Has anyone tried something like Polylith which lets you build all your code like normal functions for local dev and testing and then seemlessly pull out parts to network services as needed?
I find the threading operators in Clojure bring much joy and increase readability. I think it's interesting because it makes me actually consider function argument order much more because I want to increase opportunities to use them.
As sane of a suggestion that is, SPY index fund dropped a huge amount because the whole market dove. Not everyone can plan on suddenly missing 20% of their savings.
I see you getting downvoted which seems a bit harsh given it seemed like a very honest question. As far as I understand, a huge amount of materials and intermediate steps and companies are over in Taiwan or Shenzen or other places in Mainland China. Even if you could do some parts in the US (like final production) all the sub-components and companies who can actually produce them are dotted over Asia. This is part of the problem in raw economic terms: there is a large vendor and/or geographic lock-in. It's hard to get manufacturing jobs and work back in the US because most of these lines if production are somewhere else and moving one or even a few parts can't overwhelm the huge cost of moving stuff over the ocean potentially multiple times. I read that a brake manufacturer was particularly screwed even within North America because steel came from Canada, partial assembly would occur in Mexico and in the US, and then the finished product might cross a border yet again. Without tariffs this wasn't so bad, but if all the products have to cross a ton of borders a lot of these supply chains are in deep trouble.
One time I was flipping back and forth between directories, compiling some code, then checking it, then rm -rf'ing it. I accidentally hit up an cd ..'d one too many times. Suddenly the rm command hung and I was confused because it should be nearly instant as it was a few files. I stared in horror as I was accidentally deleting everything in mybhome directory. Luckily back then I had a .pr0n directory with a significant amount of content. A few things were lost but that .pr0n folder was luckily early enough in the list and big enough to slow down the deletion of my photos and documents. That's why I always recommend having a big "buffer" of video content for such situations, ya know, for data integrity ;)
All types don't need names exactly like the Clojure example shows. There is no "type" for the argument, likely it's a map under the hood. And maps with keywords are used broadly across Clojure projects as a method of pass groups of data and no, you don't have to name that arbitrary collection of data. Rich Hickey has an amazing presentation on the silliness of a commonly used Java Web Request library where the vast majority of object types required could have just been little nested maps instead of creating a ton of bespoke like types that make it difficult to assemble and manipulate all the bits. The interface he roasts could be completely discarded an nearly no benefits lost in terms of readability or usability. Hickey is also famous for saying he would rather a few data structures and a 100 little algorithms instead of 10 data structures and 10 algorithms per data structure.
I don't immediately see a picture of the actual chip to get a reference size bybitself, that that chip inside the tooth is pretty wild. I wonder how you would even begin to power electronics within such a space. Or perhaps this is more of a marketing level of graphics.
Isn't this sort of an inconsequential point? The commit still has one and only one author and that's almost certainly what I'm looking for so I know who to go ask questions about their code. I also use it to find the commit but less frequently.
While the reason you suggest might be a good one, the repercussions of this change are clearly large. Unknowing customers now have their voice recordings sent to the cloud whoch could be leaked, have law enforcement get a warrant to take it from Amazon, or for it to be abused directly by Amazon. Those are huge. They might just have a good scapegoat.
HI personally worked on it while I was at Pivotal Labs (RIP). I loved that the point was to get in, add some stories, aka "bookmarks for conversations", have it opinionatedly setup your sprint by averaging the last few sprints worth of points, and then you got out. I use other tools now and I cringe every time at the complexity. I don't want a swiss army knife of features, I want a tool that does exactly what I need and I don't have to spend hours grooming and configuring and tweaking. Tracker hit that sweet spot.
I know some of the lithium-sodium batteries behave differently now but my Model Y yells at me whenever I charge above 80% or go below 20%, saying it's bad for the battery. Not exactly inspiring if it causes more wear unless you sacrifice 40% of the range.
IIRC it has to do with tiling the globe. You can't correctly tild a sphere with swuares. On a local level, it seems easier to reason about squares on a flat surface, but the Earth is round and Uber is international.
Providing public APIs to such information sounds great and I think it should be done but is also probably realistically expensive and peone to security problems. The government seems to have a very difficult time in general running all sorts of web services, it's a nightmare in some jurisdictions to even pay the government money for e.g. tickets... I had a hell of a time paying a ticket in San Diego and had to visit the the courthouse multiple times and file a paper form because my paymentment became "late" when it never showed up in their online database weeks later. I was TRYING to pay them and even going to suffer through the "convenience" fees because they outsourced the website to some crappy company and they still couldn't get it right.
Anything probabilistic? There are so many interesting fields where you can assume the distribution of a dataset and the take a sample of data and assert things about it with a high degree of confidence. All of modern AI is built on so much of this. All the Deep Neural Nets are making grand assumptions about the shape of meaning of data, they literally assume convexity of the space and they have clearly very interesting results despite the imprecision of the model. Anything dealing with finance is also dealing in lack of data. So if you had a list of prices of a stock over time, you could probably start making assumptions exactly like that, tgat the probability that it doubles over a short time is so unlikely so you can subsample the data and have it still be super useful to make assumptions exactly, especially when you have intractably large data.
There are no Clojure stacktraces specifically, only JVM/Java ones. Clojure sits closely beside its host so there is no difference. People do complain about stacktraces and they've made some improvements. But they are long because the abstractions run deep and the function you passed in might be on the stack at the very bottom with the actual origination of the cause.
You are being snarky but I'll take immutability over types and interoperability and access to a vast and stable ecosystem over the simple and extremely limited and underperforming alternatives. Thanks for the segway. I also cannot understate the reasoning ease that comes with basically everything being a static, top-level function. Testing never requires mocks because you can redef any function anywhere trivially when you otherwise would reach for a mock which tends to be painful, complicated, and requires comlex mocking libraries. Even with immutability by default, you get an extremely performant platform that usually stomps on nearly all the dynamic languages you might otherwise choose like Python or Ruby etc. And you get first-class functions and a concise and extremely stable language that is a joy to program in. Thanks Rich and Alex!
If you see a red car driving down the street do you not call it red because there are many other red cars? They're adding color (pun intended) to their description of the general bias of the UK government. What you're doing is called Whataboutism - the argument that others are doing something similar or as bad in different contexts. It doesn't make what the UK is doing any less bad for citizens (and non-citizens) privacy or data sovereignty.
I think there is an implicit assumption that the top 40 apps is reflective of what all app developers should be doing. The top 40 apps are probably also all made my mega corps with enormous budgets and desires to be multi-platform right out of the gate, and also have already vetted necessary issues like "does app performance matter" versus "does developer productivity and app-parity across platforms" matter. Just because a top 40 app is high, maybe it's something like Amazon's shopping app, maybe the trade-offs are heavily towards slow-starting but very feature rich and also equivalent across platforms is necessary. If someone writes a note-taking app that takes 10 seconds to start because it's loading some JavaScript updates and spinning up the non-native code execution hamsters, I already forgot my note. Native is obviously desirable. Or a game is interesting because maybe people are patient for a slow load, but still desire very fast native performance. My main point is there are quite a few trade-offs, and it's unsurprising the leads at Expo are saying look how awesome this stack is. I've used Expo and it was awesome, but it comes with costs too.
I totally understand the desire to avoid politics on all these platforms but in some way I always expect the greater powers want to destroy these platforms and make us even more hopeless.
There is a belief that we've peaked in terms of bigger model = better results. I think it was GPT 4 is actually smaller in parameter count and better than 3.5 for instance. There is also a finite amount of useful data that people think we have hit, so adding more parameters isn't helpful if you don't have new data to train them on.
Define dramatically with numbers. From all the sources I've read, it was so significant and also run on a far more limited cluster and the results are as good as the other frontier models. Optimizations have been coming, I think the one or more they found were significantly larger.
The stock price had assumptions baked in about the number of units expected to be sold. DeepSeek cut that hardware estimate by as much as 45x. That is that absolute obvious correlation between that model being very efficient to train and NVDA dropping 18%.
I really like "jump" which I have aliased to "j". It auto learns and you can also pin shortcut names so I don't have to fuss configuring it but also can control certain ones. Great trade off between usability and configurability. Can't find the repo but "brew install jump" works :)
I love the quote "accountants don't use erasers". So many things should be modeled over time and keep track of change right out the gate. Little things like Ruby on Rails always adding timestamps to model tables was super helpful but also a little code smell. If this is obvious enough to be useful everywhere, what is the next level? One more reason Datamoic is so cool: nothing is overwritten, it is overlayed with a newer record and you can always look back and you can always also always take a slice of the db at a specific time and have a complete and consistent viewbof the universe at that time. Immutability!
I may be more qualified then, I work in fraud. I accidentally called a fake airline number to get a refund for a Delta ticket and happily gave the guy my credit card for some such fee. If requiring a credit card fee to get a refund isn't a red flag, I don't know a better one. To be fair, my sister had found the number and three way joined me in because I was helping her buy the ticket. So an extra rule: don't even trust a phone number someone you know found because /they/ might have been the first victim and passing compromised information to you.
So so true. 1Password refusing to auto fill a password has saved me multiple times in the past! Also, one of my friends has a PhD in literally rocket science (aeronautical engineering from MIT) and got scammed by someone who stole his brother's SIM card and did some shenanigans. No one is safe, no matter how smart or tech savvy you think you are! For the less tech savvy folks, I understand why they are scared, it's hard to give them even general tips to not lose the farm to fraudsters.
There isn't an end to that, what is "all the data"? Someone will always want to record more data, and then sell it to someone. How do you force people to always reveal all the data they have. I think if you start peeling back the onion on what you're suggesting you will realize that it's not really possible or practical in any sense.