaluating you but YOU are also evaluating them. It’s actually VERY costly to you too if you start working for the wrong company. If you realize after a week or a month that truly the culture, the tooling, etc basically anything but the pay does not match YOUR needs, whatever they may be, they you HAVE to pull out.

You can be polite about removing your application, as you were, but you should not feel bad. It is precisely WHY there are interview. Candidate think about it as only them being evaluated and that’s very wrong. As your title says clearly it is about self respect but not just during the interview, the whole time. If you are not a match sure it does suck, for both, but that’s again better than a forced match that will bring both down over time.

Finally regarding your last part, I recommend you edit your post to put your precise skillset and experience there. Hopefully someone can refer you to the right place.

Come on people, think of the shareholders! /$

Slop, MicroSlope, MacroSlop, it’s BS all the same.

IMHO the key aspect isn’t where you host things but rather understanding how hosting itself works.

To me the most challenging aspects are how to :

• route traffic
• start a service
• backup your data

and also ideally

• have more than 1 service on a single machine
• restore your data
• restore your entire setup

For that very first step I would say having a machine directly exposed to the Internet makes it easier. I don’t know what ISP you use but at least in Belgium where I’m currently located all ports are closed and IP are dynamic. That means if you want to show your freshly started Apache Web server to your mother in law it will challenging.

Meanwhile if you do manage to get to the last step, namely restore your entire setup, then restoring to a cloud service or a RPi is the same, you transfer your data, start your services and voila, you are back either LAN only or on the entire Internet via a cloud provider.

So autonomy isn’t as much as to where things are physically hosted and by whom as in the actual capacity to able to host there or elsewhere.

Finally if you are using a commercial ISP, as opposed to having your own AS, are you really self-hosting?

That’s actually my recommendation yes.

If somehow after a month you feel like you do want this “lifestyle”, are comfortable with setting up a VPN (if you need external access) THEN spend more and get your a SBI like a RPi and have it at home. If that’s still not enough then go up to a proper server you host, use a non commercial ISP, etc … but IMHO don’t start with a server at home if you are not familiar with all this, it’s counter intuitively harder and definitely more expensive.

Also FWIW you should still have an offsite backup regardless of how you do it.

Sure, rent a cloud server for $10/month, install Docker/Podman then all self hosted services you need. Invite people on your Jitsi Meet server, publish your videos on PeerTube, work via NextCloud, etc. It’s not easy the first time but with each (well documented) step it becomes easier. Most important : backup your data.

I don’t think it matters so much. It’s possible to test Linux literally in seconds with nothing to install thanks to virtual machines on the Web. It’s risk free.

What prevents people from migrating isn’t technical, it’s mostly FUD and marketing (not to say lies) from MicroSlop.

Honestly I don’t mind that, at all. What I mind is if it’s mandatory and only through proprietary applications.

WiFi, BT, Zigbee, Z-wave etc are not per se a problem. The question instead is who practically owns the device. If the behavior is force on you as a customer, then it’s easy, it’s not YOUR device. Consider then buying OSHW or whatever alternative you need, including potentially non connected devices that you yourself connect on your terms.

Edit: check which devices are compatible with GadgetBridge and/or HomeAssistant then reviews from actual customers. That should help you find out which devices can match your requirements.

Or the original original upload https://media.ccc.de/v/39c3-a-post-american-enshittification-resistant-internet leading to a LOT of other interesting videos.

Truly shameful to use YouTube (thus Google, BigTech surveillance capitalism) links on Lemmy.

Edit : FWIW https://lemmy.ml/post/41213924 was posted 1 day ago and there it was posted properly.

It’s always about AGENCY and power, not performances or preferences.

Right, Betamax much? It doesn’t really matter if one technology is objectively “better” on all aspects than another if the strategy to make it popular outpaces the other.

To be clear I wish you were right (even though I don’t find open source models to be free of problems) but I think the conclusion is a wish, not a logical one.

No.

Not because it’s not technically feasible but rather I would psychologically not manage to make money knowing my portfolio, either directly or via EFTs, makes me money by profiteering of BigTech or surveillance capitalism.

Full disclosure : I did have Apple and NVIDIA stocks and I did sell them not because they were not making money (there sure were) but because I felt disgusted by HOW they made money.

PS: KYC and related laws in a lot of countries demand you use your real information and declare your earnings, so again it’s not a technical problem, it’s at least ALSO a legal problem, and arguably a moral one if you believe KYC kind of laws help to curb money laundering.