Never been one to let the carrier drop

This is one of my posts that I make so I can link to it from facebook instead of repeatedly typing the same things in comments.

You’d think you guys had never watched pro wrestling.

Trump is playing a Heel. By choice, and I imagine to his great entertainment. It’s pretty clear, since he and Hillary are friends, that they decided to get together and rig a election. Nothing in the rules says they can’t do it, and I have to say, it’s starting to be a lot of fun to watch. He’s clearly having a lot of fun with his mustache-twirling villain role – now he may actually believe the things he’s saying, or he may not, but I would take anything he has to say with a grain of salt.

For me, this is a pretty winning situation. A heel is not going to win the election, and we don’t end up with a right-wing whacko running the country. But if you’re actually feeling outrage about Trump, well, I guess he’s playing his role well. What I really wonder is whether he will break kayfabe at the end or not. If he’s genuinely a force for good, he will, just to help his supporters understand how broken they are. Assuming they’re not just part of the gag as well, which – to be honest – they might be.

So, in a comment to http://www.sheer.us/weblogs/?p=3006, Alderin said “I agree that the software needs adjustment. The difficulty is that we don’t have the source code, we have to decompile and reverse-engineer it before was can being to find the bugs we need to fix.”

I think this is part of the hope of what we might gain through making a ANN the size and shape of a human.

1) It would let us find out what happens when you run our memetic software with a different instinct table
2) It would let us find out how our memetic software ends up rendered ‘on the iron’, which might be very instructive.

One of the questions that’s still very much open for debate is whether, even if we did know what to do differently for software, we’d be able to change the existing running software on the thundering herd of humans out there, or if all we could do is improve future generations. It is worth noting as a side note that, much as I loathe aspects of Christianity, I also recognize it was probably a significant upgrade from what was being memetically distributed prior to it being authored. Ideally a *really* good software upgrade would tend to be viral because the people running it would be better adapted – this I think was the plan in the fictional work Stranger In A Strange Land.

So, lately the DAW plugin market has been a bit saturated with a bunch of vendors all producing plugins that more or less do the same thing. This is great insofar as it’s driving down the price of plugins, but not so great insofar as it suggests no one can think of anything new to do.

My first and most obvious suggestion on this front is to abandon the beaten path of what’s been done before. In particular, might I suggest that there’s a huge GPU on most modern computers that could be used to play with building artificial neural networks to perform audio processing. I’m not exactly sure what the results of this would be, and maybe a GPU isn’t big enough yet to do meaningful amounts of ANN processing, but I still think it’s a neat idea.

I have memories when I was a tween of having a vision of manipulating a giant artificial neural network via a set of controls very similar to a mixing board, which controlled trigger levels and amplitudes of signals passing between subnets. I didn’t fully understand what I was imagining at the time, but looking back…

I realize that it’s become politically incorrect to say that all lives matter – that by doing so, some people somehow feel like I’m saying all lives *don’t* matter. Or like I’m saying there’s not a problem with people of color being gunned down by people wearing blue.

Look, I do get it that statistically said people have been getting the short end of the stick. However, I belong to another group of people who I would wager are a lot MORE likely to be gunned down by the cops – people with mental illness. 99% of the time I’m perfectly normal. 1% of the time I’m really not. I’m not *dangerous*, but I’m definitely way, way outside of normal. Numerous times I have had cops with their guns drawn and aimed at me, so I know of what I speak. I’ve never acted threatening towards them in any way, but nonetheless it would not at all surprise me if I were gunned down someday by a cop with had a itchy trigger finger.

But I’m not looking for special service for my group, because I think that’s part of the problem. I’m looking for general, broadband fixes that will help both people of color and the mentally ill – and anyone else who interfaces with our gaurdians:

1) Streaming video from body cameras, which can not be disabled nor easily tampered with
2) Guns and tasers that have small but powerful ANNs* and video cameras and software designed to make the ‘danger/no danger’ decision, so that even if a cop does pull the trigger, the gun will not fire. As a backup, two individuals (generally dispatch, who has streaming video from #1, and the officer in question) can override the computer to un-safe the weapon.
3) Until the day that #2 is a reality, any cop who shoots a innocent should lose their right to carry. Actually, anyone who shoots a innocent, period, should lose their right to carry. You’ve proven via the only test that matters that you don’t have what it takes to carry a gun. It’s *extra* important that this apply to cops, however, given that their job often puts them in situations where emotions are running high and split second decisions are needed.

#2 is just common sense. Humans are not really rational enough to be trusted with guns. Unfortunately we have a ton of them out there, and it’s going to be a long time before we realize this, but of all the guns you want to be safe, the guns carried by the guardians are the most important, because they might see use at any time, 24×7, they’re always loaded and always being carried.

In any case, while it may not be possible to preserve all lives under all circumstances, all lives matter. The lives of our guardians matter. The lives of people of all colors, stripes, and religions matter.

* = Artificial neural networks. It should not be that difficult, in a era when we can do face and voice recognition, to teach a computer to recognize when there’s real danger, especially once we collect a corpus of incidents where there was real danger from #1.

http://www.sheer.us/weblogs/?p=1727.

2 terabytes was actually a impressive amount of disk. Now I routinely slam 2T SSDs in things, and one system I work with regularly has a RAID array of PCIe 2T SSDs capable of 2+G/s read and write. I’ve recorded more than 2T worth of multitrack content. And so forth.

One of the things my friend Andy wanted me to do was figure out a mathematical model for measuring the impact of gratitude.

I haven’t done that yet – I’ve put in some time on it, but it’s resisting a easy solution. I have, however, as a side corollary been thinking about measuring suffering, which is not exactly the inverse of gratitude but is tangentially related to the inverse.

Measuring suffering mathematically is a important thing to be able to do in order to do triage to figure out which issues facing the human race should be solved first. I grant you that we don’t currently do this kind of triage in any meaningful or useful way – somehow the herd picks a flavor of the week to solve, but it doesn’t appear to be that connected to what’s hurting the most people.

Anyway, let me talk about my thoughts on the matter, and then we can come back to that if I still have the energy. If not, I’ll probably talk about it in a future article.

For measuring suffering, the first and most obvious thing to measure is direct impact. You measure the suffering intensity (we could arbitrarily scale this as between 0 and 1), the suffering duration, and the number of people impacted. Multiply these three numbers together and you have the suffering quotient. It doesn’t really matter what scaling you use on the three numbers, as long as you use the same scaling for all problems, since all we’re really trying to get at here is some meaningful way to measure that can be used to compare sources of suffering and figure out which ones to fix first.

Beyond the direct impact, there’s the indirect impact – the “If mama aint happy nobody aint happy” impact. For example, the police shooting innocent people has a large indirect impact – it makes everyone sad and angry – while kidney stones would have almost no indirect impact. Deaths tend to have indirect impact. This can be measured in the same terms, but is a lot more complicated to figure out what the appropriate values are and likely involves a statistical distribution of values depending on the number of people indirectly affected and how strongly the issue affects them.

The direct and indirect impacts just sum together, nothing complicated there.

There’s also the potential impact of not resolving the suffering. For the cops shooting civilians, you have a potential civil war on your hands, which has a enormous suffering quotient. For cancer and AIDS you have people dying, which is not as bad as people dying in a war zone but still bad. For some things, this third effect doesn’t apply.

Because it’s a conditional, it should really be in a separate column rather than summed with the other two.

Another question is where to get the suffering quotient from. You can get people to self-report, but outside the land of physical pain a lot of suffering quotients (fear of being homeless, fear of getting shot by the cops, etc) are really hard to quantify. It’s possible that some of this could be determined by looking at the fatigue poisons in people’s blood, and possibly neurotransmitters behind the blood-brain barrier. I don’t know to what state our science is when it comes to measuring misery, so I don’t know if this is something we already know how to measure or not.

Anyway, I do think figuring out a mathematical model for measuring suffering and using it to measure the current large problems facing us would be a smart thing to do.

This is more of me trying to get a handle on Vienna. This one is interesting in that *none* of the instruments in it have any corporeal existence – well, many of them might be based on samples, but as of now they’re all software running on a x86 😉

String Blues

http://www.umc.org/news-and-media/florida-bishop-responds-to-massacre.

I’m not a huge fan of Abrahamic religions. But increasingly, I see people growing beyond the horror and hate of religion, and embracing the idea of love, tolerance, acceptance, and the like. In the midst of the dark, fear, and dystopia, every ray of love, life, acceptance, and hope is appreciated.

So, I keep going back and forth as to whether this would be a good idea. I think I’ve talked about it before. But it would be technically feasible to add a device similar to Amazon Echo to every dorm room and frat house. The idea would be to create a ‘safeword’ – something that would almost never false positive – that would call 911 and give the location of the device and audio clips from it.

The downside is it’s surveillance, and it would also probably get abused by government which would then insist the devices could be used for monitoring. The upside is that it could vastly cut down on things like campus rape. I can’t decide if it’s a good idea or not. But watching The Hunting Ground makes it seem like it might be.

It does occur to me that there could be a easy opt-in system.. a app you’d install on your mobile phone that listens for a word and calls 911. It’d be power-hungry, you’d only activate it when going into a situation where you knew there was danger.. but it’d be useful. You could also have it call 911 if you *didn’t* say a certain word every N minutes, if you were concerned about date rape drugs, or if someone tried to disable it without disarming it first.

With a *lot* of CPU power (more than a phone likely has, you’d have to stream it to a more powerful computer), you could add voice detection so the app would know whether it was hearing your voice or not.. You could also add a monitoring center, like high end alarm systems, so rather than calling 911 you’d pay a monthly fee to have someone listen in and determine if the situation was all right. Groups of friends going to the same party could all register with it, or it could use GPS coordinates, so the monitoring system could do things like sending a message ‘Amy might be in trouble, we haven’t heard her voice in 20 minutes and she didn’t give the signoff / all clear word’

The other difficult thing would be getting people to know and understand that college campuses are dangerous places.

I also think encouraging the idea of affirmative/positive consent would be a good thing.

When I moved to the new house in Seattle I replaced the oil burner with a heat pump. This was mostly self-preservation – the inverter-drive pump I chose draws about $200/month worth of power off the line, vs $600 worth of oil.

The installers managed to take many days to install it, despite it being a basicly drop-in kind of thing. Partially, I chose a Mitsubishi PUMY, which has a computer network between all the various components and requires assigning unique addresses to them all, something that apparently was too complex for your average HVAC guy, and partially they needed some help installing the thermostat. (Yes, I installed it. They had spent a day trying to get it to hang on the wall.. the old wires were too short. A couple of quick disconnects and short extenders later, while they were out on their lunch break, problem solved.

I have come to suspect the people who I bought it from are idiots.

It’s a multi-zone system.. multiple heat exchangers with multiple fans.. because me and Gayle have different ideas about what is a comfortable temp. After my first “free” tune up, the basement heat exchanger wouldn’t turn on.

I called and they sent a tech out. The tech reported it needed a new circuit board. I expressed dubiousness, but told him to go ahead and order it. Tech went away. Later, another tech came out to install the circuit board, and after testing reported that it needed *all* new circuit boards, because a “power surge had destroyed something”.

Now, I’m dubious as anything. First of all, all the other zones were working fine. I couldn’t think of *any* failure mode that would have all but one zone working but need a new, say, inverter drive board. So, I sent him away with a “well, order whatever parts you think it needs…”

Then I got out the manuals. After perusing the relevant bits, I got out the voltmeter, and measured the voltage across the network cable to the zone that wasn’t working. 0 volts. Hmmm. Pretty sure the manual says it’s a current loop and I should see 24 volts at all times.

I do some quick tracing, and discover that they had used wire nuts barely adequate for two 24AWG wires to bind together 5 20AWG ones. I tug on the bundle of wires, and one comes loose. I go and get the proper wire nuts from my toolkit, replace them, reboot the system (after a couple of false starts, turns out you must turn off the compressor last and turn it on first.. which is in the manual, but not in the obvious place) and lo and behold, my zone works again.

I’m better at troubleshooting a AC system than *two* technicians who do it as their *full time job*?

I could forgive them more easily if it had been something that was unique to a computer driven / networked HVAC system. But this issue would have broken even a plain ol’ relays and motors system.

I’m trying to decide what a appropriate thing to do is. I don’t want to be deliberately hurtful (i.e. call them and say “you guys are idiots..”) but at the same time I feel like they should in some way learn from their mistake.

What’s more, I paid for a year’s service contract, but there is no way in *hell* they are ever touching this system again. I’m scared what they might do to it. I have no doubt that if they’d replaced every PCB in the system, it would have *lowered* the reliability. Do I demand a refund? I also feel for anyone else who might be getting their system serviced by these guys. Do I publish my story?

Even worse, I looked at the PCBs.. they have surge suppression out the ying-yang. MOVs. zeners. Snubber caps. The idea that a surge could have knocked this thing out and not damaged any of the much-less-well-built electronics I have all over the place is laughable.

Meh. I have no idea what the right thing to do socially is. There has to be somewhere between “If you can’t say anything nice, don’t say anything at all” and letting people walk all over you.