I’ve noticed Claude has been noticeably worse over the last week. For example, it told me I should pass frozen to make my Enum immutable—that’s not a thing. (It is a thing for dataclasses, but not for Enums.) That’s a pretty basic language feature it was nailing until recently. It also suggested I parse a URL using urlparse in a function that already uses urlparse. These are basic mistakes it wasn’t making before. Something seems to have changed, but I’m not sure what.
I work at a non-profit organization and part of our mission is exactly this. You can find more details here: https://indieauthorproject.com/
Another part of our mission is providing libraries an open source ebook and audiobook reader called Palace. Our project is on GitHub here https://github.com/thepalaceproject.
I've encountered frustration with Python packaging in two main areas:
1) Installing applications within Docker containers. While wheels have improved this situation, I was surprised that building a package for easy copying into a container and running without the need for installing build tools and extensions in the final container image was not straightforward, especially coming from other languages.
2) Distributing Python utilities to end users across various platforms in an easily installable manner without requiring them to follow lengthy instructions to set up all the dependencies has been another challenge.
We use Poetry and have largely "solved" #1 with a somewhat complicated Docker build. It works well now, so no one has to think about it much. That made deploying Python server-side fairly easy. However, #2 has been much more of a challenge, and I wonder if that is where other folks in this thread are feeling the most pain.
We dropped the penny in Canada in 2012 and I can’t say that I’ve missed it at all. Penny are still charged if your paying by bank or credit card, otherwise price is rounded to nearest 5c.
> If the price ends in a one, two, six, or seven it gets rounded down to 0 or 5; and rounded up if it ends in three, four, eight or nine.
> Alternative platforms already exist: one promising place to start might be the Palace Project and the associated Palace Marketplace.
It's great to see the Palace Project mentioned as an alternative in this post! I'm one of the developers working on Palace. The idea behind the project is to create a single app that aggregates all of a library's digital content and makes it readily available to library patrons.
This is really cool. The killer feature I see is to be able to have a cloud storage tier for warm data that goes off to s3, while keeping the hot storage local. Does anyone know of another option that allows this kind of hybrid local / s3 storage that also has a filesystem interface?
> Since then, he has been unable to remember almost anything for longer than 90 minutes. ... Without a record of new experiences, the passing of time means nothing to him. Today, he only knows that there is a problem because he and his wife have written detailed notes on his smartphone, in a file labelled “First thing – read this”.
I think it would be fascinating to read the "First thing - read this" file on his phone. I can see why they don't go into more detail it would probably be pretty personal info in there.
Still as a mental exercise, its interesting to think about what I would write down in that file for myself.
I still have the problem that my 16" MacBook runs really hot with a external display plugged in. There is a pretty good thread on the issue on the mac rumors forum [1]. It seems to particularly be a problem at 2560 x 1440 @ 60hz.
With Safari removing my ability to use uBlock in version 13, I've been wanting to make the jump back to Firefox on the desktop. The one thing that has been holding me back is that Firefox can't use the OSX Keychain for passwords. All my passwords are in my OSX/iCloud Keychain, and this is great. It works on my iOS devices and on my Mac and everything is lovely, but its keeping me using Safari on the desktop.
There was an extension that did this, but the API it relies on was removed in FF 57 with the switch to the extension API.
> The government may steal a dollar, but it cannot erase the idea that earned it. I wrote this book, Permanent Record, for you, and I hope the government's ruthless desperation to prevent its publication only inspires you read it—and then gift it to another.
> The court's ruling is a hack intended to circumvent First Amendment limits on what the gov't can censor. They can't (yet) ban the book, so they ban profit to try and prevent such books from being written in the first place.
> "It's a prototype for sure," said McDougall, "but in every way it's a high-tech piece of equipment, which is kind of ironic considering the airframe that it's attached to is actually one year younger than me — 62 years old."
> McDougall's flight is the first exercise in what is expected to be a two-year process to get the e-plane certified for commercial use.
For some context Harbour Air is a scheduled floatplane airline operating primarily between Vancouver and Victoria in BC. Since there is no bridge connecting the capital Victoria to the largest city Vancouver, they offer short flights on seaplanes that connect downtown to downtown as an alternative to ferries or larger airplanes that arrive at the airports further outside of the downtown core of either city.
In this constrained context it seems like an electric airplane could work really well and provide fuel savings and a quieter ride. They also seem to have some ambition with these electric aircraft to provide other short range flights.
> MagniX CEO Roei Ganzarski said Dec. 10, 2019, will go down in history as the start of the electric aviation age, and believes the e-plane will eventually revolutionize how people travel by making short- to mid-range flights more economical than driving. "It means you can stop driving for three, five, seven hours to get to a destination because there's no other way to get there," he said. "It means you can fly in a small aircraft from a small airport to a small airport.... It's faster, cheaper and more convenient than any other method of travel, including going with a standard airline."
> The government isn’t a bigger version of a startup and can’t act like a startup does. Innovation activities in government agencies most often result in innovation theater. While these activities shape and build culture, they don’t win wars, and rarely deliver shippable or deployable products.
> The very definition of a contractor implies a contract. And a government contract starts with fixed requirements that only change with contract modifications. That makes sense when the problem and solution are known. But when they are unknown the traditional methods of contracting fail.
This I think is the biggest failure in public private partnerships. These arrangements were supposed to save the taxpayer money by allowing nimble contractors to come in and do the heavy lifting on projects.
However what it really did is:
1) Create an adversarial relationship between between the government and the contracting entity, which is very much filled with even more red tape then it was were trying to replace.
2) Hollow out the talent in the government, causing more reliance on the private sector.
I'm not convinced these relationships can exist in a healthy way. If there isn't a lot of red tape and strict contracts then it's likely the government will get robbed blind and that leads to stricter contracts and more red tape: a virtuous cycle of red tape and lack of progress.
I think less emphases needs to be put on these partnerships and more on developing talent within the government that can innovate from within. This creates the possibility of empowered leaders within government departments that can run projects like agile mini-startups.
> This is the most astonishing part of the story to me. If you had a crystal ball in 1928 and you could peer into the future of any scientific discovery, you would have been screaming to the world, “Guys, GUYS! OVER HERE! This one!” [pointing furiously at Fleming/Penicillium]
> A thing that keeps me up at night is that there is almost certainly, today, something which might not be shaped like penicillin but has a similar level of impact, and which is really sweating the next thousand dollars.
I went to an engineering school in Canada for Electrical/Computer Engineering and this was an example of engineering failure we studied in a course about software quality assurance and engineering ethics. Not sure if its widely used as a case study outside of Canada, but it really stuck with me. Especially when designing critical systems.
There are two types of reactions to the Therac-25 story.
1) We need to put a lot of process in place to review and test and ensure we have proper interlocks in place.
2) They just didn't have enough talent on the team. It couldn't happen here.
You want some people with reaction 1 working on a safety critical system.
As the article mentions it seems like the show "Chernobyl" is a dream come true for the anti-nuclear crowd. Its dramatization of the event is truly horrific and actually had me questioning my support of nuclear power as a sensible alternative.
Watching it has me wondering a lot about how bad the accident actually was. I've done a little research and the number of fatalities seem to vary wildly depending on the source.
Does anyone have any reasonably good un-biased sources of information about the accident?