edub

I think 404 Media has an ethical obligation to provide Proton Mail’s response outside the article’s paywall. The word “Helped” in the headline is more sensational than stating that Proton “was required by Swiss law to provide...”

For readers who do not want to pay to read the article, the headline leaves incomplete context and creates a misleading impression of the story. That damages Proton’s reputation, and the missing context is only available if someone pays for the article, reaches out to Proton, or searches forums for substantive information.

I was able to have AI generate an image that made this, but not by diffusion/autoregressive but by having it write Python code to create the image.

ChatGPT made a nice looking clock with matplotlib that had some bugs that it had to fix (hours were counter-clockwise). Gemini made correct code one-shot, it used Pillow instead of matplotlib, but it didn't look as nice.

Using LLM to control your machine has amazing potential for accessibility.

This seems like a strong candidate for a Constitutional amendment. Twelve amendments were ratified in the 20th century—about once a decade since the end of the Civil War. However, if you exclude the 27th Amendment[1], we haven't ratified an amendment in 53 years. My favorite type of amendment is one that extends rights to people, and in this case, also to their property.

[1] The 27th Amendment took a different path compared to the other 16 amendments ratified since the Bill of Rights. It was originally proposed as part of the first 12 amendments but took 202 years to be ratified. This was largely due to the efforts of a University of Texas student in the 1980s, who, motivated by a C grade on a paper, embarked on a mission to see it finally adopted.

I can't load the article, so this is likely off topic, but the memory came rushing back when I read this headline.

I attended a wedding that got hit by a major flash flood that required a rescue operation by boat and helicopter. Thankfully no one was seriously injured. A half dozen people were swept away and rescued from trees. The rest of us got to higher floors of the building and they were concerned about us waiting out the flood because cars from the parking lot floated and rammed the first floor of the building. It was featured on an episode of I Do, Redo.

They brought in busses to transport the 61 people rescued to the local high school where they had activated the Red Cross and provided us dry cloths and food. I can't say enough nice things about the Red Cross volunteers and the staff at the high school, and same with the fire department and EMS.

However, when we were loaded into the busses, the police held the busses until they had a drug dog come into the busses to walk up and down the aisles, and only after that let the busses take us to the high school. While no one was seriously injured physically, people were traumatized from the flooding event and many attendees suffered from PTSD for years.

It was so cruel for the police department to do what they did with everyone in the mental state that we were in.

For reasons I do not understand, we were not free to leave the high school, and even the people that did not lose their cars to the flood were required to go to the high school on the busses. I can't recall how long they detained us at the high school, I'd guess 3 to 5 hours, and then they let us leave. The friend that came to pick me up (I did lose my vehicle to the flood) got to the high school not long after I did and had to wait in the parking lot for hours.

We wanted to feel safe once we got to dry land after the ordeal we went through. I did not feel safe until I got home.

I've not worked with the Adafruit PN532, but for an extra $10 you can get a Pepper C1 USB from Eccel which is very easy to work with. It is a stand-alone device, so you don't have to connect it to anything but power. Has WiFi & BT built in and has a built-in web server to configure it with, you can have it make calls via REST, MQTT, WebSocket.

Is it the same? On the Models page of the API docs it says that GPT-4 is using the June 13th which would be different than the March 23rd.

Check out Fossil SCM, from the creator of SQLite.

Many great anecdotes in this thread. Many regarding lack of access to a computer while designing code.

Mine is as a child wanting to try to work on the computer every moment I was awake, but parents insisting that it is important for me to got to bed so I can go to school the next day. My solution to continue to working on the computer when I was expected to be in my bedroom was to keep writing code from my room but on paper.

I'd often stay awake with pen and paper until 4am, and maybe sleep during class the next day. But without the computer to give you the feedback of compiling your code and being able to test it, it forced my mind to try to reason the code I wrote on paper to think through if it would work as I expected.

So maybe 15 minutes writing on paper what I thought the code should be, and 45 minutes reasoning with myself if what I wrote would work the next day when I typed it into the computer. It kinda trained me to compile code in my head, and it was a practice of mine for years throughout middle school and high school.

I'm not sure I would have had the same understanding today as I do now if I had access to the computer throughout the night and could keep doing trial and error with the computer to get my code to work. Even though I was working with BASIC at the time, I think it shaped how I think about working with the languages I use today.

I'm 5 days late to your disagreement, but regardless: I think the choice of word "dishonest" implies that CGs post was fraudulent in nature, and was a poor word choice unless RR has some reason to believe that CG knows the truth to be different from was you stated. Disingenuous might be closer to what RR meant if RR thought that CG knew that there is an imbalance but chose to focus on treating the situation as both parties have equal weight in the relationship.

I think that the original comment that CG replied to was making an important point about the balance of power between the two parties.

An employee working for an employer vs a customer of that same employer is not equivalent.

You shouldn't compare a person working for an employer with the hopes that they will receive a future paycheck for the work they are doing today (a paycheck that they likely are relying on to make good on commitments that they have) with a customer that receives a service today with the promise to pay it in the future (even though the service provider is also relying on that future payment to make good on commitments that they have made).

There is almost always an imbalance of power between an employee and an employer.

An employee might be doing identical work that a contractor could do, but as a contractor you might expect to be paid up front in partial or full, which an employee would not be able to expect.

It is more complicated with a customer and a service provider in terms of the balance of power. If you only have one option for who provides heat to your house that your children live in, then the provider has more power than the customer. If you like to drink at a bar when you have extra cash then the customer has more autonomy in that relationship than the bar.

One can come up with examples where the employee has more power over the employer, but it is more rare and not what the article in question is discussing.

The program was written in 1988. I ran the text through LZSS which was published in 1982, so was available before 1988. I used a 1989 public domain version by Haruhiko Okumura, which is after 1988, but I don't believe it is optimized to improve upon the compression level of the 1982 algorithm.

It took it from 2357 bytes to 534 bytes, which is smaller than the Xmas.c program which I counted as 917 bytes, but another poster counted 913 bytes.

The article says that you win points but there is no way to win money back or get payouts. What is legal is not the same thing as what is moral (and vice versa).

No illegal alternative necessary, the players of this app get zero odds and payouts.

gherkinnn is saying that just because it is legal does not mean that the people behind the company are free of responsibility for the damage their app creates.

I might be misunderstanding what you are saying, but if there are less than 600k homeless people in America and over 15 million vacant homes, it doesn't seem to be "caused by the physical reality of there being too few homes for the number of people".

I wasn't familiar with Kalman, and I tried to understand it by reading the primers on this website and it wasn't clicking for me. Last night I was thinking about the sample graphs, and it occurred to me that it just looked like an Exponential Moving Average, which is a much simpler concept to grasp.

I just searched Google for a comparison and found that EMA is as good as Kalman for "random walk plus noise" on a Stats Stack Exchange, and that a 2003 paper from Brown University (Joseph J. LaViola) showed that a Double Exponential Smoothing algorithm is of equal quality to Kalman (and extended Kalman) but 135 times faster, and a simpler approach.

I find Double Exponential Smoothing to be much easier to understand than Kalman, and assuming the LaViola paper is correct, I'm not going to put additional effort into understanding Kalman.

Counterfeits: In August 2014, I purchased from Amazon, shipped from Amazon (just checked my order history to verify), an Apple MagSafe Power Adapter. It was clearly counterfeit.

I contacted Amazon and told them it was fake, without hesitation they told me to throw it away and they were sending me a replacement.

It would be hard for me to believe that they didn't know exactly which company shipped them that product.

It would be easy for me to believe that they didn't notify every other customer that received that product from the same shipment.

I did gaming math in Vegas for what I believe to be the biggest gaming math consultancy in Nevada (at least at the time). The games were divided between Table, Slot, and Video Poker. I did table games, because I find slot and video poker math to be boring.

Each assignment to me I was provided the rules of the game and the payout table. It was my job to determine the optimal way to play the game and calculate the house edge if you played with the optimal strategy. When I turned in my assignment I only provided the exact house edge. The owner of the business worked independently to determine the house edge as well, and if our numbers matched then he was confident that it was correct. If we had different numbers then we had to dig in to see who made a mistake.

I found table games fun to work on because of the variety of games people came up with, and it was mostly fun to determine the optimal strategy.

The owner of the company was very ethical. He attempted to convince clients to not spend money on introducing new table games. He would tell them that there was about a 1 in 1,000 chance that they would be able to convince a casino to let them do a 3-month trial run, and another 1 in 1,000 chance that the game would outperform blackjack in terms of dollars earned per square foot per month (not necessarily because it isn't a good game, but just because people love blackjack and are more wary of a game they aren't familiar with (it is intimidating to walk up to a game you don't know how to play, especially if you can't practice online first, and you don't want to feel dumb in front of strangers)).

Slots do fine in terms of making the casino money and plenty of room for variety. They have a high house edge. Video poker has a lower house edge, but a new variant of video poker is just adding the game to the menu of a video poker machine already in the casino. There is also less risk for someone that hasn't played a slot or video poker game because it is just them and the machine and no one to feel embarrassed watching you figure out how to play. So he didn't need to warn people that came to him with those types of games.

For new table games the gaming commission requires a trial period to verify that the drop (the profit the casino made) was statistically consistent with the calculated house edge.

I wouldn't have wanted to do it full time, I did about one game a month on a contract basis. I was going to work on a game a month no matter what to satisfy my curiosity and need to do math (finding a new game in a casino). I found a new game with a mistake that gave the player and edge, shared it with the guy that ran the company and he hired me when he confirmed I was correct.

The exception to his warning about table games was adding side bets to existing table games. That doesn't take up any additional square feet, and the house edge is much higher on all of the side bets, and very easy for the player to understand how to participate in the side bet.

As far as I can recall the only new table game introduced in the last few decades that succeeded was Three Card Poker, and the other poker-variants that spawned from it.

If you want to get into this, you are probably better off working for a company like IGT or Arsitocrat that makes their own games and is constantly trying to come up with new ideas and testing them.

  SEEKING WORK
  Austin, TX
  Remote preferred

Full Stack/DevOps. Uniquely qualified in architecting new products/solutions across the full technology spectrum. Design and build MVP. Lots of experience consulting with businesses to improve internal processes by applying technology solutions: interviewing stakeholders and users, mapping out current process and pain points, design and architect an improved process utilizing technology, researching existing products to implement the improved process or building (or overseeing) a custom solution.

  Location: Austin, TX
  Remote: Yes
  Willing to relocate: Yes
  Technologies: Rust, Elixir, JavaScript, PHP, SQL, NoSQL; Linux
  Résumé/CV: on request
  Email: eric@cinque.xyz

You are a faster reader than me. The October update has around 4,700 words plus plenty of images to digest. I understand you likely mean to review the highlights, but a non-trivial amount of information is in each months update. And if you haven't been keeping up for the past 6 years, it isn't easy to catch up now. Not to mention the breadth of the marketplace.

Not yet, but it is on my todo list. There are multiple projects/companies/groups that I am the bus factor of 1 for any of the technical parts (hosting, code repository, etc). This is a real concern to ensure they continue on without me, and hit me hard one night when I had to be rescued from a flood.

I am a paying customer of Bitwarden, so that's the easiest path for me, but I like complicated things.

My plan is to use Shamir's Secret Sharing. Specifically I was thinking of using Klaus Post's Reed-Solomon (golang) which is a port of Backblazes JavaReedSolomon. One could perform an All-or-nothing Transformation first depending on the security level needed.

The primary advantage of this compared to Emergency Access with Bitwarden is that it isn't reliant on a single person surviving me. I would give my wife the emergency access, but if we became incapacitated at the same time (almost happened in the flood), then other trusted people can come together to assemble the keys to unlock the data.

Additionally I can give different people different weights. Perhaps my wife and my mom have enough keys by themselves to unlock, or maybe just a couple or a few keys short. Whereas my trusted friends have enough keys that would require X amount of them to agree to unlock my vault, and people that have an incentive to kill me have the least amount of keys :)

I would likely just store my password to my Bitwarden account, my email account, and my note-taking application. That way I don't need to update it except when I need to change the password. Which is also how I could revoke someone from holding a key, change my password and re-run RS and redistribute keys. Realistically if you gain access to my Bitwarden then you have the keys to the other places, but not necessarily the ability to pass a 2 Factor Authentication, so I could include recovery codes for 2FA.

There is no reason I couldn't have multiple vaults for different things with different levels of keys needed to open, so for a non-profit I work with it only takes a few key people to come together to unlock but only gets them access to stuff relevant for that organization.

If someone loses a key, or it gets corrupted, it just takes more people to agree to use their key to gain access.

In addition to death, something could happen to cause me to forget my master password, but otherwise I'm still capable of doing things. So it is also a backup for myself.

For those that need a reliable drive, ignore every drive listed that doesn't have a 5 year warranty and that pretty much filters out all the scams.

edit: and filter for new and use your normal Amazon smarts to not buy if there are few reviews or from a seller that has poor ratings. It would be helpful if diskprices.com showed the number of reviews and average review for the drive as well as who it is sold by and what their review count and average review is. Also a filter for shipped/sold by Amazon (I'm aware of commingling of inventory). And a filter for the number of warranty years.

Also good idea to run a tool like f3 (fight flash fraud) before putting anything on the disk.

A lot of people commenting that MS should not have went with Qualcomm on this device, although I don't think this locks Windows into Qualcomm. More importantly, I'm super curious to see what Qualcomm announces in December in reference to the next generation chips being designed by the Nuvia team (some of the people that designed the M1). ARM is suing to prevent it, but that seems shortsighted, and I assume it will net out that the Nuvia design will launch, but may have to pay some additional license money to ARM.

FWIW the M1 is a little more than twice as fast on GeekBench, but also uses twice the power (TDP). So sort of similar speed score per watt, so possibly they can be within the same ballpark as Apple in the next couple years.

My guess is that they are using what would otherwise be a marketing budget. If that is the case, then I'd say it is worth it and working. I'd bet more people have heard of POP_OS then System76, which would increase exposure of their hardware.

I'm downloading it from:

https://releases.ubuntu.com/kinetic/ https://releases.ubuntu.com/22.10/

For the use case of this post, revision/version control of notes (markdown/text), I recently started using Fossil [0].

I've wanted a reason to use Fossil for quite some time, and this is perfect since I'm the only user for this repository; Fossil is ideal for small groups of people and git is ideal for projects with thousands of contributors.

I'm just using VS Code with a Fossil extension and storing the notes as .md files in the repo. You could use the built-in wiki to store them, but that would give you less flexibility and less of the features of a source control.

Although I'm not yet, I could integrate some other features into my workflow like bug tracking, forum (to talk to myself and hold conversations over time to center myself where I left off), and technotes. I love the web-interface.

The only configuration I needed from stock was to turn on the search feature, change the home page to be my root document (README.md) and set it up to run as a daemon with systemd.

[0] https://www2.fossil-scm.org/home/doc/trunk/www/index.wiki

In the early 80's on a TRS-80 Model III, I was in whatever grade you learn to alphabetize words, and I wrote a Basic program to handle the alphabetization. I input the 20 or so words, it output them in order, and then I transcribed that to my homework assignment. My mom said it was cheating and I couldn't use it, and my dad said that if I could write a program to do it then I illustrated that I could alphabetize word lists and it wasn't cheating.

I have a suspicion that I probably found the code in one of my dads computer magazines, so it probably was cheating since I doubt I actually wrote the program from scratch. Maybe partial credit for being resourceful. :)

In addition to named functions, it also adds LAMBDA, Lambda Helper Functions (MAP, REDUCE, BYROW, BYCOL, SCAN, MAKEARRAY), XLOOKUP, and XMATCH.

My sister interviewed my dad years before he was terminally ill, so I'm very fortunate to have that. I bought StoryWorth for my mom as a Christmas present to basically interview her via email, and she hated it and wouldn't do it (my father in law loved it and we printed books for my wife's siblings). I asked her if we could do StoryCorps in person to help with prompts and recording, and she didn't want to talk about her past like that either.

Then I had the idea to ask her if I could interview her about her mother so that her mother's memory wouldn't be lost, and that worked. So it'll take time, but planting the seed for her to understand the value.

I can't square it either. Lets say that you live in a world where rich people want to control the narrative by making entertaining news shows to distract the populace from what is important, but needed a way to trick the people that sought to dig deeper from more authoritative sources that are less TV-entertainy... then I think it would make sense to tell the people that are digging deeper that it is bad for them and they should stop paying attention to the news for their own mental health and productivity.

I recognized that I was over consuming the news during the Trump presidency and needed a reset, and decided to not spend any more time reading the news than the short summaries that Axios posts and ignore all other forms (including John Oliver, although I do still watch Closer Look).

It appears that my actions and aspirations don't seem to be making the world a better place, but it is too depressing a thought to think that the best thing for me to do is be ignorant to the bad things in the world.

I plan on trying out subscribing to Delayed Gratification to see if that is a better path. https://www.slow-journalism.com/

Search Engine Roundtable reported it on July 21st (crediting Deccan Herald from the 20th) and said that Danny Sullivan is looking into it: https://www.seroundtable.com/google-search-timer-stopwatch-s...