This is the blog part

Thu, 02 Feb 2023 17:11:13 GMT

Twitter is cutting off access to Drummer soon. I already have a complete backup, with revisions, of my Drummer files because I use Roady to sync them to GitHub. But I still used Drummer's Tools → Download my files command just in case. I wish I could help Dave Winer with the transition, but I don't think there's anything I can do beyond just letting him do his thing. I am glad he wants to keep Drummer running!

Fri, 03 Feb 2023 03:44:31 GMT

Sometimes I run the roomba just because it makes it seem like there's another live thing in my apartment, and that is science fiction as fuck.

Fri, 03 Feb 2023 04:53:40 GMT

You start with no cassette decks, and you can't play any tapes. And then you have five cassette decks, but you still can't play any tapes. Everything is broken, always. Sometimes it's because cassette decks are full of super-complicated parts, all 40 years old, wearing out. This is also how software development is, all the time. Something is always broken, and you have to fix it, before you can get to the good parts. Sometimes it's bugs. Sometimes it's billionaires scraping the friction from the gears of their money creation machines, optimizing away all the stupid extra edge cases, what are those parts and people for anyway, you do not need those, because they do not make me money. Most people are not like that. We do not break things on purpose, for thousands and thousands of other people, just for our own gain. There is a type of person who will do this in a heartbeat, and say you are stupid not to do it that way, too.

Fri, 06 Jan 2023 01:45:01 GMT

The best part about AI is that it is insane, and stupid.

Thu, 05 Jan 2023 00:07:48 GMT

Introducing Dababdra. You never knew you needed it. But now you do.

Recording the new year

Mon, 02 Jan 2023 23:10:29 GMT

(11 min read)

A cassette recording is not a copy. You can make an exact copy of a digital file, and each time you load that file you will get the exact same series of bytes. A cassette recording is literally a recording of a specific event. It is a record of reality, as it exists at that moment. It's the intersection of the audio signal, measured by a specific source device such as a Super Audio CD player, and captured by imposing magnetic flux onto a particular tape by a particular cassette deck. The recording is a captured performance of specific set of conditions, each of which contributes to the final sound.

Every cassette recording is literally unique in the universe. The cassette deck is a physical device, and mechanical devices are not perfect. Tape heads, capstans and pinch rollers get dirty. The motors spin as consistently as possible, but there is always wow and flutter. As a painter may work on a stretched canvas, or paper, or cardboard, the tape itself is also a physical medium. Each has its own characteristics, such as how much signal it can take, how much noise the tape itself provides, how much bias it takes to properly orient the tiny magnetic particles.

The final sound, at playback, is another performance, created in real time by a physical device, a cassette deck which measures the magnetic flux on the tape. This signal is amplified, and used to drive a preamplifier, amplifier, and then speakers or headphones. Every time you listen to a cassette, it is a specific, particular, unique, individual experience. A performance, a rendition of a specific reality, created in real time.

This weekend I have been making recordings from my NAD 6300 Monitor Series Cassette Deck. I spent the past several weeks aligning this deck. Most of the adjustments you make are related to recording, and the pots and variable capacitors are extremely sensitive to the smallest movements. Sometimes you're just applying pressure, rather than turning. And once you've made that feather touch, you record a test signal for ten seconds, adjust the test signal to a higher frequency, record another ten seconds. Then you rewind the tape, and play it back.

Measurements are taken from test points at the input to the Dolby circuitry, at the minimum levels you can record. Input around 6 mV results in a measurement of around 40 mV. You watch values change on the multimeter, which you have set to maximum digits, and adjust the meter's rate so it gets a steady value on the first frequency, and set that as your zero dB reference. Then you watch for the change in voltage at the higher frequency. You want the difference to be as small as possible. To be in spec, it should be within 0.5 dB. If it is not, make the adjustment again, and redo the recording.

Each individual nudge-record-adjust-frequency-measure cycle takes a minute or so. There are more than two dozen variable resistors and tuning filters to be adjusted, and it can easily take a dozen or more attempts before you get one into spec. These are resonant circuits, which means they have a specific point at which they are in tune, and you can't just turn a pot until it reaches maximum. Maximum will be somewhere in between the extremes, and that point will shift as you tune the other parts of the circuit. Some of the steps affect adjustments made in previous steps, so you go back and fine-tune those, another half-dozen more adjustments to something you've already tuned.

If you are patient, you can get all the adjustments to within the specifications listed in the service manual, and your deck will perform as if it were new, fresh from the factory. In that condition, the NAD 6300 is famous for being able to make some of the finest recordings possible. If you are servicing someone else's deck, that's where you stop. It's performing to spec.

But for your personal deck, you can be extremely patient. You can get all those 0.5 dB differences down to 0.05 dB, or even less, and your recordings will be even better. (And if you're just learning, as I have been, you will do the entire sequence multiple times, just to be sure you understand what you are doing, and to ensure that repeated measurements let you know for a fact that you got it as close as is physically possible, and that you didn't stop just because you got tired or bored.)

The reason there are so many adjustments to make is that, in addition to having do do everything twice, once for the left channel, and once for the right, you have to do everything three more times, because there are three main types of cassette tape, and multiple adjustments to make for each of those.

To ensure that the magnetic particles, embedded within the thin, stretchy plastic carrier tape, take the maximum signal, the recording circuitry will add a high-frequency component to the signal. This high-frequency energy, outside the range of human hearing, is bias, which I think of as a way of overexciting the particles so they settle down more easily into the pattern we want. It establishes a baseline of energy, atop which the musical fluctuations ride.

There are three main types of cassette tape. Type I is "normal", type II is "chrome" and type IV is "metal". (Type III exists, but is rare.) They differ mostly in the kinds of particles embedded on the tape, and each requires a different amount of bias, and therefore a separate set of adjustments to the cassette deck. There is not just one "type I" tape, of course, there are different brands, such as TDK and Maxell. Each manufacturer made several variants, aimed at specific levels of performance and price. So a TDK D is different from a TDK AD, and differs from a Maxell UR, or Maxell UD XL-I, even though they're all type I tapes. Even the same type, from the same brand, can vary depending on when it was manufactured.

But you have to align your deck to an actual, physical, cassette, so you have to select one brand, and one formulation from that brand. Your deck will be aligned perfectly to record on that kind of tape. But because you will want to record on other brands and formulations, the NAD 6300 has a fine bias adjust knob on the front. If you're recording on the tape you used to align the deck, you leave it alone. But if you're recording on some other type of tape, you carefully adjust it for accurate treble reproduction.

I have signal generators, distortion analyzers, and spectrum analyzers, so eventually I'll get around to doing it right. But for now I am doing this part by ear. The other, critical, adjustment you make by ear is setting the recording level. This can be done with lab equipment, but we don't listen to test signals.

Sources vary in loudness, musical dynamics, high-frequency energy and bass. Tapes vary in their ability to take the magnetic flux. Some tapes can be recorded hotter before saturation distortion becomes audible. You want to record the maximum possible signal the tape will take, because the tape itself has some noise, a hiss, you can hear all by itself. The higher the signal you can put onto the tape, the louder the quietest parts of the music, the less hiss you can hear.

In addition to fine-tuning the bias circuitry during align meant, and the front-panel fine bias adjust, the NAD 6300 has two more features designed to maximize recording volume: HX Pro, which was an improvement to a Dolby design made by Bang & Olufsen, and DynEQ, which came from Tandberg. There aren't many decks made with both, which is one reason the NAD 6300 is unique. The result is that you can record a hotter signal to the tape, without distortion, than you otherwise can.

Cassette decks generally feature noise reduction, usually Dolby B and Dolby C. This is a process that modifies the signal before it gets to the tape, and there's a complementary process during playback that undoes the changes made during recording. The result is that the hiss, a high-pitched noise inherent in the tape itself, is pushed out of audible range.

Dolby B recordings are more common (the deck in my car can only decode Dolby B), and due to how the process manipulates high frequencies, some people like to listen to a B recording with the decoding process disabled. It makes it sound brighter. Dolby C recordings feature better noise reduction, up to 20 dB instead of the 10 dB provided by Dolby B, but you do not want to listen to a Dolby C recording without the Dolby C mode enabled, nor do you want to play it in Dolby B mode.

When you make a tape, you are recording a specific musical performance, a particular reality, as measured and imprinted by a unique combination of tape medium, capabilities, adjustments, settings and tunings, each of which you select. It is an individual act, resulting from a series of judgments you have made. You decide what music to record, onto which tape, using which device, and apply your knowledge, hearing and expertise, to create a real, physical thing, which only exists because you made it. What will you want to listen to next, the next time you play this cassette? What will you put on side one, what goes on side two? If it doesn't fit, and it never does, which songs get included, and which skipped?

I've spent New Years making some cassette recordings. They are on TDK SA-90, which is not the type II tape which I used to align the cassette deck. So I've tuned the fine bias control. I also took some time to set the recording level. The NAD 6300 is a three-head deck, which means as the tape goes by, one head erases, one records, and one plays back. This, along with the input monitor switch, lets you hear both what you're recording, and the actual playback from the tape, as you're doing the recording. However, playback can be a different volume than the source, and louder things sound better, so you still have to adjust the playback volume when switching back and forth to hear the differences properly.

These recordings are being made without noise reduction, because I discovered I could hear the difference in the brushed snares of a symphony orchestra going full blast, which is simultaneously a very loud sound, and a very subtle one. The detail was lost using Dolby C. That source was a Super Audio CD of the RCA Living Stereo 1959 and 1961 recording of Arthur Fiedler's Boston Pops Orchestra performing George Gershwin's Rhapsody in Blue. My goal is to record this performance, best I can, as if I were there, and recording the very event so I can listen to it later, as if I were there, crystallizing this experience into a single, physical, thing: the cassette.

A cassette is more than just the tape, that thing you put into the deck, which is a precision-made shell that contains the actual plastic tape, several spinning hubs, a tiny spring-loaded pressure pad, and slip sheets to minimize friction. Cassette tapes also come with a plastic case, which holds both the cassette and the J card. The J card identifies the tape and describes its contents.

In prerecorded tapes, it's a designed object, created with attention to colors, proportions, and appearance, just like the music itself, but generally by a graphic designer, rather than the musician. It's another representation of what is on the tape, in a different medium, print rather than sound.

It should tell you what's on the tape, so you can tell it apart from the other tapes. It should also tell you the settings for the tape, whether you play it back with noise reduction, and what type of tape equalization to use.

The J card can also contain production notes about the recording, such as the level the tape was recorded at, and any adjustments made to the fine bias. These can be useful for making better recordings next time.

The J card should tell you what's important to know about the tape. And it should be beautiful, in its own way, like the music. Like all graphic design, it should be pleasing to look at, and convey specific information. For these New Years recordings I chose to use photographs of the deck itself, in the act of recording, showing the positions of the knobs, printed on 4x6 glossy photographic paper, carefully trimmed and folded. I love the way these tapes sound, and I love the way they look.

In a very real sense, they literally are New Years Day, 2023. This is just like blogging, except when you're done, you've created a physical object, a cassette tape, a thing that exists, a captured musical event, a work of art.

Why Twitter 3.0 is going to be the greatest

Fri, 30 Dec 2022 01:24:53 GMT

(9 min read)

We have finally figured out how to build and run a social network that maximizes freedom, safety and also profitability. We are building Twitter 3.0, on a handful of tested principles, to be the final social network.

First, tie every account to a person. No fake people, this is important because fake people are really the worst. Bots are fine, because they belong to someone real and set a flag saying so.

Fake people are just garbage data, trying to infect humanity with something terrible, which is generally stupidity so we will all fall easier for someone or other's scheme. So no fake people.

You can make up whatever name you like, of course. You can have sock puppets for your various moods, but they’re aliases tied to the same main account. This allows for legit bots, pseudonymity, role-playing, masking for safety, avoiding being dead-named, and all the other perfectly reasonable reasons people have for not wanting to use their full legal name just to have an opinion, or create things and share things.

Then you list the bad words you can’t say. This is a pretty short list, like just a few thousand things. The key is, the list needs to be comprehensive and vague so it can apply where it’s needed. I am kidding of course, that is how you build something that exists only to deny responsibility and just get through the end of another day.

What you really do is have a sensible set of guidelines, like no fake people, and apply them where they're needed.

But how do you know where it’s needed? This is the huge thing, right here. This is what makes or breaks your social network. And we've finally figured it out.

There is one key lesson you can take away from every single successful site that deals with making user content available to other users. This is a lesson learned at interest-specific forums, Reddit, Metafilter, and many other online communities. You have to use humans.

At scales such as Facebook, it is done with subcontractors in cheap places. Smaller communities, such as Reddit, are watched over by volunteers within the community. But you have to use humans.

I can’t stress this enough, algorithms are only going to get you so far. There is no substitute. Humans are very good at figuring out how to navigate around the edges of algorithms. We use puns and misspellings and cultural references, and we mutate it all faster than software can adapt.

In Twitter 3.0, automatic anti-fake measures, combined with a properly implemented flagging system, along with straightforward, easily applied moderator tools, ensure an approach that has a well-documented history of working well in practice, to ensure quality discussion with minimum drama. All at scales that might surprise you.

Now, what differentiates a small community such as Reddit, or Metafilter, from a real social network like Twitter, is that there is only one conversation at a time. Everybody is participating in the same thread, so there’s a consensus reality going on.

You can count on people flagging things that are outside the norm of a given space, ie, this-person-is-out-of-line, is something that is outside whatever norms usually hold for a particular space. Even in a social network you have myriad simultaneous conversations, you don't need a huge number of types of reasons for flagging. The number of reasons should match the number of rules you have.

Flagging just denotes, I saw this and think this should be looked at by someone. It works no matter what kind of community you have. What it doesn’t handle is the little toxic cesspools of conversation, where everyone is in on the shitty behavior and they love it because they are getting away with it. The problem is, the little cesspools are private—and they should be.

But we all have an interest in maintaining the safety and freedom of this new social network. So you handle this in two ways.

First, you give people who aren’t members of the group the ability to flag content posted in private groups. This can involve a slight piercing of the veil of privacy, in that the content of the messages may be occasionally displayed to other safety-minded volunteers on the network. But the identity of the problematic participants can be obscured until really necessary, and after discussion with actual humans about when this is really necessary. And of course to ensure protection against retaliation, flagging naturally remains completely anonymous, as do appeal and final review.

The other way you handle these private conversations is that you provide people an incentive to flag content in their own private conversations they otherwise might not. You grant privileges, or even daily cash back, and people will really want to step up and do the right thing.

I don't have to tell you, of course this can absolutely be tied to what we are calling an "ethical NFT", which will absolutely be exchangeable for something of value at some point in the future for a completely plausible reason. You can trust this plan, because it comes from a real billionaire. Not one of those imaginary-money bilionaires.

We all love codenames for things, it helps you focus when you're building it, and it helps focus people's attention, it's like a mini brand name. Because this structure is specifically designed to maximize freedom, participation, safety and stability, we've come up with what I think is a fantastic name for this part of the project. So incredibly excited. I really like the way it rolls off the tongue. Very snazzy.

The key is stability, security, safety. Think stability, think status quo, think rock solid, think conservative, think stasis—

Stasi. It’s for you!

Now, this next part is critical to our goal of bringing social networking into the next era, beyond what's already been possible. We are going to amplify the possible, we are going to lift the possible to another level. We are going to unite the online with the offline.

We are going to maximize independence and freedom right into the real world. This amazing country was built on the power of the individual, working alone, standing astride history with hammer held high, declaring, “Here we shall build our homes and our liberty.”

Twitter 3.0 is going to be true to that spirit in ways no one has ever seen.

To maximize freedom and independence, we maximize the freedom and independence of the individual. Remember, our members are real people, with as many names as the want, readily available to preserve anonymity and maximize communication. People adopt personas for many reasons, and one of those reasons is to tell the truth to power.

You’ll be able to participate in a way that ensures that no one will ever know it was you. You’ll be able to communicate in real time with similarly-minded, freedom-loving individuals, and you won’t even need to know their real name.

I’ve already described how moderation works, so this anonymous network is also guaranteed to be completely safe. We use end-to-end cryptography and conversations are permanently available to everyone involved, so everyone can keep in touch and know what’s been going on. (Don’t worry, if it turns out later you need to, with one click the entire thing can go away, permanently, because we don’t keep any backups.)

Twitter 3.0 has also got some really great features for families, that help keep us safe, and find new things. We call this part "Theme Park Mode".

Theme Park Mode lets you share a detailed, 3-dimensional topographical map overlaid in real time with the exact location of every member of your family. Augmented reality features let you scan the horizon and immediately see visual markers telling you where your family members are—just like a quest in a video game.

Need to tell the twins to head to the roller coaster, while Dad needs to pick up some snacks and then meet somewhere else, all at the same time? It’s super easy to split out little groups like this and still keep everyone in sync.

Real time push-to-talk; live task assignment, update and completion (“make it chocolate!”); instant note-sharing; inventory tracking (“who has the baby wipes?”). And voice-activated mode for smart assistant earbuds, for the inevitable times when your hands are full. Active noise cancellation lets you hear your family even in the middle of crowds.

Naturally, Theme Park Mode works with all kinds of groups, not just families. It’s going to be fantastic for church groups and school field trips. It's going to be fanatastic for tourists, and experiential artists, or even just heading down to the mall with a few close friends.

This is another lesson we're adopting from smaller, interest-bound communities. It turns out people like to meet up, even if they’ve never met. Getting together in real life, to talk about things you care passionately about, even if you’ve never met, and may never meet again? Well, that’s a special human connection that we’re proud to help enable. No matter where it takes us all.

Along with Theme Park Mode, we’ve got big plans for a little thing that we’re calling Quest Mode. We are going to empower anybody to build an augmented reality game. You’ll be able to build your own treasure hunts, and pretend spy thrillers, secret monster hunters, you name it.

We've got full, high-resolution, 3-D maps of every street, every sidewalk, every park, and every building in major cities around the world. We’re launching with a salute to every state capitol, and major metropolitan center, which means you’ll be able to lay out your puzzles and goals using real buildings, real doors, real windows.

And because anybody can make one of these games, and anybody can participate, practically anything can happen! You can even update your game in real time, change things up to keep your players on their toes.

And our patented algorithms will help your players, your families, your friends, your neighbors, even thousands of people you’ve never met and will never meet again, all come together to participate in something amazing. Maybe even world-changing, history-making! Wouldn't that be something.

Of course, we haven't forgotten our commitment to safety. And here's where we have a unique opportunity all of human history: Tesla.

Self-driving capable cars like Teslas are just bristling with sensors, and you don’t need to be sitting in one to make use of it. People are going to love how safe this makes everything.

Even when your kid’s cell has run out of battery, you’ll be able to locate them anywhere. Even if they’ve gone behind closed doors, you’ll know just where those doors are.

Because at some point, on the way there, they walked past a Tesla. Or a Tesla drove by, looking really cool, like a Tesla. And that Tesla noticed. And told Twitter. Possibly also SpaceX.

(It definitely told the AI that remembers everything that ever happens to everybody, and looks for patterns to be exploited, in service to Its Holy Owner, which I just totally made up, and isn't a real thing a truly godlike billionaire would order built.)

Twitter 3.0 will make it so easy to send someone trusted to pick your kids up, or whoever you're looking for, and escort them right where they really belong. It's like knowing everybody around you, personally, and always being able to call on people you trust.

And because humans look out for each other, it’ll be easy to tag houses in your neighborhood so everyone can see where to get the good candy on Halloween, or who needs an extra little watching after. Never know who might need looking after. You can keep a list, subscribe to lists of people you trust. People you can count on. Count on to be there, and step up, and do what needs to be done. To whomever it needs doing.

The essence of this project, the emphasis is on mobile. Think mobility. Flexibility, maneuverability. Being at the right place at the right time—hundreds of places, all simultaneously. Staying in connection with everyone and everything around you, yet still able to focus, and move, stay in motion, stay mobile. Motion. Mobile—

Mob. Join it!

Twitter 3.0 is going to be everything you love about America, turned up to 15. Plus crypto, don't forget that.

Why again is Mastadon not just DNS and RSS?

Thu, 29 Dec 2022 05:09:47 GMT

(1 min read)

I am wondering why I should have to do anything like set up an account on a Mastodon server to be able to say, "follow me on Mastodon at my-name@my-server.com" if the blog part has an RSS feed.

There should be a DNS entry I can maintain that could point my-name@my-server.com to an RSS feed, or some other simple, plain file format that contains a list of recent status updates. I think that takes care of everything you need to follow someone, an easy way to get to a known place for a list of updates for an identity.

The rest is federation, which is moderation, synchronization and caching, which is all definitely useful. But not literally necessary to the task of saying "hello, here are the latest posts associated with my-name@my-server.com".

I guess this post is part, "why are we redoing it again", again, and part, "why should I have to do anything other than have an RSS feed to be a participant in the larger conversation", and part, "look at the virtuality friction".

Virtuality friction

Thu, 29 Dec 2022 01:58:15 GMT

(3 min read)

Some days, you spend most of it waiting for the test to fail. Finally, at the right spot, for the right reason. Now you know why it's broken, and now you can fix it.

If you are also writing the test infrastructure around that code, the code that tests the code, then it is harder to be sure about where the problem is. This is why I like test equipment. You can guarantee it will perform to spec, and as long as you maintain it, it will always perform to spec. If it doesn't, it's out of spec! It's beautiful, really, how it all works out.

Automated testing is a new thing in software development, by which I mean, it is something from the most recent 20 years of computing, rather than the first 20 years of computing.

There was a period in between where things changed pretty slowly, because performance just wasn't good enough, no matter how fast the hardware changed, you still had to rewrite your software to take advantage of it. You can think of this as friction between the performance of the intended virtuality and its ability to be expressed in hardware and software. This friction is the amount of time you spend optimizing for the particular platform, the cost of implementing in your code a feature that in later generations would be provided by the system.

Longer word lengths, and more registers, and access to large amounts of memory, and device access, and file structure, network access, graphics generation, fonts, animation, state machines, garbage handling, request handling, response generation, pipes, handles, sockets, key-value storage, structured data storage, multiple-host synchronization. Each layer reduces this virtuality friction. Each layer enables other layers.

If you need a layer, and it is not yet part of what you can count on, you are building it yourself. That is the friction. At one point, you needed to manage the contents of registers yourself. Later, you'd write your own random number routine because your game needed that.

Automated testing is the layer that allows you to know for a fact that you haven't broken anything. You can't fool the compiler. And you can't fool the test. If something breaks, you add a new test, and then fix the broken thing. Passing the test means you fixed the bug, and since it's a test, you can run it again, automatically, and know for a fact that you didn't break it. Anytime you rely on manually verifying something worked, instead of having a robot do it for you, eventually you will forget to keep verifying it, because we expect things to never break, so why keep checking? But eventually it will have broken when you weren't checking.

Now, you can create applications that are websites that are social connections that are protocols that are just literally when you get down to it, just a couple key things, a handful of short strings defining what kind of information to look for, in what place, and how to build these strings. For example, RSS or json.

The actual mathematical structure defined by these standards is extremely simple, and the entire rest of technology is built around generating and manipulating these things in all the ways we can think of. This is where we are currently paying most of the costs of virtuality friction.

As it becomes easier to treat working with these strings as fundamental capabilities, ignoring things like in-memory, on local disk, local network cache, same-host service, local network service, caching proxy to load-balanced cluster, remote server, pooled connection, database index structure, query structure, query optimizations, cache optimizations, and strategies of all sort, we can finally build other things on top of all that. That of course becomes the new level of virtuality friction.

Why are there billionaires

Thu, 29 Dec 2022 01:59:40 GMT

(1 min read)

I think this is how billionaires have always happened: They are in between the gears, like grease, making things a little smoother, but mostly extracting wealth from the changes the rest of society would have done anyway.

They always believe they are great men, who deserve to be where they are because of who they are. But they are not.

Someone as great as they believe themselves, as great as their reality has assured them they actually are, surrounded by nobody who could do anything but validate every fleeting whimsy as sheer genius wisdom, fulfill every request without considering costs, listen, and excel in one key ability: knowing exactly what to say to the billionaire, to stay close to the billionaires, someone as great as all that would have found a way to contribute to society without needing to grow ever richer, forever, always gaining power, always needing more, dreaming of breaking down society itself, to reshape it for their own use, to piss on contracts and promises, and laws and regulations and fines and criticism.

But they never do, they are always greedy, and whiny, and need more, always more, and cannot stand anybody who says no, or even hesitates. It is literally how they are built as human beings, and that type of human being naturally ends up where they are, controlling the resources they do, because we let them, because it is literally how we are built as human beings. The rich will always be among us.

(What actually would happen if we nationalized all the billionaires?)

Wed, 28 Dec 2022 06:36:26 GMT

Just realized how to add buttons to the blog part that say "MORE ANGRY", "LESS ANGRY" and have AI-generated text where the prompt consists of the original post, except angrier, or stupider, or rhyming, etc., iterated as desired. Like a knob you can turn up or down, a filter, applied to the text.

Fixing it

Wed, 28 Dec 2022 03:48:27 GMT

(2 min read)

"I can't mess with anything or else it will break." Welcome to my world, kid. Everything wants to break, all the time. Sometimes you can fix it. You never know.

Tonight we listened to music played by my other new tape deck from 1986, the NAD 6240 Stereo Cassette Deck. This lasted for as long as the kid could stand, which was not as long as I'd have liked, but longer than I think would actually cause him to drop dead from not being able to put the Nintendo Switch on the big screen. So not too long. But long enough for me to feel pretty good, because I fixed it myself.

The first tape deck I ever purchased was an NAD 6240. I am pretty sure I got it on discount because it was the previous year's model, but I selected carefully. It was the best cassette deck I could afford. I really wanted an NAD 6300, but they were several months' rent. Later I traded that deck to a friend for an NAD 6325, which I never liked as much (the deck, not the friend). I looked up the specs recently, and it turns out that deck rolls off at 15 kHz because there's no MPX filter switch. So apparently I could hear that difference.

I don't trust my ears too much, because it turns out my ears are kind of slutty. They like the sound of everything. I get used to the sound of something, and I can find something to grab onto and marvel at, no matter how high fidelity or digital or pure it isn't. You can get everything you need from music from a mono AM radio. Music is magic that way.

I fixed the playback on the NAD 6240 thanks to a thread on Tapeheads. The output was down around 6 dB in the right channel, and even more on the left. I replaced a capacitor that had gone dead, and now playback works great. Recording is another matter, it is completely broken in that way, so I have more to do. But tonight we listened to a TDK AD60, recorded on the NAD 6300, of Melos, Vassilis Tsabropoulos, Anja Lechner, U.T. Gandhi. Lots of piano and violins. I haven't done any mechanical work such as changing the belts, or lubrication, on the 6240 yet. Its wow and flutter are already basically within spec, 0.08%, so I could leave that alone if I wanted. (I will not be leaving it alone, of course.)

This is the other system, which is a pair of Alexis Sound monitors, which feature crossovers by Dennis Murphy, driven by a TU-8500 tube preamp I built from an Elekit, with power and gain provided by a Blue Circle Audio BC204 hybrid power amp. I call this the mono rig because the record player hooked up is outfitted with a mono cartridge, but it's stereo. And it sounded really, really good. The TDK AD is a Type I cassette, not the fancy chrome, and the 6300 that made the recording was calibrated to Maxell UD XLI, and I haven't done any adjusting to the bias, nor really tried to optimize recording levels yet.

But it sounded really, really good. And I fixed it myself.

Boxing day

Tue, 27 Dec 2022 06:49:55 GMT

(6 min read)

I am finally sitting down in the good chair, between the big speakers. It's been a few weeks of learning about cassette decks and how to set them up properly, and then doing it, over and over, until I get it right. I'm listening to a recording of Norah Jones from a Maxell XL II tape recorded from a Super Audio CD directly from the Oppo BDP83-SE to the NAD 6300 I've finally declared ready to listen to.

The speakers are being driven by a pair of Class A amplifiers based on the Nelson Pass F-4, which I built myself several years ago, and refer to as the Wild Side and the Weasel Side. Class A means they sound great until they get hot to the touch, at which point they sound amazing. They're my winter amps. They don't provide any gain, just current, lots and lots of current, so I have the Parasound P/LD-1100 Line Drive to swing the volts required to make the big speakers sing. They are six feet tall and just two inches thick, like great golden monoliths, and they are Magnepans, and listening to them is like wearing headphones that are six feet tall.

The audio on the Super Audio CD is recorded in DSD, which is not like ordinary PCM audio at all. You can get the digital output in PCM form, it's 88.2 KHz and sounds great, but it's not what's on the disk. DSD audio runs at a clock rate in the megahertz, and it doesn't sample the voltage, and convert it to a binary number corresponding to the scaled value, which is what you normally think of when you think of how a DAC works. DSD just records stream of ups and downs. Is the voltage rising, or falling? There's your ones and zeroes. You don't need 16-bit words. What you do need, is a special player. You cannot listen to a Super Audio CD on a regular CD player. There may be a regular Red Book audio layer for your regular CD player, but you will not be listening to the DSD version.

You also can't rip a Super Audio CD to a digital file. You can intercept the digital audio going through the HDMI output, and get that 88.2K stream, but you cannot get the DSD bits. That's because we're all dirty filthy pirates, of course. We brought it on ourselves.

So the only way to listen to these discs is on the Oppo, which they stopped making because making great DVD and Blu-Ray players was distracting from cellphones. And since the only thing you get as output from a Super Audio CD is actually analog audio, why not record that to a cassette tape, and then listen to the tape instead? This sounds insane. The entire point of digital audio is absolute fidelity to the original source. No distortion.

But actually, we like distortion. Nelson Pass has noted that about half his customers preferred a little second harmonic, and the other half preferred a little third harmonic. And because he is a genius at this, he can tune his amps to produce extremely pleasing flavors of distortion. The F4's are mainly second harmonic, which is generally considered warmer, versus the more crisp sound you'll get with the third-harmonic variants. They don't have much distortion anyway, and running them as mono blocks cuts that by another 3 dB.

Audiophiles, at least those of a certain nervous sort, focus on short signal paths, purity and low distortion. Even tone controls, your regular treble and bass knobs, are considered bad, because they change the sound, on purpose. The fewer changes, the better. This is what is referred to as the "absolute sound", which is precisely how the music literally is, which is a real thing that you can totally reproduce, or get at least keep as a shining beacon on the horizon, as you perpetually upgrade your gear, and settle down for some critical listening. Discrete circuitry, no chips, no op-amps even. The less, the better. (This basically describes the F4's actually. They are nothing but a handful of transistors, the power supply, and lots of heat sinks.)

The thing that cracks me up, though, is just how mangled the sound was on the way to that recording. There are so many op amps in pro audio. Everything's been filtered, and boosted, and cut, and panned, and impedance matched, and run through so many transformers and coupling capacitors, and resampled, and reclocked. And that's before you even get to the mastering part, where everything gets another sheen of tiny boosts and cuts and feather touches of compression, and finally it sounds like natural, pure music.

Using a cassette deck can be considered a way of adding distortion to a recording. In fact, several of the steps in the recording studio rely on the specific distortion added by tape recording, so there are digital plug-ins that apply this distortion right in the DAW, lots of them. (And, I haven't looked, but they probably all model the same handful of classic tape machines. There is magic in using the exact same equipment used to make famous recordings. Maybe the luck will rub off.)

When I get new equipment, which usually is actually old equipment, from eBay, I like to fix it up so it matches the factory specs. Test equipment generally has great documentation, with step-by-step instructions for getting it back into spec. Follow the steps, and it will perform as it should. This, for me, is one of the fun parts.

The NAD 6300 has hundreds of little fiddly parts, and dozens of fiddly steps, to get it working properly. I am still trying on the wow and flutter. Wow is slow changes in tape speed, generally below 6 Hz, flutter is the faster stuff, generally up to around 200 Hz. It's basically the analog version of jitter. You can really hear it with piano, if it's bad. This deck should be able to do 0.025% weighted JIS, which is really quite good, and it's at 0.05% right now, which is really only excellent. So there's still room for improvement there.

But the electronics are on the dot, they have been dialed in, and dialed in, and gone over, the entire 25-step process, repeated several times from scratch. The entire purpose of this device, now, is to add distortion, but as carefully and precisely applied as possible. So some steps you just do over and over, seeing if you can get a difference down to 0.01 dB, even though it was in spec back when it was 2.5 dB, ten minutes ago.

And while I already had the HP 8903A distortion analyzer, and the Meguro MK-668D wow flutter meter, and the Keithley 2015 THD multimeter, and a bunch of other really quite neat, and not boring at all, test equipment, I did not have an Information Terminals M-300 alignment gauge, nor did I have a bunch of test tapes with various precise signals recorded with precision, nor a torque meter cassette, so it was fun getting that stuff, and learning how to use it, too.

It's been several weeks of evenings and weekends, and now I'm finally sitting in the good chair, listening to Thelonius Monk, and his gorgeous piano. It sounds pretty good. I did good. I am happy.

Sat, 17 Dec 2022 21:00:58 GMT

Nearly every one of the several hundred parts that go into a cassette deck is tiny, and will go flying with a little "ting" sound if you grip it too tightly with the pliers. If you're lucky, you'll find it in the carpet after an hour of searching. If not, you keep looking. I hate little fiddly things, always have. I am not mechanically minded. I break pretty much everything, often while simply removing purchases from the packaging. I've lost this little clip a half dozen times now. But I have decided that I am going to do this. I am tired of not completing projects just because they are hard. Everything is hard.

Fri, 16 Dec 2022 15:09:59 GMT

Recent bans of journalists and non-Nazis demonstrate that if you annoy Elon, you can lose your Twitter account. I need my Twitter account because Drummer uses it for login. If I lose my Twitter account, I won't be able to update the blog part. Fuck you, Elon. You fucking asshole.

Sat, 10 Dec 2022 04:52:17 GMT

Apparently FeedLand's "my feed" isn't actually my feed, and is intended just for making comments about FeedLand. So I guess I will have to come up with some other way to make quick posts to the blog part from my phone. Disappointed.

Sat, 10 Dec 2022 04:57:10 GMT

As a proud union member who does not cross picket lines, turning off the New York Times feeds at the root level of FeedLand feels like the Musk approach. Unsubscribing from them on your own feed seems better to me. If there’s a social aspect to using FeedLand, it is users following each other’s curation judgments. Not the owner of the platform using the platform itself to express personal judgments.

Thu, 08 Dec 2022 01:19:33 GMT

Driving the kiddo to school this morning and forgot to bring any tapes to listen to, so I had the radio on. “What’s that sound?” he asks. “Is it on purpose?” He has never heard static before…

Brief updates from pmg

Tue, 06 Dec 2022 18:23:52 GMT

Experimenting with FeedLand's "my feed" feature. Added an RSS node to a scratchpad tab in Drummer, expanding it refreshes the posts from my feed. Copy-pasted the nodes here, let's see how they look:

Just got a brand new NAD 6300 tape deck from 1986. Listening to a prerecorded Guardians of the Galaxy Awesome Mix. Holy moly this thing sounds amazing.

Listening through the craziest chain. NAD 6300 cassette deck into TAC-8 Thunderbolt audio interface, using Logic Pro input monitoring, run through a Waves plugin which reproduces in binaural audio the acoustics of the actual Abbey Road studio mixing room, via Denon D7200 headphones. It’s analog, it’s digital. Have spent no time tuning, calibrating or aligning. Sounds amazing.

Now to figure out the easiest way to get these posts into my Drummer blog.

Looks like the easiest way to get the posts from my FeedLand feed into Drummer is to just subscribe to my feed in Drummer by creating an RSS node, double-clicking it, and then copy-pasting the refreshed posts into my blog. Not so bad. Getting closer to my ideal scenario of being able to update my Drummer blog with quick thoughts from the phone.

Superhero series that probably exist

Sun, 27 Nov 2022 19:29:35 GMT

Uncle Ben is actually a womanizing cat burglar and con man with a gambling problem. He's kindly to his family and friends, unless he's been drinking.

Batman is a vampire. His car has really cool red accent lighting. He fights crime but he also must feed.

Superman does manual labor, for pay.

Peter fights crime because he must atone for the things Uncle Ben made him do. And the things he makes him do now. That voice.

Aunt May is full of spiders.

Upgrading my system to maximum hassle, with a cassette deck

Sat, 26 Nov 2022 21:52:47 GMT

(4 min read)

Because I like test equipment, and I like audio equipment, and I like fiddling around with knobs and dials, and I like music, and I like learning new things, and I like the things I liked when I was younger, I have bought a cassette deck from 1986.

It is cold and dark, so we take the Jetta wagon to school instead of the fun car. This car has a CD player, but it also has a cassette player. We've been listening to the Traveling Wilburys and the Awesome Mix tapes from Guardians of the Galaxy. This has been fun, and part of the fun has been showing off all the little things about how it used to be in the olden times.

There is no song navigation, which means you can't pick a song from a list. You can't skip to the next song, either. What you get instead is called "fast forward", which is not fast, and "reverse", which is also not fast: it can take like ten or fifteen seconds to skip a song. You can never be sure exactly how long this will take, so you need to stop and play every few seconds.

Also it always picks up from where you left off. Like, if it's in the middle of a song when you parked the car, it will be on that song when you get back. Unless you hit the eject button, then it might be in the middle of some other song, because that song is on the "other side" of the tape. The songs on the other side are in backward order. If you fast forward on this side, it goes backward in time on the other side. It's a mirror universe over there, where time flows exactly opposite to ours. This is not how time travel and parallel universes generally work in movies, so it can take some getting used to.

You get like four or five tapes in the car, or as many as you like in a pile on the floor, because these things are huge and only contain a couple dozen songs each.

If you eject the tape, you have to put it back into a little plastic case, which provides no real protection, and which itself is easily damaged. It's also super fiddly, and not in good way. You do this because the case's job isn't to protect the cassette, which is actually pretty hardy if you disregard the whole "spools of flimsy, stretchy plastic" thing. The case is there to store the J card, which has all the navigation info and documentation for this release, printed in extremely small type, in white knocked out of poorly registered four-color black, which means it's fuzzy and has color halos.

Together, all these limitations of this technology, combine to produce a default scenario where you listen to one side of the tape all the way through, flip the tape to the other side, and then listen to the other side, in a loop. In the real world, you will end up liking one side more than another, or just one song on one side, so you learn to navigate the whole parallel-universe-backwards-time thing. It's fun!

I only have three tapes anyway, because at some point I got rid of all my tapes, of which I had a few hundred. I got rid of them because that is a lot of tapes.

Before I got rid of the tapes, and the tape deck, I transferred some to mp3. Except of course the dozens I didn't transfer because I got bored with the project. You have to import the songs in real time, there is no fast forward, if an album is 45 minutes long then it takes 45 minutes to transfer. You have to adjust the volume manually. You want the loudest possible signal for best results, but tapes are produced at all sorts of levels, so you have to watch the meters.

But now I have no tapes. To get tapes, I can either buy new tapes, or make my own tapes. When you make your own tapes, you can pick your own songs, and put them the order you like. This is the mix tape.

Making a mix tape takes a lot of time, because you have to get all the records out, and find the song, and get the record out of the jacket, and flip it over, and put the needle down, and press Record, and listen to the entire song, and press Pause. Then do it over again for the next song.

(Also, the Record button is often actually two buttons, which you hold down like you're playing a chord on a piano. It's a skill. You gain a lot of skills with cassette technology, is what I am saying.)

This is going to be loads of fun.

Thu, 10 Nov 2022 02:27:24 GMT

My new superpower is knowing what to do next.

Tue, 08 Nov 2022 16:07:02 GMT

I don’t follow anyone on Twitter, I’ve just been using it for the past 12 years as a place to type quick thoughts to get them out of my head. So the new Edit my feed command in FeedLand ought to be perfect for me. But, like Drummer, FeedLand is way too hard to use on the phone.

Mon, 07 Nov 2022 04:08:25 GMT

Now witness the power of this partially operational chaos factory. All the lab equipment is finally back online. Made maybe a hundred or so connections today, and sorted a dozen cubic feet of wires, plugs, adaptors, tools, screws, risers, widgets, cables, power supplies, test leads and various other things. Finally found that carriage bolt. Need more bins.

Sun, 06 Nov 2022 16:55:11 GMT

Never underestimate the importance of a well-organized workspace.

Sun, 06 Nov 2022 04:01:06 GMT

I have cut my first record! It sounds so, so incredibly terrible. I love it.

Sun, 06 Nov 2022 01:12:19 GMT

I managed to acquire a PO-80 Record Factory during the brief time it was available at the teenage engineering website at the offering price, $159 plus accessories like additional blank vinyl records, cutting needles, and the carry case. There are a couple on eBay right now for around $600. If I had any sense at all, I'd just sell it for a nice profit. Of course, instead I am going to put it together.

Sat, 05 Nov 2022 18:24:43 GMT

Feedle is a new search engine for blogs and podcasts, apparently powered by RSS. You can get an RSS feed for your search results. It looks interesting, but I can't find the RSS feed for their top stories page, and it really seems like there should be one.

Sat, 05 Nov 2022 17:02:31 GMT

I have been stupidly, ridiculously busy lately, putting together all the things so I can make all the things. These next three pieces need work, but it's better if I just post the drafts now and move on, rather than sit on them until they're done.

You feel exactly the way I do right now

Sat, 05 Nov 2022 16:17:00 GMT

(1 min read)

There's a beat that goes with this, it's stumbly. Bump baddaaddabumpbump bump dash uh uh uh.

You feel exactly the way that you do right now. It's not a science experiment, though it could be. There's a beat that goes with this.

You don't have to go out and find a beat, and make sense of what you're feeling right now. That's you. That's right.

So you don't need to do whatever it is that you do, however it is that you do, just go on doing that with your bad self. But you can make your own new beat if you like. And this experience that you are doing, well, that right there? That there is magic. It's you.

Whatever beat you want that goes with this, that is what this beat is. You feel exactly the way I do right now. And that is you right there. You.

There's a beat that goes with this. I just have to go hook up the mic.

It's stumbly. But you make your own beat, you do what you like.

This is the art part.

If it takes more time, that's how much time it takes. You get to spend a little more time doing that right there. Hopefully there's a beat that goes with it.

It does not have to be a dope song, with a hypnotic beat, going round and around and around in your head. It can just be a blog post, like this one.

But you take this idea, this part, this part, the art part, right here, and express it yourself, wherever you are, doing whatever it is that you are doing, you express that. You make your own art. And that idea that is coming through to you right now, because of the art that you are receiving, however this is being expressed to you, in such a way, that you can experiencing right now, that is the art part. And you make it yourself. And you record it, however it is that you record what it is to be you, you record that. That experience that you are going through, that's the art part. And we all make it ourselves.

There's a beat that goes with this. You'll hear it. Just listen.

What is this blog post about?

Sat, 05 Nov 2022 16:11:46 GMT

(2 min read)

This blog post exists as one specific embodiment of the larger art work that it describes. It is not the entire thing, it cannot be. It is just the blog part. The larger art work exists in a conceptual space, the hypothetical space, options. It is that space.

Reality is clumpy, it is held together and made of fractals and power laws. Making large changes requires large amounts of change. The speed of light is the speed of reality. You squish it hard in one direction, it becomes a hologram, a single change, one commit to reality, like git. The largest lumps are called kurts. This is reality, where you are, right now, all the rules and state that make up now, the present, but with all the weight of history behind it, and what comes after, up ahead.

Everything is always happening, right now, everywhere at once. Making large changes require you to know what has come before, and that transformation, that difference between just now, and what came before, that is called the cobain. It is the difference between where you are and what you are, and what everything there is around you is, and what it is going to be once this moment has passed. The cobain is a mathematically applied manipulation to reality, the exact definition of that which needs to be different, to get from here, where we are, wherever that happens to be at the moment, to there, wherever we are going to be.

Reality is clumpy, but it changes. Larger clumps of reality, what you'd call a parallel universe, movie-style, exist as perturbations of the cobain, clumped together, as a power law, as a fractal, and propagate at the speed of light. There's THINGs, which exist, movie-style, as gods, because what they are is defined literally between kurts, between the cobain itself. Also they make good storytelling.

But that description of reality, with kurts and cobains, and THINGs, and whatnot, that is not what this particular blog post is about. That's one way of viewing, experiencing, creating what we are right now. It is one embodiment of what this particular blog post is about.

This is the art part. The art part is the part where you are doing what you are doing, right now. Just like me. I'm saying it again because I like repetition, because repetition is time, it is change, it is frequency, it is happening everywhere and it is clumping. In mine, there's a beat that goes with this. It's easier to hear if you say it out loud. That's the art part. It keeps happening, we all make it all the time.

The art part is where we say it out loud.

That's what this post is, what the blog part is. It is one part of the art, one specific embodiment of a description of the way the universe is, and how to view it, and how to change it, how to explain it, how to make it happen, how to make it all happen.

That terrible thing we are all careening toward

Sat, 05 Nov 2022 16:12:29 GMT

(2 min read)

How you experience reality defines who you are, and that can be changed. We go with what's around us.

If everybody else in the room is saying insane things, we listen to what they're saying, and we agree, because that's how reality is. That's how it works. Your expectations are what has happened to you. One day, there are two sides, and the next, there is just one side, because that's just what happened, and everyone will probably get over it, because there are things to do.

I am not kidding about this, humans are capable of scary things.

And we live in a place where many, many people have all the weapons they need, more weapons than they need, weapons they couldn't possibly use in a lifetime of peaceful murder and mayhem, all waiting there, just waiting to be used and loved for their glorious murderous joyous what-had-to-happen-but-it's-all-better-now, more weapons than even that, quite frankly, I'm talking about just a stupendous number of high-powered ways to murder as many people as quickly as possible, then multiply that by a very, very large number, that kind of mind-boggling amount of firepower just itching to be used, let's say it's a wonder there aren't mass murders and political killings and showing people who's boss once and for all, happening each and every day, all the time, everywhere.

Some are just waiting for their turn, for permission. Permission to go nuts. Permission to finally use all those sweet, wonderful weapons, and put an end to everyone who is different. Twitter has been instrumental in granting that permission.

I mean, someone always has to pay. And everybody is going along with it, like it's some sort of fucking joke. Elon, you fucking asshole. I know what you're up to. We all know. All Elon cares about is Elon, just like Putin, just like Trump. Twitter wasn't a net good for society before you got hold of it, and something tells me you like that. I know what you can do to people with your money. Continue to fuck yourself with all the money and expertise and resources you have, because I know you like that, we all do, we all want what you want, the way you want it, all the time, in every possible way.

The blog part is about works of art that talk about where the world is, and how it got there, and how it's going to be next. Twitter is a sideshow compared to what is coming. Twitter is just part of how we are getting there, how we got to this point, and how everything is just going to "just happen" and suddenly millions of people are dead. I don't think any one person can change that, it's like climate change. But some of us can make it far, far worse, some of us have the ability to really fuck shit up, and they are not used to being told no.

Sun, 30 Oct 2022 03:18:23 GMT

I only noticed the FeedLand likes for all users feed today, and I like it. I subscribed to it in FeedLand, of course. But the individual items in the feed don't show where they came from in my river view. Part of my decision-making process for "should I read or trust this article" involves knowing the source. I suppose you could mouseover or something to see the article URL. But I do nearly all my reading on the phone.

Sun, 30 Oct 2022 03:27:04 GMT

The FeedLand personalized news page I use now as part of my regular reading, i.e., all day long, stopped working with an error message about an inability to get something undefined. This was unexpected, but easy to fix in the FeedLand settings dialog. Put at least one category into the News product box.

Thu, 20 Oct 2022 14:12:56 GMT

Air quality index was at 293 when I went to bed, still at 260 ("very unhealthy") this morning. Starting to agree with Frankenstein on the whole FIRE BAD thing.

Thu, 20 Oct 2022 14:21:53 GMT

I ended up unsubscribing from Hacker News on FeedLand because it updates all the time, and I think that was pushing items from less frequent sources off the end of my untabbed river page. (Not sure how many items, or how far back in time, the river goes, actually. That page has turned out to be the one I use most often, because I do most of my reading on my phone. I am probably not a typical reader for rivers of news. I should write a bit more about that at some point, because maybe there are more like me.

Mon, 10 Oct 2022 15:21:25 GMT

Apparently every FeedLand user has a news product page, or something like it. This is the one for my account. It doesn't have my tabs, and it's unlabeled, so I dunno if this is a bug or a feature.

Mon, 10 Oct 2022 15:11:10 GMT

It would be neat if I could subscribe to a feed directly from a FeedLand news product page. Right now if you see something interesting in a river, I think you need to find that user's page and locate the feed in their list of subscriptions. The link to the XML viewer page for the feed is useful, I guess, but I keep expecting it to take me to somewhere in FeedLand so I can subscribe. Perhaps that might be considered an attempt at lock-in. But some sort of "add to FeedLand" or "view in FeedLand" icon would save a whole series of clicks and round-trips, and people are used to seeing little Twitter, Facebook, etc. icons already, so there's precedent.

Sun, 09 Oct 2022 18:08:57 GMT

It would be handy if I could categorize a feed at the same time that I subscribe to it from the hotlist page in FeedLand. The "choose categories for feed" icon shows up for feeds you've already subscribed to, so once you've clicked the checkbox, you need to refresh the page and scroll back down the list to get it to appear.

Sun, 09 Oct 2022 18:16:32 GMT

I prefer the rendering of feed items in the mailbox view in FeedLand to the river view. The river view seems to omit line breaks for some feeds, and other formatting, that show up in the mailbox view.

Sun, 09 Oct 2022 18:18:54 GMT

You can reorder your tabs in the river view by changing the order of categories in your settings in FeedLand, but I can't figure out how to tell it which tab to default to. Looks like it always defaults to the second category in your list, but that seems a bit weird.

Sun, 09 Oct 2022 18:37:14 GMT

I've assigned my feeds to three categories in FeedLand. If I set one of them as my category for home page news, items from the other categories show up as well. Not sure if this is a bug or I am misunderstanding what that setting is for.

Sun, 09 Oct 2022 18:45:44 GMT

Nice, new posts seem to be showing up instantly in FeedLand now. I think the feed may need to support rssCloud, but mine does, so maybe others do, too.

Sun, 09 Oct 2022 19:00:19 GMT

FeedLand lets you create something called a "news product", a curated river of news built from a collection of feeds. Here's one Ken Smith has created for news from NASA. It seems like I ought to be able to subscribe to that news product via a checkbox. Perhaps there's an RSS feed for it that I can't find. It would also be neat to be able to see which feeds make up that river from within FeedLand, too.

Sun, 09 Oct 2022 19:26:12 GMT

Some FeedLand news products, like this one, have an XML icon that looks like it should provide a list of the feeds that make up the river. But the OPML file appears to be empty. It looks like the linked URL puts the category where the name of the user that created it should go, and uses "undefined" as the category.

Sat, 08 Oct 2022 18:38:45 GMT

Tangle Tower disappeared from the Apple Arcade recently, but the game still worked if you'd already downloaded it. Until just now. Even if you have it on your computer or iPad, when you launch it, you get a dialog box telling you it's no longer available. Extremely disappointing, and I had no idea Apple could—or would—even do that. My kid LOVES this game, and it is (was) fantastic.

Sat, 08 Oct 2022 17:53:47 GMT

I wish there was a button for "check feed now" somewhere in FeedLand, rather than having to find the URL and paste it into a dialog box.

Sat, 08 Oct 2022 17:50:34 GMT

The date used for "when the feed last updated" apparently defaults to when the feed was added to FeedLand, rather than the date of the most recent entry found when the feed was added.

Sat, 08 Oct 2022 17:44:11 GMT

The mailbox view in FeedLand will show you posts that haven't yet shown up in the river.

Sat, 08 Oct 2022 17:30:24 GMT

I was surprised to find that assigning a feed to a category in FeedLand apparently puts it back into the subscription log as a newly-subscribed feed. Perhaps that's because I created categories but omitted the "all" category, and then later added it back because I couldn't figure out where all my as-yet-uncategorized feeds had gone.