DVRC @dvrc.bsky.social · 13/08/2026A SunGX (CG6). I tried to load MAIKOCOLOR.LCOM, but it failed and goes to URAID. Does the same inside QEMU, which has CG3 instead (with the original firmware to have the 1152x900 resolution). 300
Reposted by DVRCTCRF Informalities @tcrfinformal.bsky.social · 04/08/2026A bit of a more unique post: The site's ChatGPT source prevention screen! 313568991
DVRC @dvrc.bsky.social · 13/08/2026I guess this might interest @interlisp.org , considering that some of the Schlumberger papers have been mentioned in the LispCore group 010
DVRC @dvrc.bsky.social · 13/08/2026Sure, although the first thing I do is reading the manuals. I should join the related groups 100
DVRC @dvrc.bsky.social · 12/08/2026Medley Interlisp running on my SPARCstation 10, using Sun Pixrect instead of X11 120
DVRC @dvrc.bsky.social · 12/08/2026The issue was much more subtle: I compiled for SunOS 4.0 instead of 4.1.x, which likely has some differences in the way SunWindows events are managed. Now works fine, and I'll have to learn how to actually use the system. 120
DVRC @dvrc.bsky.social · 11/08/2026Thanks, I'll try soon. Otherwise I can try on my physical SPARCstation 10. Unfortunately I need an older version of QEMU to run NeWS 1.1 due to a mysterious bug, but hoepefully I have two separated versions installed. 100
DVRC @dvrc.bsky.social · 11/08/2026Progress. Now I have to figure out why the mouse is giving issues and how to load the rest 110
DVRC @dvrc.bsky.social · 11/08/2026Managed to compile Maiko under SunOS 4.1.4 from the historical archives (used Sybalsky distribution, disabled license check due to issues), since I want to explore the PixRect version. Now I have to figure out how to load the system. 110
DVRC @dvrc.bsky.social · 01/08/2026Update: performed surgery on a flex PCB, managing to get a compression connector working again and the ESI graphic board as consequence. Now I'll need a second mobo if I want both the machines to be operative without having to put the mobo in one of the two. 010
DVRC @dvrc.bsky.social · 31/07/2026(Warning: heavy flickering near the end) My port of NeWS 1.1 to NetBSD and Linux is going pretty well: events and mouse work fine, psio communication over socket seems to work. Now overlays need to be fixed, because the trash the framebuffer inverting the whole thing 010
DVRC @dvrc.bsky.social · 10/07/2026Caveat: the SE graphic board of the Octane (I have another one, not in photo) donated the compression connector to the motherboard, so I've put it in the Octane 2 (which has a VPro V6) that came without it (it was born with a 400MHz CPU). Hoping to find a complete one at a human price... 110
DVRC @dvrc.bsky.social · 10/07/2026About 5 years ago my SGI Octane motherboard stopped working, due to bad compression connectors and (later discovered) 3 cracked resistors. Thanks to a friend who provided the stuff now it's alive and kicking again 120
DVRC @dvrc.bsky.social · 26/06/2026Rant: before starting a new project or whatever, when you choose a name be sure that is not already taken by something/somone else! Too many times happened that I'm looking for something with a specific name, and cannot find easily because someone reused the name for something else. 010
DVRC @dvrc.bsky.social · 11/06/2026First menu on the native server port to NetBSD. It is still quite buggy (no highlight of entries, got also a crash), but it's a step in the right direction. 000
DVRC @dvrc.bsky.social · 09/06/2026Some degenerate cases before the fix, just for fun and to show what something apparently trivial can have nasty consequences 000
DVRC @dvrc.bsky.social · 09/06/2026After many months I finally fixed a bug in NeWS's CScript that eluded me for quite a lot: arcs rendering. The fix wasn't obvious, but the culprit were the fract multiplication and division, plus a little thing in arcs.c. The new fract routines are basically the ones from Doom's m_fixed.c 100
DVRC @dvrc.bsky.social · 15/03/2026Icon editor from the paper "How to develop an application on the GMW window system". 000
DVRC @dvrc.bsky.social · 05/02/2026Little "game": searching computer science people in the DoJ Epstein Library. I didn't expect to see Danny Hillis and Gerry Sussman mentioned or as mail recipients. Surprisingly Patrick Naughton (the Java co-creator who did despicable things) isn't there. 000
DVRC @dvrc.bsky.social · 26/12/2025These chatbots don't have the redeeming quality of being funny as Mark V. Shaney, they're just depressing 000
Reposted by DVRCrob pike @robpike.io · 25/12/2025Fuck you people. Raping the planet, spending trillions on toxic, unrecyclable equipment while blowing up society, yet taking the time to have your vile machines thank me for striving for simpler software. Just fuck you. Fuck you all. I can't remember the last time I was this angry. 10081882202
DVRC @dvrc.bsky.social · 16/12/2025Playing with OpenWindows 2.0 and FrameMaker, since that some russian site kindly hosted some goodies... Elvis instead comes from an obscure Usenet archive that had the pieces that Google Groups didn't had. BTOL instead comes from SCN. 000
DVRC @dvrc.bsky.social · 22/11/2025Maybe it's my impression, but I feel we are slowly getting in a "revivalism age" in computer science and engineering, rediscovering forgotten ideas and technologies. My hope is that history will be studied to avoid errors and create better hardware/software in general. 010
DVRC @dvrc.bsky.social · 21/11/2025The fear is the "unknown copyright status" and that Oracle will knock at the doors of who put online the sources of the server (and Don said on HN that he couldn't put it on GitHub for this reason IIRC). If it wasn't for this I would have put public the repo I have. 210
DVRC @dvrc.bsky.social · 21/11/2025Pm me, there is a little issue, but I can send you the stuff... 110
Reposted by DVRCDavid Buchanan @retr0.id · 29/09/2025tired: it's always DNS inspired: ICANN feel it coming in the air tonight 181286265
DVRC @dvrc.bsky.social · 02/09/2025I should check if setting the menus to use the linear ones the behavior changes. Still, look at the source code... 000
DVRC @dvrc.bsky.social · 28/08/2025OFC I'm stupid and I was byteswapping again the glyph bytes back into the original array instead of using a tmp var to do it. You don't want to see what's going on inside "mem_rop.c" and its CPP macro abuse... 020
DVRC @dvrc.bsky.social · 28/08/2025Little progress done with NeWS port on BSD and Linux. 32 bit color has been hacked into in a crude way, and glyph rendering is almost fixed (although some glyphs don't render). Also wrote a very WIP wrapper for libinput (and the Linux/Wayland event model is very unfriendly towards this software). 120
DVRC @dvrc.bsky.social · 17/08/2025Reading more Schlumberger papers I learnt that they developed software on Interlisp-D, including OOP extensions (the Strobe language) and an interface builder (Impulse-86). I wonder if any of it survived, in the hope of running it on Medley Interlisp someday. 011
DVRC @dvrc.bsky.social · 04/08/2025I didn't know that Kubota Graphics put the DORÉ graphic library out in the wild. ftp.funet.fi/pub/graphics/packages/dore 000
DVRC @dvrc.bsky.social · 31/07/2025Today a SPARCstation 10 arrived. After some fighting with the CD drive, I installed SunOS 4.1.4 and NeWS 1.1. Probably this is the first time after years that it runs on Sun hardware. The next step would be making it run on a Sun-3 (emulated or hardware). 140
DVRC @dvrc.bsky.social · 27/06/2025The NeWS lore is a rabbit hole: as it wasn't enough a company called Schlumberger developed a tool called "LISPscript", which could either generate Common LISP-PS glue code or compile LISP to PS. They also had an OOP library for CL called "HyperClass". (Pics from the NeWS book and papers). 231
DVRC @dvrc.bsky.social · 26/06/2025I'll have to put all the stuff I found on some repo that anyone van clone. Anyway, QEMU 8.0.5 is the last version working for SunOS 4.1.4. NeWS 1.1 comes from Don Hopkins archives, and GoodNeWS comes from Sun User Group 89 tar from John Gilmore 210
Reposted by DVRCStop Killing Games @stopkillinggames.bsky.social · 23/06/2025Remember, we still have 5 and a half weeks in which we can still make a difference with #directdemocracy Get friends and family to sign our Initiatives, and get influencers talking about the most important movement in gaming history! eci.ec.europa.eu/045/public petition.parliament.uk/petitions/70...eci.ec.europa.euEuropean Citizens' InitiativeGive your support ! 0291256
DVRC @dvrc.bsky.social · 13/06/2025With the keyboard working, now many other things work, including the GoodNeWS environment and the PSIBER 2102
DVRC @dvrc.bsky.social · 08/06/2025And now keyboard works fine. The workaround wasn't obvious: the patch suggested by Sun works as intended, but the file "repeat.ps" is damaged. Commenting "(NeWS/repeat.ps) run" in the "UI.ps" file did the trick, at the cost of losing key repetition. At least now I can use native terminal emulators. 010
DVRC @dvrc.bsky.social · 08/06/2025After fighting with QEMU, I finally managed to compile the sources of UniPress Emacs 2.20. It's really cool to see it working, but I can't type yet due to the keyboard issue that breaks nearly all the NeWS applications. 322
DVRC @dvrc.bsky.social · 08/06/2025There are also 4 proposals to abolish part of the "Jobs Act", and to make contractors liable in case a worker gets injured/dies in a workplace of a subcontracted company. From center to right, the political class invited either to boycott, refuse the cards or to vote against. 000
Reposted by DVRCLars Brinkhoff @larsbrinkhoff.bsky.social · 18/04/2025New addition to the Emacs collection: UniPress Emacs V2.20 from 1989. 132
DVRC @dvrc.bsky.social · 14/04/2025Learnt about JXL art yesterday because of a DEF CON challenge. I made these fiddling with parameters on this site: jxl-art.surma.technology 010
DVRC @dvrc.bsky.social · 06/04/2025Font unpacking is done. Now I'll have to investigate why the colormap is wrong, why the text rendering is broken (which also is needed for the cursor). 000
DVRC @dvrc.bsky.social · 16/03/2025Had to rewrite cst_readfamily() to correctly unpack the font. It was tedious considering the format is big endian and the members of structs were 32 bit long. Wrote also a little tool to dump font infos. Further work will be needed. 000
DVRC @dvrc.bsky.social · 06/03/2025It was the wrong ioctl() (a single letter mistake). Now I'll have to investigate why fonts are not getting retrieved. 100