mikewave

I'm amazed they've been able to make a business out of this. At D-Wave, we have been able to generate truly random numbers for aeons; we use them for calibration, but they don't have any real business value presently. Truly random RNGs have been commercially available for ages, and if you need such a high level of randomness, you're probably sensitive enough that you don't want to accept them from some 3rd party over the internet.

If you want some quantum-generated random numbers, sign up for D-Wave Leap and you can trivially generate your own for free... or perhaps, run an optimization algorithm to do something a little more useful.

I don't get it. Isn't the whole point of using a VPN to hide your traffic to some minimal extent from the FAANGs of the world? Why would I run even more of my traffic through Google than I would otherwise? What on earth does this "protect" me from?

I might as well just trust an MITM Google CA and have done with at this point.

I can't imagine being surprised by this. Fundamentally, the true (teleological) purpose of carbon credit markets is threefold:

* Provide the public perception that Something Is Being Done about the looming threat of climate change while not actually forcing anyone to change anything in reality (aka "greenwashing")

* Enable a new class of speculative investment that can be used for profitable ends by the usurious class as they skim off the top

* Further shift the balance of power to larger corporations that can afford ESG sinecures and push policy so as to make it harder for smaller companies to meet regulatory compliance

The net result is not going to be much decrease in carbon - rather, if we are granting some amount of carbon credit to every area of land that presently absorbs carbon, we will just figure out a way to use it all up. Meanwhile, one can safely expect that this wealth transfer will be co-opted on the far end by all manner of corruption; the kickback schemes with this will be convoluted and inevitable.

The only actual solution to decarbonization is to actually make it make economic sense for individuals in the first world to not emit as much. This does not have to come in a form that enables countless others to skim off the top of the effort. Implementing zero-carbon electrical generation (e.g. nuclear) and getting people to switch to electricity-based energy consumption (from heat pumps to EVs) is the end goal regardless; a government with gumption and honesty would just do this directly rather than hoping some artificial "free market" will have the same result in a fashion where they cannot be blamed as quickly for failure.

Wow, this is an old document...

> To start your walk head across campus to the Shell station.

Hasn't been up there at SFU for well over a decade now, maybe longer...

Back in the day this Java applet by the famous Ken Perlin was fun: https://mrl.cs.nyu.edu/~perlin/demox/Hyper.html

Recommendation: Your brain actually has two depth perception systems: hardware (using your two eyes for parallax) and software (using visual cues to establish a 3D scene, the way you can with one eye, or when looking at 3d projections onto a 2d surface).

Use each depth perception mechanism for a different axis! In this old applet you could enable stereo projection and also thick-lines-mode which really helped this, and gave me a strong intuitive feeling for the motions of the hypercube.

From what I can tell from engaging with Flat Earthers, it's actually 50/50 people who reject everything reasonable they've been told, and people who are playing a game.

The game is fun and simple: pretend you believe in the Flat Earth to piss people off. Refute their scientific facts with questioning and amusing statements, like "I never said it wasn't a circle" or "How do you really know Australia exists if you haven't been", etc. You win the game if the other person gets mad and truly believes you believe in the Flat Earth. You lose if either they don't believe that you believe this, i.e. if you've given it away with your facial expressions, or if they manage to actually present convincing evidence (e.g. like Carl Sagan's sundial shadows) that cannot be refuted.

I have conversed with both kinds. The latter can be highly amusing people to talk to, the former are legitimately strange and I can't fathom how they come to exist. The linked post may be correct; I think some people simply have the innate urge to reject everything they've been told and to look for a countercultural answer instead, regardless of the subject matter. At least this is a safe direction to channel their madness into.

More like, ever tried to dig down more than a foot or two? Unless you have really soft ground, even a good steel spade and shovel aren't any guarantee of success. Powered augers are often needed to make holes for fence posts, etc.

Sure, if you have a billion years to wait for it to all run through the rock cycle again.

The big difference here is that CI/CD should ideally be taking away annoying work from devs so that they can more smoothly go from problem/feature -> hard thinking work -> code -> out the door, without having to deal with the highly repetitive part of the job.

The problem is that there's an expectation that this increase team velocity more than it increases quality of living.

15 years ago in IT if you could automate things, you just ended up with less active work to do. Now that automation is the norm instead of a bonus cool thing, it has become formalized and measured, and you may not have the saved time as free time, you have to now double down on doing the hard parts of the job.

This doesn't actually work in reality for most people as we can't actually think hard for more than a few hours out of the work day (10x'rs exempt, of course.)

I think it's important to note that it has sucked the humanity out of levels of worker above the bottom-of-the-pole labourer, as well; the HR people and trainers in the post are similarly labourers, though instead of using their muscles they're using their words; but they're still required to work like hands of the automation, with scheduled onboarding sessions and processes they have to follow.

Given enough time, and a smooth process, eventually this should not be so painful; but when it is painful, it is supposed to be the job of a human (especially once in HR) to act as a bridge between process and person, so as to keep morale at a good level and to provide a channel for feedback from employees up to the teams responsible for process automation. It doesn't sound like this is in place here, at all; instead we have people who are displaying all the signs of not caring about the human anymore because they have become so subservient to the machine that they cannot afford to.

> how does one make time for it all?

You truly don't - it's impossible. It's like the CAP theorem: there are fundamental limits that all stem fundamentally from a lack of time. You have to sleep, eat, wash, exercise, do your chores, etc - these are fixed costs on your time. If you have children - and I recommend it, it's amazing - those fixed costs increase significantly.

If what you want is to continue to have a really good career - well, you'll either have to find a way to keep it within work hours, or you will have to sacrifice something else. Some people sacrifice exercise, sleep, hygiene, etc. but this always has its own costs in return. Some people sacrifice all of their free time, which instead has mental health impacts.

Instead, you have to...

- Become as efficient as possible at all of the above. Cut the corners that can be cut while still fulfilling the requirements. - Accept that you will not be able to perform the same as someone who is dedicating every moment of their lives to do something. You have to accept the benefits you get from your other pursuits in life as being worth the sacrifice of not getting extreme career advancement, not getting every moment of leisure time you want, etc. - Know that you can still achieve your goals to a large extent as long as you are willing to have a long horizon (low time preference).

> As a person with eclectic tastes/interests, limited time to pursue them and a finite (atleast last time I checked) lifespan, I have found it increasingly difficult to figure out how to spend my time and seem to end up with an overall feeling of being overwhelmed and settling for the easiest activity at the time. Is it just a question of time management and prioritization?

100%. For example, I love hiking and camping, and I would love to take a week away from my young family to explore the mountains, but ultimately doing so is somewhat selfish as it would lump all the work on my wife for that period of time. If I can find a way to enable her to have her own week of leisure, then maybe that can balance out; but more realistically I am going to be on a hiatus from week-long outings until the kids are older. This doesn't mean the occasional weekend can't happen, though!

Planning in advance and ensuring unknowns are minimized for yourself and your partner is key to ensuring that you can still do what you want.

I will leave you with this:

I have no idea what the hell I was doing with all my free time before I had children. There must have been hours upon hours in the average day that I had absolutely nothing to show for other than consuming content or excess calories. Now, every moment is utilized; there's way less "downtime" but honestly I don't even want it - I just want to get to have "uptime" of my choosing, and it happens as long as I make it happen.

I know parents who are overwhelmed and never get any free time, according to them, but when I watch how they live it is clear that this is happening mostly because of their own inefficiencies and the fact that they expect downtime to look like vegetating in front of the TV, where hours slip away, instead of going out and doing something like hiking / spending time with their family on an outing / etc. No wonder they feel like nothing is getting done and that their days are drudgery! "It's all I have energy for" is a poor excuse - the more you do, the easier it gets.

Agreed. There's way too many things to do in reality. I have about 30-60 minutes a day of Me Time, and the last thing I want to do is piss it away in front of a computer when I've already spent 70% of my time doing exactly that.

Go outside, people. Talk to your friends in reality if possible. Experience the place you live in; participate in something physical to improve it materially. There's no end of socialization and minigames in reality, the graphics are better, and it's free.

> I can't imagine had I grown up busy farming and raising siblings and then children in my 20s I would have been able to accomplish anything at all.

This really depends how you frame 'accomplishment', which is very much up to you. I think that successfully raising children who can successfully raise children is in and of itself an accomplishment; forming a family and keeping it intact through your inevitable troubles, working the land and producing enough excess food to earn everything else you need... our culture would be better if we actually viewed such people as 'accomplished' instead of pretending that being a C-suite officer of some SaaS b2b griftware is inherently of more value to anyone, anywhere.

One mechanism is to design the tax to be revenue-neutral for the average taxpayer.

For example, with carbon taxes on gasoline, we can calculate what the average person consumes in terms of gasoline per year, and then calculate the outliers (people with super-efficient cars and people with gas guzzlers). Then, we establish some reasonable maximum that we think we can get away with surcharging the guzzlers, and establish a gradient. The average person is given back a tax break that corresponds to the surcharge they'll pay at the pump, so it's a wash for them; the guzzler gets the tax break too but ends up paying more, incentivizing everyone to be the efficient driver who basically gets a bonus.

I think you could do the same for any kind of tax; establish the baseline for resource consumption efficiency for a particular recyclable commodity (and it will have to be per-commodity to make any sense at all); set up the incentive gradient so that companies producing more-recyclable-than-average goods end up getting free cash for doing so, hopefully offsetting the other costs associated with this, and companies producing things that are harder or impossible to recycle end up paying more.

The end result is that the product for the consumer that is more recyclable should end up making more financial sense. Instead of pinning the gradient the way you do for gas (literally, 'what they can get away with and still get elected'), you'd pin it at a level where it incentivizes companies themselves to be purchasing recycled materials instead of new ones.

All of the above is predicated on the material in question being able to be recycled without requiring more energy input / producing a higher carbon footprint to recycle than acquiring the original raw product is. There are some materials that it's just not worth to recycle, most of the time; plastic is definitely on that side for now, like it or not.

There are companies that will drive your "smart" thermostat and purport to save you money by strategically controlling it, but in the end if you want to save money on your house's climate control, you're going to end up being warmer/colder than you would like.

> I don't care when the dishwasher runs as long as the dishes are clean by tomorrow, you know?

You can already accomplish 95% of this now with any dishwasher made within the last 20 years that has a delay timer on it. Just load it up and tell it to run in 6 hours and then go to bed.

I don't want some 3rd party company driving a huge team of middlemen sucking up a gigantic pile of data in order to determine when it might be strategically useful for my dishwasher to be on. I don't want my dishwasher on the internet. I barely want it to have any electronics at all, because I want the damn thing to last for 20 years, not the scant 5 years people seem to be getting out of major appliances these days. You wanna talk carbon footprint and recycling, making things reliable would probably save us a million times more energy than would using the internet-of-things to run this stuff at night time.

I _might_ be willing to accept a compromise where my smart power meter uses an open protocol to inform devices in my house of the current energy cost for Time-of-Use billing and then the appliance decides when to start based on a threshold I set, but even that's more implementation than is really necessary here.

Also, there are only a few appliances that can really make use of that kind of thing. As a parent, I need to run laundry all the time, non-stop, because children are filthy monsters. I can't factor energy costs into that, because laundry takes a long time to run and many loads need to be run. It's only a small number of people who can stick their one weekly load into the dryer and tell it to wait for night - and again, a timer would do 90%+ of the work spreading the load around, you don't need a gigantic network of flimsy compute doing the work here.

IIRC - fuzzy old memories, but I think you could use it with OLE to combine a few different embedded OLE sources into one document, e.g. a Microsoft Draw vector piece and a Media Player MID into one component that could be embedded elsewhere.

Very useless. Most Win 3.1 machines with 1-4MB RAM could barely even start MS Draw without churning swap like they were trying to make butter....

Precisely. If oil prices get high enough, the Canadian oil sands will begin to be a profitable operation again, and will scale back up to take advantage. At this point, while we're still weaning off of oil - for probably the next 50 years - we need the lowest carbon, least cannibalistic sources of it possible.

That's mostly because of the lack of midmarket cars.

I bought a used sub-$20k electrical car a few months ago; it can be done. The market needs pressure to create cars in that price range; right now EVs are mostly luxury cars. Tesla is not the car we should be aiming for; what we need are things in the class of the E-Golf, Kia Niro, FFE, Fiat 500E etc. that are the size of a Corolla and cost 20k CAD or less.

Until that's common on the market, the value proposition is flawed and nobody is buying an electric car to save money except for people who buy used cars that someone else has eaten the depreciation on (like me), or people who drive a LOT. Everyone else is basically buying these things as luxury items, and the cost doesn't matter, especially when you can get near-0% financing over 5+ years.....

What was he driving in '89?

I think that a lot of physicists who are now in their 40s+ learned and used Lisp in university, and have simply found ways to keep using it since then.

For new hires, the fact is, a quantum computing codebase is daunting to learn regardless of the language. Therefore, apprenticing someone on a new language at the same time as everything else they are learning is not really the same relative burden that it would be if, say, the application was a plain boring database CRUD app in an unfamiliar language.

We definitely make excellent use of macros and DSLs and the REPL in many ways. I wish I could talk about it more, but alas.

They're not the only QC company using Common Lisp ;-)

I work at D-Wave and it has been a really exciting week for us with these announcements. It's great to see a writeup at this level of detail, we definitely appreciate Ars' coverage.

If anyone has any questions for the D-Wave team, feel free to reply here and I can see what I can do to answer your question myself or get an answer from someone else on our team.

I work on the classical computers around the exotic stuff, so I'm not really qualified to comment. Either way, Aaronson's negative commentary - which is many years out of date, at this point - is not something that anyone pays very much mind to, because at the end of the day one can't allow a mere critic on the sidelines to get in the way of actually producing real machines.

> maximising the chances of producing a viable product

That's the goal. The annealing QPU is a co-processor, like your GPU, like vector processors, or a DSP, etc. It doesn't need to compete with classical compute on the things classical compute is good at; it needs to compete on the things classical compute is bad at, or at the very least, bad at without throwing massive piles of money at it. There is a crossover point for optimization problems where we will be able to show a price/performance advantage over classical compute, which we term Quantum Advantage (vs. the more divisive term of Quantum Supremacy).

The trick at this point is formulating problems in such a way as to be something you can run on our hardware, which still requires a deep mathematical skillset. This is something we're building on... perhaps pay attention to our Qubits conference coming up next week - https://www.qubits.com/ - there should be some interesting announcements!

Sorry, but I did try this, and it sounds absolutely awful. There's a ton of noise. I don't know if my Pis (and I've tried 3 of them) are different from yours, but a cheapo $20 USB sound adapter already has a much better noise floor than the built-in one. This is regardless of what I connect them to.

There's "audio woo" where you spend $1000 for a USB to SPDIF adapter that literally does almost nothing, and spending $50-100 for a decent DAC that doesn't have an audible hiss.

I work on a real, cold-as-space production quantum computer and its cloud service that you can sign up and use for free today - the D-Wave Advantage system, with 5000+ qubits, that you can use via D-Wave Leap.

The difference here is that we produce a quantum annealer, which is useful for optimization problems instead of for database searches + factoring. It's already delivering some real-world value for early applications.

While gate model machines are interesting, D-Wave took the tack of implementing the model of QC most likely to lead to actual useful applications within our lifetimes. Gate-model QC does seem to be very useful, but until it reaches millions of physical qubits it's not going to be producing any results beyond pet laboratory projects, and it remains to be seen if that's even physically possible. In contrast, quantum annealing has been able to grow at a good rate both in terms of qubit count, degree of connectivity between qubits, and also in terms of reaching lower noise and better quality results.

We also have hybrid solvers that combine the state of the art in classical algorithms with QPU sampling to get the lowest energy state possible with a much larger graph than we can do in hardware.

I think it's a very interesting field to follow, there are huge investments being made and real progress is happening on a number of fronts. Our competitors are trying to bring live systems to market, too, but it's harder to see them being much more useful than simulators for the foreseeable future.

Slick, thank you!

Any recommendations for the DAC? I'm using a USB sound card right now on my Volumio Pi but haven't been thrilled about it. What I really want is SPDIF coaxial or optical output, on the cheap.

For anyone else who wants to do this, just set up the absolutely amazing Volumio on your RPi or any other computer.

https://volumio.org/

The web app is great for use in your house, the Spotify plugin is first class - it's supported as a target in your Spotify app, or you can browse straight from the webapp, which means even a guest on your wifi can easily pop on and help pick the music.

> Spotify is trying to be the one that owns it.

If you're interested in pushing back on this, the Podcasting 2.0 community welcomes you. Check out https://podcastindex.org/ for details.

Podcast 2.0 apps have more metadata, chapters, cover art that can change like a slideshow during an episode, etc.