Their timestamps don't show those ages, unfortunately, as they have been transferred from machine to machine and medium to medium over the decades.
Here's one of my own COBOL files:
0130* INSTALLATION. Giacomo Software, P.O. Box 584, Hamilton, 3300
0140
0150 DATE-WRITTEN. 29 MARCH 1984.
It started life as a plain .txt on disk that I'd manually copy over from computer to computer, first via floppies and eventually null modem cables and ZIP disks and thumb drives.
When email got popular I started emailing it to myself as a way of storing it in the cloud. Eventually it made it into Gmail and I could store it there, versioned, across generations of computer platforms and ISPs.
Then Google Docs came around and it's lived there ever since. I still go through the version history there sometimes. It doesn't go all the way back, but does have some interim license keys if I ever need an older version (like some gray market version of Windows 7 for a refurb laptop).
While even now with the filesystems that support them they're tied to that instance of the filesystem (ie: non-modifiable, unlike Windows), which has always puzzled me given the need to variously restore from backups (or just have an identical copy in a destination) where one would desire such info*.
* When I last looked into this I saw some quite creative workarounds, such as a script loop where for each file to be copied it changed the system clock to the date of the timestamp, so it was recreated in the destination.
I really should go through some old 3,5" disks to see if they still work to see if there is anything there.
[1] https://en.wikipedia.org/wiki/Ghost_goal#/media/File:Ghost_G....
In terms of actual timestamp, the oldest I found was "Aug 18 1994", a text file with some notes from a concert that I wrote to send to friends.
-rw-r--r-- 1 uid uid 21714 Aug 18 1994 concert
As for files I created myself, probably emails from 25 years ago, as I lost pretty much everything I had from the '80s and early '90s in a fire (and all my older emails migrating from older ISP/university email accounts, and didn't really do anything worth saving in the late '90s).
Stretching the definition, I did recently uncover a video tape containing an award-winning media fair project I did with a couple friends in the mid '80s that I hope to digitize just as soon as I obtain a VHS deck I'm certain won't damage the tape.
This was on a HP benchtop gas analyzer where I had added the HP Basic option which was needed to handle the complexity of petroleum data like only companies having mainframes had been able to, up until 1979. Most other research operations did not need Basic since the underlying expert system was adequate for less data-intensive work. Those who had a bare-bones analyzer were way ahead of most of the world though, which was still going with analog chart recorders and simpler mathematical calibration techniques that had been well-established.
Interestingly, there are still plenty of non-petroleum gases and other analytes where the raw analog data is so simple that no computer has ever been needed at all, some needing not a calculator nor even a slide rule. But people naturally have all-computerized systems in the modern world anyway, including the complex software that for such simple analysis provides more room for error than the analog days. In theory and in practice as directly observed.
But the chemical assays were not related to the gases, I just liked having a personal desktop system that would compute, and wrote a couple hundred lines that would get it done in a way that was completely auditable. The workflow included manually entering data that had been gathered from bench work in the non-gas labs.
Slightly different syntax allowed my app to run on the P-E equivalent to the HP, and I managed to keep at least one P-E running until 2014 so I could operate the vintage analyzers in my own lab. But stopped running my old chem app on them in 1993 when my employer got their first office PC with Windows 3.1 and GW-Basic. So that was the first actual DOS version of the same old thing, which is the one still in use on my FAT32 partition when I boot to DOS, as well as run from the command line when booted to 32-bit Windows.
When it comes to timestamps a couple old EXEs that have been carried forward on the FAT32 volume are Tetris and Battle Chess from 1989 which I can still run any time since.
That looks like it for files that are actually useful still.
Years later I managed to find another disk of the same model, but I can't find the board-less disk with the data... did I throw it away?
It very slowly and inefficiently compares hashes of files in its directory (and subdirectories) and deletes duplicates. I don't think I've actually used it in years, but it invariably gets restored along with some old family photos that happened to be in the directory it lived in.
Ironically it is older than any of the files remaining there.
The drive was always a secondary bulk storage drive that just kept moving into each new computer I built. It finally died a few years back and I still have a dump from it. Nothing super interesting that far back, just schoolwork and some photos. I didn't get into programming until much later in life.
Outside of that, the oldest file in my home dir is "de01file.cpy" from 1994-01-06, something I saved for reference from my old Job.
> Oldest thing that I remember was perhaps my parents dropping me off to pre-school before leaving for work and me crying - circa 1985.
41493 Feb 1 2004 linus-says-linux-english.au
I thought it was funny that it is an audio file of how Linus says Linux 8-)
1969-12-31+19:00:00.0000000000 ~/.cache/paru/clone/aws-cli-v2/src/awscli-2.16.0/PKG-INFO
* 1980 STOMP.MOD music file[1]
* 1987 - hitchhik.exe the HitchHiker's Guide to the Galaxy text-based adventure game.
* 1988 - QuickBasic 4.5
* 1990 - a bunch of Screamtracker S3M/Mod/XM music files.
* 1991 - Moneky Island original SCUMM game.
The oldest which I created is some QuickBasic code which says 1999 but I think it was a couple of years older. A drawing and some Python and C# from 2002. Looks like my code from the mid 1990s and any school essays of the 1990s are long gone. About the oldest emails I have are 2002 as well.
[1] "stomp da diko-tek #hoffman / mono formerly known as.. dreamfish! / mono this is true minimal techno. this track started with a sample from retro dna - mix 2. i took a 2 beat loop, contured, distored & filtered it to create a clanky hard-edged sound (sample 3). the rest you'll have to work out yer self. surf the mono website www.scene.org/mono". That dates back before I was born.
Oh! And some slightly less ancient with original timestamps still on my server home dir (now Raspbian, but from some mixture of SunOS etc)...
% ls -alrt ~/ | more
-rw-r--r-- 1 dhd dhd 1981 Jul 20 1989 .rootmenu.old
-rw-r----- 1 dhd dhd 56 Jun 9 1990 .XLog
-rw-r----- 1 dhd dhd 37 Dec 21 1991 .forward.example
-rw-r----- 1 dhd dhd 6019 Apr 6 1992 SCI_arrives.z
-rw-r----- 1 dhd dhd 42 Apr 9 1992 .mailrc.OLD
-rw-r--r-- 1 dhd dhd 934 Apr 9 1992 .defaults
-rw-r----- 1 dhd dhd 0 May 28 1992 .pnewsexpert
-r--r----- 1 dhd dhd 615 Jul 28 1992 .bprofile
I don't really use it myself, it's leftover from an IT project.
This was from an industrial facility where I had done field training back in 1982. Their mainframe had a "public" terminal where contractors had access to their inventory control system, so whoever was on-site had to wait for their turn at the green monochrome dumb terminal, then continue to wait for our printouts to appear afterward.
Apparently sometime between then and 2019, I would imagine they might have replaced the mainframe with a PC-based server, but it would be difficult to be sure. Expected steps could have included first simply replacing the dumb terminal with a PC on the desk for contractor use instead, no differently than the rest of their office people would want.
But for all I know they very well could have had the latest modern server gear back there in a rack somewhere. You would never know the difference.
Regardless, by 2019 there had been no more public terminal for a while. Each contractor had their own laptop to access the facility system, still running the same desktop Dataflex program since 1983 on a DOS platform. Laptops weren't around back then, so that's a clue.
Without that, a contractor was useless, so the incentive had always been there since the first person figured this out in the 21st century.
At this employer, when we wanted to get in there, our IT guys had no luck at all. The plant couldn't help much, they cloned their own systems in a convoluted way rather than revive from decades-old files and that was it. I was provided with a Windows 7 laptop that successfully ran Dataflex to work from, but more than one person had probably never known how to prepare these to begin with. You definitely had to be able to take your data with you and print it later on a non-DOS printer.
First made it run on bare metal, then hammered away until it ran self-contained on a Windows 10 laptop and connected properly to the system when on-site, just like everybody else had who treated the old laptops having Dataflex like gold, since it was like their ticket in through the gate of the facility.
* Castle Wolfenstien for MS-DOS (1983-6-29)
* Lisa OS Source Code (1983-6-29)
* Classic Mac Disks from the Boston Computer Society (1984-12-24)
* Atari 7800 Ms. Pac Man Source Code (1988-12-24)
* Pyroto Mountian BBS files (1990-10-5)
* Jumpman Lives Source Code (1991-04-13)
* Delightful AU sound files and TIFF images from Sun and NeXT systems (1992-2-29)
* Tim Berners-Lee's WWW Browser Source Code for NeXT (1993-6-21)
* C64 Disk Images (1994-6-17)