Is this dude's version of C actually an improvement? I'm too much of a pleb to be able to really judge, but considering what he's managed to accomplish I'd have to imagine he knows what he's talking about
Is this dude's version of C actually an improvement? I'm too much of a pleb to be able to really judge...
Caleb Campbell
Other urls found in this thread:
en.wikipedia.org
scs.stanford.edu
web.stanford.edu
en.wikipedia.org
homes.cs.washington.edu
lazowska.cs.washington.edu
users.cs.duke.edu
pdos.csail.mit.edu
pdos.csail.mit.edu
ocw.mit.edu
pages.cs.wisc.edu
os161.eecs.harvard.edu
faculty.salina.k-state.edu
xinu.cs.purdue.edu
wiki.osdev.org
twitter.com
Samuel Smith
>considering what he's managed to accomplish I'd have to imagine he knows what he's talking about
He might, I do not know; but his accomplishments are not really testament to that. There is nothing particularly technically impressive about what he made; lots of people could have done the same. The reason nobody does it is that it's a project absolutely doomed to never go anywhere, even as you spend 20 years of your life on it. It's an impressive show of autism, and worthy of salutations on those grounds, but not a real sign of technical mastery.
Nathaniel Myers
>making your own c compiler
>building your own OS
>using your own c compiler in your own OS
>nothing particularly technically impressive
Kys
Leo James
wow you get easily impressed
Jason Foster
Everyone who goes through an undergraduate CS program did all of those.
Anthony Rogers
user, it's really not. I am absolutely certain I could do both without breaking a sweat, I have done both such projects and gotten a fair while into either, and I really don't particularly count that as a technical masterpiece of mine. And so have many, many others.
The reason I didn't finish it, and everyone else also doesn't finish it, is that it would take vast amounts of work before it reaches the level where the system can run a self-sustaining bash shell; and far, far more work than any one person could do in a lifetime before the system becomes more practical to use than a 1995 linux distro.
I could write any part of such a system. I really could. But I can't in 80 years write every part of such a system. There's just too much of it. And it would accomplish nothing practical in the real world, which means there are more useful things I could spend my time on. Better to contribute to a part of modern software where I can improve one component and make it better than it has ever been before, than spend 20 years building an entire software distribution that can *almost* but not quite compare to 1995 redhat.
Doing that is an awesome display of dedication, and I salute him for that. But there's really not much in it that lots of other people also know how to do.
Andrew Phillips
Any idiot can do it, but not any idiot can do it well, and TOS's literal madman design made it noteworthy before everyone knew about the author.
IIRC, Terry's coworkers considered him gifted before his mutt genes kicked in and he lost his mind.
Isaiah Clark
>Any idiot can do it, but not any idiot can do it well,
Indeed, and I see no indication whatsoever that Terry did it particularly well. It was certainly eccentric, but that does not quality make.
Evan Evans
>making your own c compiler
A hobby tier level.there is ShivyC, nwcc and probably more, it's just hard to find hobby compilers.
>building your own OS
Again, hobby tier, There are OSes on Assembly, having C OS is nothing impressive.
>using your own c compiler in your own OS
What else is the point of compiler and OS if you have build them other than use them in tangent?
Nolan Campbell
No, he was a good engineer and has a good understanding of fundamentals that most people don't have. His coding ability is not really something special and most people who know how to program at a professional level could do it, it would just take a long time and since most people are not schizos who can dedicate years of their life to such a thing it's really more about time management than programming ability.
Anthony Roberts
dude said he was telepathic and could talk to elon musk. guess he know's his shit.
Easton Collins
>paranoied of CIA being after him
>adopts prng with proven NSA backdoor
Elijah Allen
Anything other than standard C is an improvement.
Samuel Cooper
>lived long enough to see Jow Forums being salty about terry
Now I've seen everything.
Carson Bennett
I just got an old latitude from 2004, this thread made me decide to put templeos on it.
William Nguyen
Lmao at these downplaying dunning-kruger retards xD Building an OS is one of the hardest things in computer science. OS require the skill to write multi threaded code
Joseph Walker
>Building an OS is one of the hardest things in computer science. OS require the skill to write multi threaded code
Angel Gutierrez
Well memed my friend :^)
Xavier Rogers
iirc the dude used to make operating systems for ATMs or something before he Syd Barret'd himself with the strongest acid known to man: DNA
Isaac Rodriguez
>i could totally do it, i’m just too intelligent to actually try
That’s great user, keep telling yourself that.
Cameron Wood
He doesn't need too, that's what mummy is for.
Thomas Martinez
there are hard and easy things in building OS, terry mostly took the easy
e.g. building something like FreeRTOS with SMT in about month of work after taking a solid undergrad OS course
what templeos would have over such thing are the compiler, graphics and user-facing interface.
I've never done such graphics from scratch myself and dunno how terry implemented them - pure sw rendering or some intel graphics stuff? (tell me is you know).
Matthew Lee
>pure sw rendering or some intel graphics stuff?
TempleOS rendering is all done on the CPU. Terry didn't like the idea of writing drivers for every different piece of hardware since God said he needed to follow a 100k LoC limit.
There are some multi-threaded 3D games included in the OS though so he did know a little bit about graphics.
Christian Lee
I wouldn't use it personally, but I understand where he was coming from. Something has gone deeply wrong with software and Terry could see this. Modern software is not fun anymore.
Thomas Howard
I've seen it in rationale for the master/slave model. Everything runs default on core 0 unless requested otherwise, e.g. rendering in those games should run on separate core as they are intensive to compute and sensitive to latency and there are no priorities in scheduling.
Caleb Walker
this
Terry is the Martin Luther of computing
Henry Stewart
It's not so much that he did something difficult, he did something that took an incredible amount of work.
Like people who make giant murals out of beads or something. It's not hard, just a huge amount of work.
As a CS student you make a tiny toy compiler, and a minimal OS. You don't make a full C compiler or an OS with a GUI and 3d games.
It's a mammoth amount of work, not hard work, but raw hours.
Cameron Walker
>Is this dude's version of C actually an improvement?
If you want to talk to God, this is the best.
Christopher Peterson
for curious ones who want to learn on OS devel, few resources
en.wikipedia.org
scs.stanford.edu
web.stanford.edu
en.wikipedia.org
homes.cs.washington.edu
lazowska.cs.washington.edu
users.cs.duke.edu
pdos.csail.mit.edu
pdos.csail.mit.edu
ocw.mit.edu
pages.cs.wisc.edu
os161.eecs.harvard.edu
faculty.salina.k-state.edu
xinu.cs.purdue.edu
wiki.osdev.org
my attention will go to HelenOS in future as I'm interested in microkernels, but more for userspace design around microkenel that's not targeted for embedded
Jack Young
Speaking as someone who is currently putting in the "80 years", it's definitely not for everyone. If you feel better working on an existing system together with other people, then go right ahead. There's plenty of space for both types of passion :)
Lucas Perry
>this is what glowniggers actually want you to believe