I thought about that recently while researching how VCRs work before attempting to fix one. I didn't even think about seeing the actual video signal, I was just curious what the diagonal lines and control pulses on the tape look like. There are many other things as well that would be interesting to look at (all kinds of tapes, all kinds of floppies, hard drive platters, magstripe cards), but unfortunately I don't think there exists a technology capable of visualizing magnetic fields with enough precision.
This is 1970s tech, so of course there was no compression. It's basically a composite video signal etched into the medium. Every revolution of the disc is one full frame (or two fields). So when you look at it, the adjacent grooves represent the same lines of video in different frames.
Now imagine if you picked a Y coordinate and extracted the corresponding row of pixels from every frame of a video file, and stacked them vertically. If there was scrolling text, or if the camera moved vertically, you would see a meaningful image, the same way a scanner obtains a complete image by moving a single-row sensor across whatever you're scanning. This is the same effect, just arising from the way the video signal is laid out on the disc.
It blows my mind they used AI to generate a couple paragraphs worth of rather nonsensical text. It would've taken them about as much time to just write something sensible by hand.
For the Mastodon Android app, I also sometimes see crashes that make no sense. For example, how about native crashes, on a thread that is created and run by the system, that only contains system libraries in its stack trace, and that never ran any of my code because the app doesn't contain any native libraries to begin with?
Unfortunately I've never looked at crashes this way when I worked at VKontakte because there were just too many crashes overall. That app had tens of millions of users so it crashed a lot in absolute numbers no matter what I did.
Well, vendors' randomly modified android systems are chock full of bugs, so it could have easily been some fancy os-specific feature failing not just in your case, but probably plenty other apps.
Usually I'd just look at clusters of crashes (those that had similar stack traces) but sometimes when you're running a very small % experiment there's not enough signal so you end up looking at everything. And oh boy was there a lot of noise.
In an app with >billion users you get all kinds of wild stuff.
So does robbing a bank. But it’s far from the only option. Plenty of indie developers thrive without any VC funding, and I thank every one of them for it. VC funding is essentially a guarantee that if the software isn’t shit now, it’ll be in the future, and that the creators care more about the money than doing something good. Case in point, the deterioration of 1Password.
Sometimes you need to pay the people who made the software. You can't steal during all your life. At some point you have to pay the others for the work they did.
Nobody is saying don't pay your developers. Just that VC funding creates perverse incentives within your business where you are pressured to do what is best for your investors, rather than your customers. But there are other business models where one can earn money and still pay the people working on the product.
Definitely that, a finite scope is good and finished software is beautiful.
But also, most of the modern software is in what I call "eternal beta". The assumption that your users always have an internet connection creates a perverse incentive structure where "you can always ship an update", and in most cases there's one singular stream of updates so new features (that no one asked for btw) and bug fixes can't be decoupled. In case of web services like YouTube you don't get to choose the version you use at all.
I moved to Obsidian after Evernote increased their subscription prices beyond the point I could justify and I think Obsidian is heading down the same path Evernote did. They keep adding more and more features to it when I wish they would call it complete and move it into maintenance mode.
For me, the turning point for Obsidian was their Canvas feature. That was a big move beyond the initial design of it being an excellent editor for a directory or markdown files that supported links and all the other cool things you can do with a basic directory of files and a few conventions. Nothing proprietary, nothing much beyond the directory of files aside from a preferences store. IMHO, Canvas and beyond should have been a new product.
If Obsidian was open source I would have been tempted to fork it at that point.
Joplin is open source but isn't half of Obsidian, it's slow and grouchy, a fairly poor Electron app, but at least I can turn off the plugin that does freehand drawing. It's also has self-hosted sync for free. If Obsidian was open source, I'd have switched already even if it needed a subscription to sync, especially if I could toggle off unwanted features.
The issue is that a finite scope is paid once (and often pirated, so not even that), while subscription SaaS are paid perpetually, so there's more money there.
We should have a system where we continue paying for software that we keep using even if it doesn't change, but that's never going to happen, we will start wondering how we will replace it and cut costs.
I myself started making the same distinction when I talk about these things in English, except it's "social media" vs "social networks". Though I have no idea how to make that distinction in Russian, social "media" never caught on as a term there.
An extra annoying problem about social media for me is that while I can make most of the platforms give me a chronological feed of content authored only by people I follow, most other people see mine in an algorithmic feed. This includes people I have zero social connections with. For example, I just gave up trying to discuss politics on Twitter, because every time I post anything political, that tweet ends up in the feeds if hundreds of people who hold the radical version of opposite views, with predictable results. And there's nothing I can do. I can't opt out of being recommended.
Sure you can. You can not post political things on social networks. They're not doing any good anyway. They're not changing anyone's mind. They're not providing depth or width to the discussion. I don't say this to be insulting, but rather a realist.
My point is that I just want to be able to discuss any topic with my followers without self-policing lest a bunch of anonymous accounts butts into the conversation and completely derails it.
What you're probably looking for is closer to a closed discussion group or mailing list than "social media", which is presently universally-readable, algorithmically-targeted, feed-based, advertising-supported, and increasingly, saturated with AI slop (which itself has replace clickbait and ragebait).
Which reminds me of Kitman's Law: Pure drivel tend to drive off the TV screen ordinary drivel.
I want my posts universally-readable and universally-interactable (that's why I don't like the idea of locking my accounts). I also want to be able to explore the social graph — looking at who follows who, what that friend of a friend posts, etc. It all forms an integral part of what social networks are.
What I absolutely do not want is the platform having any of its own agency. I want a social network that ideally works as a dumb pipe. I especially don't want my content surfaced in front of the kinds of people who would've never found it through their own exploration.
It should come as no surprise, then, that I have a lot of faith in the fediverse.
My evolved view is that there's a time and place for various types of interactions. That's after being a long-time fan of universal readability.
Truth though is that today's Internet is vastly different from that experienced in the 1980s (when I first came online), '90s, aughts, or even the teens. Scale is a huge piece of this, though broadband, mobile devices, advertising, attention merchants, clickbait, and AI have all had their impacts. The Internet (or proto-Internet) of the 1990s and earlier was very limited in access, with soft-but-imposing barriers to entry (selective research universities, some government agencies, some tech firms), which made the experience both "open" and closed. Yes, there was exposure to a large audience, particularly as contrasted to immediate physical space or mass media of the time (print, including early small-scale copiers, amplified audio, radio, television, and telephones). But the total online population would be considered a minuscule social network by current standards --- a few thousands to a few millions of souls in the 1980s and 1990s.
I continue to use some smaller networks today (HN, Mastodon, Diaspora*), and find that they tend to retain at least some of the feel of the forums I was familiar with in the 1980s and 1990s: small, intentional, generally motivated. Ironically, their limited size and the fact that those who are there want to be there is something of a feature. A significant problem isn't so much people leaving as dying, which seems to happen with regularity. (An older population amplifies this, though I've noted previously that mortality at FB/Google scale is likely on the order of tens of thousands of accounts daily.)
The platforms I mention also largely lack agency, which as you note is quite refreshing. I'll note that HN is somewhat an exception, but it's mediated mostly by humans (member flags, moderator actions), as well as some automated rules, though those are largely guided by HN's mission of "intellectual curiosity" rather than attention-mongering.
Factors other than scale alone include broadband (enabling graphics, audio, video, and interactive content, all of which have considerable downsides), mobile devices (making for more distracted and far less nuanced discussion, as well as quite brief responses contrasted with physical keyboards), and the pernicious first and higher-order effects of advertising, manipulation, algorithms, AI, and the like.
I've toyed with the notion of a set of interrelated scopes, some limited and personal, some more widely open, though arranging that formally and as part of a designed system has yet to emerge. I have hopes for that though.
There's also the distinction between a pure social graph and a highly-curated specific discussion or forum. I've tried the latter from time to time with stunningly good results, especially at modest size (< 50 participants generally).
(This comment, as most of mine, was composed at a keyboard, and edited several times.)
Politics is a complex topic. If you want to learn more, social media is not the way to do it. Well reasoned books and essays are. If you want to convince others of your positions, social media is not the way to do it. Personal relationships in real life are.
Again, you seem to insist on an ulterior motive, completely discounting the value or pleasure of conversation. In contrast, reading is a solitary activity. Have you heard of book clubs? People read books, and then they get together to discuss the books.
Hacker News itself is all about reading articles, and then discussing the articles with others. "If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity."
reply