1000 contractors in India working on the Just Walk Out technology (Amazon) serve as a vivid reminder that AI is not always what it seems.
We are waiting to see what will happen with Apple Intelligence

1000 contractors in India working on the Just Walk Out technology (Amazon) serve as a vivid reminder that AI is not always what it seems.
We are waiting to see what will happen with Apple Intelligence

Today, let’s talk about the very interesting Norwegian artist Ingebjørg Frøydis Støyva (Ingebjørg Frøydis Støyva, 1978). The first painting with a horse in the closet is called ‘Can I keep him?’. Some people have skeletons in their closets, while others have a whole horse. I really love this artist. Generally, artists who embrace their quirks and create from that place are always fascinating.












from the series “unexpected photographs”
Quentin Tarantino at Pasternak’s grave in Peredelkino, 2004

In all this buzz about AI integrated into operating systems, what really doesn’t concern me is privacy. Rather, it’s the fact that overengineered software begins to devour hardware faster than the hardware can evolve, and eventually, I start contemplating a switch back to Linux, where things are much more transparent.
Just look at this. My Mac’s advertised battery life is 21 hours. In other words, you turn on your laptop at 8 AM, start streaming something from YouTube, and the battery should only run out by 5 AM the next day.
But in reality, that’s not what happens. Indeed, it does last significantly longer than any other laptops I’ve had before, but sometimes the battery drains in just a few hours. Why? That’s unclear.
Why? Because the OS, for example, might find an unindexed unpacked archive, and the corespotlightd process kicks off to index it. This process can’t be stopped—you can only turn it off forever, but then the search function won’t work. It’s possible to exclude indexing in Documents (which I’ve already done). But then another process wakes up due to some signal or schedule, and it too starts consuming the battery or CPU.
Still, it’s fair to say that this doesn’t really cause any major issues. Things run, they heat up the air, might be useless, but specifically the M3 Max never lags.
For instance, among the processes is the Apple Neural Engine Daemon (aned). It periodically wakes up and consumes resources. With new functionalities, such a process will awaken more frequently and use up more resources. Or something like com.apple.NRD.UpdateBrainService decides it needs to update some neural networks. And the more software you install on the computer, the more such instances you’ll encounter. Just Intellij Idea alone drains my battery and processor faster than anything else. I’ve made it a rule—when on battery, shut down Idea.
Ideally, having AI on a device should indeed heat the chip and drain the battery more actively, and most likely “just in case,” since not all users need all these AI features. I suspect that Apple will employ a trick: measuring battery performance without a configured iCloud and Apple Intelligence account, and we’ll see all those 21 hours of autonomous work. But as soon as the computer switches to working mode, it will need charging more often, and the office will be slightly warmer.
Such a great interview! Andrey is an amazing guy, very proud to know him and hoping it counts as friendship since around 2005 or so Andrey Anischenko
P.S. And, oh, when you decide to conquer North America, head north, ideally along the East Coast and preferably through Washington!
Everyone else – tune in, watch, Andrey really shares some interesting insights about the journey and the edtech market.
https://youtu.be/XWZ8f9RxUmw?si=bmnkUbSQzYISxnDT
An intriguing artist, Terry Miura (born in 1964), originated from Japan and relocated to New York, where he pursued his education and furthered his artistic journey.
As I gaze upon his paintings, I am reminded of Richard Schmid’s assertion in his current book that nature contains no definitive lines. Thus, when translating visuals onto canvas, line-oriented thinking may prove counterproductive, even obstructive. It’s more useful to consider swathes of color. Another compelling concept he mentioned is prioritizing the observed over the known; should a conflict arise, depict what you see, for painting the known risks portraying an absence instead of presence—a fascinating notion indeed.
This perspective harks back to the ethos of earlier painters (and some of their modern counterparts): viewing objects more as people and items within a context than as visual constructs shaped by light necessitates painting based on known attributes—what other references could they rely on? Understandably, the more these early artists knew about their subjects, the more adept their renderings. Thus, depicting what was expected involved extensive study and memorization due to the sheer expanse of subjects. Naturally, this led to specialization: one might focus on animals, another on architecture, and a third on portraiture.
Even today, such artists (including many contemporaries) are often easily recognized as much for their thematic focus as for their stylistic and period distinctions.
I find this methodology compelling. It’s not novel, and Richard Schmid articulates it effectively. This becomes particularly evident when attempting to paint a modern car seen from a few meters away. It appears straightforward: four wheels, some curves. Yet, each car’s curves are unique, lacking clear rationale, and knowing the ‘anatomy’ of a Toyota RAV4 offers little aid in depicting, for example, a Chevrolet Corvette. Furthermore, proximity amplifies these curves’ perceptual distortions, potentially overwhelming the mind. It proves simpler to bypass sketching and commence with color patches and shapes, refining them subsequently.
Clearly, Terry Miura shares this perceptual framework. For him, there are no figures clad in red or white dresses; there are only patches. Compositionally effective and balanced patches are all that need to be accurately placed—and misplaced where not needed.
Note that related posts are consolidated under the hashtag #artrauflikes, and the “Art Rauf Likes” section on beinginamerica.com lists all 79 (unlike Facebook, which omits nearly half).











Baltimore. 7 eleven. No more than two children/teenagers allowed. Otherwise, measures will be taken.

Another self-taught artist from Britain/France, Julian Merrow Smith, is known for his still life paintings. Most of his works are still lifes, but each one is a masterpiece. His use of color, composition, realism, and simplicity is exceptional. However, the realism is not achieved by meticulously placing fifty “pixels” per square millimeter. Upon closer inspection, all his works are relatively quick. And yes, after viewing his paintings, you will likely crave something juicy, sweet, ripe, and vibrant.
He has a website called Postcard from Province (shiftinglight) — just Google it. Currently, there are 3697 paintings posted there, one for each day since 2005. Remember I wrote about Duane Keiser? He should have patented “a painting a day.” This is exactly how one should hone their skills.
I remind you that similar posts are grouped under the hashtag #artrauflikes, and on beinginamerica.com, in the “Art Rauf Likes” section, you can find all 78 posts (unlike Facebook, which forgets (ignores) almost half of them).












If you’re planning to buy a power bank, here’s a lifehack to get a better one for the same money:
Firstly, pay attention to the type of battery – Li-ion or polymer Li-ion (LiPo). The latter have a higher energy density (yielding more charge for the same weight), and they are safer.
Secondly, look at the ratio between weight and stated capacity. For example, mine weighs 436 grams and is labeled 40000 mAh at 3.7V. Convert this into watt-hours by multiplying 40000 mAh by 3.7, which gives us 148 Wh, resulting in 148/0.436 = 339 Wh/kg.
The thing is, such a density in batteries does not exist. For LiPo, the range is 150-250 Wh/kg. And 250 Wh/kg is for the most advanced, expensive types.
For a battery weighing 0.436 kg and a voltage of 3.7 V, the capacity range would be approximately from 18000 mAh to 30000 mAh with an energy density of 150-250 Wh/kg. It’s more likely between 20000 mAh to 25000 mAh. Which is quite good, but definitely not the 40000 mAh as listed.
In other words, take the device’s weight in grams, multiply it by a number between 40 and 65, and you get a very probable real capacity in mAh. I would use 45 for certainty. But it’s very likely that you should use 40 if you’re also trying to get it for the lowest price.
Next, look at the charging time. On mine, it’s written that with a 30W power source, it should take 6 hours to charge from 0 to 100%. Typically, it’s about 9 volts (common for 30W), though it’s not specified. So, the average charging current would be 3.33A (30/9). The battery capacity can be calculated by multiplying the current (3.33) by the time in hours (6). That gives us 3.33*6=19.98A*h=19980 mAh. This is another hint that the battery is nowhere near 40000 mAh, but rather around 20000 mAh.
Is 20000 mAh a lot? The instructions, and on the battery itself, state that for the USB-C port, the charging current is 3.1A at 5V, 2.22A at 9V, 1.66A at 12V, and the same for the Lightning port at 5V. If all this holds true, a full charge from 0% to 100% of my iPhone 15 Pro Max with a 4400 mAh battery should take a maximum of 2 hours. You could charge 4.5 phones like mine with this power bank, or it should nearly fully charge a laptop (70Wh) to 100% and then be depleted. So at first glance, even 20000 mAh is not bad.

They told me “Better practice… on cats”. Done! I don’t know if I should finish up or get a new cat
