Disclaimer: I've read this article at least 10 times and I have never posted it to Hacker News.
That being said, even though it is a repost (and should have the date added to the title), with all of the "interesting" articles on HN about the latest theories in start up-i-ness, like another company who shows everyone their employees salaries by name, or once again how bad/good/awesome bitcoins are, this is definitely worth another read. A lot can be learned about how to succeed in technology by studying Bellard's methods:
-- Incremental changes to under served problems
-- Solid techniques that stand up to the test of time
-- Great documentation
-- Building communities around problem sets
-- Knowing your strengths and sticking to them, only jumping when it makes sense
The list goes on and on. Any person who wants to be successful in this business would do well to take the 10 minutes it takes to read this post.
Would love to read a good interview with the guy, in the spirit of "Coders at work", say, which was really good at this sort of thing. There are some nice bits in this article, but in the end it's a bit like reading a CV.
Impressive. I think these sorts of articles are hard to come by because many (if not most) super productive programmers are doing their work for a company where the outside world won't find out about what they have done. Fabrice is notable because he's got a lot of high profile work.
Might anyone be able to comment on his employment in all these years?
Every time I've come across this article it's seemed to me that the work he's done, whilst notable, is an entirely reasonable and very achievable amount to have completed over the course of the at least 20 years that he has been developing.
If he has held a full-time job that takes away the majority of his time or something similar (as many here have) then I could understand these codebases being particularly notably huge. Otherwise, as has been commented by others here, this level of productivity, whilst admirable, seems to strike me as similar to many, many programmers working around the world - in their cases simply behind closed doors, instead and not working on their pet projects, but on those chosen by their employers.
I have to disagree. Most programmers don't get to write something like ffmpeg or qemu in their careers, regardless of the time invested. More likely, most of us are writing the same amount of lines but in simpler, mundane projects, with less impact in the whole ecosystem.
But to the layman, we can do magic, just like Bellard.
I agree with you, but I think that the reason that so many don't get the chance to work on projects such as these has less to do with proficiency or 'productivity' (in many, but far from all cases) and more to do with employment and the chance work on self-directed projects. Which is why I wonder about his employment status during that time.
Whilst it could be argued that those employed in simpler programming for their day-jobs could come home and work on these sorts of things in their spare time, this of course neglects to take account of the difficulty of maintaining such effort past the 5-8+ hours of concentration one typically puts in.
TL; DR: I think there are (have come across) a fair few frustrated programmers out there who could build an achievement record like this if they could put in Bellard's amount of time whilst being paid and otherwise supporting themselves but are 'stuck' in jobs building much simpler, less interesting systems and failing to dig themselves out.
As such, I'd like to know how Bellard has been supporting himself.
Absolutely again. In both cases, people who had read this article commented saying they were glad to see it and re-read it. This is one of two posts that I would be fine re-posting year after year.
I've not seen it before. There are others who haven't either. If it's getting re-posted and staying visible, that means enough people think it's interesting to not fall into obscurity.
That being said, even though it is a repost (and should have the date added to the title), with all of the "interesting" articles on HN about the latest theories in start up-i-ness, like another company who shows everyone their employees salaries by name, or once again how bad/good/awesome bitcoins are, this is definitely worth another read. A lot can be learned about how to succeed in technology by studying Bellard's methods:
-- Incremental changes to under served problems
-- Solid techniques that stand up to the test of time
-- Great documentation
-- Building communities around problem sets
-- Knowing your strengths and sticking to them, only jumping when it makes sense
The list goes on and on. Any person who wants to be successful in this business would do well to take the 10 minutes it takes to read this post.