Brian Cugelman’s "12 Emotional Journeys of Color Psychology" breaks down how color influences user perception and behavior through a structured, neuroscience-informed framework. Rather than textbook color meanings, it defines 12 emotional journeys people take in digital experiences and shows how color can support those transitions in context. It’s a practical lens for product and UX folks who want to move beyond ‘blue = trust’ and tie color choices to real emotional motivators and reinforcers.
As a non-American, I wonder how effective these measures are. Way too often I hear that "ICE abducted ... an American citizen". Are these individual cases amplified by the media to attack federal government's reputation or are these a significant percentage of all arrests? In which case, why is ICE doing it, they have absolutely no way of turning such arrests into a positive for themselves? How easy is it for an American or a legal alien to produce documentation proving their legal status - can it be that the fact(?) that many legal residents don't possess the necessary documentation at all times leads to the majority of such arrests/detentions/"abductions".
I know that many here will question ICE's mandate to act even against immigrants who don't have their paperwork in order, but this is another question.
Thoughtful observers recognize that the real issue is the lack of accountability transparent to the American public. When regular people mess up at their job, there’s a proportionate response to that mistake. When ICE / CBP murders, rapes, disappears, tortures, silences, intimidates, etc. etc. there are no consequences. There is no accountability. In fact, there is encouragement! “You’re doing a public service!” they’re told.
This is by design. You’re asking the wrong questions out of some imagined “higher tier of thinking” and it’s only beneficial to those in power.
Why would any of this be effective? They drastically reduced the duration of training ICE agents and drastically increased numbers. You should assume they are all incompetent.
It's rebounding on them, support for ICE is now negative
They're going after the weak and easy targets and dont care about citizenship as Miller has set them daily targets
This is just the standard precision/ recall tradeoff with recall being prioritised at the extreme (which is usually the wrong way to go for legal field).
A standard situation of 1% positive rate and even if your diagnosis method having 90% accuracy (for both sensitivity and specificity), more than 90% of the positive detection would be false positive.
Net migration to the US is negative for the first time in 50 years [1] so seems way more effective than whatever was done before. As for arresting citizens - federal agents arrest ~100K people per year[2], if the often cited in these threads research to be believed then most of them are citizens as illegal migrants don't do nearly as much crime. Being a citizen does not make one immune to the federal law enforcement, even the Department of Education has armed agents, which arrest citizens, even if they don't go to school [3].
This is like complaining that your screwdriver is bad at measuring weight.
If you really need an answer and you really need the LLM to give it to you, then ask it to write a (Python?) script to do the calculation you need, execute it, and give you the answer.
That's a problem that is at least possible for the LLM to perceive and learn through training, while counting letters is much more like asking a colour blind person to count flowers by colour.
Why is it not possible? Do you think it's impossible to count? Do you think it's imposing to learn the letters in each token? Do you think combining the two is not possible.
The main point they are making is that open source software is creating public good/value and needs to be recognized as a civic service. The problem I have with it is that in the majority of cases it doesn't create any value at all, there are plenty of open source projects that are not illegal but net harmful for public interest. Knowing how Germans usually distinguish good from evil (paperwork/bureaucracy/official review and certifications), this will likely end up with yet another verse of Vogon poetry one would need to engage with to get a tax break. The recipients of benefits will likely be those who will go for the trouble of proving that they are "good open source". It is highly likely that actual net benefactors of society will not condescend to do this paperwork and. end up not getting the proposed benefits.
I've been thinking about similar setups with lots of batteries as I have excess energy I can generate during the day and a lot of otherwise barely usable space in my attic. Fire hazard is my major concern. How do you make sure old (or even new) batteries don't do their usual thing, especially if they are in an environment with unstable temperatures and scarce monitoring?
Old laptop cells are going to need a lot more babysitting than a new LiFePO4 rackmount battery. Those generally do pretty well with the internal monitoring that they bring. Semiconductor fuse as overcurrent protection, passive balancing and monitoring of individual cell voltages seems to get you pretty far. Get your 48V or 51.2V battery in 5 kWh, 10 kWh or 15 kWh cell size from one of the Chinese outlets in dubious quality, or from real brand names like Pylontech. RS485 is usually present and there are kludgy home-assistant gateways, but they're not required for safety.
External monitoring is made by Victron, who initially did electrical solutions for boats. Their inverters are also very popular and pretty great.
Also, brick walls are kinda nice I have to admit, along with an exhaust for any fumes.
yep, I have brick walls downstairs where space is more scarce, the attic has lots of space but also lots of wood lying around.
I guess another possibility is to have the battery pack outside, attached to the brick wall, and ideally away of direct sunlight... but then no wooden roofs. It looks like no DYI is really feasible except maybe a concrete bunker as others have suggested. And even then it's probably uninsurable which is considered a no-no where I live.
This is my main problem with having a battery in the house. Once sodium based batteries and compatible inverters get on the market I might switch them out to lower the risk of catastrophic failure and losing the whole house.
Almost exactly the same story here. 6-8+ cups a day, cold turkey with no replacements. Two weeks of misery and headaches and a beautiful new beginning. I drink one decaf or mate a day from time to time now. And I drool like a bloody St. Bernard when I pass by a coffee stand where they serve the real stuff. So I am still a coffee junkie, my body is no longer addicted but my mind is. But it was definitely worth it to quit.
To be fair, Pumped Storage is really dependent on Geography. It has downstream affects on environment due to having reservoirs where none have been and all.
It's powerful, cheap and long lasting, but not easily available.
Well, this thing also depends on geography. There aren't many places where such a construct would be easy to build: seismic situation, static stability for these enormous weights, winds (it's actually quite funny to see cranes side by side with wind turbines in their commercial)
Have you looked at Lektor? https://www.getlektor.com/ . We use it for in-house needs and we have also built an extension, Lektorium https://github.com/sphericalpm/lektorium that allows us to create hosted instances of Lektor for content managers who aren't technical enough to run Lektor locally.
After ~30 minutes of "connecting..." I restarted Telegram (Linux app), won't log me in, says "Too many tries, please try again later". Hope they'll reset try counters when they are back online.
I appreciate Telegram's latest features and understand that stuff happens when you change things... hope they'll fix it soon.
