fbdab103

Yet it is a captive market. If I am in a building, I only have access to a singular type of elevator. Why not always give the smoothest ride possible unless it is $0.12 cheaper for the installer, so everyone has to suffer forever.

Joke is on me, because while I have a disconnected TV, the Roku I use for streaming likely tattles on everything I watch.

  After that it removes (or replaces) Unicode punctuation, performs a SHA1 hashing, and uses the first 8 bytes for deduplication comparisons (paragraph level)

Is taking the first few bytes that much faster than comparing the entire hash? Or something else? That is one of those performance optimizations I would go back and forth on endlessly wondering if I lost something by trying to shave off a few cycles.

"You should really be ashamed of yourself for doing this" is an inappropriate response for basically anything but kicking puppies.

Super frustrating. From the mailing thread:

  However, this is not good as it remove under-powering range too far. I was getting only about 7% less performance but 90W(!) less consumption when set to my 115W before. Also I wonder if we as a OS of options and freedom have to stick to such very high reference for min values without ability to override them through some sys ctrls. Commit was done by amd guy and I wonder if because of maybe this post that I made few months ago(business strategy?)

I have certainly crossed a threshold where hardware seems good enough. I do not need the 99th percentile performance if it comes with a non-linear increase in power consumption.

  As one might expect, these tapes eventually began to take up space — a lot of it. By investing in Apple stock, Stokes saved enough money to buy as many as nine additional apartments, which she used as storage units to hold all of her tapes. 

Why apartments vs storage units? More secure/humidity/pest controlled? Also, starting in 1979, I assume the raw price of the tapes was exorbitant.

Any notable highlights for a consumer of Numpy who rarely interfaces directly with it? Most of my work is pandas+scipy, with occasionally dropping into the specific numpy algorithm when required.

I am much more of an "upgrade when there is a X.1" release kind of guy, so my hat off to those who will bravely be testing the version on my behalf.

At least for the lower power draw models, can you disable the fan entirely? Or cut the RPMs in half?

Time is money. You might be a lab with long queues to train, leaving expensive staff twiddling their thumbs.

Utility solar is not being installed on expensive land.

Even if you did review it, a motivated attacker is not going to have an exfiltrate_user_data(). The xz backdoor exploit was incredibly sophisticated, and one key of the design was sneaking a "." into a single line of a build test script.

A cursory audit of primary dependencies has almost zero chance of catching anything but a brazen exploit.

I think it is worth separating the Python ML ecosystem from the rest. While traditional Python environment management has many sore points, it is usually not terrible (though there are many gotchas still-to-this-day-problems that should have been corrected long ago).

The ML system is a whole another stack of problems. The elephant in the room is Nvidia who is not known for playing well with others. Aside from that, the state of the art in ML is churning rapidly as new improvements are identified.

There have been many subsequent studies which generate evidence in support of the underlying idea.

As they say, it only takes one result to disprove a theory, but nobody has been able to do so to date.

Part of the problem is that studying the brain is hard. We have lots of in vitro or mouse models that tell a story, but nothing is conclusive. There have been many orthogonal studies which can replicate some of the associations, but nothing that can precisely point out underlying cause and effect. Yet people are suffering today, so researchers are following the best leads we have.

I think the fundamental problem is that if you want to talk to databases, you have to speak SQL. Like Javascript on the web, you are stuck with what the platform provides. Both languages have significant deficiencies, yet is/was the only game in town.

Sure, there are some languages which can compile down to SQL, but like Typescript, every once and a while, you find some edge case where the transpilation fails you and you might as well be an expert in SQL.

I love all of the ideas of PRQL, but I do not know if I would be painting myself into a corner adopting something that will go the way of CoffeeScript.

Limited by VRAM is a huge constraint for me. Even if it is slower, being able to load 100GB+ into RAM without any batching headaches is worth a lot.

Unless cudf has implemented some clever dask+cudf kind of situation which can intelligently push data in/out of GPU as required?

The aspect of Ikea furniture I like - it is typically stylish. To get an equivalent look somewhere else usually invokes large price increases for seemingly no change in quality.

It is definitely not the best quality, but I have moved several Ikea pieces over the years without much ill effects. Would I like to upgrade to some "adult" furniture one day? Maybe, but I fear much of it is like other designer goods - identical crap construction with a reputable label stitched on the front.

Absolutely not would I give Apple a free pass either. They can say all the nice things they want about protecting my privacy, but I do not trust any commercial entity will act in my best interest. Especially when they all have government requirements to hand over my data when a cop asks nicely.

We are speed running into a neuromancer dystopia where tech companies control every facet of our lives. Why would I be ok with them making it easier to monitor my every keystroke?

Supposedly chess layers burn excessive calories playing. From one popular culture article[0]

  ...The six-game championship in 2004 left Rustam Kasimdzhanov 17 pounds lighter. And in 2018, one company tracking the heart rate of grandmaster Mikhail Antipov concluded that he had burned 560 calories in just two hours. Sitting still. Playing chess. (An average-sized person would need two hours on a treadmill to break 500.)
  Last week, ESPN looked more closely into the toll the stationary sport takes on the body. ESPN spoke with Robert Sapolsky, a Stanford University researcher who studies primates. Sapolsky explained that some chess players respond to the game like any elite athlete, burning upwards of 6,000 calories a day during tournaments, due to tripled breathing rates, elevated blood pressure, and muscle contractions. That means during tournaments they can lose two pounds each day.

I have done several fasts, and I am quite dubious of the supposed mental improvements. My brain has spent 99%+ of its existence no more than 12 hours away from a steady supply of glucose. I personally feel physically and mentally sluggish. While probably erring much on the side of caution, I avoid driving during a 5+ day fast because I feel my reaction speed is diminished.

Furthermore, the brain is the single most energetically expensive organ in the body. If there were a shortcut metabolic pathway, it seems like it would have been exploited long ago.

[0] https://www.menshealth.com/fitness/a29144951/chess-players-c...

Management seems determined to mandate RTO. I am not seeing a flood of alternative perks being offered with suddenly being forced to waste 5+ hours of my life each week commuting.

I really enjoyed how he was struggling to balance the laser with a lightbulb. That seems like one of the worst household objects available for the task.

Article really would have helped if I could have understood how much cheaper/faster it is to make one piece instead of three.

At some point, you would think maybe he would quit the sport of his own accord?

Were I to break a single digit number of bones in a foot or hand, I am done.

I know nothing of PHP, but I am sold based on just the mascot and the tag lines.

Given the damage is already done, why not leave it in place? Encase it in concrete and come back in a decade or two when the worst products have decayed?

Cautionary tale? Some excerpts from Toyota's press page[0]

  - Toyota is projected to be the number one retail brand for the 12th consecutive year
  - Electrified vehicle sales – battery, hybrids, plug-ins and fuel cells – surge 30 percent and make up nearly one-third of total sales volume
  - Projected to be the number one seller of passenger vehicles for the 12th consecutive year, and 20 of the last 21 years
  - 2023 electrified vehicles sales of 565,800 represent 29.3 percent of total sales volume
  - Corolla projected to be the number one compact car in America
  - RAV4 best-selling SUV in the U.S. for 7th consecutive year

You can also just look at Toyota stock killing it vs SPY[1] over the past year

[0] https://pressroom.toyota.com/toyota-motor-north-america-repo... [1] https://www.google.com/finance/quote/TM:NYSE?window=1Y&compa...

I get the trick, I am just disputing that you have any deniability if caught.

As to the text dumping -I have no idea. I could believe some of the HR screening tools select all text and dump into some structured fields for review. Probably not, and the trick will go completely undetected.

I must be missing something - what is the plausible deniability? If someone does a raw text dump of a pdf and gets the hundred new keywords, surely someone would recognize what you did?

Or do you just mean, it will silently pass the white text test?

Could be the read that the article had essentially zero details on what the technology promised.

Is there any kind of grid-scale seasonal storage technology on the horizon? Batteries seem right out from the gate unless the cost drops to 1/10 of current. The only practical one I see is synthesizing a fuel (methane, hydrogen, whatever) which can be stored ~forever.