In the 1980' I read with passion articles about the Unix operating system in Bell System Journal.
I discussed this with some colleagues, and to my horror, they strongly refused to accept that Ken Thompson used a more or less abandoned computer to work on a pet project.
For them, what I said was pure fantasy; they thought no innovation could emerge without some master plan from upper management.
YC Application - Bell Labs Unix Project
Dennis Ritchie & Ken Thompson, 1970
What is your company going to make?
We're building a new operating system called Unix. It's designed to be simple, elegant, and portable - running on different types of computers without major rewrites. Think of it as a clean slate approach to computing that treats everything as files and emphasizes small, composable tools.
What is your company going to make? (continued)
Unlike the complex, monolithic systems dominating today's market, Unix follows a "do one thing well" philosophy. We're also developing a new programming language called C to write Unix in - making the whole system much more maintainable and portable than assembly language implementations.
How far along are you?
We have a working prototype running on a PDP-7 at Bell Labs. The basic kernel, file system, and shell are operational. We've ported it to a PDP-11/20 and are actively using it for our own development work. Several colleagues have started using it for text processing and software development.
How will you make money?
Initially, we see licensing opportunities to computer manufacturers and universities. The real value is in consulting and support services as organizations adopt Unix. Longer-term, we believe this architecture will become the foundation for a new generation of computing - from minicomputers to whatever comes next.
What do you understand about your business that others don't?
Most people think operating systems need to be complex to be powerful. We believe the opposite - simplicity and elegance create more robust, maintainable systems. The industry is moving toward smaller, more affordable computers, and they'll need operating systems that aren't resource hogs designed for room-sized machines.
Who are your competitors?
IBM with their various OS offerings, DEC with their systems, Multics (which we worked on previously). But honestly, we're not trying to compete directly - we're creating something fundamentally different. A system that's simple enough to understand completely, yet powerful enough to grow with computing needs.
What's the most impressive thing you've built?
The file system design that treats devices, processes, and files uniformly. Also, the pipe mechanism that lets you chain simple programs together to create complex workflows. It sounds simple, but it's surprisingly powerful - like building with Lego blocks instead of carving monoliths.
Had the concept of an “operating system” crystallized enough in 1970 that someone would know what was meant by a “new one”? Or did they basically invent the concept?
We develop a text-processing system, with patent applications as our stepping stone into electronic publishing program with full typesetting. The time-sharing core is extensible to other domains. We support the PDP series now, with long-term plans for hardware portability.
Of course, they wouldn't have applied to YC. The Bell monopoly provided a lot of funding, and the market was small as only a handful of companies needed a computerized patent application system, and as we saw, Bell Labs funded them internally.
I discussed this with some colleagues, and to my horror, they strongly refused to accept that Ken Thompson used a more or less abandoned computer to work on a pet project.
For them, what I said was pure fantasy; they thought no innovation could emerge without some master plan from upper management.
(I worked at the French Telecom operator)
Of course, they wouldn't have applied to YC. The Bell monopoly provided a lot of funding, and the market was small as only a handful of companies needed a computerized patent application system, and as we saw, Bell Labs funded them internally.