...like on the Xerox CP/M machine I used to use. It was the 1st computer I used with 1-megabyte 8-inch floppies, a unbelievable amount of storage at the time. The drives made this loud "clunk-clunk-clunk" every time the heads moved. On many drives of the period, it was not uncommon for the drive to lift the head off the disk before moving to another track; supposedly to reduce head/disk wear.
NEC had a CP/M machine that had a color screen, but no graphic capability. Very pretty.
I eventually did most of my CP/M work on an Apple ][e with a blindingly fast 6-MHz Z-80 card. It was kinda the best of both worlds as the entertainment and graphics oriented software could run on the Apple side, and I could run business apps on the CP/M side of a single computer.
I worked at a computer store in the day of "baggie wrapped" software on casettes hung on the pegboards.
I had considered a degree in computer science, but the school I started at was still using punch-cards and mostly teaching COBOL (already starting to fade in the early '80s), so I studied accounting and economics instead; a wise decision that was to benefit me far more. People with accounting backgrounds who could program were far more valueble than people which CS degrees who know little else.
I remember resisting the 1st IBM-PCs; the first configurations were little more capable than the Apples and CP/M machines of the time. I had Apples and was using mostly CP/M (writing dBase II applications). Apple's biggest mistake was not coming up with a successor to "Apple DOS" soon enough. The Apple ][ operating system simply was unable to be adapted to larger disks. To use a hard disk with Apple DOS you had to partition it out in 160k blocks, of course meaning that no single file could be larger than 160k. This pretty much made the Apple useless for database work. By the mid-'80s Apple came out with "ProDOS" which addressed this problem. But by then, it was too late and Apple was focusing its future on the MacIntosh.
Remember that "PC-DOS" was not the primary first OS; you had a choice of buying USCD-P System or CP/M-86. I don't think anybody outside of academia used USCD-P, and even then not for very long. CP/M-86 lasted a few years, mainly because there were so many relatively mature 8-bit CP/M business applications that could quickly be ported over. A big part of our business for several years was having several machines connected via RS-232 or model, and converting people's legacy data between the half-dozen or so formats of the time.
PC-DOS eventually won out. When the first hard disks showed up, PC-DOS 1.1 was not up to the job. There was no hierarchical directory (or folder) structure yet, so all of your files for everything (OS, programs, and data) were in one giant directory with several hundred files; your word processing, database, and accounting data all living together in a single directory! This of course meant that you could not have more than 1 file with the same name. This made managing data and backups very difficult. Fortunately, programs were much smaller back then.
Before Lotus 1-2-3, there was a system called "Context MBA" (I still have a copy around somewhere) It was far more advanced and integrated than Lotus was. I vividly remember the 1st demo I saw of it and was amazed at its capability. It was also expensive and slow. It was developed on a P-System minicomputer and emulated on the PC. This made it excruciatingly slow. Lotus, on the other hand, was written on the PC and was comparatively crisp and fast, even running off of floppies. It won hands-down, and was responsible for selling millions of IBMs, just as "Visicalc" had sold millions of Apple ][s a few years earlier.
DOS 2 solved the hard disk organization problem with the "tree" structure, and shortly thereafter the more capable and stable DOS 3.1 arrived and became the standard for several years. By this time, it was clear that the PC had won the battle for the business platform for the foreseeable future. Then came the "clones", and the rise of Microsoft.
Keep Up with TechRepublic