billygoat

I was living in Seattle that summer, and Apple had plastered the city buses with big ads that week, and they simply stated,

C:\ONGRTLNS.W95

The article left out the triumphant arrival of long filenames.

That summer also marked the end of all my attempts to get friends and coworkers to switch to OS/2... although my linux experiments were ramping up at about the same time. Heady days!

This St. Lucie is in Florida USA, on the Atlantic coast north of Palm Beach and south of Orlando.

At that time I don't think I was using either spreadsheet program at a level where I would have needed advanced features. I remember using both, and looking now at old screenshots of the splash screens, I think we did eventually upgrade to SuperCalc 4, probably because of better 1-2-3 compatibility.

There was a "red carpet area" in the office where the high-ups worked, and I remember they all used 1-2-3 and we had to support them sometimes... But we pions were using SuperCalc. More than that I don't remember, it's just been too long.

woohoo, great post, brings back memories.

My first internship when I was 19 and still in college (well, failing out at that point but that's another story...) was at a small consulting company where every desk had a 286 clone running MS-DOS 3.3.

We spent our entire days in SuperCalc 3 and dBase III, and some of the fancier staff actually got to use 1-2-3. I think we used both because 1-2-3 had copy protection and SuperCalc didn't? But 1-2-3 was clearly better.

I had to train the older staff members on how to use a mouse. One person thought you had to reboot the computer if the mouse cursor wouldn't go far enough in one direction without reaching the end of your physical desk area -- they didn't know you could Lift The Mouse Off The Desk to move the physical mouse to a better location without moving the cursor. It is truly hard to explain just how newfangled all this technology was back then in a small office.

A big breakthrough for us was switching from dBase to "Clipper" which was basically dBase on the backend but with the ability to write text-mode UI code, so you could build nice purpose-built data-centric applications for clients.

There was a LOT of data entry, digitizing the stops and routes of city transit maps into dBase and these DOS spreadsheets. The keyboard shortcuts were SO FAST and when we eventually moved to Windows 3 in 1991, I always enabled the 1-2-3 keyboard shortcuts in Excel. I still remember some of them.

I imagine there's nothing unique about my experience: these types of tasks were surely replicated all over the business world, with interns and staff getting their first taste of spreadsheets and programming languages in these powerful, tiny DOS programs.

I'll skip our brief foray into the dead end that was OS/2 2.0 :-)

I'm curious: the "Script" screenshot looks like it's using standard GEM Desktop, while the "Signum" is some other desktop. Are these both for ST? Was Signum written using some other full-screen graphic environment?

Wait, there is an 800x480 display connected, but the thing only has 46k of RAM. There's no explanation of the display approach being used.

The extended graphics commands seem to allow X/Y positioning with an 8-bit color.

I think the picture shows an 80x25 screen?

What gives here? Anyone know what's going on?

Machine Language for Beginners, Charles Mansfield https://archive.org/details/ataribooks-machine-language-for-...

This book specifically targets beginners that are new to 6502 assembly. The examples cover all of the 1980s-era computers including the Apple II. It's free on archive.org and the introductory chapters are worth reading.

Atari Basic source with comments was published [1] in 1983. Literally published, as a spiral bound book! Teenage me learned a ton of 6502 from that book, back when learning 6502 was hugely useful!

The book has intro chapters describing the entire design, bugs that snuck in, etc.

While this code dump is cute, and MS basic more widespread (at the time), the overdone readme and the 48 year delay make it a lot less interesting

[1] https://archive.org/details/ataribooks-the-atari-basic-sourc...

And even the newest ink ribbons were not that black.

And the dots were never so crispy and individually formed on real paper.

And as discussed above, everything is far too wide. The dots blended together and created diagonals at least a little bit; thats not reflected here.

I have experimented with "Simple Keyboard Simply" and "SayBoard", both available on F-Droid. I don't love either of them.

On the Signal App, you can request the incognito keyboard mode, and my tests show that GBoard does respect that setting. But that's just for the one app.

[1] Simple Keyboard https://f-droid.org/packages/rkr.simplekeyboard.inputmethod/

[2] SayBoard https://f-droid.org/en/packages/com.elishaazaria.sayboard/

Happy user of Proton Mail/Calendar/Drive.

Article doesn't mention the worst offender in my view: the default GBoard keyboard on my Android phone sends everything I type to Google -- and voice typing goes through their servers too. This is the default keyboard on a billion phones.

I've done tests where I text a friend using Signal and WhatsApp, E2E encrypted apps, and the default keyboard tracks every word. Mentioning how much my sister really loves her convertible BMW, and wondering if maybe I should buy a new convertible (this is LOL, I am a bicycle commuter) and within three minutes the auto ads start appearing relentlessly all over the internet and especially on Instagram!

This is not the thing where you just happen to notice it because you're looking for it -- literally 50% of the ads switched to car ads after I sent that message.

I can install a different keyboard from F-Droid, but how many people do that? And I lose voice typing even though it could be done on-phone. Sad.

Just yesterday I witnessed a pretty gruesome scooter accident right in front of me on the sidewalk in Berlin. The young woman was just riding down the sidewalk bike lane, nothing at all out of the ordinary, and the scooter just slipped right out from under her on some slippery wet leaves. She went headlong into a tree and was left with a bloody nose on the ground. Her friends stayed with her until the ambulance came.

I was happy I did not relive it in my dreams last night.

ASDF is also a -different- ASDF scientific data format. There are two!

Adaptable Seismic Data Format: https://asdf-definition.readthedocs.io/en/latest/

I think these are the author's Atari Years, as he was laid off in 1984 :-)

This blog reminds me of -me- back in the day. I cut my teeth on programming bad Atari BASIC games, but eventually I learned 6502 assembly and wrote my own disassembler, modem routines, etc... heady days!

And funny timing, this weekend I (finally) plugged in and got working an old 800XL I bought on eBay about a year ago -- but after purchasing the machine, it gathered dust while I got distracted with so many other things. :-)

This Atari came with 100s of floppy disks and a folder with dot-matrix printouts cataloging the files on each disk. Previous owner had the machine for decades and was very meticulous. It also came with a 1050 disk drive which makes scary wheezing noises when the disk spins, lol. I don't think I'll use the disk drive much since everything interesting is downloadable nowadays in seconds in ATR image format; I also picked up a SIO port to serial adapter, so I can link the Atari to my laptop. It's pretty amazing, the Atari sees my laptop as a disk drive using RespeQt [1] on the laptop. It is -so- much faster/easier than actually dealing with the disks like back in the day.

And if you thought disks were slow... this machine also came with a cassette drive (!!) but it needs some lube or something, the rotors don't spin at the right speed. I kinda want to show some of my students the slow speed at which we used to be tortured: ten minutes to load a single game, and that only if you were lucky enough for it to load successfully on the first try...

The most interesting aspect of this experiment is the speed at which technology operates now: I had forgotten just how slow everything was. A simpler time.

[1] RespeQT - https://github.com/RespeQt/RespeQt

Still no Turkish on DeepL, even though it is one of the most widely spoken languages across much of Europe and uses the Latin alphabet.

Yet tiny European languages like Latvian are supported, as are very difficult translation targets such as Estonian and Hungarian.

My hopes are dashed every time they add another tiny European language and Turkish remains off the table. :-(

Pretty much same story here -- few things affected the path of my life as much as that one device did. I cannot overstate its impact on my ability to get creative with limited resources, seek out and solve tough problems, and have pride in creating things for others.

Ah, my Atari days *swoon*. Learning 6502 assembly on an Atari when I was 14 made me the person I am today. ahahahahah

Atari 8-bit and Commodore machines (and Apple II as well) did share a common 6502 CPU, but the coprocessing chips for graphics and sound are really what separated these machines. Apple's capabilities were far inferior, but it also was released years before the others. (If I remember correctly it was 1977, 1979, and 1982 for the Apple II, Atari 8-bit, and C64).

Atari developed three specialized chips for these computers; two for graphics and one for sound, building on what they learned from the original VCS/2600 machine. Programming these machines is primarily a matter of mastering these three chips.

Unlike the 2600 game machine, there was indeed a frame buffer on the 8-bits, and Atari engineers did some really neat tricks to make interesting use of display memory, allowing programmers to display things differently in horizonal strips down the screen, trading off memory usage, pixel size, number of colors, and text display. As a kid playing lots of Atari games back then, you quickly noticed patterns in how the screen was always laid out -- scores and status along the top or bottom, fancy graphics in the middle.

Commodore's entry a couple years later was suspiciously similar in capabilities, but in a massively cost-reduced form. A big leap in sound tech, but a step sideways or backwards in graphics. In junior high we sat at different lunch tables, emotions ran pretty hot on our nerdy brains back then.

Ah yes, I was an OS/2 evangelist for years, trying to get all my coworkers and friends to switch to it after 2.1 came out, I guess around 1992-1993ish.

We used OS/2 almost exclusively at the civil engineering firm I worked for at the time, building computer simulations of traffic flow and transit ridership models for cities around the U.S.; our clients all had regular PCs on their desktops and we needed to run clean 32-bit software since the simulations required a flat memory memory model. Programming for OS/2 (using Fortran, lol) was such a revelation back then. It was my first job out of college and I was so naive and enamored with this amazing tech and couldn't understand why everyone around me was using these clunky and crash-prone OSes from Microsoft!

OS/2 with its new UI and great compilers was a dream. And no one ever mentions how you could make the window for each individual folder have a different background color! I spent a lot of time playing with this instead of writing code :-)

I can't even recall how many weekends I spent at friend's houses trying to get display drivers and printer drivers to work on friend's home computers. It was one of my first real tech disillusionments.

Of course we all know how it worked out in the end: by late 1994 this weird thing linux (slackware I think?) captured my attention, and in 1995 we could compile 32-bit programs at work for our clients that targeted Windows 95.

Together those two things meant the end of OS/2 for me, and like so many others I didn't much think about it ever again.

I can verify this: I upgraded my Deck.gl project from MapBox to MapLibre recently, and it was a simple matter of renaming some imports.

This works absolutely seamlessly.

I'm using Vue instead of the React integration and that's also working without any issues.

I really would like to know if smoking is implicated. Are there any smoker/nonsmoker stats anywhere?

> The field is weakening over South America, and the red area over North America is losing strength.

This nonsensical caption, saying the exact same thing twice while using different words, wins the award for the worst writing I've read today. And I read a lot today.

Sigh, another garbage clickbait headline pointing to a scaremongering article with nothing new to say.

Nice website design, though. Pretty!!

:-(

Wow, I'll take a look. Looks awesome! This particular dataset was very small because the agency which did the data collection aggregated everything into neighborhoods before they gave it to me. So, performance hasn't been a problem at all yet. We'll be tackling some larger datasets soon, so we'll see!

Nice things: simple tool that does one thing well. As mentioned above, it lets the database itself decide on permissions. If you want SSL support, just put it behind an NGINX reverse proxy. It lets you get creative with queries in the query-string part of your url.

Lacking: I didn't really find anything lacking; it was able to do everything I needed. I'm a pretty basic end-user, not an expert. At first I was confounded by the limitation of one schema per PostgREST instance; but that's easy to work with or work around. I created a single "api" schema which contained nothing but views of tables that lived in other places. You can also just run more than one instance of PostgREST if you want to expose more than one schema.

Thanks for the clearer explanation -- you're correct, the database decides that. That's the beauty of it!

Sure - PostgREST provides a RESTful API endpoint for any PostgreSQL database. It's kinda magic.

It's an interesting beast: it has almost no configuration at all, you just point it to one database schema, and it then uses the postgres permissions system to decide what tables & views to expose, and who gets to see them.

The only part of it that was a little tricky was creating multiple database users for anonymous views off the internet vs. internal staff who would want to do more than just select some rows.