davidy123

They glossed over the NeXT roots which is telling, the ideas were only half-heartedly baked into the OS. It should be strictly enforced that everything on-screen has a relationship to everything else. But that would have been hard work, and they realized eye candy was an easier route when you can just announce something else every few years. This is why desktop evironments really haven't changed since the first commercial releases and Apple is just another PC vendor. Too bad because imagine if the relationship based operating environment was well developed for their spatial headset.

Could come in handy one day.

And you think the reddit API protests are fierce.

To the replies; I think one feature of llama.cpp is it can handle models with more RAM than VRAM provides, this is where I would think more cores would be useful.

I'm a bit surprised by the numbers. It's "only" a 2× speedup on a relatively top-end card (4090)? And you can only use one CPU core. With 16+ core CPUs becoming normal and 128GB+ RAM being cheap, that seems like leaving a lot on the table.

[edit] realized it's relative to the merged partial CUDA acceleration, so the speedup is more impressive, but still surprised by the core usage.

This is why I like the "open case design." Your "PC" is just a frame to hang components on, with nothing to enclose it. And the nice thing is these days people care a lot about quiet computers, so you can have an open frame case with three GPUs that is just making a not-unpleasant whooshing sound at full bore.

Strapping one of these on your face is undeniably quite nasty, and the video eyes just kind of make it creepier. And their shiny reps going on about how everyone can participate, meanwhile it's Apple only is really the icing.

I think though that the dream is for it to get smaller and more human adapted until you get all the benefits, with normalized cues that make it more social, perhaps more social than the now-ubiqitious smartphone and headphones. But that is a long ways away.

The other thing that I think is not emphasized enough is that this headset is really a two way interface for personal AI, where it could be very compelling. Apple prefers to sell it as an entertainment/work device, but with all those sensors and upcoming multi mode AIs, it would be a whole new category. I think Apple realizes others have taken the "AI" flame, but they know it's important, so instead for now emphasize machine learning, but this will eventually become more like the interface for AI assisted imagineering and self guided entertainment.

I saw "semantic model" and got excited, but that's at the "grain" level.

The idea of applications has been stale for a long time, it just takes a long time for people to get there.

Microsoft doesn't want it to happen quickly because they like selling Office as a package, but they're slowing decomposing it. OLE was a micro step, now with a cloud graph model and AI they're going much deeper.

Ultimately, your "table" can be generated from data in a "spreadsheet" or "database." And your AI can interact with this data precisely. At a certain level, you really shouldn't need to care where something was generated or viewed. It's data. But the more strongly it's "typed," the more accurate everything becomes.

Yes there's some hype over AI, but it is just incredibly useful, even the most basic and stumbling step of importing your documentation to an app or browser based AI to answer questions. Anything that doesn't include it as a first class consideration is last epoch, in the same category as Visicalc, 1979. Unfortunately, that includes the vast majority of Open Source. KDE was way ahead of this, but they stumbled. I hope they can catch up. Local, Open Source AI needs to leap ahead or computing is going to collapse.

So the first step is to give all the data a schema, ideally a domain-relative one so it's interoperable. See Linked Data.

Thanks, I didn't know that backstory. I will check out that repo.

Sorry, but

  (base) vid@kk:~/D/ai/tortoise-tts$ python3 -m venv venv
  (base) vid@kk:~/D/ai/tortoise-tts$ . venv/bin/activate
  (venv) (base) vid@kk:~/D/ai/tortoise-tts$ python -m pip install -r ./requirements.txt
  Collecting tqdm
    Using cached tqdm-4.65.0-py3-none-any.whl (77 kB)
  Collecting rotary_embedding_torch
    Using cached rotary_embedding_torch-0.2.3-py3-none-any.whl (4.5 kB)

…

    × python setup.py egg_info did not run successfully.
    │ exit code: 1
    ╰─> [8 lines of output]
        Traceback (most recent call last):
    File "<string>", line 2, in <module>    
    File "<pip-setuptools-caller>", line 34, in <module>    
    File "/tmp/pip-install-i7ubxxkc/scipy_4d5af4f3e2094adca3313ccb41a6d5ff/setup.py", line 196, in <module>    
      setup_package()    
    File "/tmp/pip-install-i7ubxxkc/scipy_4d5af4f3e2094adca3313ccb41a6d5ff/setup.py", line 147, in setup_package    
      from numpy.distutils.core import setup    
        ModuleNotFoundError: No module named 'numpy'
        [end of output]
  
    note: This error originates from a subprocess, and is likely not a problem with pip.
  error: metadata-generation-failed

  × Encountered error while generating package metadata.
  ╰─> See above for output.

  note: This is an issue with the package mentioned above, not pip.
  hint: See above for details.

  [notice] A new release of pip available: 22.3.1 -> 23.1.2
  [notice] To update, run: pip install --upgrade pip
  (venv) (base) vid@kk:~/D/ai/tortoise-tts$

I'm sure you could eventually help get this working, which is kind of you, but the point is the "supposed tos" don't work either. It needs to be comprehensively fixed if python really wants to be approachable. Maybe it doesn't. It's also just not a good citizen when it comes to heterogeneous apps on the same system.

This isn't the first time venv didn't work for me, then there's anaconda, miniconda, and a bunch of other things that add env and directories. I don't really know what any of them do, and -I don't want to- I'm not an expert on every app on my system, but I can use nearly all of them without pain. (remember this is about ease of use)

Oh yeah, and python 2 vs python 3. <rolls eyes>

It's very much the "works for me" experience from the old days. There's no good learning from it, except dependencies suck and python systems aren't good at them.

I think when releasing anything that includes dependencies that span the operating system, it's just good engineering to use a container approach. Otherwise you're just causing a lot of discomfort in the world for no good reason.

It's funny because chatgpt would give me an answer to this in a few moments, but I'm locked out for a week because it can't figure out where I am.

Now I'm spending my Sunday morning setting up a dockerfile for tortoise-tts. At least I will learn something reusable from that. I guess I will create a PR for it, though it seems the author isn't tending the repo anymore.

Granted I'm not coming from the python world, but I have tried many of these projects, and very few of them install out of the box. They usually end with some incompatibility, and files scattered all over the place, leading to future nightmares.

  ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
  sentry-sdk 1.22.2 requires urllib3<2.0.0, but you have urllib3 2.0.2 which is incompatible

Just for fun, here's the result of python -m pip install -r ./requirements.txt for tortoise-tts;

…many many lines

          raise ValueError("%r is not a directory" % (package_path,))
      ValueError: 'build/py3k/scipy' is not a directory
      Converting to Python3 via 2to3...

…

  /tmp/pip-install-hkb_4lh7/scipy_088b20410aca4f0cbcddeac86ac7b7b1/build/py3k/scipy/signal/fir_filter_design.py
      [end of output]

  note: This error originates from a subprocess, and is 
  likely not a problem with pip.
  error: metadata-generation-failed

I'm not asking for support, just saying if people really want to make something 'easy' they'd use docker. I gather there are better python package managers, but I gather that's a bit of a mess too.

Someone is thinking "this is part of learning the language," but I think it's just bad design.

I think Google is now thoroughly infected with Big-Company-itis. A couple departments would like to and know how to comprehensively use AI across many services, which the consumers would love (though some would be confused by it). But legal, marketing, and some guy in a department called "Annex B?" are preventing it. So then the people in those departments get bored and go somewhere else and their perspectives and skills are lost.

Oh, I know all about mute.

30-ish years ago, DVD players supported alternate soundtracks so people could offer a soundtrack with music and without. I don't think I've ever seen that basic affordance on a web video. That would probably lead to more diegetic sound, and the tools to simulate it.

I cannot stand the music, or music in any video of this type. I don't need or want its loaded suggestions. I would love to be able to hear even simulated wind whistling, and sounds from the scenes below.

I tried KDE a few times over the years, but it always felt a bit off. However, six months ago I tried it again, and really like it. As others say, it has a lot of useful customization. What worries me more is the switch to Wayland, when essential apps like synergy/barrier don't work well there yet.

It's a shame they didn't mention Hydra. https://www.hydra-cg.com/spec/latest/core/

This is designed to let data be exchanged richly between sites, and comes from the earliest ideas of the web, it is the HATEOAS part of REST.

I think I'd rather find a way to enjoy developing tech for universal friendly anonymous healthcare workflow identifiers. But if anyone can write supporting music, please do.

I've dabbled in making music, but it makes me bump against an existential crisis. I could make nice sounding music, but what does it mean? Often, it's just an excuse to buy more gear.

https://www.youtube.com/watch?v=ylXWTN1XTtI

You can run it with CPU or GPU in the VM, in the latter case you'd need to pass the GPU(s through, which depends what kind of VM you're using.

Yes, but Apple could have introduced it as an open design on introduction. That would effectively have been two products; the functionality and the open-handed, wide-ranging ecological solution. They chose to offer a different second product, an implied proprietary, exclusive design. Now, three years later, it's an open standard, in the meantime there is probably thousands of tons of e-waste of interim products.

Yeah, I've stuck with Google for a somewhat more open version of the best of tech without trying to do everything myself, but I also run Linux everywhere, my own mail server, node-red, etc and would love to one day either free myself from the Google, or see them fully accept and work with the open world.

I would include wireless magsafe in the bad decision category. The idea is good, but the implementation… they should have found a way to make it backward compatible, at least for the first generation of devices, and to be a really good company, introduced it as a standard from the get-go. But that wouldn't be very Apple. So now we have a few generations of wireless charger e-waste, which is a lot more material than a cable.

The dock is a main feature of this device. You can even buy additional docks to place around the house for charging/speakers. If you don't need the dock, I guess it's not for you (though maybe they will release a dockless version later… and reduce the price). And unless you are really into the Google ecosystem (which can be compelling, but also annoying when they abandon products or gut them) it's not well priced in any market if you're just looking for a tablet.

My theory is Google doesn't want to be too competitive, they mainly want to enable the Android market. If they're too competitive, Samsung, Motorolo, Oppo, etc will change to a different base. Google benefits from Android anyway by its embedded core of Google services.

However, I think with the -a series (7a, etc) they've turned to filling a gap with a reasonably priced high quality device, because Samsung &c low priced handsets are really terrible.

From the study: Based on the official COVID-19 death reports, it is estimated that vaccinations have prevented 14.4 million excess COVID-19 deaths worldwide between December 2020 and December 202139. Thus, vaccination is the most effective method for preventing the spread of SARS-CoV-2.

Nightlight on the colour & ambient bulbs I have don't go nearly dim enough. I haven't tried the hipster bulbs, but I don't want a lightbulb that calls attention to itself.

I wish there were a way, though maybe not at the hardware level, to get these bulbs to go dimmer. In many circumstances, especially at night when avoiding glare off windows, just enough light to prevent tripping over things post eye adjustment would be ideal. But the lowest setting is far brighter than that.

I guess as soon as a kit is available for this purpose that doesn't require advanced knowledge (say, `aisupport4 my-repo/`) and runs on mainstream-ish hardware and doesn't require a centralized service (even running in the browser via eg transformers.js), things will change considerably.

I keep thinking there should be a way to train a copilot against just one set of code libraries. I know LLMs require training against a lot of text to get their smarts, but is there a way to set this up so a model can be created for a specific library by anyone, so it could provide open source support via a transformer + model? Maybe this would be a better approach than a jack of all trades, master of none.

Hmm, I don't think Inkscape on Mac supports Automator &c. And you can run it as a local server to mitigate privacy concerns. There are ways to support stylus features in browsers, though it's an area that could use more support.