jpetso

Higher resolution means higher power consumption. If you follow laptop reviews, you'll see that getting a 4k panel will set you back at least an hour of battery consumption compared to standard 1080p, perhaps two or three depending on panel and battery size.

I like high-res screens. But in general, the best course of action for a laptop is to get the lowest possible resolution that still works for you. Personally, I have a hard time seeing how an upgrade from 1440p-ish to 4K-ish makes any discernable difference on a 14" and how it justifies cutting into my power efficiency.

More or less, yeah. Not just as API for apps, also as offline-capable executable for embedded devices.

There's an investors page now, with the PR blurb and an informative webcast: https://investors.soundhound.com/

Counterpoint, we travelled this year and got two flights for the price of one. The refunded flight from early COVID days literally paid for both. Even after somewhat expensive COVID testing costs, still a significant saving.

Sure, now that travel restrictions are causing less uncertainty and demand is ramping up, this might not be repeatable. Perhaps hotels are skimping, not sure as we stayed with family. Either way, there's always some subjectivity in use cases and that can easily tilt the scale one way or the other.

I think Firefox and Chromium are good examples of what direction can be expected of mainstream open source projects that end up being steered by a company with commercial needs.

Both got forked a bunch, but the mother project remains the popular one due to user stickiness and, frankly, sheer amount of (paid) effort invested compared to small-scale volunteer efforts. Users can still choose to go with Iceweasel, disabled Pocket integration or Ungoogled Chromium or whatever fork de jour, but generally they don't because "the ecosystem keeps moving". They generally get more desirable things added than annoyances. Unless a company gets it wrong and you get a WhatsApp-level backlash to make another competitor viable.

With constantly moving goalposts, commercial interests can keep the upper hand as the community (and much other competition) can't keep up fast enough. That's why Google keeps pushing the web into ever new directions, and why standardized APIs got replaced by ever-changing Electron apps with mandatory auto-update.

If the popular distributions ship with ads, I'm not afraid as long as there is competition out there that does without. I can always do Arch or Debian instead of Ubuntu and I know they'll do the right thing.

Where it gets dangerous is when the base technology is controlled by commercial interests. Microsoft owning secure boot certificates, Android not getting drivers upstreamed, Apple deciding who gets to compete by owning the App Store. Linux may end up in this spot at some point if a particular vendor (or consortium of commercial interests) becomes too powerful compared to the community and smaller competition. Once it gets there, the only escape is having viable alternatives.

If you rely on commercial software, your alternatives come with huge switching costs. Chromium to Firefox is a little tedious, Mac to Linux is a major disruption. The more you rely on a particular piece of commercial software, the more they can control you.

But competition generally does the trick. I don't mind games being closed source, there are so many quality alternatives that the game studios simply can't lock most users in. Now, app stores for games? That's where you want to be cautious, because they control what competition looks like. Operating systems and drivers? If they turn on you, better start swearing and leaving investments behind. Chip-level DRM? Zero leverage, ask your government for help and if they're not willing, you're screwed.

So really, commercial software isn't that much of an issue high up in the stack. The lower you get, the more careful you want to be about end user communities retaining control. Let the companies make apps, let them sell apps, don't let them grow platforms. The platforms should belong to users.

If a company comes along and makes a commercial platform out of Linux, run.

Phoronix is running some benchmarks again which should be coming out in the next days.

Canada had a more or less total stop to immigration until recently, and despite that Toronto managed to have another crazy housing rally earlier this year before vaccines were widely available. Before that, the 15% Non‑Resident Speculation Tax caused a brief scare until the market realized that it can rally just fine with bona-fide residents.

I'm not buying the "wealthy foreigner as the main culprit" story. Surely they add to the existing demand and it's always nice to have an external scapegoat, but there's clearly something bigger going on than just foreigners. I'm thinking continued difficulties to build housing without going through multi-year permit processes, most of the city still being NIMBY'd into low-rise SFH zoning, and the sizeable number of millenials both getting wealthy enough to buy into housing as well as getting FOMO from continued price appreciations.

But it's always nice to have a plausible explanation for why the innocent locals aren't at fault.

To be fair, Lenovo is now pushing out firmware updates via fwupd, employs at least one engineer to work on Linux support for ThinkPads (Mark Pearson, does he have a team?) and apparently also asks their component suppliers for Linux compatibility now. They might not do all or even most of the work, but even the requirement to certify your devices with Ubuntu and Red Hat stickers is a positive change compared to the early days or vendors like ASUS who just ignore Linux entirely.

There is F-Droid, which is a terrible client app but provides (non-Amazon, non-pirate) open source software such as NewPipe, Simple Notes or DavX.

My take would be that actively working against both equal opportunity and equality is unethical. By giving your children a giant leg up, you're taking from all of the others.

Imho society should be liberal in letting people acquire and enjoy their wealth, and strictly limit the passing down of wealth to prevent dynasties and having someone's opportunities decided by their parents' means. This would solve a lot of compounding problems. However, people are generally pretty terrified of not being able to let one's own kids inherit the entire fortune and sharing it with other random kids too. We'll likely continue along those lines for a long time coming, whatever your stance on ethics doesn't stand a chance against family bonds and their voting power.

You must be a fast learner. After one year of learning a new language, I personally would not feel comfortable speaking it well enough to use as examples for others.

Unfortunately, there's always a trade-off. You want both quality data for your use case, but you also want lots of data so it generalizes well. Those are conflicting goals.

Fortunately, splitting models into separate accent-specialized variants and helping them out with language model training will often help in case the model doesn't cope well enough with the cognitive dissonance.

You're not wrong, but with so many chips incorporating some sort of dedicated "AI" or "tensor" functionality, perhaps the issue will resolve itself for most portable devices in a few years. Plus there's always the option of optimizing a little more and/or abusing other available hardware such as DSP chips to get the real time factor down. Anything over 1 isn't great, but it's not a bad start.

Which, it must be said, isn't always as bullet-proof as it could be. There's a not insignificant amount of transcription (or pronunciation) errors in those datasets and Mozilla might want to find ways to increase the quality of already-released data over time.

For Mozilla Common Voice, it looks like even Bokmål isn't listed as dataset yet. Language packs have the advantage that a single dedicated user can come up with the entire thing, but for voice collections you need a large variety of different people and ideally tons of them. For any language with a small native speaker population, even a rich one like Norway and especially a fractional subset like Nynorsk, getting enough speakers to participate in open source collection efforts will remain a challenge. Purportedly, even for commercial companies it's hard to find enough Norwegians willing to speak a few sentences for a nominal payment unlike most other countries.

Luckily, speech recognition research is making some good progress on dealing with low-resource languages so hopefully we'll see some acceptable models made from the little available open data that's out there.

I have high hopes that the apparent future popularity of the Steam Deck will make a serious dent into this shortcoming within the next year.

Rent-seeking is when profits massively outstrip investment. Perhaps infeasible in practice, but if government protection of licenses (patents, even copyright) could be capped at a maximum that's proportional to money & time invested, rent-seeking could become less of a problem.

As in, allow entities to make a handsome profit, far outstripping their original investment. Just don't allow them to profit off of it forever. Patent and copyright expirations are a proxy for this, but they don't take into account the actual money & time invested, instead they replace them with a hard-coded magic number which will work for some products and fail for others.

The difference is lack of competition.

In housing, that's zoning for single-family houses which limits affordability and provides exclusivity vs. apartments or boarding houses, except in downtown cores which aren't enough supply to really move the needle (and expensive for other reasons).

In US healthcare, it appears there is regulatory capture from insurance companies and private health care providers, both successfully ganging up on the government to minimize real competition or strict price regulation.

Without real competition for a product that consumers can't feasibly opt out of, no heavyweight market participant interested in driving the price down, that's what you get. But yeah, testing is good too. Get the government to try out 10 years of broadly available at-cost housing competition and public healthcare, then evaluate :-P

> It isn't "society" that's absorbed absorbed increased worker productivity, but employers and those who own the automation.

What society has "absorbed" is an expectation that stock markets will continue rising at a pace that significantly outstrips inflation and (related) wage increases. Pension funds and individual retirement savings (401k, RRSP) heavily rely on employers pocketing the bulk of productivity increases for themselves and continuing to exploit labour. It'd be interesting to see the chaos that would arise if governments decided at some point that workers should get an equal share of that.

I love it when company representatives reply meaningfully to all questions of a thoughtful request instead of just cherry-picking the convenient ones. Thank you for setting a great example.

> Unless you have a strong civic identity that aggressively assimilates and ensures respect of those civic norms, you have an inevitable clash that just becomes a question of scale.

I'd like to suggest multicultural cities such as Toronto as an example that preserving cultural backgrounds and integrating with the rest of society aren't necessarily mutually exclusive. You have a broad mix of people from (among others) European, East Asian, South Asian and Middle Eastern backgrounds here and while all of the usual discussions are alive and well (left vs. right, urban vs. suburban, immigration vs. bubbling), people's cultures are just not the main issue dividing people. You'd think that with a lot of Muslim immigrants, Canada would have larger problems with value differences, but guess what, most people are just people and want a good life for themselves and their families. Nothing new here, move along.

The point is that full-on assimilation isn't necessary. Celebrate shared values. Celebrate diverse backgrounds and differences, too. I don't have to tell a Muslim woman that she can't wear religious clothing, just like I don't like it if her husband tells her the opposite. I can just grab food with the two of them and discuss tech, ethics, religion and whatever, just like with regular people, because they are regular people. The kids just see a lot of different ways of people living their lives and learn that it's okay to be different.

> In your hypothetical, how exactly would this situation [Islamic Law / autocratic dictatorship] come about?

Recent history suggests that there is a template: a politician gets voted into office, gets drunk on power, finds a common enemy and sells it to the people, changes elections and eventually the constitution to remain in power, kills the independent press, persecutes opposition leaders and intellectuals with dangerous ideas.

Turkey and Hungary are well on their way. Russia has arrived. Iran and China went the revolution route, that works too, although probably requires more of a flawed system to begin with.

Having written this, I think it's worth pointing out the main flaw that I see in nationalism: it's that it uses culture and nations as a proxy for principles, as opposed to digging for the actual principles underneath.

As an example, I stand for freedom of speech, in the form of a diverse, largely independent set of publishers and authors that allows me to make up my own mind by exposing me to different viewpoints. Many Western countries have a better track record at this than authoritarian countries. But this doesn't mean that I'm on Team "The West" - if my country drops this value, I'd rather drop my country than my values.

If my country thinks that skin color or wardrobe of a person is more important than what that person is saying, I'd rather drop my country than team up against that person.

If my country decides that Islamic Law or autocratic dictatorship is a better form of government than democracy, I'd rather drop my country than abandon my principles.

It's great to be in a shared space where you agree with everyone. The question is, when you eventually disagree (because you don't get to choose what your neighbors think, and because nationalism is just a proxy for actual values), then are you going to follow your principles, or do you let your nation and culture dictate what your principles should be?

I find that despite my national background, there are tons of people in my country whose ideas and principles I don't share, and lots of people from other countries/cultures that I would team up with anytime.

Drawing a line primarily along cultural lines means ignoring or downplaying all of the other lines that other people might find similarly important or more so.

Why would I bias my decisions about people over other people just because they grew up within a few hundred kilometres of me? Why would I prefer helping a bunch of deadbeats from my own country over a bunch of promising bright people from halfway across the world? Because they'll stand up for me in return?

Because it's always one team against another? Fuck teams. There are my immediate friends, and outside of that I'll fight only for principles, not arbitrary teams. Everyone's a different person in their own right. Everyone deserves to be treated as who they are, not where they're from. (Cue Backstreet Boys.)

More important than lives, no. Important nonetheless.

So you're basically saying that Germany's a sucker and Greece benefited from Germany's generosity. Sounds great in theory, but my impression was that despite all the debt haircuts, it was much less fun to be in Greece than Germany at the time.

If you win at (maintaining) debt, aren't you supposed to improve your lot in life? Just being a "deficit black hole" doesn't necessarily give you an edge, just like credit card debtors and beggars on the street always run the risk of getting hit hard at the slightest surprise, and often are. Whereas not being in debt is perhaps the nominally less efficient way, but it prevents you from getting messed up by your loan sharks again and again. In that sense, trade surpluses are a risk avoidance strategy and as such, make a good match for traditionally risk-averse Germany.

On the other hand, if you work fewer days, there's more time for home cooking :P

My guess would be that this allows developers to distribute applications on Apple's App Store without much of a headache, where otherwise the user has no (LGPL-guaranteed) way to swap out libraries otherwise. IANAL, and haven't spent any significant research on this fwiw.

"It's complicated": https://www.python.org/dev/peps/pep-0394/#for-python-script-...

With packaging technologies such as Flatpak and Snap, app makers now have the option to distribute a bundled-everything version of their software to Linux users.

This has drawbacks too though, in that it's now up to app makers to take care of keeping supporting libraries up to date and secure. That's not necessarily their top priority though, which is a risk for the end user. Also, unnecessary duplication of libraries increasing memory use, when the distro is able to share most of them. Plus the poor middleman has an incentive to set user-friendly, privacy-preserving defaults that I wouldn't trust as much when the package is provided by a commercial company with different incentives.

It's great to have the option, but overall I would still prefer to use distro packages except in the rare special case.

Nuance spun out their automotive business into Cerence in 2019, so they presumably have zero business with carmakers right now. There should be conditions not to compete with Cerence in the automotive market.

Left-leaning European here, I think he gathered enough demerit points that wouldn't usually fly with any review team, except he was a high-profile political leader which made it relevant regardless. Now that he's not POTUS anymore, by all means let him play by the same moderation rules as the rest of us serfs.