khaki54

Daniel is a fantastic writer, his command of English and humor are fantastic. I was following his troubles getting a US visa... it was terrible, but after a few years he finally got it with no explanation.

This looks promising, not sure how it stacks up to Transkribus which seems to be the leader in the space since it has support for handwritten and trainable ML for your dataset.

Complete nothingburger. The malicious action would be to subtly change key datasets to promote some kind of biased narrative. It wouldn't be the removal of less than 1% of the datasets.

Comical to think the Trump admin would be the ones "scrubbing" data when they passed the OPEN Government Data Act as part of the Foundations for Evidence Based Policymaking Act at the beginning of 2019. The law that requires agencies to publish their data to the site!

Trump must really be playing 4d chess if one of the first actions in office is to "scrub" a few obscure and yet unidentified datasets from data.gov.

New technical interview question just dropped

Are they really blaming it on that or is that what the typical (inferred) explanation is? Clearly things started going wrong economically around that timeframe.

I think it's clear they aren't really sure; putting up a bunch of quasi related charts points to a desire to facilitate a discussion.

Balance bikes are super cheap, very light, and the kid uses it for like a year or more until you are ready to buy the proper bike

Is this new or interesting? I mean my 10 year old car has a feature that shows realtime traffic cams wherever I am. I imagine these things are widely available and you could run a plate reader on those feeds.

Seems crazy to steal a design for something like this. Wouldn't you want something original and innovative? And when the author offered to design them something new, why turn that down?

Ultimately they probably plagiarized from something else.

I love when people do random stuff like this. I couldn't even suss out his reasoning for taking this project on. Normally there is at least a notional but absurd use case. Cool project though, and I'm sure he had fun.

I wrote a shell in C for Bruce Molay's Unix System Programming class. I thought the hardest part was getting the globbing, special characters, and escaping to work right. The other element of Unix system programming in C (that some may not be aware of) is that you are expected to make sure every line is 80 char or less. At the end of the day, I was proud of what I accomplished, but I think I only got a _C_ on the assignment.

Jaguar and Land [Range] Rover are notorious for their electrical problems. I wouldn't trust an ICE vehicle from them let alone an EV. I doubt that even with new ownership or investment (or whatever is going on here) they can overcome the culture of unreliability and poor quality that has been built over a century.

Decent gag that baits my click every time

I can't see the plaintiff gaining anything really. Maybe he's just fed up and decided to do something about it.

I don't think the FDA should be destroyed.

Sometimes you have to question if they are in the wrong too frequently, for instance recalling 80,000 lbs. of butter (when it's already more than $7 lb. in most grocery stores) because the butter packaging didn't say "Contains milk". It's a recent example of something they did that only caused harm. Notify the company to fix their packaging, sure, but under no circumstances should they be recalling the butter.

https://www.cbsnews.com/news/costco-butter-recall/

We're doing something like BM25 with a semantic ontology enhanced query (naive example: search for truck hits on Ford F-150, even if truck never appears in the doc) then vector based reranking. In testing, we always get the best result in the top 3.

This is one of those "lying with statistics" moments. The implication is that "Unwanted telemarketing calls" have gone down significantly, maybe even "more than 50 percent since 2021" and the second order implication is that it was the efforts of the FTC that made the drop happen.

The reality is simply that reports to the FTC have gone down which is going to be from myriad other reasons besides the implied ones.

  1. Since many spam/scam calls utilize some form of spoofing, many savvy people aren't going to waste their time reporting the number.  
  2. Since many phones / carriers have the ability to report spam/scam, people aren't very likely to spend their time reporting the number to the FTC.  
  3. Many carriers and phones have services that come standard (or at least paid) that automatically filter or flag low reputation numbers, leading to fewer reports to the FTC.
  4. People that have tried reporting in the past and have gone through the trouble of associating their email and phone number in yet another government database have given up reporting after no appreciable effect.

Frankly, I don't think calls have dropped much if at all, and many of us have stopped accepting calls from unknown numbers unless we are expecting it, so those "ignored" ones certainly don't get reported either.

Not sure I would want something non-deterministic in my data pipeline. Maybe if it used GenAI to _develop a ruleset_ that could then be deployed, it would be more practical.

Kind of shows the echo chamber. Looks like the Jordan Peterson account, which has never posted anything, is around number 14.

How did they get these engineers to get through their whole presentation without saying "um", "uh" or look like they are reading. It really seemed off the cuff too. Very talented folks

Well, you would read their methodology which they have posted and then decide if it's reliable. Opinion is naturally ranked downward in their methodology, so in a sense it is separated.

The chart has been around for a while and is generally believed to be the leader. It's widely used in academia. It's a good tool but not bulletproof, and they admit the bias of their editors freely.

Somewhat depending on writer, The Guardian is highly biased and fairly inaccurate. I don't think they are filling any type of niche in reporting that we will miss. https://adfontesmedia.com/interactive-media-bias-chart/

Generalizing here, as kids in the U.S. we used to be taught to never mention or ask about someone's disability, no matter how curious we were. I definely remember a time later (with PSAs and all) where some of those who were disabled lamented that they were actually sad no one asked them about it, and everyone looked at them stone faced and grimly; but in fact they would rather answer questions and didn't mind a joke or two.

The relevant wisdom here is "sticks and stones..." people who are different are often teased but sometimes celebrated. Sorry you are wheelchair bound (likely due to no fault of your own) but dwelling on a throwaway comment that has no ill intent behind it is 100% within your control.

I was thinking the same thing, fully knowing pinewood derby is mostly concerned with friction and center of gravity.

Haha pull request on the Readme and add this link!

Great but "infant in the womb" and similar phrases go all the way back to antiquity. If anything, dehumanizing the fetus is a recent invention by pro abortion advocates.

I presume you could subscribe and still use this tool? People use automation tools like this to download things that they already pay for because it saves them the effort of logging into 5 different apps depending on which walled garden it's in.

Gross. What's wrong with you?

Does it always just say pending regulatory approval and get stuck that way? Is that the joke?