dopu

That idea of a sidepanel showing a feed of potentially related notes via fuzzy search or something would be such a game changer. Tempted to learn some clojure just to implement this in logseq…

This would make sense -- you can get the mac mini at the base M1, upgrade it to an M1 Pro, or jump up to a Mac Studio for M1 Max and above.

I initially thought the Mac Studio was the "pro" version of the Mac Mini. But no -- they're still selling the i5/i7 mini. This makes me wonder whether they're going to update it at all. Maybe they'll just quietly drop it at some point down the road?

I've been holding out on replacing my i7 mac mini until Apple announced its replacement -- and wow that mac studio looks good. The fact I can put linux on it due to the Asahi team's efforts makes it even more attractive.

I doubt they would ever endorse the use of a third party interface to Signal.

I'm a pretty happy logseq user. I've found that I still need some degree of hierarchy in my notes, as opposed to just having organic, non-hierarchical linking between everything. So I end up making subnotes: [[Topic A/subtopic x]] is a pretty common occurrence in my knowledge database. As an example, all projects I'm currently working on get [[ProjectA/log]] and [[ProjectA/todo]] subnotes. Keeps my TODOs segregated by topic. You could of course do most of these things with typical note-taking apps that rely on the hierarchical file system motif. But the fact that it's up to me to explictly construct my organizational structure in logseq forces me to be more intentional about how I organize my notes.

Perhaps the greatest benefit I've found in using logseq is in organizing the scientific literature I'm reading. At any point while writing, linking a paper I've read (the PDF itself, as well as my notes on it) is a few keystrokes away. This linking allows me to externalize the context with which I know a paper. I just need to browse the backlinks to see all the ways in which I've made reference to it over the entire history of my notes.

It should be noted that Tarn would probably agree with a lot of the criticisms leveled here regarding the UI. The focus so far has been on getting the systems working (and there are many more to go and be further fleshed out: magic, the organic development of economies in the game, world history, warfare, etc). The Adams brothers essentially want to construct a general fantasy simulator capable of constructing rich worlds all on its own. Wouldn’t make much sense to pour tons of energy into making and refining the UI over and over again when the game isn’t even halfway done yet.

I’ve tried mechanicals (Cherry switches), various rubber domes, the IBM buckling spring keyboards, and the Apple scissor switches. Then I was recommended the Niz by a friend, and it’s the closest thing to ideal I’ve ever used. To be honest, owning mechanicals made me realize how gimmicky they are. They distract me from my own writing, and can be a huge pain in the ass to coworkers or close ones trying to sleep. Unusable during work calls. No such issues with these nice electrocapacitive boards. I anticipate I’ll stick with them for a long time.

Totally agree here. Having a functional menu bar on the level of Unity or macOS is an excellent design choice, imo. I've always felt like Gnome just wastes that space, causing everything to be crammed into an application's title bar in an inconsistent way or into a hamburger menu.

> We’re rolling out the Firefox Multi-Account Containers extension with Mozilla VPN integration. This lets you use a different server location for each container.

This is really cool. I wonder how difficult it would be to allow users to wire this up to arbitrary user-specified VPNs e.g. through WireGuard.

I guess. Nothing’s stopping you from remapping ‘ls’ to exa. That’s what I did.

I'm actually surprised by how much snappier it feels. I'm on an i7 2018 mac mini.

This is great. By the way, one of the things that has often discouraged me from filing bugs is not really knowing how to do so. Sometimes, looking at issues on github, I find people provide super structured and clean descriptions of bugs. Then there are the other kind, that the author describes: “it crashes.”

Is there a standard way of describing bugs? What resources/standards have people found useful to refer to when reporting one?

I actually used Athens for several months and then switched over to logseq once I found it. They're quite similar in functionality, though logseq is a bit further along in my opinion (e.g., the task management and time tracking is more advanced in logseq). Which I was fine with not having in Athens. What ultimately pushed me towards switching was the fact that logseq stores notes in plaintext, whereas Athens notes are stored in these difficult to parse .bkp files. Having the ability to easily write scripts to do whatever with my notes is important for me. Also, not a technical point, but the developer of logseq seems like a really nice guy! :) (see https://www.youtube.com/watch?v=DUGV8HL-poQ)

Yep, and you choose where logseq files are stored, so you can use anything you'd like to for syncing. I will say though, I've tried using Dropbox for syncing with logseq, and it ended up creating a lot of duplicate copies due to conflicts. Have had better luck with iCloud/Syncthing (also, many people actually use Obsidian + logseq, so I believe you can take advantage of Obsidian's sync if you choose to). In any case, having better file syncing is something the logseq devs are currently working on.

Logseq does not currently have templates integrated the way Notion does, but it wouldn't be difficult to programmatically create them since they're just plain markdown files. There's a plugin ecosystem for logseq now, so I think it's just a matter of time before someone writes one to generate templates.

I’d check out logseq [0]. Closer to roam than obsidian, but GPL, operates on plaintext files, and supports Org-mode style formatting as well as markdown. It’s getting to be fairly fully featured by now, and the development team moves fast. It’s a joy to use.

[0]: https://logseq.com/

Looks pretty good! I will say, as someone who programs in their day job and has been trying for ages to get into game dev as a hobby, love2d [0] has been excellent for getting started. My github has a few repos of previous attempts at making simple games (in .cpp, .rs, etc) which I abandoned from the amount of work it took.

If you're in a similar boat, I would recommend checking the framework out. Lua's a pleasure to program in and you can focus on the game development itself instead of getting bogged down in the details of rust / cpp. In fact I've been thinking lately about how easy it would be to use it for things other than games -- quick prototyping of graphical simulations, psychophysics experiments, etc.

[0]: https://love2d.org/

I signed up to NYT using a virtual CC, knowing how much of a pain it was to unsubscribe (I'm not in California). When I was through with it, I just paused the card. And I was amazed at how many times they tried charging it before giving up and actually cancelling the account. 13 times, now that I'm looking at it. The Gray Lady isn't as well-mannered as one might think.

> Every qualified person I've heard talk about this said the opposite. "Top-20 school" isn't how professors or large industry labs evaluate a PhD, they will look at the advisor's reputation and the specific lab.

While I think this is technically true, academics care a great deal about pedigree, in the “he/she went to $FAMOUSLAB, they must be a great scientist” kind of way. The most productive, well-funded, and well-known labs tend to be at the top institutions. I wish it wasn’t this way, but that’s how it is. If you want to maximize your chances of being successful in academia post-PhD, then you have to go to one of these places. For industry, I don’t think it matters as much.

It absolutely is. It's just a pretty narrowly-defined backdoor (which could always be subject to widening).

It used to be that iOS was the recommended phone OS if you were looking for the best combination of privacy and security. Even Daniel Micay (the lead developer of GrapheneOS) thought so, 2 years ago [0]. But these ROMs are looking much more mature these days. Anyone have thoughts on how CalyxOS and GrapheneOS compare to iOS in the present day?

[0]: https://www.reddit.com/r/GrapheneOS/comments/bddq5u/os_secur...

This article (https://madaidans-insecurities.github.io/firefox-chromium.ht...) by madaidan goes into quite a bit of detail regarding the security of the two browsers. Things are improving on FF, though.

Excited that there's more progress being done in implementing proper site isolation. Would be great to not feel like I'm losing something in sticking with FF over Chromium-based browsers.

After searching around (and trying Roam, Obsidian, a bit of Tiddlywiki, etc) I ended up sticking with Athens [0][1]. It's essentially an open-source version of Roam, and it's YC funded. I've found it to already be mature enough to be a good replacement for Roam.

[0]: https://www.athensresearch.org

[1]: https://github.com/athensresearch/athens

This is why I wish Marx's thoughts were actually gone over in standard philosophy/economics/history college curricula. For having had such a deep impact on the world, the average level of comprehension about what Marxism even was among well educated people is so frightfully low. The fact that so many conflate it with central planning and state capitalism is a clear example of it. If we ever hope to understand, in a thorough and critical way, why Marxism in the 20th century was a failed project, we need to actually learn what it was.

Both good questions. It could very well be that low dimensionality is simply a byproduct of the fact that neuroscientists train animals on such simple (i.e., low-dimensional) tasks. This paper argues that [0]. As for your first point, it is known that auditory cortex exhibits tonotopy, such that nearby neurons in auditory cortex respond to similar frequencies. But much of cortex doesn't really exhibit this kind of simple organization. Regardless, technological advancements are making it easier for us to record from large populations of neurons (as well as track behavior in 3D) while animals freely move in more naturalistic environments. I think these kinds of experiments will make it clearer whether low-dimensional dynamics are a byproduct of simple task designs.

[0]: https://www.biorxiv.org/content/10.1101/214262v1.abstract

Agreed, it's really cool :). A lot of this is very new -- it's only been in the past decade and a half or so that we've been able to record from large populations of neurons (on the order of hundreds and up, see [0]). But there are a lot of smart people working on figuring out how to make sense of this data, and why we see low-dimensional signals in these population recordings. Here are some good reviews on the subject: [1], [2], [3], [4], and [5].

[0]: https://stevenson.lab.uconn.edu/scaling/ [1]: https://www.nature.com/articles/nn.3776 [2]: https://doi.org/10.1016/j.conb.2015.04.003 [3]: https://doi.org/10.1016/j.conb.2019.02.002 [4]: https://arxiv.org/abs/2104.00145 [5]: https://doi.org/10.1016/j.neuron.2017.05.025

If I’m recording from N neurons, I’m recording from an N-dimensional system. Each neuron’s firing rate is an axis in this space. If each neuron is maximally uncorrelated from all other neurons, the system will be maximally high dimensional. Its dimensionality will be N. Geometrically, you can think of the state vector of the system (where again, each element is the firing rate of one neuron) as eventually visiting every part of this N-dimensional space. Interestingly, however, neural activity actually tends to be fairly low dimensional (3, 4, 5 dimensional) across most experiments we’ve recorded from. This is because neurons tend to be highly correlated with each other. So the state vector of neural activity doesn’t actually visit every point in this high dimensional space. It tends to stay in a low dimensional space, or on a “manifold” within the N-dimensional space.

Sure, but the neural activity is actually low-dimensional (see Extended Fig 5e). By day 4, the first two principal components of the neural activity explains 75% of the variance in response. ~3-4 dimensions is not particularly high dimensional.

Your comment illustrates another way in which these institutions maintain exclusivity -- they hire and admit people that are already in them. I did my undergraduate in a run-of-the-mill liberal arts college. I managed to get into a post-baccalaureate program at a HYPS-level institution, and I'm now doing my PhD at a HYPS. I find it extremely difficult to believe that I would've been evaluated the same way in grad school applications if I hadn't done the post-bac where I did.