Hello HN,
I was clearing my old house and got a box with a 100+ 3½ floppy disks. All except one working!<p>Among others I got DisplayWriter 4.0, OS/4 Warp Slovenian, dBase IV, Clipper, PowerPoint 4.0, Netscape Navigator 3.0, etc.<p>I found also some files with .CHT and .IMG extension which I can't find any software to open them. I know I made them, but unfortunately I can't remember which software was used. There is my family tree and chart of my height growing (which I'd like to compare now with my kids).<p>Some files I uploaded here: <a href="https://github.com/twooclock/KdajBi/tree/main/cht_img" rel="nofollow">https://github.com/twooclock/KdajBi/tree/main/cht_img</a>
Files timestamp is 1990 and 1991.<p>Please help me find appropriate software!<p>So far I tried (does not work):<p>-ABC Flowcharter 3.01<p>-IBM Storyboard Plus 1.01 (5.25)<p>-Harvard Graphics for Windows 4.0<p>-Harvard Graphics for Windows 2.0 (3.5)<p>-HarvardGraphicsViewer<p>-PowerPoint 98<p>Googled, checked file.org, WinWorld, ... loosing hope... really makes me think about digital preservation.<p>Thank you!
Try using binwalk, file(1) or strings(1) to get some information about the contents of the files ... try searching found strings on Google, etc. Good luck!<p>--- Edit:<p>Those four files are really small and I can't see any person's names in them when just opening them as plaintext files, but LINEN contains some instructions for solving math systems of equations: "sistem dveh linearnih enačb z dvema neznankama. Možne rešitve". Other files seem like a listing for a music album (MARLEY: 01. No woman no cry) and possibly lyrics or something in BEATLES (eight days a week she loves you, ... Maybe something for a karaoke program). KASETA is also really short and contains no text - maybe some chiptone music, but that's just a silly guess.<p>To correct myself: binwalk, file and strings do not help you here. It's not even ASCII text.
I would guess the CHT files are commands for text output on a dot matrix printer, or a format pretty close to that.<p>* The MARLEY songs are laid out in two columns (side A, side B) from left to right then top to bottom, which suggests its pretty close to a matrix of characters, rather than something more abstract/structured.<p>* Every character is taking up two bytes, which could be some Slovenian encoding but I doubt any European language was taking up 16 bits in the early 1990s, which makes me think every second character is a formatting for the character that precedes or follows it.<p>* The MARLEY title has a repetitions like "BBoobb MMaarrlleeyy" which might be the kind of overprint mode I remember dot matrixes having, where the printer prints a character then goes back and reprints it, to get a bold effect.<p>I would dig around for old dot matrix standards, Epson and so on.
I will try to message this directly, since this thread is old now:<p>I solved this, figuring out the file structure and encoding from the examples. I was totally wrong about the printer idea I suggested in another reply.<p>MARLEY.CHT, for example, looks something like this:<p><pre><code> ╔════════════════════════════════════════════════════════════╗
║ ║
║ ║
║ ║
║ ║
║ ║
╠════════════════════════════════════════════════════════════╣
║ BBoobb MMaarrlleeyy && TThhee WWaaiilleerrss ║
║ BBoobb MMaarrlleeyy && TThhee WWaaiilleerrss ║
║ LLeeggeenndd ║
╠═════════════════════════════╦══════════════════════════════╣
║AA:: ║BB:: ║
║AA:: ║BB:: ║
║- - - - - - - - - - - - - - -║- - - - - - - - - - - - - - - ║
║1.No woman no cry ║1.Three little birds ║
║ ║ ║
║2.One love/people get ready ║2.Buffalo soldier ║
║ ║ ║
║3.Is this love ║3.Waiting in a vain ║
</code></pre>
etc.<p>You can see all the files and the code that made them here:
<a href="https://github.com/econandrew/hn_mystery_files" rel="nofollow">https://github.com/econandrew/hn_mystery_files</a><p>This was a strangely satisfying task. I hope I can help the OP recover their more important files too.
Netscape! The pulsating throbbing N is back. I think you need to lock that disk in a box with a sprig of sage and seal it with wax.<p>I think the CHT files are dbase databases. IMG could be anything, lots of programs used that extension. You’ll need to get in with a hex editor and look for clues.
CHT, other than Harvard ChartXL, looks like some printers support it. Could be worth trying to raw print it?<p>Now maybe that floppy disk has corruption. (You could check if you get the same byte-for-byte output using another drive or give a run of GRC SpinRite worst case)<p>All the repetitive '31 02' in both CHT and IMG files make it look like it comes from the same app suite or it's just data loss somehow.<p>KASETA has interesting patterns in a raw pixel viewer, RGB32, 15 pixels wide, but LINEN doesn't seem to be an image.<p>Tried opening this with Corel Draw, Ventura Publisher or else ?
MARLEY seems to have more in common with KASETA, even though they have different file endings. `AA::` and `BB::`, a large section of `- - -`... and second byte alternation are present near the start of both, as well as the large section of `1`s (assuming this is all in ASCII). Are they all the same file format?<p>I'm guessing they're all from a word processor at this point - Text602 is a shot in the dark from a vague link on Wikipedia - and I think the file is structured as pairs of bytes, with the first being some data, and the latter being a presentation attribute. For example, I'm guessing from LINEN that `0x06` indicates superscript.
A shareware (freeware?) program called Graphics Workshop was popular in the 1990s and available from many BBSs. For example, you can get it at <a href="http://files.mpoli.fi/software/WIN32/GRAPHICS/" rel="nofollow">http://files.mpoli.fi/software/WIN32/GRAPHICS/</a>. I mention it because it supports an IMG image format.
Do you recall what kind of machine and operating system these disks were originally written with? You might have some luck with a local retro computing group. I don’t know much about digital preservation and old school storage media, but maybe there was a quirk with how the data was originally written on that specific disk that you just need a retro computer to untangle?
I don't know if you found the program for your family tree or if you inferred that from the filename, but you could try this one: Sierra Generations Family Tree, which uses the CHT format. Do you recall using that piece of software?
OP here. Both filetypes cht and img are "images", because I remember printing them out. Cht should be somekind of chart with boxes, lines and text. Img should be an image? It's way far back so I don't trust my memory.
My first programming environment was dBASE II! Would have been cool if these were dBASE databases and I could have helped, but then again I have neither the software anymore nor knowledge of the raw binary format.
It might be a long shot, but if you do think that some of those files might be images, you can try dumping their contents as hex and then use an online hex-to-image converter and see if this produces anything useful.
Your IMG might be from Gem Paint? I remember that extension from my first computer, and that's what we used. It's about the right age.<p>EDIT: Ugh, maybe not, I looked at LINEN.IMG and there's a lot of text in there.
The files you posted hold no information about their type in the headers. So it's impossible to figure out.<p>BEATLES and MARLEY and simple lists of songs.
LINEN.IMG contains information in Slovenian about methods of solving a graphical algebraic table.
KASETA is more obscure, doesn't contain readable information I can extract.