Hacker News Reader

Comment by tinkerteller | Hacker News Reader

(only including people who wrote massive amounts of important and complex code themselves and quite possibly dozen median engineers wouldn’t have cut it)

John Carmack

Dennis Ritchie

Ken Thompson

Bill Gates

Donald Knuth

Jeff Dean

Brian Kernighan

View on HN

I would strongly disagree on the Bill Gates part. The early Microsoft compilers (yes, Microsoft started out as a compiler company) were by all accounts not that great. The really successful and technically impressive Microsoft projects have always been acquisitions. The technically best (or the only decent, depending on who you talk to) Microsoft product is SQL Server, which started out as a Sybase fork.

As I understand it, Microsoft's success is largely because they poached an operating system contract. Basically the CEO of their competition was on a two-week vacation and there was no way to get in contact with him. So the contract went to Microsoft by default. Then Microsoft purchased an existing operating system and hacked it for their client. And the rest is history.

Original interpreter Gates wrote was pretty hard. Imagine flipping switches to enter a program to do entire BASIC interpreter. You are literally entering machine code, not even Assembly. In those days doing any compiler or interpreter was a monumental amount of work and there was a reason Gates was able to sale his work to establish fairly profitable company standing on just one product: BASIC interpreter. He is not in this list for his later successes as businessman or leader but the fact he was able to code this almost single handedly.

I was in high school at the time, and Gate's interpreter was buggy and lame. I could have done better, and knew it at the time.

Fortunately the quality of his software had nothing to do with his success. It was about marketing and business timing.

People took his code and re-distributed en-mass. BASIC paid the Microsoft bills for almost 4 years. Are you saying people stole and/or paid for unusable code?

Buggy, lame code still let you do something with your shiny new computing device. Early adopters will put up with anything. Which Gates understood very well. It was about timing, not quality.

He wrote his interpreter on a larger machine (PDP-11 I think), and tested it on an emulator they wrote. When they got to MITS to show off their code that was the first time it ever ran on real hardware.

They did not develop BASIC by flipping switches.

It was a PDP-10 at Harvard, and it was not just Gates, it was Gates and Paul Allen working on it together. Another interesting thing is that the PDP-10 was being paid for by Department of Defense grants, and when an auditor discovered how much computer time Gates and Allen (who was not even affiliated with Harvard) had used up, there was a disciplinary hearing, after which Gates left Harvard.[1] In his "Open Letter to Hobbyists"[2] Gates claims they used $40,000 worth of computer time. So basically Microsoft started with $40,000 stolen from the government.

[1] This is documented by Noam Cohen in The Know-It Alls. [2] https://en.wikipedia.org/wiki/Open_Letter_to_Hobbyists

BillG claims Paul did most of the emulator work and he worked on the interpreter core. With the floating point routines by Monte Davidoff who they hired. It was a team effort of course, but BillG was adamant to take credit for the interpreter specifically.

From the source code itself:

  PAUL ALLEN WROTE THE NON-RUNTIME STUFF.
  BILL GATES WROTE THE RUNTIME STUFF.
  MONTE DAVIDOFF WROTE THE MATH PACKAGE.

https://www.pagetable.com/?p=774

It's a combination of finding someone really talented, putting them in an area where their talent gets well used, and giving them lots of freedom to implement as they wish.

Being there at the right time helps as well, there were lots of low hanging fruit waiting to be picked.

This list is not about talent finders/leaders/managers. It's about people who wrote ton of code to accomplish something by themselves that would have been very very hard for a good manager to accomplish by simply lining up big team of median engineers.

Reply on news.ycombinator.com