Hacker Newsnew | past | comments | ask | show | jobs | submit | bitwize's commentslogin

Bodybuilding does not have a linear progress bar. Sometimes you train and train and train and see little or no progress, and you need your coach to come make adjustments to your diet, routine, or supplements (incl. anabolic "supplements") in order to break the logjam.

Source: Not a bodybuilder but know enough bodybuilders to have heard this story many times.


Don't be surprised if you're paid a visit by the SCP Foundation: https://scp-wiki.wikidot.com/scp-079

(edit: change url)


Qemu can run SPARC binaries and emulate a full SPARC system enough to run Solaris 2.6. Qemu can emulate Loongarch as well!

I know, but I would love to have it transparently, similar to some IOCCC guy that wrote a userland emulator to run Unix V4-v7 and up to BSD 2.1 binaries seamlessly. Similar to FreeBSD and NetBSD's compat(8) approach.

So you want CPU emul and a kernel personality. Kind of a tall ask, especially since you'll need much of Solaris anyway to run those vintage binaries.

AOL was an "online service", internet connectivity was provided starting in the mid-90s but the real value was in their proprietary members-only services. An online service was more or less a large BBS you dialed into via modem, the first of which could be accessed with an ordinary telecoms program but later ones like AOL and Prodigy required special client software that could decode the proprietary protocol allowing the service to send graphics, menus, and other information to present a UI to the user.

AOL is not the internet we lost. It's literally what bigtech is trying to turn the rest of the internet into. Online service as centralized marketing platform for bigcos. Remember "Visit us at www.blah.com or use AOL keyword 'Blah'"? The real internet was much more amorphous and chaotic.

Pretending that AOL represents the pre-enshittified internet is like pretending that Clippy represents the idealized past of desktop computing when he was a prototype of what Microsoft is building now.


This. A decentralized, 90's-style internet would probably involve home servers, and there have been attempts to do that: Sheevaplugs would be a good example from the late 2000s. I don't think anyone ever managed to make self-hosting at home frictionless enough, though, with secure defaults and turnkey startup. Open-source, sadly, tends to overestimate the amount of fiddling the average user will tolerate; Apple does a better job of making difficult things simple.

It can be both. For many, AOL was their onboarding platform for the chaotic internet at large. It served that purpose well.

Strong static typing is a must for large scale software engineering. And by "large scale" I mean anything longer or more complicated than like a utility shell script.

These type extensions let you get some of the benefits of static typing for projects already written in Python or Ruby.


One of my childhood computers was a Tandy 2000. This was a 186-based PC-incompatible computer available from Radio Shack. It was more performant than an AT, could access more base memory (due to a disk-based rather than ROM BIOS), and available at a lower price so it was a real contender before it was clear that the IBM standard would be used by everybody.

Not only could it run Windows 1.0, Microsoft used the Tandy 2000 internally for Windows development because in the early 1980s it was the only x86 machine out there that could do hi-res (640x400) color graphics. So, getting Windows 2.x backported to the 2000 is definitely feasible.


Hi, the article's author here~

I just checked the Tandy 2000 Windows pre-installation - it has the drivers unpacked, which means you can just get the Slow Boot Windows 2.0, and put the drivers from this floppy to it. And the fonts, of course. Definitely do not check this bad pirate website that has it: https://winworldpc.com/product/tandy-2000-ms-windows-pre-Ins...


> "[...] in the early 1980s it was the only x86 machine out there that could do hi-res (640x400) color graphics."

Mindset Mindset II.


Which was also a 186-based PC-incompatible.

Could the Mindset do 640x400 noninterlaced?


Neither Mindset nor Mindset II. Mindset II had the 640 x 400 four-color mode upgrade. Double-buffered but interlaced. Could be genlocked to another Mindset.

The Tandy 2000 could do 640x400 8-color noninterlaced. It was pretty useful as a CAD and DTP workstation, predating even the EGA by a year, which is why it was chosen to develop and test color Windows.

The Mindset machines were (also) advertised as Video Production System, with custom VLSI chips and modules for graphics and video workflows and assorted applications (e. g. Lumena). So a bit of a different market.

I wonder if there's some top-class Japanese x86 machines from the "IBM workalike" era.


PC-9801 could do 640×400 in 8 colors, and is supported by early Windows.

Indeed. At the time, Tandy was trying to break into the business market with its computer offerings, and the 2000's software library reflects that, including MultiMate, Basic Four, Lotus 1-2-3 and Symphony with hi-res graphic/charting add-ons, a version of Ashton-Tate's Framework, and the best desktop version of AutoCAD available at the time. The Mindset and Amiga lapped it in terms of video, audio, and in particular NTSC- and composite-compatible output capabilities, but machines that were targeting those workflows were doing something very different from the generic business market (which Windows, at the time, was aimed squarely at).

A composite compatible output addon, called the TV/Game Adapter, was planned but never released for the Tandy 2000; if released it might not have supported genlocking or really much beyond getting 16-color 320x200 or so video onto a TV. Until recently the only graphical game I recall actually being released for the machine was a special version of Flight Simulator; but recently on Facebook I saw a photo of a 2000 running some sort of video poker or other card game. It was unlikely to have been very fast paced and may even have been written in GW-BASIC; I don't know much about it.


I believe it because it's a plausible variant of what I call the "Fus Ro Data Loss" vulnerability: shouting at hard drives causes them to resonate in a way that affects their ability to access data.

https://m.youtube.com/watch?v=tDacjrSCeq4


x86 based small computers are just so much easier to work with than most second- and third-string ARM vendors. The x86 scene has had standards in place for a long time, like PCIe and the PC BIOS (now UEFI) for hardware initialization and mapping, that make it a doddle to just boot a kernel and let it get the hardware working. ARM boards don't have that yet, requiring per-board support in the kernel which board manufacturers famously drag their feet on implementing openly let alone upstreaming. Raspberry Pi has its own setup, which means kernel support for the Pi series is pretty good, but it doesn't generalize to other boards, which means users and integrators may be stuck with whatever last version of Ubuntu or Android the vendor thought to ship. Which means if you want a little network appliance like a router, firewall, Jellyfin server, etc. it often makes more sense to go with an N150 bitty box than an ARM SBC because the former is going to be price- and power-draw-competitive with the latter while being able to draw on the OS support of the well-tested PC ecosystem.

ARM actually has a spec in place called SystemReady that standardizes on UEFI, which should make bringup of ARM systems much less jank. But few have implemented it yet. I keep saying, the first cheap Chinese vendor that ships a SystemReady-compliant SBC is gonna make a killing.


> I keep saying, the first cheap Chinese vendor that ships a SystemReady-compliant SBC is gonna make a killing.

Agree. When ARM announced the initiative, I thought that the raspberry pi people would be quick but they haven't even announced a plan to eventually support it. I don't know what the hold up is! Is it really that difficult to implement?


The Pi boots on its GPU, which is a closed off Broadcom design. Likely complicates things a bit.

Apparently Pine64 and Radxa sell SystemReady-compliant SBCs; even a Raspberry Pi 4 can be made compliant (presumably by booting a UEFI firmware from the Raspberry's GPU-based custom-schmustom boot procedure, which then loads your OS).

Ah, Autolisp. Based on an early, early version of XLisp, it's zombie software: not quite alive—Autodesk really wants you to use its COM and .NET integration to program AutoCAD—nor can it truly die. It was also the first Lisp I ever programmed, and back in the mid-90s there was a trend of putting a Lisp, or Tcl, into all the things (Emacs, GIMP, frickin' Abuse...) and Autolisp had prepared me for that world rather well.

That mid-90s trend of embedding Lisp/Tcl everywhere was special. Everyone understood: give users a real language, not just config files. Emacs, GIMP, Abuse, AutoCAD - all bet on "let users program it." Now we have YAML and JSON configs. Progress? Part of why I built this: if AutoLISP becomes truly undead inside AutoCAD, at least the workflow and playground can live on independently in the browser.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: