AGI Milestone: Maximize Programmer Resources part 2: about SPEL

You might think they are all different at a superficial glance. However there are linguistic universals, which are shared by large quantities of languages, and larger still are the similarities when you take away word-order and whether something is an adposition or an affix.

So it is quite easy to use the most common and regular grammar, which was most likely spoken by mitochondrial Eve near the great lakes of Africa. Post-positional SOV is the kind of grammar we are so innately attuned to, that deaf people, never having heard spoken language develop it all on their own (https://en.wikipedia.org/wiki/Al-Sayyid_Bedouin_Sign_Language ). Even people that speak an SVO language natively do better with a new SOV language vs an SVO one http://langcog.stanford.edu/papers/TFJ-cogsci2011.pdf

Anyways, I don’t expect many people to learn a new language, that why the ideal grammar is mostly reserved for the core-language or interlingua which is also the bytecode. Instead there are formal variants of human languages which share vocabulary and word order of the native language — even though they are more regular and have a more limited dictionary.

Advertisements

AGI Milestone: Maximize Programmer Resources

Aachen, Technische Hochschule, Rechenzentrum
For a full fledged AGI, we aren’t talking about some small project, that can be done by a small team of geniuses, like the Manhattan project.

We are talking about fully automating all of humanity, since a full AGI would be able to do the same tasks as any human. Some people have calculated the development costs (in San Francisco programmer wages) as reaching Trillions of dollars. So one of the main necessities is lowering cost of programmers. Currently only 0.3% of humans are computer programmers.

Only about a quarter of humans speak English, so about 75% of the possible human brains we can harness are untapped by primarily English programming languages (Python, C etc). Also of all humanity only a small percentage are visual-spatial and math-savvy enough to do computer programming with visual-symbolic languages (context-free (math grammar) languages). Whereas the literacy rate of the planet is over 80%, so human grammar languages have a wide adoption rate.

Of course in due time GPU’s will be competitive with human brain power, so they could also be harnessed for programming. As Google and several others have recently demonstrated (with the 800 GPU’s they used being roughly human equivalent).

So to maximize the amount of programmers that can work on AGI, it is best to have a formal computer programming language based on human grammar, that is also compatible with machine programmers.

I’m dreaming of a Cold Christmas.

Christmas, doesn’t feel the way it used to. Four hundred years ago, it was cold on Christmas. I’d snuggle up with my family under our blankets, and we’d hovel near a small fire, as we hardly had enough money to keep it going. Yes, it was cold, and wet in Britain, but we had each other. Some hot beverages, and our sooty selves.

Now with forced air heating, meh,
it’s just so hot, like all year long!

In Toronto, tenant rooms are legally required to be at least 21C throughout the year.

I don’t know what’s happened to humanity, but I guess people like being comfortable. It should always be the temperature it was 130 thousand years ago in the great rift valley of Africa, right?

Sorry, just feel kinda spoiled with warmth.
I’ll have to go do more swimming in a frozen lake,
or rolling around in snow, to help myself feel better.

You have till Dec 15 to tell Canada that lowering national cyber security is a BAD IDEA

cybersecurity_canada


Canada wants to lower encryption standards, put in backdoors, force
keeping of arbitrary user data, so it would be easier for foreign
governments and exploiters to take advantage of Canadians, steal their
personal information, emails which may include financial records, and
other private and personal information enough to steal identities.

We have until December 15 to tell them that this is a BAD IDEA!
(link to Canada's Online Consultation at the bottom)

Motivation:

They think it is a matter of National Security that they be given easy
access to the above information, but they forget that it is a National
Cyber Security risk, for that information to be easily available.

My suggestion was that Canada needs a Cyber Security task force, with
high end GPU or ASIC hardware for password cracking, this will give
them all the access they need to whatever they want. Human minds
aren't very good at remembering passwords, so they might just forget
when interrogated.

On the other hand 80+% of people have weak passwords,, so should be
easy to crack.
https://blog.dashlane.com/study-reveals-extremes-people-go-online-protec
tion/

Also there should be a "building code" for cyber security, minimum
standards before any sensitive user data is allowed to be kept.

Because there are over 70 thousand hacked servers available cheap
online, some of them government and otherwise
https://securelist.com/blog/research/75027/xdedic-the-shady-world-of-hac
ked-servers-for-sale/

Basic Subscriber Info ( name, address, telephone number and email
address) can be enough to get into many accounts, and with a few other
pieces (which may be stored alongside) would be sufficient to steal an
identity.

For example once a criminal has that basic information,
It only takes 6 seconds for a criminal to get accurate credit card
information and security code of a target:
http://www.telegraph.co.uk/news/2016/12/02/hacked-just-six-seconds-crimi
nals-need-moments-guess-card-number/

Encryption is essential for securing everyday interactions on the
internet. For instance Hillary Clinton didn't encrypt her email
exchanges, and largely because of it lost the election and almost went
to prison.

Internet exploiters are getting increasingly more sophisticated and
even took down the East Coast recently,
https://www.wired.com/2016/10/internet-outage-ddos-dns-dyn/ So having
good Cyber Security is essential, weakening it to make it easier to
spy on people is a VERY BAD IDEA.

Canada's Online Consultation:

https://www.publicsafety.gc.ca/cnt/cnslttns/ntnl-scrt/thm09-en.aspx

I'm also planning on scheduling a meeting with my MP about this, I
recommend you at least write to them.

Working on Draft for “Green Buddhism: A Programmer’s Religion”

If you’re wondering why I haven’t been posting in a while,
it is because I’m channeling my posting energy into writing a book.

If you would like to follow along it’s progress.
Can do so via ZeroNet or Github.

ZeroNet

  1. download ZeroNet and start it up
  2. then click on Green Buddhism.html or GreenBuddhism.pdf

GitHub

  1. download the package from elspru/green_buddhism
  2. Install texlive, latex and all dependencies
  3. compile from source using ./compile.sh
  4. then can read either the green_buddhism.pdf or green_buddhism.html file

Idea: Robot Representative Government (Party)

Representatives that work for constituents, regardless of actually being elected. I say (party) in brackets, as there isn’t a “platform” other than representing the constituents current opinions as best as possible.

For example, if there are at least 25 people in support of something, then can petition the House of Commons in Canada. Thus it should be acceptable to petition any lower body, such as a an MP, MPP, county, mayor, or district representative.

The idea is that the robot representative would lobby the current government representatives, on behalf of it’s constituents. It can do so via email, phones, letters, and physical presence at law drafting. This would in fact be a service to those representatives, as it would notify them about their constituents’ live opinion on whatever topic it is.

Historically small parties get rather few votes, such as around 100, this is about as many as one candidate could realistically form a personal connection with — all their friends.   In order to get into the thousands range, you need friends of friends to be involved in campaigning, and-or some kind of advertising.

Though for the robot representative party, even a hundred or so people would be enough.  Don’t even need to have an official election to start working. The Secure Electronic Signature Regulation [1], says that a cryptographic hash of the document is an acceptable electronic signature.

So for instance a GNUPG signature should suffice, for instance if the constituent sent an email reply with the bill contents, the PGP hash from it could be used as their signature, and could be verified.  The hardest part would be making this easy for the constituents to use.

\section{App}

Can have a mobile and desktop app, which people use to view, deliberate and vote on various petitions.

\section{liquid democracy}

As it stands, liquid democracy isn’t supported by the current definition of a petition, however if we have at least 25 regular signers, can likely include how many additional votes they may represent via the liquid democracy approach.

Liquid democracy is when someone delegates their vote to someone else, who generally shares their opinions, or whom they trust to make a wise choice. This may be someone in their social group who is interested in politics, and has the time to read through the various bill proposals, and sort out whether to vote for or against, perhaps even coming up with potential amendments that would make a for vote viable.

This is of course retractable, if for instance you want to vote on something yourself, or you feel someone else is better suited to represent your interests.  It may even be encouraged to select a liquid democratic representative, so your vote doesn’t go to waste, by sending a reminder every parliamentary session (approximately every year or two).

For me, I might select some keywords for the kind of legislation I want to personally suggest amendments on, and vote on. While otherwise delegating to someone that holds views of the Pirate and Green Party but is pro-Israel.

henceforward reference to voters will include liquid-democratic representatives, who may count as multiple voters due to representing other people.

\section{Bill information}

The next major issue, is getting the actual bills which are up for proposal.

Openparliament.ca seems to be a good source for federal bills, it seems to have a fairly simple layout, so it should be viable to harvest the data from it. ontla.on.ca while having slightly more complex interface is also navigable and has accessible bills.

\section{Live Presence}

If a certain riding has enough campaign contributions, it can afford to have a physical robot, which can attend committee meetings open to the public, and both record them as well as voice the opinions of the members who have logged in for the committee meeting.

For instance if a quorum (25) of voters vote to go to the meeting, then the robot will go to the meeting.  At the meeting if a majority of the logged in voters wish to say something, then the robot will say it at the next given opportunity.  For instance there could be some kind of LED that lights up, or a hand that is raised, which the facilitator can acknowledge when there is time for the robot to speak.

If using state of the art technologies of today, could have a 360 degree recording, so people could don their virtual reality glasses and be virtually present at the meeting. When looking down where the robots body would be, a person would see the console or chat interface of voters, so can work on amendments and voting for what the robot will say next.

For accessibility there can be a chat only interface, and the facial and body expressions of the various people in the room could be added to the chat via feature recognition software.
When the robot speaks, it could be of the form “representing <x number of voters>, <statement>”. That way the people at the committee meeting would know how much weight the statements carry.

In the transitional period, where there isn’t a robot that can get itself to a committee meeting, a person could substitute,  though they would have to be paid enough to cover their time, and ideally would have a live stream video connection on their laptop.

Then the person could raise their hand, and say what the majority of those logged in wish to have said.
Ideally this would also be someone who has many liquid democracy votes, as if it is above quorum (25) and the majority at the meeting then they could speak on behalf of all the voters present.

\section{Conclusion}

In the same fashion, it is actually possible for a robot to do the work of a political representative.  Or at least the work that is accessible to any member of the public, though which most members of the public don’t pursue as they have busy lives — like attending committee meeting.

If I could virtually attend the local municipal, provincial, and-or federal committee meetings virtually I would. At least those of them which are important to me — even though some of them are hundreds or thousands of kilometers away from me.  Of course I would also be happy if there was some app or email service, which allowed me to vote yes or no to on real bills being passed right now, and then have my opinion matter enough to be included on a petition to the local representative.

Eventually if it works well enough, these robot representatives could go all the way to the top.  Though I understand MP’s and MPP’s have a variety of tasks which are not policy related, those could also be automated if sufficiently described.

[1] Secure Electronic Signature Regulation http://laws-lois.justice.gc.ca/eng/regulations/sor-2005-30/page-1.html#h-2

Paying with Watt Hours #altcoin #currency

Imagine you could pay with your computer resources. So there might be something you’d like to buy, such as a course.

You happen to have solar panels, or other forms of renewable electricity, or perhaps you simply don’t pay an electricity bill.  Or you are in a place or living situation where it is difficult to acquire fiat currency, but you have a computer.

In this case you would download a special app, it would operate in a sandbox, where it can’t see anything on your computer or internet connection, but can do it’s own thing either compiling or evolving programs for example, until it reaches to wattage limit you set, or gets your account into good standing.

The app or electricity based currency exchange would be pegged to some reasonable average of global electricity prices.  So it may even be lucrative for you to run this app if you happen to be in a country or have electricity costs which are lower than average.  It would also make sense to run this app if you are off-grid and have surplus electricty which can’t be stored,
in which case it can be assigned to and used by the apps.

Of course, for an average 100w computer, you’d only be spending 2.4kW, which if international prices are around 20 cents per kW, then 40.8 cents a day.  However humans don’t have to work 100% of the day to get a living wage. Most working-class humans get roughly a 60-80% bonus (they don’t have to work for 60-80% of the time and still meet their bills).  So it makes sense to give the same bonus to these computers.

Now that would be a $1 per day, or $30 a month, or $365 a year,  enough to keep the computer in good repair.  If it had a powerful GPU card, that performed at an estimated 500Watts, then that is almost $2,000 a year, enough for some serious upgrades, even a whole new computer.
If the online course only cost $30, then could pay it off within a month, simply by keeping your computer on.

Various cloud services could be hosted on these computers, especially those without high performance compatible GPU units.