6 ms·

I have watched most of the Mill videos, and while they are certainly impressive and innovative, I believe Ivan Godard is a perpetual non-finisher. He comes up

by pdq 12y ago

I have watched most of the Mill videos, and while they are certainly impressive and innovative, I believe Ivan Godard is a perpetual non-finisher. He comes up with grandiose complicated ideas, which turn out incredibly difficult to implement, debug, and close. And by the time he makes much progress, he has another idea that takes his focus, and starts working on that. He has these massive goals, which are certainly noteworthy, but possibly does not realize it's likely not possible to do them in one step. You have to release incrementally, get something out the door, and then move to the next. Not "Let's design and architect a complete ASIC from scratch, based on an architecture from scratch, using a prototype simulator from scratch, with a toolchain built from scratch, with a new ABI for Linux, and a complicated debugger. Oh, and this is actually a family of processors (low perf/high perf), not just a single processor we are building." It's the opposite of MVP. If he really wanted to make progress, he would find a way to incrementally build one piece of his suite of technology IP that he can license to existing CPU manufacturers and get into production. Then you have revenue, and can rinse and repeat. Otherwise the Mill as vaporware is a virtual certainty.

jodrellblank 12y ago
tl;dr: Molyneux'd.
- oneweekwonder 12y ago
  Populous, Dungeon Keeper and Black & White as well as Theme Park[0]. [0]: http://en.wikipedia.org/wiki/Peter_Molyneux#Games http://en.wikipedia.org/wiki/Peter_Molyneux#Games ps. I understand he hypes, but I feel people tend to look to much into negativity and forget the games that made their childhood great.
  - eterm 12y ago
    B&W was pretty meh though, all hype over actual gameplay, it's the first title of his that made people step back and go "hang on, what was all the hype about?"
- MBCook 12y ago
  He may have over promised but his games were generally actually released.
  - jodrellblank 12y ago
    Oh? How is Godus? http://www.rockpapershotgun.com/2015/02/13/peter-molyneux-interview-godus-reputation-kickstarter/ http://www.rockpapershotgun.com/2015/02/13/peter-molyneux-in...
yuhong 12y ago
I wonder how it compares to IA-64 for example.
ohazi 12y ago
I believe the appropriate term for someone like this is "architecture astronaut" Edit: Apparently this is where I heard the term: http://www.joelonsoftware.com/articles/fog0000000018.html http://www.joelonsoftware.com/articles/fog0000000018.html Obviously this was written from a software engineering perspective, but it seems at least marginally fitting, as the topic in question is literally computer architecture.
- noblethrasher 12y ago
  "Researcher" is also an appropriate term that also happens to have the virtue of not being pejorative. Even though they're not popular in the business world, we do also need people who will change their goal when confronted with new ideas, rather than just the ones who filter ideas based on their service to a goal. We even gave a Turing award someone like that a few years ago.
  - lambda 12y ago
    "Researcher" is appropriate for someone who actually produces academic output. It may not be immediately practical, but it consists of new ideas that have been formalized and tested sufficiently to pass peer review of other researchers. Has the Mill architecture done that? Are there any actual papers about it? Anything beyond marketing fluff? It seems like there are some ideas behind the Mill that seem interesting when you hear about them, but I haven't seen anything rigorous enough to even be reviewable. And it doesn't look like that is the intent, either. Rather, from all appearance, it looks like they are trying to drum up interest for investment.
    - noblethrasher 12y ago
      This may just be a semantic quibble, but I'd say that someone that produces academic output (usually in the form of original research) is an "academic", and the academy has it's own peculiar rules for deciding which output has merit. The Mill stuff is just an example of research taking place outside of the academy.
peterfirefly 12y ago
I think you are very wrong. I also think you don't know how wrong you are. You might even be Dunning-Krugery wrong, actually. You don't build bridges like that, do you? Or skyscrapers? What's the minimum viable product for something like this: http://en.wikipedia.org/wiki/Great_Belt_Fixed_Link http://en.wikipedia.org/wiki/Great_Belt_Fixed_Link http://en.wikipedia.org/wiki/Millau_Viaduct http://en.wikipedia.org/wiki/Millau_Viaduct You do need feedback, lots of it, but not necessarily from customers. It is not a social media platform on the web using the hottest new framework. Your feedback mostly consists of numbers from measurements. That's why they built a simulator first (and it has apparently been running for quite some time). That's also why they did work on a compiler that they did get something useful out of (they have since switched their compiler approach to making an LLVM backend instead). But getting to a point where you can run it in silicon (even just as an FPGA) is a long, tough slog. Especially if you want decent/realistic speed. Getting good numbers on power use is also both hard and expensive. And it's all worth nothing if they don't have the patents. And many of the innovations need the other innovations to be useful or even just make sense. I met both Ivan and a German guy who works on the compiler late last year and have corresponded a bit with both. What I have seen so far fits very well together... but of course there are still many pieces that they haven't shown (and that I haven't seen either).
dang 12y ago
This feels uncomfortably close to a personal attack. Not everything can be built via MVP-and-iterate. How much further could a project be from the MVP-and-iterate sweet spot than a brand new CPU architecture? People should be free to try wild new things, whether they succeed or not. A strong visionary, once at speed, won't be stopped by nitpicking comparison to less ambitious things. But I worry about how many nascent visionaries we miss out on when the culture trains us to disparage those impulses in ourselves, and so many people tell you you're doing it wrong.
- igodard 12y ago
  Thank you.

by pdq 12y ago

jodrellblank 12y ago

tl;dr: Molyneux'd.

oneweekwonder 12y ago
Populous, Dungeon Keeper and Black & White as well as Theme Park[0]. [0]: http://en.wikipedia.org/wiki/Peter_Molyneux#Games http://en.wikipedia.org/wiki/Peter_Molyneux#Games ps. I understand he hypes, but I feel people tend to look to much into negativity and forget the games that made their childhood great.
- eterm 12y ago
  B&W was pretty meh though, all hype over actual gameplay, it's the first title of his that made people step back and go "hang on, what was all the hype about?"
MBCook 12y ago
He may have over promised but his games were generally actually released.
- jodrellblank 12y ago
  Oh? How is Godus? http://www.rockpapershotgun.com/2015/02/13/peter-molyneux-interview-godus-reputation-kickstarter/ http://www.rockpapershotgun.com/2015/02/13/peter-molyneux-in...

yuhong 12y ago

I wonder how it compares to IA-64 for example.

ohazi 12y ago

I believe the appropriate term for someone like this is "architecture astronaut" Edit: Apparently this is where I heard the term: http://www.joelonsoftware.com/articles/fog0000000018.html http://www.joelonsoftware.com/articles/fog0000000018.html Obviously this was written from a software engineering perspective, but it seems at least marginally fitting, as the topic in question is literally computer architecture.

noblethrasher 12y ago
"Researcher" is also an appropriate term that also happens to have the virtue of not being pejorative. Even though they're not popular in the business world, we do also need people who will change their goal when confronted with new ideas, rather than just the ones who filter ideas based on their service to a goal. We even gave a Turing award someone like that a few years ago.
- lambda 12y ago
  "Researcher" is appropriate for someone who actually produces academic output. It may not be immediately practical, but it consists of new ideas that have been formalized and tested sufficiently to pass peer review of other researchers. Has the Mill architecture done that? Are there any actual papers about it? Anything beyond marketing fluff? It seems like there are some ideas behind the Mill that seem interesting when you hear about them, but I haven't seen anything rigorous enough to even be reviewable. And it doesn't look like that is the intent, either. Rather, from all appearance, it looks like they are trying to drum up interest for investment.
  - noblethrasher 12y ago
    This may just be a semantic quibble, but I'd say that someone that produces academic output (usually in the form of original research) is an "academic", and the academy has it's own peculiar rules for deciding which output has merit. The Mill stuff is just an example of research taking place outside of the academy.

peterfirefly 12y ago

I think you are very wrong. I also think you don't know how wrong you are. You might even be Dunning-Krugery wrong, actually. You don't build bridges like that, do you? Or skyscrapers? What's the minimum viable product for something like this: http://en.wikipedia.org/wiki/Great_Belt_Fixed_Link http://en.wikipedia.org/wiki/Great_Belt_Fixed_Link http://en.wikipedia.org/wiki/Millau_Viaduct http://en.wikipedia.org/wiki/Millau_Viaduct You do need feedback, lots of it, but not necessarily from customers. It is not a social media platform on the web using the hottest new framework. Your feedback mostly consists of numbers from measurements. That's why they built a simulator first (and it has apparently been running for quite some time). That's also why they did work on a compiler that they did get something useful out of (they have since switched their compiler approach to making an LLVM backend instead). But getting to a point where you can run it in silicon (even just as an FPGA) is a long, tough slog. Especially if you want decent/realistic speed. Getting good numbers on power use is also both hard and expensive. And it's all worth nothing if they don't have the patents. And many of the innovations need the other innovations to be useful or even just make sense. I met both Ivan and a German guy who works on the compiler late last year and have corresponded a bit with both. What I have seen so far fits very well together... but of course there are still many pieces that they haven't shown (and that I haven't seen either).

dang 12y ago

This feels uncomfortably close to a personal attack. Not everything can be built via MVP-and-iterate. How much further could a project be from the MVP-and-iterate sweet spot than a brand new CPU architecture? People should be free to try wild new things, whether they succeed or not. A strong visionary, once at speed, won't be stopped by nitpicking comparison to less ambitious things. But I worry about how many nascent visionaries we miss out on when the culture trains us to disparage those impulses in ourselves, and so many people tell you you're doing it wrong.

igodard 12y ago
Thank you.