So, you are saying the judicial system should only use evidence depending on how severe the crime is? That is a very poor precedent.
366 karma · joined May 22, 2018
So, you are saying the judicial system should only use evidence depending on how severe the crime is? That is a very poor precedent.
Let's make another scenario: someone's home was burglarized, and the family who lives in the home submits DNA so the police/investigators can flag finger prints/DNA not left by the family. It turns out the father of the house raped and killed several women 30 years before. The DNA submitted by the family has effectively convicted their father by volunteering his DNA for crimes he committed 30 years ago. Would you be equally upset that this rapist and murderer has been caught?
This does not undermine the fact this person was raped, and that is horrible, BUT justice will be carried out in that case. The justice system cannot overlook a reported crime when there is evidence to persecute because a criminal involved in that crime has been hurt.
- Understanding of the passage of time
- Independent activity, without prompts
- Has self understanding of it's current inputs, like a camera
I don't think that's a full list, but to me the transcription really indicates that LaMDA responds to user prompts with reasonably relevant responses, but it gives me no indication that the system has a self awareness.
I think a good test would be to ask it the same non-mathematical question several times in a row, and see if there is any awareness of the same question, or if the answer change. Like if I asked LaMDA it's favorite color, can it name a color, and tell me that answer consistently? Does it know that I've already asked it, and give answers that seem frustrated after a while?
- Every player can use their tablets or phones, and as many as they'd like
- Each device can be assigned with different controls
- Wasn't sure if this would be decided before the game starts or if some player could control it, or if anyone can just add in controls as needed
- The game plays only on the ship bridge, there would be no first-person anything
- One device acts as the viewer screen
Scenarios would be things like escorting a ship through hostile space, delivering cargo, peace treaties, or search and destroy. One thing I wanted is the game to favour fun over realism. Like a player could go rogue and navigate the ship anywhere they wanted, or start firing on a friendly ship. Controls should be easy, like navigation could be as easy as scanning for nearby locations and then picking it, letting the computer plot a course.I liked the idea of players all hanging out in a living room, connect the "viewer" player to a TV, and just have fun. Some scenarios and situations would involve teamwork, like having an engineer reroute power from navigation (ie nav has to slow down to make power available) to weapons for more powerful shots.
I thought it would be fun to build, but I just didn't have the time to develop it myself.
It's strange to read that a secular news site like this refers to biology as "design" rather than ask why their biology evolved that way.
- How to be decent human beings to each other, and especially how to not kill each other
- The media has too much power
- Government over-reach
Probably other things, but I don't know, these seem like way better hot-button issues.
I only brought up trusted vendors because it seemed that you also want repositories of customization scripts, and I don't disagree, but it shifts the trust from the people making your operating system to you reviewing a bunch of scripts to get the system you want.
I'm making a point that people don't want that at all. That will absolutely cause analysis paralysis, there would be too many possibilities, and it would continue this idea that Linux is hard and elitist when it really shouldn't have to be that way.
I'd say in terms of consistent UI, OSX probably is the biggest winner. Both Linux and Windows are the losers in UI consistency. Yeah, you can customize it, but customization, in my opinion, can be a real time-suck and it doesn't make you any more or less productive, it's just visual enhancement.
Unix is the real commandline winner, because piping and good underlying tools, and as a result, OSX and Linux do very well here. I'm not familiar with the powershell equivalent, but at the very least, it's just a copy of how unix works.
Application installs is a very mixed bag across the board. Every distro of linux has their flavor of installs in commandline, and some have graphical. I've never had a lot of success mixing the two, it always puts some package database in a corrupted out-of-sync state. OSX is primarily graphical installs, but there are cli tools like homebrew - neither seem to fight each other, though. Windows is download/installer based. There is a store, but the store is mostly paid applications, and the Windows ecosystem has a large amount of free installer tools that will almost certainly never be on the store.
The article doesn't even talk about hardware. My typical experience is hardware "just works" in OSX and Windows, and will often/mostly "just work" in Linux, but also there can is usually is some tinkering (as well as do I want the free/open source drivers vs the binary blobs in cases like graphics cards).
So I guess in summary, Windows doesn't suck, in the same way that Linux and OSX don't suck. Every OS has a mess in some capacity, and I would say Linux's biggest mess is there is too much room for customization that will always lead to rabbit holes. For context, I use a company-given OSX laptop for work, and use Linux for my personal computer/daily driver, but have a Windows partition for gaming.
- Take breaks between tasks. Your flow should mostly allow 5-10 minute breaks often (like once an hour or so) without anything or anybody blowing up.
- It's healthy that your client (or bosses) will always want more work, especially if you and your team do good work.
- Always instill in your client (or bosses) that work will get done when it's done. They are paying you to know how long something _should_ take, and to identify obstacles that could make something take longer, as well as the time you need to plan/prepare to keep you and your team sane.
- Your best working state is when you're relaxed. People have told me they work best under pressure, and I really refuse this to ever be true. You can certainly work under pressure, and even work fast, but I honestly believe your best work is the work that you enjoy and care for, not just the work you spit out fast.
- You can have fun doing just about any task if it's not stressful. Sure, there are a lot boring tasks in software, like waiting for compiles, or doing slow-moving dev-ops work, but if there is a clear path with no major obstacles, it really won't be that bad.
The biggest thing for me is having me and my team take breaks often. It's a good time to reflect on the work that's been done, think about a problem, or just take your mind off a larger problem for a few minutes. Depending on the industry, your client, your boss, etc., your mileage is going to vary, but as a software consultant, it is your job to advise on what time you need to get something done right.
So I guess I agree with the article. Not that you shouldn't regularly give 100%, but you should always be buffered so in high stress, high stakes, you and your team have capacity to go above and beyond the usual, but your average work load isn't maxing you out all the time.
The short/simple answer is carriers don't care, because they make money when a call is placed on their network. There is also a difference between what is a valid number (digits are correct) vs a real number (someone owns a number). It is cheap for a carrier to check validity, but not "realness" - to check a real number, a carrier may have to do some sort of data request to any number of carriers to determine if the number is owned.