Why do we presume our OSes will never be 'big'?

[Therx]

Just wondering. Something that newbies are always told is that their OS will never become widly used, but why not? Below is the post by Linus Torvalds regarding Linux.

Hello everybody out there using minix -

I'm doing a (free) operating system (just a hobby, won't be big and
professional like gnu) for 386(486) AT clones. This has been brewing
since april, and is starting to get ready. I'd like any feedback on
things people like/dislike in minix, as my OS resembles it somewhat
(same physical layout of the file-system (due to practical reasons)
among other things).

I've currently ported bash(1.08) and gcc(1.40), and things seem to work.
This implies that I'll get something practical within a few months, and
I'd like to know what features most people would want. Any suggestions
are welcome, but I won't promise I'll implement them

Linus ([email protected])

PS. Yes - it's free of any minix code, and it has a multi-threaded fs.
It is NOT protable (uses 386 task switching etc), and it probably never
will support anything other than AT-harddisks, as that's all I have .

It seems that he'd writen an OS, single handed, which could run gcc and bash in 5 months (that post was in August). This doesn't happen anymore, why not?

Pete

[TheChuckster]

Well. First, many people do not have the knowledge or skill to write a big OS, so our goal is only to write something we're capable of. Also, many people give up. I am close to giving up right now from the frustration of writing a floppy driver.

[Solar]

It seems that he'd writen an OS, single handed, which could run gcc and bash in 5 months (that post was in August). This doesn't happen anymore, why not?

What Linus did was writing a well-understood kernel architecture providing a well-understood interface on a well-understood hardware using lots of "previous art". He has not writen an OS, but "just" a kernel, and he didn't design anything. (Spare the flames, I know I'm exagerating, but only a bit.)

Even this doesn't happen anymore because:

1) the "entry hurdle" to OS development has become so low - thanks to everyone having a computer, a compiler, and access to the internet - that people who don't even understand CS basics try to roll an OS of their own, reducing the signal / noise ratio to infinity so good projects just don't show up on the radar;

2) most projects today don't aim at "just" reimplementing yet another POSIX kernel but try their hands at something all-new, increasing their time to maturity quite a lot;

3) somewhat related to 1), they don't build them anymore like they used to. Kids (and students!) today start learning about computers at such a high level that their "l33t skillz" usually revolve about using P2P software and fragging each other in Quake 3 or whatever. I believe that geeks like us are a dying race.

[Solar]

Let me add some:

4) Most OS projects are started by students, and usually die when the student graduates and finds himself quite short of the excessive spare time he enjoyed in the university. (I don't flame, I was a student myself.)

5) Many people get really excited about the idea of writing their own OS, but get disenchanted once they realize there's so much more to it than "writing a bit of kernel code and then design the greatest GUI ever".

I started my project fully knowing how little time my job leaves me, and fully knowing that, in best case, it would be 10+ years of hard work to get anywhere near Linux v1. I hope this knowing gives me some advantage.

[Tux]

Let's see, a person probably devs for 3 hours in a 5 day period. Let's say there are 4 weeks in a month. 5*3=15. 15*4=60, 60*12 months=720, 720*10=7200 hours needed to write a full OS.

Hmm, add lots of caffeine, and:

17 hours a day in a 7 day period, 17*7=119, 119*4 weeks=476, 476*12 months=5712,5712*1.3=7425.6

So it will take the sad fellow 1.3 years to make an OS.
Side effects, insanity... (7425.6 hours)

[Therx]

I think you miss read the post that was Linus Torvalds' post in August '91 announcing Linux which could already run bash and gcc and he'd started in April. April to August is just 5 months in which he wrote most of an OS

[Tim]

I agree with Solar. Linus didn't write an OS; he cloned Unix. He had Minix code to work from, and a small set of target functions (Posix).

[Therx]

Still quite a job for 5 months and his post said:-

it's free of any minix code

I don't know if that's true but it proves what anyone can do if they really put their mind to it

[TheChuckster]

I've been "putting my mind to it" for the last 3 months. And I really lack motivation now since I've used up all my positive attitude. Putting your mind to it may be important, but skill is the key.

[Pype.Clicker]

whether your OS'll become good or not is a matter of patience and skillz. Whether it'll become "big" (i mean linux-like) or not is a matter of luck and marketting (yes, again). Would Linux have become what it was if The Hurd had been ready on time ? or if Linus owned a Mac rather than a PC ? or a Vax ? ... who knows ...

and does it really care, after all ... we all dream of becoming big without daring to admit it ... we say "we don't plan to become big" as an excuse for our lack of design or releases irregularity ...