BSD Cafe Billboard

@scruss Hmmm i ran some Linux on one in a car computer, a way too early 2005 or so. https://boingboing.net/2006/06/28/tom-jennings-car-mp3.htmlThe url has changed, try hostname sensitiveresearch.com, maybe add Archive. Pages and info still exist. Im on a phone on vacation else id look on the machine for you. Bit it was pretty neato. It did "war driving" for open APs, and ehen pulled into the driveway, did sutomatic wifi update of sustem and music collection. The hacked radio was exquisite. Crude velocity sensitive band/song selector. I wish id saved the control head.

The Early Days: A History of Interactive Computing https://lobste.rs/s/tvfhvm #historical #retrocomputinghttps://obsolescence.dev/interactive-computing-history.html

@stefano Donald Duck's Playground. I guess that ages me!

@bat3000 It was our amateur movement that did it all. Parallels on rhe cp/m side. All stolen by corps. Apple became one; ibm took over the other.

Fixing a Sun Type 7 Keyboard Recently, not quite sure why, I decided to start looking for a Sun keyboard. Originally I was on the hunt for a keyboard that would have vim bindings printed on the keycaps, but as I was browsing around, I stumbled on a great deal on eBay -- fraction of the price, but there was no USB cable. That's a relatively easy fix, so I got one. The keyboard arrived in good shape, and apart from the insane amount of fluff in it and the faded / yellowed plastic of the keyboard it was in decent shape. Yes, the cable was cut. But it was easily accessible: two screws on the bottom and further three underneath the top keyboard cover. Once undoing the two longer screws on the bottom I popped off the top keyboard cover; this is held on with tabs and if you work your way around the left, bottom, and right side, it hinges at the top. I decided the first job would be cleaning. I used a soft brush to get rid of the debris under the keys after opening (but not fully disassembling) the keyboard, and then took cotton wool balls dipped in rubbing alcohol to the surface to clean. This took a while but it worked OK. Next was connecting up the USB cable. On the PCB the USB connector is probably some kind of a standard SMD connector, with standard USB cables coming off it: black, white, green, red. These are GND, D-, D+, 5V, usually. There is a fifth ground cable going to the USB cable, which is the "shield ground". I disconnected the USB connector to the PCB and chopped up a USB connector. My initial plan was to use a USBC female to USBA 3.0 male connector's USBC end. But USB3.0 has, I think, 8 pins rather than 4, and while it should be possible to only hook up the 4 "standard" USB lines, I did not have much success with this. Otherwise the soldering job was fine, although with these thin cables, rather than stripping them, I took to melting back just enough of the cable sleeve on each to expose a workable amount of wire. Just apply some hot solder on the tip of the soldering iron and it works well. Second attempt involved cannibalising an old MiniUSB cable I really had no use for anymore. This has the "correct" amount of wires and this time it worked! I was delighted. OpenBSD recognised it really well as a Sun Type 7 USB keyboard: uhidev1 at uhub1 port 4 configuration 1 interface 0 "Fujitsu Component Sun USB Keyboard" rev 2.00/1.05 addr 4 Then I fired up xev to see whether the most exciting bits, the extra "Stop" and "Again" keys do something. And they do! xev reports the following mappings by default: Left hand extra column of keys: -Help: DEAD. Stop: keycode 145 (keysym 0xff69, Cancel) Again: keycode 146 (keysym 0xff66, Redo) Props: keycode 147 (keysym 0x1005ff70, SunProps) Undo: keycode 148 (keysym 0xff65, Undo) Front: keycode 149 (keysym 0x1005ff71, SunFront) Copy: keycode 150 (keysym 0x1008ff2f, XF86Sleep) * Open: keycode 151 (keysym 0x1005ff73, SunOpen) Paste: keycode 152 (keysym 0x1005ff74, SunPaste) Find: keycode 153 (keysym 0x1008ff17, XF86AudioNext) * Cut: keycode 154 (keysym 0x1005ff75, SunCut) Top right hand extra row of keys (above numpad): keycode 160 (keysym 0x1008ff12, XF86AudioMute) keycode 174 (keysym 0x1008ff11, XF86AudioLowerVolume) keycode 176 (keysym 0x1008ff13, XF86AudioRaiseVolume) keycode 222 (keysym 0x1008ff2a, XF86PowerOff) Bottom row: Compose: Menu So as you can see, it mostly works okay. There are two funky key mappings by default, "Copy" doesn't copy but wants to put the computer to sleep; "Find" wants to skip to the next playing song. But these should be remappable. The "Paste" and "Cut" keys do what one expects them to under X11, this is really really neat. I really like not having to press "Ctrl-Shift-V" for "paste" but ... "Paste". The Compose key will be remapped as a Compose key I think, this is doable with xmodmap and .XCompose I think. I want to dig into whether, similar to SunPaste, there ought to be a SunCopy key that I could map to Copy. For "Front" I want to set up a custom macro; for 9+Front I want to launch drawterm. Pity that "Help" doesn't do anything. It simply doesn't register as a keystroke. But you can't have it all -- besides, I haven't come up with anything funny to map it to. #unix #OpenBSD #sun #solaris #retrocomputing

@alter_kaker Probably easiest way is to just clip it out, figure out where the connections go (ID + and -) and solder longish wires, mount the battery elsewhere. That's what i do. Then you can use a holder for two AAA's or whatever is convenient.

@markmoxon Marvellous idea, that.

@manawyrm Photo of my desk, taken in March 1988. This is a Mac 512K which was upgraded to 2MB of RAM plus SCSI. It's sitting on top of a Jasmine 20MB hard drive. Next to it on the desk is an 800K external floppy drive. Off to the left is a Practical Peripherals 2400 baud modem. To the right, partly hidden under papers, is an ImageWriter II.#retrocomputing #VintageMac

This is why #AI is such a siren song: instead of going back to first principles and rebuilding our technology stack without all the harmful complexity, we're adding something we're even less capable of understanding on top of it. Disconnecting ourselves further from the tools we depend on.And #FOSS doesn't solve this problem at all: GCC, LLVM, and Firefox are all horrendously complex beasts nobody really understands. Even the organizations that maintain them only partially understand them; they substitute by implementing tests, and then when a test fails some human has to go in and learn that one particular area to figure out why the test broke. And then they'll soon forget again.

@gloriouscow The Western Digital chips allowed you/required you to specify the format bit pattern, in slow machines like cp/m usually built in ram first. There's a lot of margin to fiddle with; i routinely squeezed a 9th 1024 byte sector on a track of 8 inch DSDD drives, another 77 K per side. Im sure i did terrible things to reliability and interchange but hey.

@amoroso Also, that writeup! And somewhat, the temporal distance. Solid writeups of projects like that, fail or success, helps us think about what we want and need.

You know you want to read a paper with "Feudal Internet" in the title (and a cameo of @tomjennings Not The Internet, but This Internet: How Othernets Illuminate Our Feudal Internet. It draws on the history of computer networks to analyze how the technical features interact with and shape the social and economic structures of wide area networks.https://tidsskrift.dk/ashcc/article/download/21200/18686/48329#internet #retrocomputing

Curious machine. It's a mysteriously rebadged Sharp PC-4602 and I found some data on it here:https://forum.vcfed.org/index.php?threads/the-mysterious-sharp-mz-100.9582/#post-1312894I have a suspicion where Mom bought it back in the day, so I'm looking to see if that has anything to do with the mode number. #retrocomputing #msdos

I should probably give CompuGlobalHyperMegaNet a go soonish. Cleaning out my tabs on the phone and had this one open to check out later. I'm not entirely sure I get the point of it yet, but lfe is pointless in general so who cares Anyone on it?https://cghmn.org/#RetroComputing

@dinotron my main idea with the 4x16k banks was convenience for user programs. With 4 banks, the program can leave the top and bottom banks in place to not mess up the os, but can switch the two middle banks around at will to access lots of memory. At least that's the theory.

@lmemsm #alpine / #postmarketos ?

@tomjennings Things were indeed a bit babel-ish back in the day. The many incompatible Lisp dialects is what led to Common Lisp.

Things currently in the mail--Some 74HC595s, along with an Arduino Pro Micro -- for the purpose of flashing EEPROMs.A 90W charger for my newly repaired and battery-upgraded Lenovo ThinkPad X230.I still need to post more GOBLINS.BAS zines too. We will be running out of copies soon, may need to print and fold more!I am either going to do the 6502 build I have long discussed, or a 6809 build, or a Z80 build.Something about 8080 assembly has clicked with me.#retrocomputing

@Kroc @Lacey

@BastilleBSD oh, this is no personal account, i apologize. still, the info's valid.