I'm uneducated on how distillation works at more than a basic level so forgive me if this is a stupid question.
Isn't "distillation" of another provider's model exactly how these models got training date in the first place: Massive amounts of the written word + Prompt -> Answer. Why wouldn't distillation produce similar "reasoning" in the new model? It's just inputs and outputs.
> To tell the truth: I do not even know how to use the pen. I have never needed to do this in my life.
I truly do not understand how one could have T1D for 27 years and still be alive with this attitude. Don't get my wrong, I'm happy she is! And maybe that's just my own limited experience of managing my foster daughter's T1D for the past year talking but... Man, this year has been a rude education in how many ways all of these necessary-for-life pieces of equipment and medications can go wrong and how easy it is to get into a life-threatening situation.
Going more than 30 minutes from home without extras of everything (Dexcom, glucometer, fast-acting, long-acting, glucose tabs, glucagon) is completely unthinkable.
> Find an online community that has self-selected for some kind of criterion
I’ll add onto that: find real-life friends/acquaintances who are both not brain dead and read books. Frequently ask them “what are you reading lately?” Not only does this lead to good conversations and deeper friendships, it results in an endless stream of book leads.
Most of the good books I’ve read for the past several years have been curated for me by two friends who are prolific readers and do all the work for me of finding new books. I occasionally find something they haven’t read, but they certainly do most of the heavy lifting.
I switched from the Roku ecosystem to Nvidia Shield about a year ago because of Roku’s increasingly user hostile ad and tracking decisions and have been pretty happy with it. Android TV gives me enough control with a custom launcher that I can prevent ads.
I would guess that the number of people listening to this media on anything higher quality than the built-in speakers on a 65-inch TV is minuscule. They’re optimizing for sound on an iPad, not a full surround sound setup.
Backend with a lot of existing code that's still in .Net Framework using tech like WCF and Windows Services.
I could get away with doing dev directly on the Mac if I was _only_ working in .Net Core on our newer applications and if I was willing to use Rider. Neither of which are true.
I'm one of that tiny percentage, but all of my development is done by remoting into a cloud Windows VM.
I tried directly coding on the Mac when I first got it, but quickly gave up on that. (I hate Rider, which seems to put me in minority as far as most .Net devs go.)
I've had the same experience. Played multiple times, once for almost a year. The only thing that kept me playing that long was the good group of people on voice comms. I quit for the last time when I realized that life is too short to waste on a game I wasn't even enjoying (and that, frankly, stressed me out.)
Pure speculation: With so many subs now private, they're hitting a performance wall searching for enough content to fill the front page. If I'm logged in and go to the site root, it throws an error. If logged out, it works. If I go to a specific (not privatized) sub, it works.
> It's just surprising that audiobook listeners haven't developed a distinct taste apart from book readers.
Sample size of 1, but: my taste in audiobooks does not perfectly overlap with my taste in dead-trees books. My audiobook library skews much more to the fiction end of things, while my dead-trees library contains more nonfiction. I also use audiobooks to re-read books I've already read.
I enjoy the performance aspect of the audiobooks in its own right. The narrator makes or breaks my experience of the book, to such an extent that I have returned Audible books that _I have already read in dead-trees form and knew I liked_ because the narration was bad. I also suspect some of my favorite audiobooks would have ended up in the "donate/throw out" pile if I'd read them first on paper.
I've been thinking of doing this exact thing, although with a smaller 10x12 single-level shed, just because I want to learn timber framing. Would you do it again? Any specific book recommendations?
I just wanted to thank you for this. This is the most succinct and easy-to-understand-for-non-investors explanation of the concept I've seen and now I think I finally understand it.
Not advocating for this point of view, but I think the thinking goes that even if <insert industry of choice here> wasn't any less concentrated in 2017 vs. now, the pandemic provided a lot of opportunity for price increase that might have otherwise caused regulatory scrutiny.
At least half of people are involved in the housing market, not just immediate buyers. Because you’d better believe if residential property prices skyrocket, so will rental rates.
They benefit by preventing easy access to an export of the user’s data that could then be importied into a competitor’s software, thereby (presumably) locking in a non-trivial number of their own user base who don’t want to deal with the hassle of re-entering all of their financial data in a new product.
Isn't "distillation" of another provider's model exactly how these models got training date in the first place: Massive amounts of the written word + Prompt -> Answer. Why wouldn't distillation produce similar "reasoning" in the new model? It's just inputs and outputs.