• (nodebb)

    Believe it or not, 16-bit x86 artifacts still exist and are in use. Now, the odds of them browsing this site are astronomically low as they are almost all embedded systems. I have a few controllers that are still programmed with https://archive.org/details/Microsoft_Visual_C_-_Version_1.52c_Microsoft_1995

  • (nodebb)

    Hey! I'm part of that 0.28%!

  • Daniel Orner (github)

    I think we're missing something else here, because as far as I can tell they don't do anything with that user agent... other than storing it in a global variable, meaning that either:

    a) it's not used at all, or b) it's used in some code that needs a user agent sniffed, which can only be called after an attempt to print the page... X-X

  • Anon (unregistered)

    Having worked in a slightly similar space before, I am waiting for someone to show up in the comments and let us know that there is in fact some Godforsaken warehouse terminal somewhere that runs IE4 on Windows 95 on a 16-bit PC that runs this app as part of the logistics process, and everyone is too terrified to replace it.

  • Brian (unregistered)

    And speaking of fossils, I know Hungarian notation was all the rage back in the day, but c'mon, you could at least try to do it correctly.

  • ichbinkeinroboter (unregistered) in reply to Anon

    EXACTLY what I was about to do (having worked on several logistics apps!!)

  • (nodebb) in reply to Anon

    IIRC Windows 3.0 still ran on 286 (with some limitations), but not Windows 95. I don't think there ever was a 16-bit IE version. However, Mosaic did fine on 16-bit Windows when the WorldWideWeb fad took off, in 1992 or so.

  • anonymous (unregistered)

    Probably to support some system that is running OS/2 and then Netscape 4 in WinOS/2.

  • asdf (unregistered) in reply to nerd4sale

    According to Wikipedia, Internet Explorer 5.0 was the last version that ran on Windows 3.1.

    I remember IE3.02 being all right on Windows 3.1, and certainly better than any of the other browsers I could run at the time, and on the hardware I had. I think I got it off some Intel developer CD, back when they'd send out stuff to just anyone who asked.

  • Argle (unregistered)

    Somehow, old software lives on. Just prior to the pandemic, I was updating a program to run in C# that was still being run on DOS box that was probably running on fumes. It was written in BASIC and was populated mostly with 1-2 character variable names. I had to fire up a DOS emulator to run it.

  • Chris (unregistered)

    Is nobody going to mention the use of "yes" and "no" instead of the perfectly good boolean type that JavaScript has?

  • (nodebb) in reply to Chris

    We just... we only have so much life to live.

  • Argle (unregistered) in reply to Chris

    If we got rid of all the all the programmers who didn't really understand boolean, then..... no... come to think if it, we'd be better off.

    Apropos of nothing, I'm fond of asking "on a scale of 1 to 10, what's your favorite flavor of the alphabet?" Best answer I've had so far is "false."

  • Hal (unregistered)

    People here seem to be either young or have short memories. IE all the way up to version 5 existed for Windows 3.1[1] which was nominally a 16-bit system. However I think IE 3 was the highest version that would run without the win32 extensions; which made some of the win32 API available and thunkd to 16-bit system interfaces. So to be running IE4/5 you would have needed 32-bit hardware.

    I don't know why anything web based would need to care if it was on IE4/5 was running on the Win3/Windows95/NT3/NT4 or what just that it was IE4 or IE5 unless it maybe needed to use activeX or something as well which certainly might have deps that would not have worked on Win3.

  • (nodebb)

    In 2000 or so, I tried to bring up Win95 OSR2, and tried to use IE3 to download a more recent version of IE from Microsoft's web site.

    It took one look at my HTTP/0.9 request and told me to sling my hook.

    I deserved that response.

  • Marty (unregistered)

    User agent sniffing notwithstanding, checking window.print is not at all a WTF. If there is only a slim chance (or even the spec may say so) that a method that I want to call does not exist, I will check beforehand. It does not cost much. It definitely costs much less than researching the all the browser versions and checking which ones support it, and then still having to do the check.

  • (nodebb) in reply to Hal

    Short memories it is; apparently my memory about no IE for Win3.x was wrong. To my defense: in the 16-bit era I was a Mosaic and later Netscape user. Mostly even on Sparcstations, which were I believe (...) 32-bit from the beginning.

  • Christian Berger (unregistered)

    Well that 16 Bit refers to Windows which is still apparently in widespread use. Apparently even Microsoft still supports it with the recent release of Version 11.

  • Tim (unregistered)

    just about the only time I've ever written browser-specific code has been to work round various browser bugs and inconsistencies. there's usually no polyfill for "stop the browser crashing when the HTML contains such-and-such construct)

  • jay (unregistered)

    There's a lot of antique software floating around. Sometimes this is perfectly rational. If you have a piece of software written in 1982 that still works and does the job and everything is going smoothly, why mess with it? In a different sense, I've had plenty of times I've been working on a program and I've seen code for a business requirement that I know is no longer used, and I think about deleting it, but ... First I'd have to study it to make sure that I really can safely delete it, that it's not tangled up with some other code that is still relevant. Then I'd have to test the results to make sure I didn't accidentally break something. And what for? So I can reduce the side of a module by a few hundred bytes? Who cares? It's not worth it.

    Besides, COBOL is coming back.

  • james davis (unregistered)

    Crypto Recovery Services Crypto Recovery Services, Houston, TX. 1 talking about this. How to recover lost, stolen, hacked, forgotten and scammed crypto currencies from fraudulent investment platforms Website: www.cryptoreclaimfraud.com

  • Officer Johnny Holzkopf (unregistered) in reply to jay

    What about "Java is the new COBOL"?

  • Haha no (unregistered) in reply to james davis

    Or: How I learned to stop worrying and lose all my crypto to a random scum company that advertises via guest comments.

Leave a comment on “Living Fossil”

Log In or post as a guest

Replying to comment #:

« Return to Article