Hacker News new | comments | show | ask | jobs | submit | jasonwatkinspdx's comments login

http://www.helionenergy.com plans to do direct conversion. It's a follow on to previous university work, dunno if they tested/measured that part.

Hughes Hoppe has done a variety of research that depends on finding interesting parameterizations of meshes. His papers aren't exactly lightweight on the math, but you may find some ideas there:

http://research.microsoft.com/en-us/um/people/hoppe/

http://research.microsoft.com/en-us/um/people/hoppe/proj/vfd...

http://research.microsoft.com/en-us/um/people/hoppe/proj/lap...


Awesome, thanks. I'll check that out.

Great! Hughes Hoppe's work is amazing!

Hugues Hoppe, unless he has changed his name.

Shit. Thanks for your correction, I can no longer edit the above comment to reflect it, sorry.

Life isn't an action movie. Killing (in?)famous people without getting caught or creating repercussions isn't a simple sure thing, unless you're bombing shepherds with drones.

While I agree life isn't an action movie, perhaps the long list of Iraqi scientists being assassinated or the even longer list of Russian, North Korean or Chinese dissidents being done away with seems to be a very simple thing for those oppressive, murderous regimes to do without much interference.

And what has anybody ever done about it? Maybe sent out a few strongly worded memos, if anything at all.


None of those people have anything remotely like the global profile Snowden created with his actions. It's not really comparable to a no-name nuclear scientist getting assassinated.

> Aircraft that are more expendable and can be upgraded with new tech quickly and at less cost.

I'm just a putz in an armchair, but I've been reading a lot of the think tank docs about this stuff recently. We could get a ton of value out of a simple platform. My guess of what it looks like:

  * 2 seat, enough cabin for long manned missions
  * optional autonomous or remote piloting features
  * primarily subsonic, but perhaps with a dash capability
  * no expectation of high speed high g maneuvering
  * oversized electrical generation
  * bomb truck for 16+ SDB's
  * missile truck for enough AIM-120's, etc to defend itself
  * open platform designed for continuous upgrade of radar, EW, EO, etc
  * only low cost stealth features
  * enough fuel capacity to loiter in theater
Something like this could be pretty cheap to fly and provide a lot of flexible capability as a truck for state of the art electronic systems and the emerging generation of smart munitions. It'd still be relevant in an increasingly drone dominated future as an optionally manned node in the "flying network". Leave the missions that require more extreme capabilities to more specialized aircraft. Do all the basic stuff with something cheap.

I’m not 100% sure how I’d tackle this problem, but the plane you described could play a roll. It’s not poorly thought out by any means. Generally, I’m a fan of a diverse set of assets for survivability and cost. Focus on specific missions instead of trying to do everything. For Air-to-Air the Mig-21 Bison is an interesting example of upgradable lower tech aircraft.[1] Currently, I think the ability to go fast is still an asset to defeat missiles and would be a severe limitation of a slower platform (okay if it’s unmanned and cheap though). High tech stealthier fighters with lower tech fighters and drones in support. Mainly all working as missile trucks or targeting platforms.

For a mission such as Close Air Support the perfect tools are the AC-130 [2] and the Super Tacano. [3] For Strike I think drones, cruise missiles, and JSOW [4] are the main way to go. As for Air-to-Air, a mix of assets is critical so defeating one doesn’t defeat others, and they can’t all depend on one commonality such as GPS. Would probably keep a few high tech and more lower tech manned bombers available and ready too.

Complicated, interesting problem with strategy, logistics, and politics all involved, but the bankrupt the country super weapons definitely is the correct course in my opinion.

[1] http://forum.keypublishing.com/showthread.php?121375-MiG-21-...

[2] https://en.wikipedia.org/wiki/Lockheed_AC-130

[3] https://en.wikipedia.org/wiki/Embraer_EMB_314_Super_Tucano

[4] https://en.wikipedia.org/wiki/AGM-154_Joint_Standoff_Weapon


Sounds like a next gen A10 warthog

The navy tried to pull out of the program at one point, Robert Gates yanked the leash hard to keep them in line.

The navy has a much better case for drones. They need to attack at a very long range to avoid land based missiles. They are likely to strike targets that are strategic like airfields or ships, which are easier to identify.

Still probably a good idea for carriers to have a mix of manned and unmanned aircraft.


Interesting. Where did you learn this?

It's downvoted because the Fergusun protests were not "because people didn't pay their bills", and it's clear the poster simply has no understanding of what life is like when you're forced to chose between paying a bill and feeding your kids.

You feel bad writing what you wrote, because you likely have a bit more awareness of how badly everything is stacked against the poor from the moment they're born, but you prefer to focus on pointless semantic debates along the lines of "color blind" and decrying "political correctness" that are old and tiresome.


A couple responses:

Even with the hardware, google can't shrink the window past 7ms or so, based on published reports.

To preserve consistency there are situations where you need to wait out the clock uncertainty.

Spanner still uses paxos for replication because ordering is only part of the problem consensus solves.


Check out the inverted pendulum on a cart problem, or go balance a couple bats on your hand. When near vertical the inertia of the pendulum mass helps you out, it's the inertia of the cart the motors need to overcome. The better the control loop the fewer and smaller the corrections will be, tending towards zero.

-----


> if you start with any prior, will enough additional pieces of evidence always let you converge on the truth?

That's the general trend: pooling data tends to make priors converge. However, converging priors isn't quite the same thing as everyone converging on the truth.

Statistical expressions themselves are an incomplete explanation, we routinely use assumptions about the direction of causality that aren't captured in them. See Judea Pearl's "Why I am only half Bayesian" paper for a discussion as well as an intro to how his framework approaches such independence assumptions.

-----


1. It's clear you did not read the paper and are ignorant that the author is high profile physicist and cosmologist famous for developing measurement methods.

2. The view of science you are parroting is naive and mistaken. Theorizing and experiment are both necessary and inform and constrict each other.

-----


There are plenty of great scientists who have nutty ideas about things outside of their specialty. I've only read the OP, but that's the vibe it gives me.

-----


Which is good, isn't it? I mean, how nutty was the idea that matter could bend space(time)? What about time dilation? Black holes? Particle/wave duality? The brilliant dots in the sky being immense balls of burning gas? Big discoveries begin with a nutty hypothesis.

-----


1. I don't care who he is, and nor should you, this is science and not the middle ages. I will not bow to a Barron, and I will not accept what an MIT professor says without applying my critical facilities to it.

The paper is extremely difficult to read and extremely long, very few people will have the time or comprehensive knowledge required to read it completely. This is a fault of the paper, not the readers.

2. Theorizing is great, ideas are free, but that is because their value is very low without proof.

-----

More

Applications are open for YC Summer 2016

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

Search: