The discussion is about whether the western media is paying insufficient attention to the Ebola outbreak simply because it’s in DRC, and DRC/Africa doesn’t matter.

The post you responded to is suggesting a different hypothesis: that the media is paying limited attention because it’s in a country quite a long way away, on a different continent. In line with this hypothesis, it’s not unreasonable to question how much attention the press in countries a long way away would focus on a viral outbreak in a European country.

So I’m assuming I’ve done something wrong along the way, but I’ve not had time yet to explore it.

For comparison, I just ran a couple of quick benchmarks (default settings) with llama-bench:

Qwen3.6-35B-A3B at Q6_K_XL gave 858 t/s pp512 (prompt processing) and 43 t/s tg128 (token generation).

Qwen3.6-27B at Q4_K_XL gave 103 t/s pp512 and 8 t/s tg128.

Huge swathes of the economy basically runs on Microsoft Office - internal and external business communication is via Powerpoint in meetings, internal and external documentation is via Word, internal analysis (big and small) is via Excel, collaboration is done reasonably well via Sharepoint, and they have the network effect that everyone else uses it too.

The reality is that the alternatives just don't stack up. Google's suite is great for collaboration and okay for limited work, but falls over completely when faced with large documents (imagine thousands of pages for a regulatory submissions) or spreadsheets with large amounts of data. Other options (Libre Offce, Softmaker Office, etc.) may excel in some domains, but offer a steep learning curve, and/or may be unreliable with Microsoft Office format compatibility, and/or are weak on the collaboration side.

Aside from that, it was always pretty solid and IMO better than the typical legacy manufacturer offering.

> Being able to push prints and use the printer with direct local connection, while simultaneously having remote monitoring and remote printing when cloud/internet works and is available.

So isn't an obvious approach to just cut Bambu out altogether and just create a FOSS cloud alternative, supporting the remote aspects that the users want to retain?

> This is not the case of "wanting to have their cake and eat it too", as there is nothing mutually exclusive about these things.

Nothing technically mutually exclusive, but isn't this exactly the choice that Bambu is enforcing? Which is crappy corporate enshittification behaviour, but something they can do if they so choose? (I'm not arguing in their favour - just trying to fully clarify the situation.)

> What users want is to "have their cake and eat it too;" they want the local token authentication _and_ the cloud authentication enabled at the same time. This isn't actually possible, so this plugin approximates it by emulating the interface to the cloud authentication to make the "Bambu Network" cloud RPC calls from a local slicer (one of these calls is a local_print call, so ostensibly this allows you to send prints without running them through the cloud, although with all of the online functionality still enabled and required, this seems like a pretty brave thing to trust).

AIUI Bamba has made cloud access all or nothing: you either use local mode, with local slicing, and no cloud feature access at all, or you use cloud mode, with cloud slicing and access to all of the cloud features.

Can anyone explain what the cloud features that people want to retain are? Is it just app control of the printer, and print monitoring? Or are there other things to miss out on?

I'm there too on this. Something I've found helps is listening to certain types of audio. Some audio books can work, but if they're too engaging or interesting it's counter-productive.

My current solution for this is a particular YouTuber; I noticed a long time ago that watching his videos in bed (when I already given up on trying to sleep immediately because my brain was too alert) seemed to help me relax and feel drowsy. Now it's almost a switch - start a video, phone face down, and I rarely need a second one.

Anecdotally, we have an air monitor gadget and the highest I've ever seen (small home office, fairly well-sealed, winter, me working there all day with no ventilation) was around 1100-1200 PPM. I get that two people in a small sealed bedroom could push it higher, but 3350 PPM?!

2) It's probably not the time/place to trouble-shoot your "consumer grade server" LLM experience, but if you're running on CPU (you don't mention a GPU) then yeah, your inference speed will be slow.

3) Counterpoint: my consumer-grade Macbook Pro (M1 Max, 64GB) runs Qwen3.6-35B-A3B fast enough to be very usable for regular interactive coding support. (And it would fly with smaller models performing simpler tasks.)

> I'm not assuming that. If you get a paycheck at all you can impulse buy a $65 thing.

Not to buy the membership, but to gain meaningful benefit from it afterwards by buying in bulk in the long term, despite the greater upfront cost, motivation, planning, etc. that that requires.

> “But Local Models Aren’t As Smart”

> Correct.

> But also so what?

> Most app features don’t need a model that can write Shakespeare, explain quantum mechanics, and pass the bar exam. They need a model that can do one of these reliably: summarize, classify, extract, rewrite, or normalize.

> And for those tasks, local models can be truly excellent.

And for buying in bulk (which is basically what Costco offers) to work, you need a larger amount of available cash up front, and a home with sufficient safe space for storage.

And of course you need some sort of transport to bring your bulk purchases home from the Costco, often built on cheap land away from dense housing.