orm

I like your distinction between contrarian and independent.

Also. The part on being good at identifying the right experts is interesting...

Haha. Ok by some. As far as shows go.

interesting, can you point me to some examples?

The pervasive culture in the west seems to elevate pessimism as intellectual, while optimism is stupid or naive. This bias is interesting considering how much lives have improved by many important quality of life metrics in the last 150 years, partly due to technological progress, which has enabled social progress as well. Yet most people systematically mis-predict a lot of the quality of life metrics toward whatever side is negative (the book Factfulness makes a good case for multiple results of this kind). In high school, people read Brave New World, and it seems the thinking gets stuck there. It is often a signal of education to reference these books in conversation, and you see it. Does our collective thinking get stuck at the level of an emo high-schooler?

Along the same lines, most current western sci-fi I am aware of (mostly mainstream) is dystopian. Black mirror the show is a good example, also considered intellectual by many. I feel this tendency is even stronger in Europe than in the US, at least when I visit.

In contrast, China is somehow more optimistic. Consider AI right now. Some recent poll shows Chinese people are far more optimistic or positive about AI than western people, an interesting perspective considering their own government has made computer vision surveillance and face recognition far more widespread than it is in the west.

Do these kinds of popular sentiment matter for our collective decision making? are they a needed precaution to ensure things keep getting better? or do they create self-defeating collective behaviors? wdyt.

The video was interesting. Seems like a nice way to start a shopping search if you have a picture with something you want where the look matters. Eg, cars, furniture. etc.

I have not compared it. Have you had issues in this front?

If you are asking in human terms, its instantaneous to start, it simply creates a few files etc and starts a process. If you were using Sqlite, there is no extra associated process. But the processes are simply waiting for your input, so normally this is not really extra work.

In benchmark terms, the way I'm using it startup time is not a huge issue, instead being able to use postgres is helpful (has extensions and you know you can eventually move to a hosted postgres if you want)

If startup really matters, the startup can be amortized (if multiple processes want to access the same file, they get a handle to the same already started process, so only the first one would have waited for startup).

In terms of ongoing resource consumption, I can easily have a bunch of separate servers running for different files (it depends on your workload how much work will happen behind the scenes.) I'm not sure how much extra work postgres needs to do vs sqlite beyond what is kind of inherent for a workload (eg index creation, saving a bunch of data etc).

Not directly an answer, but similar what if thinking got me wondering: wouldn't it be cool to be able to use postgres much like how you use sqlite with python? I implemented this idea as a pip-installable python package, https://github.com/orm011/pgserver, and your feedback would be great :) I use it for my projects.

I've done both, and been on the receiving end of both, but hadn't thought of this similarity. I think it's a good analogy.

Hm, not exhaustive but I think these are potentially useful to you:

The deeplearning.ai math basics for deep learning, seems self-contained. MiniTorch repo (implement your own tiny torch) seems also helpful to understand what goes on during training. MinGPT repo (to understand a basic version of GPT model structure) Dive into deep learning (textbook avail online, more focused on practical DL)

How did you get access to the model?

I have been waiting for weeks, and am still waiting, to get access to Llama2 (released a month+ ago), and access to this model goes through the same form, so I'm not very hopeful. Are you getting it from other methods?

Does anyone know of good tools to practice tone-listening and recognition for cantonese? or tools for drilling jyutping more generally,

I've used https://chaaklau.github.io/cantorocks/, but I would like to also try others, as the audio is not great sometimes and I'd like to listen to different voices.

When studying Mandarin I found that getting the phonetics right early on really helped me, it can be done in a relatively short time and its hugely motivating.

If you have basic Mandarin knowledge, you could try Duolingo's Cantonese course for Mandarin speakers. I think it is helpful to get familiar with how words are pronounced and what expressions people use in vernacular Cantonese so far. As with the Mandarin course, there is little explanation, you kind of have to complement it eg with some drilling for tones.

I really like this idea, though I haven't tried it yet. Spotify has lyrics for many songs, but for characters the first hurdle is you cannot really sing along even with the lyrics unless you know them.

Is there some easy way one change the font used by Spotify to be this font somehow?

I'm not sure the author will ever read this, but I am also a PhD student, have been one for far too long, and in the past have struggled with similar `time management' issues, though I'm not sure that's really the problem. As time passes, especially when things go wrong, projects and writing become very emotional. Having memories of looking at something a few years ago and looking at it again a few years later to try to resubmit it tends to trigger guilt, anger etc in me. I've found some strategies to help me, because I'm close to finishing and think the marginal benefit/marginal cost makes sense (benefit is large, finish your phd in 2 years, 1 year, 6 months, depends on where you are).

One tool I use to help me work even when I'm not feeling it is Focusmate (google it). I've found that sometimes, working out my issues is useful , but sometimes the emotions are overpowering and no amount of working them out will make me not regret something. That used to make me stop. Nowadays if I use Focusmate, I tend to think less of such `large picture' issues and instead work on just a little bit, and get a little done. then after a day, a get more done, and so on. I do sometimes get upset and avoid it altogether, but I'm definitely much better at `time management' this way. Also, I don't want to leave a focusmate in the middle of a session, i just need to stick around 25 minutes, and the feelings can pass and i can even enjoy some of the craft of the specific thing I'm doing. I hope it helps someone.

I should mention focusing on doing a few interesting problems, rather than many not so interesting ones, is also one way to help yourself understand more deeply.

3Blue1Brown videos seem like a good resource to use along any book. My experience as a math major (in the distant past) is that the kind of visualization the author shows you is also something you want to imitate in your head when you are learning new concepts. I find things I learned in this level tended to stick in my head 10+ years later, other stuff less.

Instant Pot. Can be used as a rice cooker, slow cooker, pressure cooker (mostly known for this last use)

Though probably not as good for rice specifically as an actual high-end rice cooker, it greatly increased the range of foods I eat.

Makes it easy to make nice one pot meals overnight for multiple days, you can make really good broth soups from chickens etc quickly, or slowly if you prefer.

If you like to try tougher cuts of meat, this is also a good reason to get it. The fact it doesn't occupy one burner is also helpful.

After the cold war ended (last three decades), I'm not sure of any case where America has been coup'ing countries in South America. Do you know some cases?

I'm shocked they missed a pun opportunity in the headline.

It's a strange article to illustrate inflation. It mentions there is a drought that has made a local river dry, which was a tourist attraction driving a lot of traffic through the town. This meant a lot less customers, a lot less pay available. It Didn't seem to try to disentangle the inflation from this other pretty big effect... which seems hard to do... But then why pick this town or this headline.

Have you checked out the Ray library? I use it for this type of scenario and it works well, without as much effort as it would take otherwise.

I can relate to this quite a bit, I personally think it is my biggest weakness. I have managed to do a few things despite this but just as an example, a few weeks ago I could not bring myself to work on a 5 minute presentation about work that was already done months ago because the feelings related to the work were very strong. Instead I 'winged' it, and it went ok (stakes were low, as I mentioned before, I had already done some work for this for an unrelated presentation), but really it didn't convey the work that was done as well as possible. It also didn't need to be as hard as I made it ( I spent the day before fretting about it, looking at it for a few minutes, but not doing much about it ). The irony is that the worst case scenario of showing up without much prepared was not nearly as bad as my anxiety made it seem the day before, and that anxiety made it hard to prep... So the anxiety was completely off and just made my life harder than it needed to be.

I'm pretty sure this anxiety/avoidance pattern that I manage to some extent has in the past cost me years because I know it has delayed my engaging with the things I need to do to move forward. I love working, thinking, reading, prioritizing, but I do think I have trouble regulating my emotions in certain situations more than other people I know.

My limited advice since I haven't figured it out: 1. Some amount of therapy. I have done some CBT. Lately I have been reading a bit about ACT, seems related to this specific problem more so than CBT. It is ok to have bad feelings sometimes, they pass, don't worry about worrying. (Easier said than done.) This is an ongoing process and can help you think through patterns like this. 2. I find that sometimes too much thinking early on can make things worse, and things are done faster if you don't think too much before jumping up on something. 3. Focusmate.com: you tell someone what you'll work on for the next 30 mins, and you do it. Somehow having someone else there that you're accountable to, as well as who doesn't know you, makes some of the emotions recede for me a bit, enough for me to get started somehow. 4. Articulate to yourself, as if you were explaining to someone else, what makes the thing hard. Sometimes I feel calmer after this. 5. Reframe the situation: maybe you can find some other situations where you are successful against this pattern and try to remember them. Can you make the same happen here even in a small way? Can you fight back a bit? Short term stress is not that harmful to you and you may want to see it positively, see some positive properties of it in your work. Eg. sometimes when things are quite critical I somehow manage to break from the avoidance and do something. 5. Perfectionism is an enemy, and guilt follows. I find guilt over previous avoidance is a secondary effect that makes things even worse. Eg. when I finally get to something I have avoided, I feel guilt over how much time I avoided it and how it really isn't so bad. Checking your own patterns may help you realize this is a familiar pattern and just that. 6. Work with other people or run your thoughts through other people. This is not always possible, I'm a PhD student so it is especially hard to do this (your work is ultimately your own and no one will drive it forward but you, but you can try running your thoughts through someone else), but I find in a team (even just 2 people) I find myself very mindful of not imposing any of my perfectionistic tendencies on the people I work with. This in turn helps me regulate them more. 7. Get good sleep. It helps me be able to step away from myself and think of myself more in the third person than when I don't sleep enough. 8. Get some exercise. I find it is helpful in breaking through some of the rumination/avoidance cycles. 9. Share some of these emotions/situations with other people (what I'm doing right now ;)). It is easy to also start avoiding people in these anxiety/avoidance loops, I certainly feel that impulse, but you may find other people can relate and even have some advice. If your feelings can change this way you may be able to get some work done.

Iron age Greeks used the word cyclopean also to refer to the ruins of Mycenae nearby, dating from the bronze age only ~500 years before their time. It is comforting for me to know it was already a thing back then to resort to ancient aliens to explain stuff like this. They were coming up with these explanations far before any kind of modern curriculum or industrialization, and when their ability to make physical objects was still very much there, but somehow the technology and thinking had shifted. Ironically it appears it was their own partially forgotten ancestors who built those ruins, as the decoded script (Linear-B) from the previous era appears to still a Greek language.

Megalithic seems more fitting than cyclopean to me.

GPT-3 is trained on text prediction, and there's been a lot of commentary about the generation aspect, but some of the applications that excite me most are not necessarily about the generation of text, instead, GPT-3 (and also other language models) create very useful vector representations of natural language as a side effect that can then be used for other tasks with much less data, or with too much extra data. Using the text prediction task as a way to supervise learning this representation without having to create an expensive labelled dataset is very helpful, and not just to language tasks. See for example the CLIP work that came out recently for image classification, using GPT-3 and captions to supervise training. There is other work referred to in that blog post that also exploits captions or descriptions in natural language to help understand images better. More speculatively, being able to use natural language to supervise or give feedback to automated systems that have little to do with NLP seems very very useful.

Well, first, some schools are going online only for the latter half of the semester. But my understanding of the current guidance is that this is not enough to meet the in-person requirements, and so students may have to leave anyway and miss that half. Second, intl. travel is highly disrupted already and will continue being so. Some countries will fare worse under covid than others, and so students may get stuck, even after their school states are doing better. No guidance was given on students from travel ban countries, who were already here after the travel bans. Third, student visas expire. Visa politics are very volatile lately. Summing up, many people are now subject to gratuitous uncertainty about whether they'll be able to come back in at all over the next 12 months. Suppose you have to leave, a lot of people have already signed leases, etc.

Finally, the guidance so far is quite ambiguous. It is not clear who is able to remain and who won't, and my department sent an email basically saying they are waiting for extra information to know what's going on. That already is bad in itself.

Yeah, to add to your comment, this link has some more detailed explanations regarding the wheel+axle combo: https://www.livescience.com/18808-invention-wheel.html

It is also worth pondering why wheel+axle was invented so late even in Eurasia (after bronze alloys were invented and adopted, only around 5.5 kya, many thousands of years after agriculture).

This fact has changed my perspective on how much infrastructure already must exist to enable large wheels that rotate smoothly around an axis. Knowing that, it is no longer shocking that one might make toy wheels, and never scale them up to large ones.

functor is another one.

I think the form would benefit from having the calendar year as well, as these numbers change enough over a few years.

A lot of matrix concepts sound a lot more natural to me if I first think of what they mean in terms of vector spaces.

For example, I think one can derive associativity of matrix multiplication less tediously from the observation that every matrix corresponds to a linear transform from a vector space to another, multiplication corresponds to function composition, and function composition is associative.