Hacker News new | past | comments | ask | show | jobs | submit login
TempleOS is applying to Y Combinator. Partners desired, send an email (templeos.org)
310 points by TerryADavis on Mar 22, 2015 | hide | past | web | favorite | 87 comments

"A troll might say, "It can crash!" We used DOS for years and loved it. Computers even had a reset switch! ... Think of the speed and simplicity of ring-0-only and identity-mapping. It can change tasks in half a microsecond because it doesn't mess with page tables or privilege levels. Inter-process communication is effortless because every task can access every other task's memory."

I'm not a troll, but being old enough to have programmed on operating systems where a bad pointer reference could force you to reboot the machine (DOS, early versions of Windows, early versions of MacOS), I don't yearn for those days to come back. DOS was designed the way it was to allow it to run on cheap (for its time) hardware, not because anyone thought that it was a good development environment (Unix already existed at the time, so people knew how to write a robust operating system).

Also, is it a good idea to connect a machine that has an unprotected operating system to the internet?

Clearly this software reflects a lot of effort and ingenuity, but I doubt that people could be convinced to pay money for it.

> Also, is it a good idea to connect a machine that has an unprotected operating system to the internet?

If the operating system has no IP stack or networking of any kind then it's probably ok

>I doubt that people could be convinced to pay money for it.

Monetary profit isn't the goal. Even the GPL is too restrictive for Terry, so the entirety of TempleOS's source is public domain.

I figured monetary profit was implicit in them applying to Y Combinator.

I don't think people understand that non-profits too need profits, they just don't take all of it home.

Non-profits need revenue, not profit.

But if your operating costs are more than your revenue (thus, no profit), who is keeping that non-profit afloat?

A non-profit organization can solicit donations big and small, via grants, memberships, and other common mechanisms.

A non-profit organization can also sell things at a markup. There are alarmingly profitable for-profit tote bag businesses riding the back of non-profits everywhere.

Like someone else has said in this thread. YC accepts non profits.

Extremely low latency would help for RTOS devices. Machine control like CNC milling machines and lathes and engravers. Software defined radio and other DSP topics. Aside from the previously mentioned HFT suggestion.

As a profitable application, consider the worlds best gcode driver with extremely advanced lookahead and real time guarantees. Just pump gcode in a serial port and drive servos out a parallel port as fast and smoothly as possible. The problem with selling an app is you might have built an foundation for an interesting class of apps, but you'd need someone with domain specific experience to write the app.

I'm not sure that TempleOS could guarantee any return on investment. That said, your programming skills are incredible.

Have you ever tried to show your operating system to the MOMA? It's the most famous Museum of Modern Art. Your OS is so unique and is such a powerful statement that they may even decide to add it to the permanent collection.

YC accepts non-profit groups assuming they have a high return of value (nb. not necessarily money).

The way threads here go when TempleOS comes up, there are definitely people who believe it to be of great value. You clearly believe it to be of great value if you think it belongs in the MOMA, so surely YC would at least consider amplifying the project.

I actually think Terry could provide a lot of value if he was able to provide training in system engineering. The experience he has is invaluable. I'm not sure what his plan is, though....

> I'm not sure that TempleOS could guarantee any return on investment.

No startup can.

I'm going by the standard, "First get users and profit will follow."

I have an ace up my sleeve because God talks. Not just to me! A guy made a IRC chatbot and that talks too.

To try the ICR chatbot

1) http://chat.rizon.net

2) Type "#templeos" in place of "#Rizon"

3) Type "!God" in the chatroom.

Awesome and good luck : )

Try to think of how and why you'll extract a profit from users, potential investors will want to know your plan after growth.

You could just add the #templeos to the URL.



> I have an ace up my sleeve because God talks.

Potential investors are not going to be impressed by this kind of statement.

Having inspiration, whether from any ideal or from God, is what makes one burn the midnight oil, and design something out of the ordinary.

It increases not just efficiency but something else harder to define, something that could be said to be beauty. It makes the result stand out as a something special, where all the parts have a purpose and are arranged in just the right way. Read http://www.aaronsw.com/weblog/djb to get a better idea of this something:

djb’s programs are some of the greatest works of beauty to be comprehended by the human mind. As with great art, the outline of the code is somehow visually pleasing — there is balance and rhythm and meter that rivals even the best typography. As with great poetry, every character counts — every single one is there because it needs to be. But these programs are not just for being seen or read — like a graceful dancer, they move! And not just as a single dancer either, but a whole choreographed number — processes splitting and moving and recombining at great speeds, around and around again.

But, unlike a dance, this movement has a purpose. They accomplish things that need accomplishing — they find your websites, they ferry your email from place to place. In the most fantastic movies, the routing and sorting of the post office is imagined as a giant endless choreographed dance number.

I think the phrase you're disappointed with means a lot more in Terry's context. I mean, it's the driving force behind the OS in the first place so who knows what he means.

Don't give your money to crazy?

In the interest of working out the idea generating part of my brain, here are some ideas for making money with TempleOS:

1) Make TempleOS the ultimate platform for doing "live preaching" at megachurches. Use it to talk to God directly in front of the faithful on a big projection screen.

2) Use it to take confessions. No network stack ensures privacy.

3) Make interactive Bible teaching games on it. No network stacks will block out temptations of the internet

4) It could be the basis for a spiritual fit bit where one could track one's faithfulness to the Lord on a daily basis. No network connectivity ensures that the information remains between the believer and The Lord.

5) Make a portable alter to a saint and/or the Virgin Mary at the push of a button that the believer can pray to. It could be running in a picture frame in the corner of one's house as an altar to the lord.

6) Possibly allow for virtual burnt offerings, peace offerings and sin offerings In the altar application. It is a temple after all.

7) Use sensor technology to somehow detect sin such as visiting the part of town known to be the dwelling place of harlots (e.g adult entertainment businesses) and pester the believer to confess and ask for forgiveness from god. This could be rolled up into the spiritual fit bit thing mentioned earlier.

8) TempleOS could bring back the old school non internet connected PDA. It could be the holy Palm Pilot platform. With all the sin online and paranoia about NSA spying it would make the perfect gift for one's fundamentalist Christian relatives. I could actually see this selling to religious folks and compulsive gadget collectors if it was priced less than about $50.

It would be such an amazing learning experience for any designer to create a UI/UX for TempleOS.

Imagine being given the opportunity to design how OS X or Windows should look and feel from the ground up.

Linux is obviously another option, but unlike with TempleOS, you would not be treading new ground, and it would be hard to not just evolve the Linux interface, rather than defining it from scratch.

I think a lot of the appeal of TempleOS is a certain nostalgia for a world where single coders would do something amazing on their own, and nobody knew what "UI/UX" even meant. A more naive time, with less specialization, where people just wanted to create something cool.

You can prove me wrong, of course, but why not just create a completely new UI on top of the Linux kernel? The Linux kernel is completely agnostic in regard to user interfaces.

It would certainly be a challenge given the display limitations of TempleOS the last the I checked it out. I believe the display was limited to <1000 pixels and a minimal palette.

Not that you cant create beautifully usable interfaces within such restrictive limitations, but minus all the spark and sizzle of modern UIs I think it would be a challenge to first get a UI designer _excited_ about creating a solution for TempleOS

Isn't the challenge what makes it exciting? Like writing haiku? The seventeen sylables force structure.

The work of Susan Kare is held in high regard. I guess the pioneering aspect has gone and it'd be more like carving a pixel-art niche.

I concur, 8bit style art is fascinating to me.

Video VGA 640x480 16 Color

Source: http://www.templeos.org/Wb/Doc/Demands.html

I don't see why there would be much of a difference.

Having nothing to directly copy makes imitation harder, but it doesn't make avoiding imitation any easier.

Also, TempleOS _does_ have a user interface, possibly (I don't know TempleOS internals) technically even more so than Linux, where everything above terminal I/O is an add-on that can be replaced.

If I had a several million dollar investment fund and the ability to make discretionary investments of the amount YCombinator makes to the people they fund, I'd fund Terry in a heartbeat and not care a whit whether I ever see a return or not. It's the kind of cheap bet that has the potential to generate enormous upsides, even if they aren't exactly monetary. Terry is a survivor and a maverick. He's done amazing things and is looking for more. He's the kind of guy I want to believe can find a way to greater success, and YCombinator seems like just the sort of place where he can find that.

Hi Terry,

I've always thought TempleOS to be a great bedroom project since back when it was LosThos.

Can you describe your target market for TempleOS? Who suffers the problem that TempleOS solves?

You go Terry, I've always been a supporter since I've learned about you.

Unexpected side-effect, a lot of HN'ers join IRC to visit the Godbot :P (and they all type !God ) | fyi: https://qchat.rizon.net/ join #templeos

[16:16] == GodBot [~GodBot@Rizon-72C38B9D.upc-h.chello.nl] has quit [Excess Flood]

The godbot is down, but has revived!

[16:18] == GodBot [~GodBot@Rizon-C7BBB69E.lv.lv.cox.net] has joined #templeos

"Excess Flood?" I thought He promised Noah not to do that again.

No. 2nd coming is intense.

Impressive, that took three days last time.

[Apologies in advance, I have read the FAQ, I am aware of the meta-culture regarding one-liners, but I failed to restrain myself anyway. My profuse apologies to anyone upset or offended.]

Is it Easter already?

> Excess Flood

How apt

I've always though TempleOS would be perfect for open hardware, like the Raspberry Pi. Would you ever consider porting it to ARMv7/8?

The RPi is most certainly not "open hardware", it's even more closed than a standard PC. On the other hand, there are plenty of truly open ARM platforms out there, like the OLinuXino series:


That said, TempleOS on anything but standard PC hardware would be interesting to see.

I doubt whether OLinuXino's GPU is as open as RPi's VideoCore IV (among the GPUs used on ARM boards VideoCore IV (used on RPi) is the most open one - unluckily this is no praise for RPi :-( ), but convince me that I'm wrong.

That is an interesting idea. But even more so, I would certainly pay for a full MOOC on low-level OS development from scratch derived from it. Terry seems to have a rare and very useful skillset.

This is a great idea. I'd play with it on the Pi.

It would be a lot of work, but it's possible. I made it wed to x86_64.

@dang why was this thread censored?

If you really want to join YCombinator, your best chance probably is to join as a technical co-founder on another project.

TempleOS features HolyC: "In other operating systems, I hated learning one language for command line scripts and another for programming. With TempleOS, the command line feeds right into the HolyC compiler, line by line, and it places code into memory it MAlloc()s. The compiler is paused at the command line, waiting for input. Naturally, you #include a program to load it into memory and, usually, start it. During the boot process, many files get compiled before you have access to the command line. (Don't worry, booting takes only a couple of seconds.) All the header declarations for the operating system are compiled and are available for use in your programs without needing to #include them. Everything is truly compiled to native x86_64 machine code, nothing is interpreted and there is no byte code."

TempleOS is impressive. The only thing I could say after watching the youtube presentations was "wow". I see great value in it, at least as a teaching tool - a bit like Minix.

But there's more to it. Another potential use would be for hardware intensive tasks, when you want to squeeze the last bit of juice you have. Maybe for nodes doing distributed computation where GPU are not the solution, to remove the cruft of a full OS. The compiler would have to be quite optimized too, and some network features would be needed. Then there are some scaling issues, but still, if TempleOS could be say 10 to 20% more efficient than a standard Linux, there could be a usecase. Even for a smaller percentage, I believe some guys in finance would like that advantage for high-frequency trading.

I will follow your progress with great interest. Best of luck!

I think there might be potential for TempleOS in the embedded hardware markets - specifically for music-making devices. If it were possible to demonstrate the use of TempleOS for high-performance music making - say as a host environment for a DSP processing system, or as a synthesizer or effects machine - then there could be a real opportunity there.

I'd encourage anyone interested in TempleOS to consider this aspect of the project before writing it off, offhand. It could be a chance for a new class/generation of creative applications to host themselves on the unique and powerful TempleOS core ..

Here's an article about Terry:


Can you tell what you want to do with TempleOS, who is supposed to use it, and what kind of skills you are looking for?

> TempleOS is a modern, 64-bit Commodore 64.

This seems nice to me, perhaps someone could build a super-simple machine in the style of the C64 that boots straight to TempleOS. Something like the TOSBox...? :)

I wonder how cheaply something like that could be built, like the Pi only x86_64 and with its own OS. Potential use in schools etc.

I'd like to find a use for TempleOS, without access to networking it's tough to come up with ideas. Any suggestions?

What do you seek to gain out of YC?

does it support systemd and docker?

I'm Terry Davis.

My email address is tdavis@templeos.org .

Hi Terry, I've seen your work on HN and Reddit in the past and love your ambition. You have already accomplished more on TempleOS than most of us will produce in our entire careers!

I wish you all the best with YC right now, and look forward to seeing your genius and art for years to come!

  [15:32] <@TempleOS> stop talking to God for a minute
  [15:32] <@TempleOS> God has flooded

Have you tried making the operating systems then use in the movies?

It looks like the classic hacker's OS.

Maybe you could sell forks of it to hollywood.

Your company could also create UIs in computer games like in Deux Ex or system shock.

Or what about repackaging it as a Retro Game Engine, to compete with Unity or Unreal but on the other end of the spectrum.

Think of GOG.com using it instead of DOSbox for games made for it.

I thought you were shadowbanned! Good to see that was undone.

This is a new account that has not yet been shadowbanned.

Proof: https://news.ycombinator.com/submitted?id=TerryADavis and https://news.ycombinator.com/threads?id=TerryADavis


For atheists probably yes:

> http://motherboard.vice.com/read/gods-lonely-programmer

'He still considers himself scientifically minded. "Today I find the people most similar to me are atheist-scientist people," he says. "The difference is God has talked to me, so I'm basically like an atheist who God has talked to."'


Some people have broken legs. Some people have diabetes. Some people have cancer. Terry has a mental disadvantage. Come off your throwaway and say this if you really want to rip on Terry.

Terry, ignore the hate as you've done in this thread so far. There will be a lot of naysayers, and that's okay. You're an incredible, incredible talent.


Would you treat someone with autism any differently than you're treating Terry right now? If so, you're encouraging a cruel, cruel world.

While Terry may suffer from Schizophrenia, it is merely a character flaw that he endures. But he is more than that. He is Terry, a human, a man, someone's son, someone's friend, someone's brother (?), he is a genius programmer.

He just happens to suffer from Schizophrenia. He is not his illness.

People like you give people like Terry a bad name. Marvel at his achievement, because I'm willing to guarantee it's more than what you've done with your life unless you're secretly Sergey or the second-coming of Jobs or something.

Take a moment to appreciate what he's done with his disease instead of looking at him disease-first. People with cancer can't control it, just like Terry can't.

> TempleOS deserves to be in the Museum of Modern Art? Give me a fucking break.

I didn't write it out of compassion. I really think his project could be showcased there and appreciated by art critics.

a number of historically productive geniuses had schizophrenia, as well as other mental illnesses.

Previous discussions on this topic have involved debate on whether or not Terry Davis is racist, the defence being that his use of the word "nigger" is an (extremely) unfortunate side-effect of a schizophrenic condition. I don't pretend to have an in-depth understanding of schizophrenia so, although I'm dubious, I'm minded to give Terry the benefit of the doubt. However, Terry, you could resolve this once and for all by clearly explaining to us whether or not this is the case.

I don't really understand how you're capable of looking past his incoherent rambling about the CIA actively persecuting him for his cooperation with IRAN to produce BigBang weapons and instead focusing solely on the n-word sprinkled throughout said ramblings.

Sure, out of context, Terry is completely racist. He's also helping Iran build weapons of mass destruction.

Well I have no idea if Terry really does have schizophrenia. I believe he has stated he does but I also remember reading that he was not official diagnosed? I might be wrong, please correct me.

If he is mentally ill in some way I would presume his use of the word nigger, etc. are related to that, perhaps in a similar way that somebody with Tourette's will use such words?

Can you explain why you downvoted an objective, calm, not-inflammatory request for information?

He is not his disease. Bringing it up every time he tries to talk isn't going to help anyone.

He's passionate about his work. His work has done more for him personally than I think anyone can even imagine. It's given him so much.

Like I said in another comment, people with cancer can't control it, just like Terry can't help but have his disease.

Fair enough, I understand all of that. If his language and behaviour are solely as a result of his disease, then I fully agree. All I want is clarification of whether or not that is the case. Just because someone is schizophrenic, doesn't mean they can't be a racist. And many companies would find it impossible to work with someone who genuinely holds those kind of views. Heck, if ycombinator effectively endorses those views, without knowing if they are genuinely held, many will feel uncomfortable working with ycombinator.

i almost don't want to get involved in this, and i don't intend to take sides, but wouldn't many companies find it impossible to work with someone "disingenuously" racist as well?

i am hearing that incontrollable spouts of racism should be overlooked if the person is "not actually racist". something doesn't add up to me. are racist jokes ok?

i'm just trying to say this is complicated. taking either side of this argument seems short sighted.

Who cares if he is racist? I don't. It is the most irrelevant thing to talk about. An operating system cannot be racist. Consider a career in HR if you like being so ridiculously petty.

> Consider a career in HR if you like being so ridiculously petty.

Personal attacks are not allowed on Hacker News. Please follow the rules:



you didn't think about your question very hard, did you? investors, for one. no one wants to fund a company whose founder is publicly racist (sans other racists). at the very least we can agree it is a PR nightmare?

more though, i'm surprised to hear someone call racism "petty". there is a first time for everything i guess.

i'm curious as to why aren't you posting on your real account?

It's a PR nightmare because money-grubbing journalists realize the amount of money they make is a linear function of the amount of "controversy" they can produce. Take Ferguson, for example. I would fund anyone with a good product because I don't care for grade-school gossip about what type of inflammatory posts someone might write on obscure forums in his free time - which, by the way, is why contemporary racism in the West is petty. As much as the media's incessant race-baiting would like to convince you otherwise, the only legitimate expressions of racism seem to be on obscure online forums. If we could reset the clock 50 years racism wouldn't be petty, but racism as a concept has been so diluted by the media that it is hard to take seriously any more. Calling someone a racist is just a way to make money now, and the public, especially the younger generation, is so forcefully anti-racist that is hard to care if one or two people are racist, even harder to see how it undermines one of their technological accomplishments.

This is my real account. I joined a week ago. People seem to dislike everything I say.

"If you smell dog poo everywhere you walk, check your shoes".

You said the code is limited to 100k lines then said it has 120k lines at the end.

No he didn't.

He said that he capped the lines of code at 100,000, has written a total of 120,923 lines of code during the duration of the project, is using 80,037 lines of code in the current release.

Somewhere, 40,886 lines of code were deleted/replaced. They were still written though.

Applications and Demos don't count.

The 120k lines is from http://www.templeos.org/Wb/Home/Wb2/LineRep.html#l597, but 23k of that is demos, 3k is a tour, etc.

I'm pretty sure the 80k quoted in the beginning is the actual code making up TempleOS; the other 40k is extras.

He could very well have written 120k lines, but there's only 80k in source control. 40k lines being written could be replacing other lines.

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact