tensility

Two years later and this app is still rather ... lite, eh?

Well, it is open source. Perhaps its primary developers are amenable to major feature PRs.

Currently, there are over 7500 casks available in Homebrew. A significant part of the value of a GUI such as this to me is as a means of discovery via browsing, and this tool fails at that by not providing a means of browsing the entire catalog, even if that catalog is currently subsetted to cask only.

I'd like to see an "All Casks" category that allows all apps to be browsed rather than just the curated casks appearing in the existing category panes. Presuming that all formulae eventually get supported, then an "All Bottles" category would also make sense.

Of course, something like the mini-NAS that Geerling describes in the following article IS a great fit for an N1x0 machine. These are really all just a spectrum of tools we have available, many of which are better fit than others for particular niches. A NAS seems like a perfect fit for N1x0. So, yeah, my initial response was dumb, but so is dismissing the RPi simply because you're trying to apply it to the wrong kinds of use cases.

<https://www.jeffgeerling.com/blog/2025/mini-nases-marry-nvme...>

The thing is, I don't use a RPi for desktop-like experiences, so I don't want all of the accessories you've padded the price with just to have a cheap unix node somewhere that I need a small amount of compute or, especially, physical presence. For a desktop UI experience, I have and prefer much better machines to drive full desktop UIs than either a Pi or an N1x0. For that, I use one of my Macs.

Of course, I already know we don't have similar needs or desires for a desktop experience, because I haven't found a good reason to run Windows for a couple of decades (since I'm not a PC gamer type and that seems to be the primary interesting reason left for running it), other than to occasionally boot up the newest version and try it out to see if I'm actually missing much.

Also, for servers that get real load, I'd often rather run them from (unix-based) containers in the cloud anyway, if only for maintainability's sake. Heck, I already have NUCs depreciating out in storage that I don't bother to use because a RPi is usually a more solid and energy-efficient choice, unless I want a decent desktop experience or need a high-load server, for which purposes an N1x0 isn't usually the machine I want or need, either.

Now, that said, if you do want a cheap minimal desktop experience machine and you don't mind or even want the Windows experience, I suppose a used N1x0 probably is a great choice.

¯\_(ツ)_/¯

As more waves of layoffs happen due to AI, there will likely be a growing market for "No AI" products, especially given how nasty some of the C suite are being about that short-sighted "unemploying humanity" objective they are being explicit about.

VI runs EMACS?! I knew the opposite to be true, but...

;-p

At twelve years in, you are expected to be a senior engineer and to be able to bootstrap yourself into a new dev environment without someone leading you to do so. How you go about doing that, especially in a chaotic environment with aggressive schedules, will become your image with your peers.

One technique I would advise is to use 'git blame' (and/or conversations with your engineering and product managers) to identify authors of components (or even just chunks of code) that you are struggling to understand, and then set up short 1-on-1 meetings with those authors to discuss the system architecture from the perspective of the component(s) that they have written. Rinse. Repeat.

Remember, as you go about this, that what you are really doing here is building a living network of the tribal knowledge graph within your company, so be kind and respectful. Further, consider starting making a guide book of sorts, both for yourself as well as for subsequent new hires, especially more junior engineers (who you may need or want to mentor).

Don't worry. You've got this. Best of luck!

Why does running 'help' require tokens?

On my M-series Mac devices, I get a warning for missing executable when trying to launch after installing. I've had Steam re-verify its payload files with the same result on two different systems. Any ideas?

It was already demonstrably foolish to trust anything Elon says before he made his explicit turn towards techno-fascism, but that certainly seals the deal for me.

Did they ever actually install many of those solar tiles?

This is more or less the advice that I was going to give. I would add on to it by recommending that, if you don't know what to build, go find an open source project with open pull requests and start fixing broken things. Keep track of your submissions and use them to start creating a portfolio of the professional work that you've done this way a gratis.

In addition to doing volunteer work this way, also look for various "bug bounty" situations for accumulating similar "professional micro-experiences" that you can also use to show that you have been crafting and delivering functional code into one or more projects.

That's more or less also what Google does with Chrome in order to force folks to use it and only it. I know that well because I have Chromium installed for this reason alone.

Solidly enforcing citizens' constitutional right to privacy seems a far better approach to addressing the problem of Google at its root cause.

Yup. Steaming is the way to go.

Boil your eggs by steaming them for excellent results based purely on timer and no need to use anything to cook with other than water since the temperature of steam is extremely consistent. This works very well for results ranging anywhere from barely soft-boiled to quite hard-boiled. Of course, after removing from steaming basket, you'll still need the normal cold shock techniques for removing the shell.

In the spirit of constructive criticism, I'd suggest that the GPU section of the summary feels like a fish out of water and should probably either be removed or supplemented with a bit more context from the actual paper. Cheers.

Telling folks that they are being sold out for peanuts (or almonds) doesn't diminish the betrayal; it only makes it worse.

I'd agree with your recommendation while pointing out that a big part of the benefit from doing this is getting high-quality audio ADC/DAC hardware in the final mixed-signal interface. This can be achieved well with either kind of device, but the external device will be usable longer.

That said, my biggest gripe about external sound cards is introduction of drivers rather than usage of standard bus class protocols, which has a tendency to create "complicated upgrades" problems due to the manufacturer dropping support for the driver over time, causing an unnecessary forced obsolescence for otherwise good external hardware.

Not if the audio circuitry has been properly canned (i.e. Faraday-caged) and if the power supply produces clean input power; however, one or both and especially the latter often do get compromised in poorly integrated Wintel PC systems.

Mixed signal components are hard to do well and have significant impact on the quality of final analog audio I/O. So, it matters which components have been used regardless of whether it's on on-motherboard solution or an external add-on solution (whether internal over PCIe or external over another appropriate bus, such as USB); however, the economics of onboard solutions make it quite likely that this is where quality will be skimped since many consumers won't care much about quality differences in this area, whereas the focused external solution will live or die in part based upon the quality of such components.

Note, however, that this concern really only applies to whatever in your system is performing the final digital to analog or initial analog to digital conversions since that interface with the "real world" is where the quality matters. In a world with bluetooth speaker systems, where that mixed signal phase actually happens in the speaker hardware, it is relevant to consider where this kind of quality even matters in the full audio system you're deploying.

I think that the posted example doesn't even work correctly, dropping most of the article's text after the first several paragraphs. Why did you post something when you didn't apparently even bother to check the results for correctness?

Oh, wait. I'm starting to understand that you fed it through a bullshit generator rather than just stripping it of the various pieces of unrelated content and providing just the original text, which might've been interesting to me.

As with any new role/team/etc, the best course forward is to set some "evaluation time frames" at the start and then follow through with the process of deeply reconsidering whether it is actually a good fit at each checkpoint. My rough set of early such milestones happen at one month, three months, six months, and one year. If it is not, then the best way forward is to simply start looking for a new role/team/etc either internally or externally.

Why has this 4 year old article bubbled to the top of HN?

It's partially a product of Amdahl's Law. Coding and related textual activities are only a fraction of the work required for product design, implementation, and maintenance.

As does Apple's Darwin kernel, when used as intended. :-)

Tip #7 is my favorite. ;-p

Normal folks get fired for "mistakes" like this. Oligarchs don't.

Crisis of trust in Altman reflects back on Y Combinator. They have significant financial motivation to downplay his behavior regardless of any moderators claims.