Help wanted, the real story on that

For the newbs to ask questions. Note there really isn't a question/answer section other than this. The forums in general are supposed to be practical knowledge shared by experts, not custom free tutoring for all - we simply can't do that - there aren't enough of us. So, new guys can post here and if you're good, our moderators will "promote you" to being able to post on the rest of the site. It you're a bot, totally inactive joiner, or just a luser - this is where we find that out too and say bye.
Forum rules
REAL NAMES ONLY. This is a legal attribution issue. I don't care what nick you use everywhere else. Obviously fake names will be removed immediately.
If qwerty1337 posts an idea, anyone on earth can claim to be that guy and take credit - is that what you want? Or worse - qwerty can claim it was really his idea more easily. Feel me here? This is for adults.

This is a place for newbs to find their feet. Most of this board is for answers, not questions - read the title, it's a place to share knowledge, not a custom tutoring site - we don't have resources for that, and it destroys the signal to noise ratio with almost the same questions over and over from people who won't be bothered to read first. Posts on this forum will self destruct in 20 days unless a moderator wants to keep them around (and maybe move them elsewhere here).
IF you have a suggestion for where you think we should put your post in "the book" part of the forums, put that on the first line of your post. We might read it, you never know, and if you put up a couple of good ones, we might make you a "Real" member here with full privileges. Most of this board is post-grad kind of info, not grade school. That kind of thing will mostly stay here in playpen.

Help wanted, the real story on that

Postby Doug Coulter » Tue Jan 30, 2018 9:23 am

I've been getting various offers to help in various ways (and thanks, but see below). Sadly, most of them wouldn't be helpful, or fail to understand the situation.

For example, I really don't need to be told I can raise money this or that way on the web. I'm sure that if I put up a request on one of the websites dedicated to that, some decent sum of money would come in, maybe enough to pay a lab grunt for a year? Even if everyone who viewed that Vice video gave us the price of a lunch...it's an amount of money - not a huge amount, or an ongoing rate that would fund human help. Having been a CEO I understand this stuff, and I even know how to romance venture capital and so on. Not going there. Every entity on earth that will provide a rate of money attaches strings and/or gets impatient. Haven't seen an exception yet. Yet 100% of people one hires, expect an ongoing rate of pay for their work, and deserve it usually.

In my eyes, that's an immoral act waiting to happen. So, we hire a guy(s) with no way to pay him after a year - he pulls up roots and relocates here so as to be able to help - and then we strand him high and dry if web interest fails (this is not a place with a ton of job availability - heck, it's 26 miles to the beer store, much less a place with real jobs). Like a startup only with higher chances of failure and less of a big payday - yeah, if we get it, it's huge, but pretending it's in the bag would just be deceitful. People suggesting that seem to think that some amount of money is a rate of income, and it just is not (especially with today's depressed interest rates). An amount that looks really good on one of those sites wouldn't pay a real scientist for a year...I can't fund other people out of my limited bucks - I just scrape by and eat a lot of cheap noodles in order to afford the lab gear. Most workers kinda want and even need a real paycheck. And I have some help there as is - BillF has funded - or somehow acquired - quite a lot of the basic equipment we're using, as well as a lot we did use, but don't anymore - it's the nature of science to try some stuff and not all of it works, so you wind up buying more gear than winds up in the final work. And one reason we can even afford this is that old gear depreciates faster than a crashed hot rod. You can't make your money back selling what you didn't wind up using. And even trying takes time and energy away from the real mission. Bill even picks up a screwdriver now and then, and asks the best "dumb questions" a scientist could want (this is not a joke, it's an important function and I'm lucky to have that help).

As it turns out, in largest part, we are fairly fat on lab gear at the moment - we'd accept donations in some cases, but the kind of stuff that'd actually help no one gives away (for example, my turbo pump is going to wear out - it's merely a 512 liter/second turbo-drag pump costing...$23 grand for the latest, slightly smaller capacity model on Kurt Lesker. I'm not holding my breath that one's gonna fall out of a tree anytime soon. If someone's feeling that generous, well, I'm not hard to find, but like I said, I'm not holding my breath. Most of what we need either isn't available off the shelf at all - no one's needed "that" - or is but is so expensive it's out of our league; how about a broadband RF amp that covers some octaves, at the kW power level, and will drive output voltages in the 10s of kilovolts, and not mind a fairly variable output impedance? I've seen broadband amps at the power levels - in the several 10's of K$ - but not at the high impedance we need, and any RF guy will tell you that a 1000::1 stepup transformer that's broadband (say 60 to 1 frequency range) is not happening in this universe. So maybe that's a "there is no such thing even possible" so we have to instead build one narrowband thing at a time and just test endlessly till we find the right set of conditions. Which is what I'm doing now, slowly and laboriously - something that fixed that would help - a way to setup something that could test things in an automated way without having to build stuff for every pass would indeed help us get there far faster - computer code to try things and do "gradient descent" has been do-able for quite some time, despite the AI guys thinking they just invented it (multidimensional complex gradient descent and simulated annealing have been around since before those guys were born). But...

A _real_ mathematician and serious physics theorist would be nice to have around (we only need the standard model, but need prediction rather than description), especially if he was an ace programmer with experience in programming simulations with all the optimization tricks. We have a "worse than NP complete" problem here if we want to simulate what goes on in that tank - which is a huge part of why we have the tank - it's faster and more accurate than is achievable with computer simulation at this point, according to experts in that field, and yes, we've been in touch and they throw up their hands and mutter "heat death of the universe" kinds of things. Still, you always hope there's a trapdoor into the answer to this specific set of conditions that someone with insight and super skills will notice. Otherwise "the universe is its own best simulator and runs in real time, once you set up the simulation conditions" - thing is, that's descriptive, not predictive - it tells you what is, not what might be.

As I said, to do this predictively is tantamount to solving the 3 body gravitational problem feedforward, only we have 1020 things, not three, they don't all have the same inertia to gravity coefficient, and they don't even all attract - some repel. It could be possible to get a good enough subset to be useful, but in general, math has let us down and is no longer our Queen. Like I said, worse than NP-complete. And yes there are things worse, and this is one of them, practically speaking.

A data mining expert who could create a GUI that would let us look through our data - which is numeric in a SQL dababase, and also a few sync'd video and audio streams - in variable loop length displays and with various parameters plotted. I've got all that "the hard way" - eg find some interesting event on some plot, note its timestamp, open the other streams at that time...maybe edit video/audio into loops (or just use my brain to remember what I saw in that window while looking at some other data from the same period in another), adjust the things plotted via parameters and which program to use (I wrote all these) - but to really examine an interesting event, it now takes all day to really grok what happened in a few milliseconds - and what led up to the event - because if it was "interesting" we want to be able to cause it to happen on demand, not wait for a lucky accident. Which we've had, and which is why we think we're onto something - one thing we have that previous workers, including Philo Farnsworth and crew didn't have - good test equipment and data acquisition. It's how I know he wasn't talking pure trash on some of his more outrageous claims.
We saw it too - replication - but we also caught a clue as to why, just not a good enough one - yet.

Sadly, this leaves out pretty much all students and even more limiting, people who can't afford to do this for the love of doing it. That stinks, but what else to do? If I demote myself to "chief money grubber" than the project loses it's skilled workers - (me), who did all this and wrote all the code, made the stuff, turned the screws, and so on, and it also loses the integrity of being just about the only guys out there that are not and cannot be a scam - we're not taking money so how could we be conning anyone? So far, all alt-energy guys begging money have either been "true believers who are wishful but can't deliver" or outright thieves. I'd rather not be painted with that brush - I'm neither, or so I hope (haven't delivered yet!).

Yes, I know how to raise bucks. I'd need some serious convincing to go there, or even use the services of someone who just did that, so I could still do what I do. I don't want to go down that path and take bucks for nothing in return - I'm doing the best I can, but I won't tolerate the strings that almost always come with money, or make promises I'm not sure I can keep - honor is big deal with me. The fact that lots of people do it means nothing. Lots of people do things I would never do.

So, if you want to help - and we could certainly use real help - you need to be pretty competent at something we need, be well off enough to have time to do it for love, and be willing.
There's not a ton that can be done remotely, fond as I am of telecommuting myself. If you're one of the aforementioned genius theorists with a practical bent, it might be worth sending a backup of some of our database so you can learn to mine it. A normal competent team of maybe 10 could do that in maybe a year or few. A real hotshot might do it in a month. But also, it'd take me a week to put the source data together and get it sent - and I risk wasting that week.

See the quandary? I'm plugging along, not as fast as any of us wish, but to get to the next speed is a huge hurdle.

Note to mods - I put this here because it's off topic everywhere else, and will remove/change it as the situation indicates.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Return to Playpen

Who is online

Users browsing this forum: No registered users and 1 guest