You can engage a lawyer to sue someone no win no fee and that's fine if you know who is infringing. However how do you find out? If I have a patent on faster superscalar instruction execution for example, I'd need buy a bunch of chips, remove the package with acid, photograph the die, and then work out out if their design is infringing. And then I'd need to go to court and fight them over it. And they'll presumably counter sue. If you're Intel or AMD you probably have people doing competitor research they could do the reverse engineering part. For the average inventor that's not viable.
And for a lot of things it's not just a couple of companies building something, you've got loads. Some you've never heard of in China. And in each case you're going to have to reverse engineer the hardware and software to work out of they're violating your patent.. Some are bound to slip through.
And securitising IP is what makes all this viable. If invent something I talk to a bunch of IP holding companies and see who'd pay most for the IP. I sell it to them, they collect royalties from anyone infringing. If I couldn't sell IP I couldn't transfer it to them for a lump sum and let them handle working out who is infringing and collect royalties. In fact I bet they'd just ignore patents from small companies in that case - they'd know that those small companies don't have the resources to enforce.
Except the original justification for patents is so that the people who own the factories and printing presses have to pay the people who do the inventing or writing.
No copyright means the publisher keeps all the money and the author gets none. No patents mean the factory owner keeps the money and the inventor gets none.
In January 1842 Charles Dickens and his wife, Catherine, traveled to the United States. Dickens wanted to see the sites, learn about the country and do research for a future series of articles.
While on tour Dickens often spoke of the need for an international copyright agreement. The lack of such an agreement enabled his books to be published in the United States without his permission and without any royalties being paid.
This situation also affected American writers like Edgar Allan Poe. Poe's works were published in England without his consent.
Dickens first realized that he was losing income because of the lack of national in international copyright laws in 1837 when The Pickwick Papers was published in book form. At times the novel was reprinted without his permission and sometimes even imitated.
Lobbying by people like Dickens and Poe is what created a system where the US respected the copyright of UK authors and vice versa. Before that if you bought a copy of a book by an author outside your country, the author got nothing. And Dickens was apparently driven crazy by people launching copies of his books with the endings changed!
Of course these days if you got rid of copyright and patents you'd find that the factories and printing presses would all be in China or some other low wage/minimal worker's rights jurisdiction.
IP now means even when an iPad is made in China, most of the value stays in the US
The chart shows a geographical breakdown of the retail price of an iPad. The main rewards go to American shareholders and workers. Apple's profit amounts to about 30% of the sales price. Product design, software development and marketing are based in America. Add in the profits and wages of American suppliers, and distribution and retail costs, and America retains about half the total value of an iPad sold there. The next biggest gainers are South Korean firms like Samsung and LG, which provide the display and memory chips, whose profits account for 7% of an iPad's value. The main financial benefit to China is wages paid to workers for assembling the product and for manufacturing some inputs-equivalent to only 2% of the retail price.
Get rid of IP and you end up with a dystopia where no one invents anything, you've got no idea if the device you buy is genuine and the only people making money are the people who own the factories in China making the hardware. All the businesses in the US, UK and EU which depend on license fees disappear, and so do all the jobs.
And if you do invent something either you keep it a trade secret, or you hand it over to someone to manufacture. However if you do the later they don't need to pay you a penny! In fact allowing people who invent things to work with people who manufacture them without losing all rights is the original justification for copyright and patents. Trademarks are there so people can tell if they have a genuine device.
Also the GPL depends on copyright. If copyright didn't work, companies wouldn't need to release their changes back to the users. So all open source would disappear and each company would maintain an incompatible fork, keeping their changes from competitors.
It's complete hypocrisy. They want to pay cheaper bills by forcing the government to regulate the telcos so the telcos treat every packet as equal.
Meanwhile Google and Facebook ban, censor or demonetize people whose opinions they dislike. Google turned a blind eye to abuse of their Youtube by paedophiles that eventually resulted in advertisers boycotting it.
And Google openly mess with the ranking on Youtube and Google search so people they agree with rise to the top and people they disagree with are hidden from search.
I.e. they only support net neutrality for telcos because it will save them a few pennies on their broadband bills. Their platforms are not neutral politically in the US and outside the US they're happy to offer walled gardens which are not neutral either.
If they actually had any principles they'd want both them and the telcos to be common carriers. And not launch non neutral services outside the US.
Well I want the script to run on macOS, Windows Cygwin and Linux. I dunno if I trust either/dev/random or/dev/urandom on all of those. And openssl and bash are installed on all of them. I trust the openssl guys more than I trust the OS vendors not to have some backdoor or a cryptographically crippled/dev/[u]random.
If Intellectual Ventures didn't exist then you'd be screwed - you'd have spent money on a patent and you need money to run your business but you don't have money for lawyers to collect royalty fees.
Now you can make an argument that the patent system is flawed but this transaction is a valid use of it.
It's like how ambulance chasing lawyers are widely reviled as parasites but if a fear of getting sued makes a business take extra precautions to avoid killing or injuring people, that makes me think the ambulance chasers are doing a societally beneficial thing. Essentially having middlemen to make the information passing more efficient can improve society. And in the US the information passing is done by lawsuit.
I remember reading about a portable computer design in the very early 80's. It was designed in the UK and run off AA batteries. Because running it of non rechargeables was prohibitively expensive they decided to run it off Nickel Cadmium rechargeable cells. And they had a very simple circuit to trickle charge them when it was plugged in. So when you unplugged it and ran off batteries those batteries were guaranteed to be fresh. Problem is, non rechargeable batteries get hot if you try to trickle charge them. So in the UK they put a sticker on the back to say 'rechargeable batteries only'. When it came time to launch in the US they were advised this was a bad idea - if someone ignored the sticker, put in non rechargeables and burned themselves they'd sue and maybe win. However they went back to the engineering team who pointed out that there was an easy fix - stick a thermistor in the battery compartment and shut off the trickle charge if it got warm. This seemed to me to something rather profound. In the UK if you're dumb you get injured and that's just tough. In the US you have a right not to be injured even if you are dumb. In a sense the system in the US had the intelligence, not the individuals. And all the information passing was done by lawsuit, or more accurately people making subtle improvements to minimise their chances of being on the wrong end of a lawsuit. This seems to me to be a more scalable system.
Of course patents can produce a comfortable cartel of insider companies who've sued each other and cross licensed their patents but who can use those patents to discourage new entrants to the industry. And it's clear from Intel's patents on SSE that x86/x64 won't be patent free for a very long time. AMD has a patent licence but is not profitable. Intel might end up with a monopoly and technology might stagnate.
And you can make a case for Tort Reform in the US to reduce the worst case damages in a personal injury lawsuit.
But I think getting rid of patent trolls and ambulance chasing lawyers completely is a bad idea. You just need to fine tune the regulation to maximize their incentives to do socially valuable things and minimize their incentives to do socially harmful ones. Of course in the US's hyper partisan, clickbait haunted world everything is boiled down to "Those evil $(OTHER_PARTY) are up to Pure Evil again. Share this and vote for us to stop them!"
openssl rand n does use/dev/urandom on a Unix like OS, but it doesn't just directly read n bytes from it, it does a bit of munging, like reading a few bytes at startup and using that as a seed for its own PRNG.
And of course openssl works on OSs that don't have a/dev/urandom at all - e.g. on Win32 it calls Win32 crypto function to get a seed and then uses that for its own PRNG.
Basically the openssl guys don't seem to trust/dev/urandom to be random. E.g. the stack exchange source says
You do not want to trust that the random source is random./dev/urandom in particular is not trustworthy because in guaranteeing that it wiill not block on low-entropy conditions, it fails to guarantee that the output is actually random. The manpage man 4 urandom has more information on this, including a cryptic allusion to an attack some government body may or may not have predicated on this condition. Suitability tests can be conducted on this data, or it can be mutated in some way to concentrate entropy, or different sources can be combined.
A read from the/dev/urandom device will not block waiting for more entropy. As a result, if there is not sufficient entropy in the entropy pool, the returned values are theoretically vulnerable to a cryptographic attack on the algorithms used by the driver. Knowledge of how to do this is not available in the current unclassified literature, but it is theoretically possible that such an attack may exist. If this is a concern in your application, use/dev/random instead.
And essentially running the output of openssl rand through sha512 a couple of times does a bit more munging. And running it through sha512 a random number of times and then encrypting with a per device RSA private key does even more munging.
As with most things in life you have to decide what level of munging you feel comfortable with and code to that level. Personally I use the maximum level, though of course it's debatable whether this is overkill or not.
It may well be that just reading from/dev/urandom on your OS is good enough.
Anyhow the judge's order was later declared unconstitutional. That didn't stop Hogan from sueing them and bankrupting them. And now Gawker is Thiel's bitch. They pissed him off by outing him as hay while he was on a business trip to Saudi Arabia.
I.e. get 4096 bytes of random data from openssl. Hash it to sha512 a couple of times to shuffle the bits. Then encode to base64. Remove any characters that might cause problems leaving [A-Za-z0-9] which most sites allow. Then pick the first n chars (where n defaults to 32 and is passed as an argument)
E.g.
randpass
To get the default 32 char password or you can override that on the command line if you know the maximum password the site supports. E.g
randpass 64
Gives a 64 char password
I store 'em in a text file in a small TrueCrypt volume as a backup and sync that across devices. For things like imgur.com I let the browser remember them because there's nothing too critical about the account. The passwords has 62^32 combinations which is hard to brute force and even if someone did brute force it, who cares? All they'll learn is that my password to other sites needs another 62^32/2 tries to brute force which is probably not worth their time.
And if you know the site allows longer passwords, just tell randpass on the command line. E.g
randpass 64
Actually once I got this to work I invented a more elaborate version that encrypted with a per device private RSA key, randomized the number of sha512 hashing stages and so on. But that only matters if you think an attacker can work out what openssl rand returned on your device, which they probably can't.
I've got a degree, but I got in the UK before tuition fees and in a subject where it's kind of necessary, i.e. engineering. Still it's harder to justify asking for a degree in other subjects, like the arts. For a lot of programming jobs like web development it's also arguable that you don't need a degree to be able to do it. Even writing embedded code like I do isn't something that actually needs a degree - I've met a lot of people who do it well and don't have one, and a lot of people who do it badly who do.
On the other hand if you have more applicants than jobs why not pick the best qualified ones? And of course if employers demand a degree, more people will do one.
Now since I got mine, more people are getting degrees, prices have gone up and they're doing them in subjects where it is less necessary.
I think it's a classic case of inflation. If employers can eliminate candidates on the basis of graduate/non graduate, of course they'll do it. And if they do it, more people decide they need to get a degree. In the UK the percentage of graduates has increased enormously, tuition fees have risen because the government can no longer fund all those graduates to study without them paying and so now doing a degree means taking on a lot of debt. In the US of course it's probably a fair bit worse.
Of course it's hard to get out of this trap of people only needing a degree because employers demand one, and employers demanding them because they know they can find someone with one. Meanwhile most people are doing degrees in subjects which aren't really helping them do their jobs. Because if you're force to do a degree you don't actually need why not do it in a subject you like.
One option would be to allow people to default on their loans. But that would burst a bubble way worse than subprime mortgages
It was $1 trillion in 2014 and has a high delinquency rate
The total outstanding student loan balance is $1.08 trillion, and a whopping 11.5% of it is 90+ days delinquent or in default. That's the highest delinquency rate among all forms of debt and the only one that's been on the rise consistently since 2003.
It's easy to see why. If you borrow $100K or more to get a degree and end up in a bad job, you're going to end defaulting.
Maybe if the government only offered degree loans in subjects which have some justification economically they could deflate the bubble because it would forces universities to drop the price of degrees in everything else. Of course telling academics their subject is no longer eligible for loans will cause a massive shitstorm and accusations of philistinism from self interested academics.
Another option would be for the government to get out of the student loans business. Bursting such a large bubble of bad debt will have dire economic consequences however it is done though.
OEMs have a long history of treating AMD like the redheaded stepchild.
Back when I used to care about gaming and GPU performance I got mostly Intel/NVidia systems. When Athlons came out and they were competitive and ATI had a similarly decent generation I decided on a AMD/ATI build. Got all the parts mail order in the UK. Put it together and it was unstable. I swapped some parts - luckily the vendor I used was very understanding. Eventually it turned out the SDRAM I was using couldn't support the FSB it claimed. I ordered an Athlon with a slightly faster FSB (333 MT/s) than the norm and DDR1 SDRAM PC2700 DDR-333 to match. The board autoconfigured to use the 333 Mhz speed as expected, and it turned out that was optimistic. If I manually underclocked to the slower speed (IIRC 266Mhz) it became stable. ATI's drivers were more bloated than NVidia's - the control panel app used.Net for example, while NVidia's was straight Win32.
So I went back to Intel and NVidia. It seemed like Intel's and NVidia's Bios and drivers were more mature. E.g. I've never had to mess with an Intel board's memory speed configuration.
I remember Michael Dell ranting about how AMD CPUs were a pain to install because they didn't have a heat spreader back in the Athlon XP days - so you could crack the die installing the heatsink. I don't remember this being an issue with my Athlon, but then I wasn't installing them in volume. And hey, who can object to a CEO knowing about stuff like this - he'd clearly experienced the problem first hand.
So I can see why OEMs are a bit biased. Plus of course Intel probably do everything they can legally to keep OEMs from using anything but Intel CPUs.
Of course even if I buy Intel and NVidia it's still pretty important that AMD CPUs and GPUs are competitive. You can see that with Intel's recent generation of CPUs. For ages Intel were making very minor performance gains with each generation because that was enough. AMD became competitive with Ryzen and suddenly Intel produce a generation of chips with a big performance gain.
So I'd like to AMD to be competitive. Right now it seems like AMD is competitive for desktop machines, not so much on mobile. Which is a shame.
It could well be AMD's problem is primarily a software one actually. It certainly used to be.
Your original post parodying him was actually pretty funny.
This is less good
It was just another right-wing YouTube freak pretending to be persecuted by the SJWs, like the guy who sprayed himself in the face with Axe body spray, pretending it was Antifa attacking him with bear spray and then laid down on the ground pouring milk on his face and crying like a wee bitch.
So because some dude lied about being maced, that means that AntiFa violence is not a problem? I've seen loads of videos of AntiFa macing people, or hitting them over the head with a bike lock. Not all of them are fake.
And look at Styx. He started off, as far as I know, an apolitical gamer. GamerGate politicised him, but GamerGate was just a bunch of people who play video games getting in a fight with tech journalists. I don't have time to play video games these days, but I've met a few tech journalists IRL and I've never had much time for their opinions. So GamerGate's revelation that these people are a cabal of liars wasn't much of a surprise. Nor of course does it make much difference to society. As someone once observed on the relative importance of music and video games 'A society with no video games is easy to imagine wouldn't be all that different from the one we have. A society without music would be horrific'. Video games are just not that important.
But then something strange happened. Trump's popularity wasn't despite media criticism of him, it was because of it. Suddenly something very like GamerGate - essentially a rebellion against the media - happened and it had real consequences. People voted for an outsider whose candidacy was seen as a joke by smug media insiders because they hated those smug media insiders more than they hated Trump.
That's the reason Trump supporters don't care about the Roy Moore allegations. They know similar allegations exist against Democrats but the media simply decided allegations against Republicans are credible and ones against Democrats are not, because the media is ruthlessly partisan.
And it seems the US left side hasn't really learned from this. You're just doubling down on the criticism of both Trump and his supporters. Youtube and Facebook are censoring them, and people who claim to believe in Net Neutrality excuse this as 'a private company censoring people doesn't violate the First Amendment'. Which is true, but irrelevant.
My point is that this is more likely to make them double down on their opinions than to change them. And there are worse people out there than Trump waiting in the wings to take advantage of a disillusion with politicians and the media. There's AntiFa commies on the left and Alt Right fascists on the right. It's fair to say these people are not really joking when they meme about using force against their political opponents.
Back in the old days the US's political system essentially made sure extremists had no power. Someone on the extreme right wouldn't vote for the Democrats - they'd have to vote Republican or abstain. So the Republicans could nominate someone centrist and then tell them that the alternative, a Democrat was even worse. And on the left the Democrats could do the same - nominate a centrist insider safe in the knowledge that their extreme left couldn't vote for anyone but them. It was all pretty corrupt, but it had the advantage of keeping the really dangerous people marginalized.
Look at the last election though - Trump as an outsider got the GOP nomination. Sanders would have got the Democrat one in a fair contest and probably would get it if he run again. It's easy to see that people to the right of Trump could get the GOP nomination or to the left of Sanders could get the Democrat one in future.
Pushing people like Styx out of polite society is pushing them to the support the far right, and yet the media both old and new seems intent on this. So is excusing violence from AntiFa and the media seem intent on that too.
Something very nasty is happening to America, and you're not helping to stop it.
In order to satisfy those people with a single IGP, AMD has to build a GPU that's competitive in both domains. That's a challenge, to say the least.
I remember back in the Windows 2000 days reading that you could get by with a linear framebuffer. Adding acceleration had diminishing returns - much of the benefit came from a hardware cursor, line drawing and BitBlts. This was just to get the GUI running without lagging across the PCI bus. I know some systems which had a secondary monitor displaying data from an embedded system and allowed Windows to layer GUI elements on top of that and they were pretty much pure 8 bit greyscale frame buffers.
Of course most GPUs even then did a lot more than that, but the argument was that if you wanted to really strip things down you could get by with that. Most GPUs were labelled 'VGA compatible'. What that meant was that if you had a full screen DOS box Windows would release control of the graphics chip to NTVDM which would allow a Dos application to do register access. So theoretically old Dos games could get decent performance. I say theoretically because in practice they'd suck really badly - the CPU would be too fast, there was no emulation of a sound card, NT's NTDOS.SYS wasn't 100% compatible with MSDOS.SYS, NTVDM might not support your Dos extender and so on.
Now I bet most graphics cards don't support this now - 64 bit Windows doesn't even have NTVDM or any 16 bit support anymore.
Modern GPUs also did quite a bit of 3D acceleration even then - they supported Direct3D for example so WIndows games could render textured polygons to the screen with hardware acceleration. Of course back then the GUI didn't depend on any of that.
Problem is Windows has moved on quite a bit since then - I reckon you'd need to support a fairly wide feature set just to get the Windows GUI to run acceptably with compositing and for videos to play GPU accelerated.
Maybe AMD could do the equivalent of a big.LITTLE design. I.e. one small, low power GPU core to run the desktop and one large high power one which wakes up to do anything more complex. Or even a small low power GPU core and a bunch of modules to accelerate video, 3D etc.
Funny thing is NVidia's Optimus is basically a big.LITTLE equivalent. An laptop with NVIdia uses the Intel IGP for non demanding stuff and then switches over to the discrete NVidia graphics chip for demanding stuff. It still seems like Optimus isn't perfect though - laptops with it have worse battery life than a pure Intel IGP system even when not playing games.
Theoretically AMD should be able to beat Intel and NVidia at this. Intel and NVidia are not on good terms and have both sued each other whereas AMD bought ATI and thus controls the GPU design team - i.e. rather than having to cooperate between two companies to switch between two complete GPUs with separate, binary drivers, it should be possible for AMD to go for the modular approach if they control the CPU, GPU and all the driver code.
In fact that possibility is presumably why they bought ATI in the first place.
Still looking at this review they're not there yet.
If you look at his latest video he said 'PopeRatzo's brilliant satire of me on Slashdot, various liberal elitists calling me deplorable and YouTube pulling my channel have caused me to rethink. I will now devote my life to promoting Social Justice, making videos defending Bill Clinton, Robert Menendez, Harvey Weinstein and Al Franken from unproven allegations. I've decided to become vegan come out and live as a gay man, like Kevin Spacey did. And like Kevin Spacey, I hope that will shield me from any and all accusations of wrongdoing in the past'
The CPU performance of a Ryzen 5 2500U is better than a i7-7600U but worse than a i7-8550U or an i5-8250U
The GPU performance of an Ryzen 5 2500U/Vega 8 is worse than a i5-8250U/Geforce MX150 but it's faster than integrated Intel HD620 in an i7-8550U
The power consumption is clearly worse than a either an Intel IGP or even the GeForce MX150. E.g.
We noted that the Acer Swift 3 with a Core i7-8250U 8th Gen CPU and GeForce MX150 pulled about 9 Watts at idle and 13 - 16 Watts under the light duty load of our HD video loop test. The HP Envy x360 15z with Ryzen 5 Mobile pulled about the same 9 Watts at idle and with similar panel brightness, but under the load of video playback with VLC, pulled 20 Watts with peaks to 30 Watts in spots. We also quickly tested CPU utilization whether running VLC or the Windows 10 video player, and saw Ryzen 5 2500U CPU utilization oscillated at a low 4 - 12 percent. So, it appears at least with respect to VLC and video playback, that Ryzen Mobile with Vega 8 graphics is more power-hungry or perhaps has a bit more driver maturity to undergo to be fully optimized.
Generally PC laptops have two major customer groups
1) People who don't care about GPU performance but do care about battery life, price, power consumption etc
2) People who do care about GPU performance.
People from group 1) are going to get a machine with an Intel CPU and use the integrated GPU.
People from group 2) are going to get a machine with an Intel CPU and a discrete GPU. And not a GeForce MX150 either - more like a Geforce 1050 Ti.
In which case where does the Ryzen 2500U/Vega 8 combo fit in? It doesn't have enough GPU performance for groups 2). It's not low power enough for group 1).
If they'd managed to build something which had more performance than a MX150 they'd be fine. If they could beat Intel IGPs for power they'd be fine. But something with less performance than a MX150 and more power usage too isn't going to do well.
Now maybe some of this could be fixed with a driver update. Still based on current performance we're going to see these machines being sold a deep discount. And if they're not commercially successful, why would AMD spend time optimising drivers?
It's a shame really. AMD Ryzen CPUs on the desktop are actually pretty competitive with Intel. It's a shame the mobile stuff has failed to find the right market niche.
Your body can't synthesize souls of the animals meat eaters consume unless you eat meat. This is why meat eaters age more slowly. It's the reason vampires don't age at all - they consume full sized human souls, not stunted animal ones.
If that $49 deal for an iPhone that AT&T has been running for months didn't entice you to buy one of Apple's smartphones, how does a free iPhone sound to you?
AT&T's 'Free' iPhone Will Cost at Least $1,355.76. The much-maligned carrier is offering new and old iPhone users alike an 8GB 3GS iPhone with a two-year contract for free. Even shipping is free. AT&T's voice contracts range from $39.99 a month with 450 anytime minutes to $69.99 a month with unlimited minutes. Data plans range from $15 a month for 200MB of data to $45 a month for 4GB of data and mobile hotspot support.
If you want to text on your phone, it'll cost 20 cents per message or 30 cents per video or image, or you can get unlimited messaging plan for $20 a month.
Add all the costs together, plus a $36 activation fee, and you'll be paying AT&T $2075.76 over the life of the contract, if you stay under the talk, text and limits in your plan.
I.e a 'free' phone that costs at least $2K over a two year contract, or about $86 a month even if you stay inside your limits.
I think the flaw in AV for a leadership contest is that it is incompatible with humans' monkey brains.
Consider primates. The best primate - and best is probably some mix of size, charisma and popularity - becomes the alpha male and the other primates accept this.
AV is explicitly designed so that this isn't always the case. The second 'best' primate wins and that is the wrong result - as happened in the Ed vs David Miliband case.
Primate leadership contests are inherently analogous to FPTP where number one wins and that rule has been wired into us by evolution for a very long time.
In fact there have been studies that size is an important predictor of presidential success.
Thus it is plausible that FPTP is actually a formalisation of the primate leadership contest based on same criteria - size, charisma and popularity. Successful US politicians certainly seem to behave like the dominant male monkey. It explains Clinton and Trump and all the sexual shenanigans that male US politicians seem to get up to.
Politicians elected by other, inferior, systems are simply not as alpha and in the long run that explains continuing US hegemony despite usually electing people who seem - at least to an outsider - as extremely questionable individuals.
Slashdot Mirror
Gee, I wonder why.
https://en.wikipedia.org/wiki/...
No mod points, but well played.
You can engage a lawyer to sue someone no win no fee and that's fine if you know who is infringing. However how do you find out? If I have a patent on faster superscalar instruction execution for example, I'd need buy a bunch of chips, remove the package with acid, photograph the die, and then work out out if their design is infringing. And then I'd need to go to court and fight them over it. And they'll presumably counter sue. If you're Intel or AMD you probably have people doing competitor research they could do the reverse engineering part. For the average inventor that's not viable.
And for a lot of things it's not just a couple of companies building something, you've got loads. Some you've never heard of in China. And in each case you're going to have to reverse engineer the hardware and software to work out of they're violating your patent.. Some are bound to slip through.
And securitising IP is what makes all this viable. If invent something I talk to a bunch of IP holding companies and see who'd pay most for the IP. I sell it to them, they collect royalties from anyone infringing. If I couldn't sell IP I couldn't transfer it to them for a lump sum and let them handle working out who is infringing and collect royalties. In fact I bet they'd just ignore patents from small companies in that case - they'd know that those small companies don't have the resources to enforce.
Except the original justification for patents is so that the people who own the factories and printing presses have to pay the people who do the inventing or writing.
No copyright means the publisher keeps all the money and the author gets none. No patents mean the factory owner keeps the money and the inventor gets none.
Which used to happen in the US. E.g.
https://www.charlesdickensinfo...
In January 1842 Charles Dickens and his wife, Catherine, traveled to the United States. Dickens wanted to see the sites, learn about the country and do research for a future series of articles.
While on tour Dickens often spoke of the need for an international copyright agreement. The lack of such an agreement enabled his books to be published in the United States without his permission and without any royalties being paid.
This situation also affected American writers like Edgar Allan Poe. Poe's works were published in England without his consent.
Dickens first realized that he was losing income because of the lack of national in international copyright laws in 1837 when The Pickwick Papers was published in book form. At times the novel was reprinted without his permission and sometimes even imitated.
Lobbying by people like Dickens and Poe is what created a system where the US respected the copyright of UK authors and vice versa. Before that if you bought a copy of a book by an author outside your country, the author got nothing. And Dickens was apparently driven crazy by people launching copies of his books with the endings changed!
Of course these days if you got rid of copyright and patents you'd find that the factories and printing presses would all be in China or some other low wage/minimal worker's rights jurisdiction.
IP now means even when an iPad is made in China, most of the value stays in the US
http://www.economist.com/node/...
https://archive.fo/UlRbF
The chart shows a geographical breakdown of the retail price of an iPad. The main rewards go to American shareholders and workers. Apple's profit amounts to about 30% of the sales price. Product design, software development and marketing are based in America. Add in the profits and wages of American suppliers, and distribution and retail costs, and America retains about half the total value of an iPad sold there. The next biggest gainers are South Korean firms like Samsung and LG, which provide the display and memory chips, whose profits account for 7% of an iPad's value. The main financial benefit to China is wages paid to workers for assembling the product and for manufacturing some inputs-equivalent to only 2% of the retail price.
Get rid of IP and you end up with a dystopia where no one invents anything, you've got no idea if the device you buy is genuine and the only people making money are the people who own the factories in China making the hardware. All the businesses in the US, UK and EU which depend on license fees disappear, and so do all the jobs.
And if you do invent something either you keep it a trade secret, or you hand it over to someone to manufacture. However if you do the later they don't need to pay you a penny! In fact allowing people who invent things to work with people who manufacture them without losing all rights is the original justification for copyright and patents. Trademarks are there so people can tell if they have a genuine device.
Also the GPL depends on copyright. If copyright didn't work, companies wouldn't need to release their changes back to the users. So all open source would disappear and each company would maintain an incompatible fork, keeping their changes from competitors.
It's complete hypocrisy. They want to pay cheaper bills by forcing the government to regulate the telcos so the telcos treat every packet as equal.
Meanwhile Google and Facebook ban, censor or demonetize people whose opinions they dislike. Google turned a blind eye to abuse of their Youtube by paedophiles that eventually resulted in advertisers boycotting it.
And Google openly mess with the ranking on Youtube and Google search so people they agree with rise to the top and people they disagree with are hidden from search.
And Facebook launched a non neutral net in India
https://bits.blogs.nytimes.com...
I.e. they only support net neutrality for telcos because it will save them a few pennies on their broadband bills. Their platforms are not neutral politically in the US and outside the US they're happy to offer walled gardens which are not neutral either.
If they actually had any principles they'd want both them and the telcos to be common carriers. And not launch non neutral services outside the US.
Well I want the script to run on macOS, Windows Cygwin and Linux. I dunno if I trust either /dev/random or /dev/urandom on all of those. And openssl and bash are installed on all of them. I trust the openssl guys more than I trust the OS vendors not to have some backdoor or a cryptographically crippled /dev/[u]random.
He owns the Cocteau cinema which has a replica of Robby. Maybe he stumped up the cash for the original.
https://www.tor.com/2013/08/13...
https://www.ranker.com/list/ge...
Consider.
You invent something and patent it. You want to get cash to finance your business and don't want to go after people who infringe it. So you sell it to Nathan Myhrvold's Intellectual Ventures. Now Intellectual Ventures pay you for your patent and do the licence fee collecting themselves.
If Intellectual Ventures didn't exist then you'd be screwed - you'd have spent money on a patent and you need money to run your business but you don't have money for lawyers to collect royalty fees.
Now you can make an argument that the patent system is flawed but this transaction is a valid use of it.
It's like how ambulance chasing lawyers are widely reviled as parasites but if a fear of getting sued makes a business take extra precautions to avoid killing or injuring people, that makes me think the ambulance chasers are doing a societally beneficial thing. Essentially having middlemen to make the information passing more efficient can improve society. And in the US the information passing is done by lawsuit.
I remember reading about a portable computer design in the very early 80's. It was designed in the UK and run off AA batteries. Because running it of non rechargeables was prohibitively expensive they decided to run it off Nickel Cadmium rechargeable cells. And they had a very simple circuit to trickle charge them when it was plugged in. So when you unplugged it and ran off batteries those batteries were guaranteed to be fresh. Problem is, non rechargeable batteries get hot if you try to trickle charge them. So in the UK they put a sticker on the back to say 'rechargeable batteries only'. When it came time to launch in the US they were advised this was a bad idea - if someone ignored the sticker, put in non rechargeables and burned themselves they'd sue and maybe win. However they went back to the engineering team who pointed out that there was an easy fix - stick a thermistor in the battery compartment and shut off the trickle charge if it got warm. This seemed to me to something rather profound. In the UK if you're dumb you get injured and that's just tough. In the US you have a right not to be injured even if you are dumb. In a sense the system in the US had the intelligence, not the individuals. And all the information passing was done by lawsuit, or more accurately people making subtle improvements to minimise their chances of being on the wrong end of a lawsuit. This seems to me to be a more scalable system.
Of course patents can produce a comfortable cartel of insider companies who've sued each other and cross licensed their patents but who can use those patents to discourage new entrants to the industry. And it's clear from Intel's patents on SSE that x86/x64 won't be patent free for a very long time. AMD has a patent licence but is not profitable. Intel might end up with a monopoly and technology might stagnate.
And you can make a case for Tort Reform in the US to reduce the worst case damages in a personal injury lawsuit.
https://en.wikipedia.org/wiki/...
But I think getting rid of patent trolls and ambulance chasing lawyers completely is a bad idea. You just need to fine tune the regulation to maximize their incentives to do socially valuable things and minimize their incentives to do socially harmful ones. Of course in the US's hyper partisan, clickbait haunted world everything is boiled down to "Those evil $(OTHER_PARTY) are up to Pure Evil again. Share this and vote for us to stop them!"
openssl rand n does use /dev/urandom on a Unix like OS, but it doesn't just directly read n bytes from it, it does a bit of munging, like reading a few bytes at startup and using that as a seed for its own PRNG.
https://security.stackexchange...
And of course openssl works on OSs that don't have a /dev/urandom at all - e.g. on Win32 it calls Win32 crypto function to get a seed and then uses that for its own PRNG.
Basically the openssl guys don't seem to trust /dev/urandom to be random. E.g. the stack exchange source says
You do not want to trust that the random source is random. /dev/urandom in particular is not trustworthy because in guaranteeing that it wiill not block on low-entropy conditions, it fails to guarantee that the output is actually random. The manpage man 4 urandom has more information on this, including a cryptic allusion to an attack some government body may or may not have predicated on this condition. Suitability tests can be conducted on this data, or it can be mutated in some way to concentrate entropy, or different sources can be combined.
Which I think is a reference to this
https://linux.die.net/man/4/ur...
A read from the /dev/urandom device will not block waiting for more entropy. As a result, if there is not sufficient entropy in the entropy pool, the returned values are theoretically vulnerable to a cryptographic attack on the algorithms used by the driver. Knowledge of how to do this is not available in the current unclassified literature, but it is theoretically possible that such an attack may exist. If this is a concern in your application, use /dev/random instead.
And essentially running the output of openssl rand through sha512 a couple of times does a bit more munging. And running it through sha512 a random number of times and then encrypting with a per device RSA private key does even more munging.
As with most things in life you have to decide what level of munging you feel comfortable with and code to that level. Personally I use the maximum level, though of course it's debatable whether this is overkill or not.
It may well be that just reading from /dev/urandom on your OS is good enough.
Thiel should sell Nick Denton into slavery with David Miscavige.
You can't ignore a court order because you think it's unconstitutional. And this was a sex tape, not the Pentagon Papers
Also Gawker's sister site Jezebel angrily denounced people for refusing to remove Lawrenceâ(TM)s naked photos.
http://libertyviral.com/this-o...
Anyhow the judge's order was later declared unconstitutional. That didn't stop Hogan from sueing them and bankrupting them. And now Gawker is Thiel's bitch. They pissed him off by outing him as hay while he was on a business trip to Saudi Arabia.
Or falls. Buy on the dip!
http://gawker.com/a-judge-told...
https://en.wikipedia.org/wiki/...
It doesn't matter if your password is stolen if you only use it for one site.
I ended up doing something like this. Put it in ~/.bash_profile
I.e. get 4096 bytes of random data from openssl. Hash it to sha512 a couple of times to shuffle the bits. Then encode to base64. Remove any characters that might cause problems leaving [A-Za-z0-9] which most sites allow. Then pick the first n chars (where n defaults to 32 and is passed as an argument)
E.g.
To get the default 32 char password or you can override that on the command line if you know the maximum password the site supports. E.g
Gives a 64 char password
I store 'em in a text file in a small TrueCrypt volume as a backup and sync that across devices. For things like imgur.com I let the browser remember them because there's nothing too critical about the account. The passwords has 62^32 combinations which is hard to brute force and even if someone did brute force it, who cares? All they'll learn is that my password to other sites needs another 62^32/2 tries to brute force which is probably not worth their time.
And if you know the site allows longer passwords, just tell randpass on the command line. E.g
Actually once I got this to work I invented a more elaborate version that encrypted with a per device private RSA key, randomized the number of sha512 hashing stages and so on. But that only matters if you think an attacker can work out what openssl rand returned on your device, which they probably can't.
I've got a degree, but I got in the UK before tuition fees and in a subject where it's kind of necessary, i.e. engineering. Still it's harder to justify asking for a degree in other subjects, like the arts. For a lot of programming jobs like web development it's also arguable that you don't need a degree to be able to do it. Even writing embedded code like I do isn't something that actually needs a degree - I've met a lot of people who do it well and don't have one, and a lot of people who do it badly who do.
On the other hand if you have more applicants than jobs why not pick the best qualified ones? And of course if employers demand a degree, more people will do one.
Now since I got mine, more people are getting degrees, prices have gone up and they're doing them in subjects where it is less necessary.
I think it's a classic case of inflation. If employers can eliminate candidates on the basis of graduate/non graduate, of course they'll do it. And if they do it, more people decide they need to get a degree. In the UK the percentage of graduates has increased enormously, tuition fees have risen because the government can no longer fund all those graduates to study without them paying and so now doing a degree means taking on a lot of debt. In the US of course it's probably a fair bit worse.
Of course it's hard to get out of this trap of people only needing a degree because employers demand one, and employers demanding them because they know they can find someone with one. Meanwhile most people are doing degrees in subjects which aren't really helping them do their jobs. Because if you're force to do a degree you don't actually need why not do it in a subject you like.
One option would be to allow people to default on their loans. But that would burst a bubble way worse than subprime mortgages
It was $1 trillion in 2014 and has a high delinquency rate
https://www.forbes.com/sites/h...
The total outstanding student loan balance is $1.08 trillion, and a whopping 11.5% of it is 90+ days delinquent or in default. That's the highest delinquency rate among all forms of debt and the only one that's been on the rise consistently since 2003.
It's easy to see why. If you borrow $100K or more to get a degree and end up in a bad job, you're going to end defaulting.
Maybe if the government only offered degree loans in subjects which have some justification economically they could deflate the bubble because it would forces universities to drop the price of degrees in everything else. Of course telling academics their subject is no longer eligible for loans will cause a massive shitstorm and accusations of philistinism from self interested academics.
Another option would be for the government to get out of the student loans business. Bursting such a large bubble of bad debt will have dire economic consequences however it is done though.
OEMs have a long history of treating AMD like the redheaded stepchild.
Back when I used to care about gaming and GPU performance I got mostly Intel/NVidia systems. When Athlons came out and they were competitive and ATI had a similarly decent generation I decided on a AMD/ATI build. Got all the parts mail order in the UK. Put it together and it was unstable. I swapped some parts - luckily the vendor I used was very understanding. Eventually it turned out the SDRAM I was using couldn't support the FSB it claimed. I ordered an Athlon with a slightly faster FSB (333 MT/s) than the norm and DDR1 SDRAM PC2700 DDR-333 to match. The board autoconfigured to use the 333 Mhz speed as expected, and it turned out that was optimistic. If I manually underclocked to the slower speed (IIRC 266Mhz) it became stable. ATI's drivers were more bloated than NVidia's - the control panel app used .Net for example, while NVidia's was straight Win32.
So I went back to Intel and NVidia. It seemed like Intel's and NVidia's Bios and drivers were more mature. E.g. I've never had to mess with an Intel board's memory speed configuration.
I remember Michael Dell ranting about how AMD CPUs were a pain to install because they didn't have a heat spreader back in the Athlon XP days - so you could crack the die installing the heatsink. I don't remember this being an issue with my Athlon, but then I wasn't installing them in volume. And hey, who can object to a CEO knowing about stuff like this - he'd clearly experienced the problem first hand.
So I can see why OEMs are a bit biased. Plus of course Intel probably do everything they can legally to keep OEMs from using anything but Intel CPUs.
Of course even if I buy Intel and NVidia it's still pretty important that AMD CPUs and GPUs are competitive. You can see that with Intel's recent generation of CPUs. For ages Intel were making very minor performance gains with each generation because that was enough. AMD became competitive with Ryzen and suddenly Intel produce a generation of chips with a big performance gain.
So I'd like to AMD to be competitive. Right now it seems like AMD is competitive for desktop machines, not so much on mobile. Which is a shame.
It could well be AMD's problem is primarily a software one actually. It certainly used to be.
Your original post parodying him was actually pretty funny.
This is less good
It was just another right-wing YouTube freak pretending to be persecuted by the SJWs, like the guy who sprayed himself in the face with Axe body spray, pretending it was Antifa attacking him with bear spray and then laid down on the ground pouring milk on his face and crying like a wee bitch.
So because some dude lied about being maced, that means that AntiFa violence is not a problem? I've seen loads of videos of AntiFa macing people, or hitting them over the head with a bike lock. Not all of them are fake.
And look at Styx. He started off, as far as I know, an apolitical gamer. GamerGate politicised him, but GamerGate was just a bunch of people who play video games getting in a fight with tech journalists. I don't have time to play video games these days, but I've met a few tech journalists IRL and I've never had much time for their opinions. So GamerGate's revelation that these people are a cabal of liars wasn't much of a surprise. Nor of course does it make much difference to society. As someone once observed on the relative importance of music and video games 'A society with no video games is easy to imagine wouldn't be all that different from the one we have. A society without music would be horrific'. Video games are just not that important.
But then something strange happened. Trump's popularity wasn't despite media criticism of him, it was because of it. Suddenly something very like GamerGate - essentially a rebellion against the media - happened and it had real consequences. People voted for an outsider whose candidacy was seen as a joke by smug media insiders because they hated those smug media insiders more than they hated Trump.
That's the reason Trump supporters don't care about the Roy Moore allegations. They know similar allegations exist against Democrats but the media simply decided allegations against Republicans are credible and ones against Democrats are not, because the media is ruthlessly partisan.
And it seems the US left side hasn't really learned from this. You're just doubling down on the criticism of both Trump and his supporters. Youtube and Facebook are censoring them, and people who claim to believe in Net Neutrality excuse this as 'a private company censoring people doesn't violate the First Amendment'. Which is true, but irrelevant.
My point is that this is more likely to make them double down on their opinions than to change them. And there are worse people out there than Trump waiting in the wings to take advantage of a disillusion with politicians and the media. There's AntiFa commies on the left and Alt Right fascists on the right. It's fair to say these people are not really joking when they meme about using force against their political opponents.
Back in the old days the US's political system essentially made sure extremists had no power. Someone on the extreme right wouldn't vote for the Democrats - they'd have to vote Republican or abstain. So the Republicans could nominate someone centrist and then tell them that the alternative, a Democrat was even worse. And on the left the Democrats could do the same - nominate a centrist insider safe in the knowledge that their extreme left couldn't vote for anyone but them. It was all pretty corrupt, but it had the advantage of keeping the really dangerous people marginalized.
Look at the last election though - Trump as an outsider got the GOP nomination. Sanders would have got the Democrat one in a fair contest and probably would get it if he run again. It's easy to see that people to the right of Trump could get the GOP nomination or to the left of Sanders could get the Democrat one in future.
Pushing people like Styx out of polite society is pushing them to the support the far right, and yet the media both old and new seems intent on this. So is excusing violence from AntiFa and the media seem intent on that too.
Something very nasty is happening to America, and you're not helping to stop it.
In order to satisfy those people with a single IGP, AMD has to build a GPU that's competitive in both domains. That's a challenge, to say the least.
I remember back in the Windows 2000 days reading that you could get by with a linear framebuffer. Adding acceleration had diminishing returns - much of the benefit came from a hardware cursor, line drawing and BitBlts. This was just to get the GUI running without lagging across the PCI bus. I know some systems which had a secondary monitor displaying data from an embedded system and allowed Windows to layer GUI elements on top of that and they were pretty much pure 8 bit greyscale frame buffers.
Of course most GPUs even then did a lot more than that, but the argument was that if you wanted to really strip things down you could get by with that. Most GPUs were labelled 'VGA compatible'. What that meant was that if you had a full screen DOS box Windows would release control of the graphics chip to NTVDM which would allow a Dos application to do register access. So theoretically old Dos games could get decent performance. I say theoretically because in practice they'd suck really badly - the CPU would be too fast, there was no emulation of a sound card, NT's NTDOS.SYS wasn't 100% compatible with MSDOS.SYS, NTVDM might not support your Dos extender and so on.
Now I bet most graphics cards don't support this now - 64 bit Windows doesn't even have NTVDM or any 16 bit support anymore.
Modern GPUs also did quite a bit of 3D acceleration even then - they supported Direct3D for example so WIndows games could render textured polygons to the screen with hardware acceleration. Of course back then the GUI didn't depend on any of that.
Problem is Windows has moved on quite a bit since then - I reckon you'd need to support a fairly wide feature set just to get the Windows GUI to run acceptably with compositing and for videos to play GPU accelerated.
Maybe AMD could do the equivalent of a big.LITTLE design. I.e. one small, low power GPU core to run the desktop and one large high power one which wakes up to do anything more complex. Or even a small low power GPU core and a bunch of modules to accelerate video, 3D etc.
Funny thing is NVidia's Optimus is basically a big.LITTLE equivalent. An laptop with NVIdia uses the Intel IGP for non demanding stuff and then switches over to the discrete NVidia graphics chip for demanding stuff. It still seems like Optimus isn't perfect though - laptops with it have worse battery life than a pure Intel IGP system even when not playing games.
AMD have an equivalent, though I've never used it
http://www.dell.com/support/ar...
Theoretically AMD should be able to beat Intel and NVidia at this. Intel and NVidia are not on good terms and have both sued each other whereas AMD bought ATI and thus controls the GPU design team - i.e. rather than having to cooperate between two companies to switch between two complete GPUs with separate, binary drivers, it should be possible for AMD to go for the modular approach if they control the CPU, GPU and all the driver code.
In fact that possibility is presumably why they bought ATI in the first place.
Still looking at this review they're not there yet.
I'd definitely like to see more reviews, but I can't really see what market segment this is aimed at.
If they can get the power consumption down below Intel IGP but deliver better performance at a lower price they might be onto something though.
Don't worry he got his channel pulled
https://slashdot.org/comments....
If you look at his latest video he said 'PopeRatzo's brilliant satire of me on Slashdot, various liberal elitists calling me deplorable and YouTube pulling my channel have caused me to rethink. I will now devote my life to promoting Social Justice, making videos defending Bill Clinton, Robert Menendez, Harvey Weinstein and Al Franken from unproven allegations. I've decided to become vegan come out and live as a gay man, like Kevin Spacey did. And like Kevin Spacey, I hope that will shield me from any and all accusations of wrongdoing in the past'
Well done!
The CPU performance of a Ryzen 5 2500U is better than a i7-7600U but worse than a i7-8550U or an i5-8250U
The GPU performance of an Ryzen 5 2500U/Vega 8 is worse than a i5-8250U/Geforce MX150 but it's faster than integrated Intel HD620 in an i7-8550U
The power consumption is clearly worse than a either an Intel IGP or even the GeForce MX150. E.g.
We noted that the Acer Swift 3 with a Core i7-8250U 8th Gen CPU and GeForce MX150 pulled about 9 Watts at idle and 13 - 16 Watts under the light duty load of our HD video loop test. The HP Envy x360 15z with Ryzen 5 Mobile pulled about the same 9 Watts at idle and with similar panel brightness, but under the load of video playback with VLC, pulled 20 Watts with peaks to 30 Watts in spots. We also quickly tested CPU utilization whether running VLC or the Windows 10 video player, and saw Ryzen 5 2500U CPU utilization oscillated at a low 4 - 12 percent. So, it appears at least with respect to VLC and video playback, that Ryzen Mobile with Vega 8 graphics is more power-hungry or perhaps has a bit more driver maturity to undergo to be fully optimized.
Generally PC laptops have two major customer groups
1) People who don't care about GPU performance but do care about battery life, price, power consumption etc
2) People who do care about GPU performance.
People from group 1) are going to get a machine with an Intel CPU and use the integrated GPU.
People from group 2) are going to get a machine with an Intel CPU and a discrete GPU. And not a GeForce MX150 either - more like a Geforce 1050 Ti.
In which case where does the Ryzen 2500U/Vega 8 combo fit in? It doesn't have enough GPU performance for groups 2). It's not low power enough for group 1).
If they'd managed to build something which had more performance than a MX150 they'd be fine. If they could beat Intel IGPs for power they'd be fine. But something with less performance than a MX150 and more power usage too isn't going to do well.
Now maybe some of this could be fixed with a driver update. Still based on current performance we're going to see these machines being sold a deep discount. And if they're not commercially successful, why would AMD spend time optimising drivers?
It's a shame really. AMD Ryzen CPUs on the desktop are actually pretty competitive with Intel. It's a shame the mobile stuff has failed to find the right market niche.
If you're Chief Mutant and you're not driving around in some sort of pimped out vehicle then you're doing it wrong frankly.
Your body can't synthesize souls of the animals meat eaters consume unless you eat meat. This is why meat eaters age more slowly. It's the reason vampires don't age at all - they consume full sized human souls, not stunted animal ones.
It's basic science.
Last time I was in California AT&T did. Something like this
https://www.pcworld.com/articl...
If that $49 deal for an iPhone that AT&T has been running for months didn't entice you to buy one of Apple's smartphones, how does a free iPhone sound to you?
AT&T's 'Free' iPhone Will Cost at Least $1,355.76. The much-maligned carrier is offering new and old iPhone users alike an 8GB 3GS iPhone with a two-year contract for free. Even shipping is free.
AT&T's voice contracts range from $39.99 a month with 450 anytime minutes to $69.99 a month with unlimited minutes. Data plans range from $15 a month for 200MB of data to $45 a month for 4GB of data and mobile hotspot support.
If you want to text on your phone, it'll cost 20 cents per message or 30 cents per video or image, or you can get unlimited messaging plan for $20 a month.
Add all the costs together, plus a $36 activation fee, and you'll be paying AT&T $2075.76 over the life of the contract, if you stay under the talk, text and limits in your plan.
I.e a 'free' phone that costs at least $2K over a two year contract, or about $86 a month even if you stay inside your limits.
More recently
https://www.cnet.com/news/veri...
Free iPhone 6 in 2014 with a $50 contract. I'm guessing the TCO will end up being more.
Of course iPhones have got quite a bit more expensive since then.
I think the flaw in AV for a leadership contest is that it is incompatible with humans' monkey brains.
Consider primates. The best primate - and best is probably some mix of size, charisma and popularity - becomes the alpha male and the other primates accept this.
AV is explicitly designed so that this isn't always the case. The second 'best' primate wins and that is the wrong result - as happened in the Ed vs David Miliband case.
Primate leadership contests are inherently analogous to FPTP where number one wins and that rule has been wired into us by evolution for a very long time.
In fact there have been studies that size is an important predictor of presidential success.
https://archive.fo/gAVPV
Thus it is plausible that FPTP is actually a formalisation of the primate leadership contest based on same criteria - size, charisma and popularity. Successful US politicians certainly seem to behave like the dominant male monkey. It explains Clinton and Trump and all the sexual shenanigans that male US politicians seem to get up to.
Politicians elected by other, inferior, systems are simply not as alpha and in the long run that explains continuing US hegemony despite usually electing people who seem - at least to an outsider - as extremely questionable individuals.