Joho the Blog » google

June 8, 2014

Will a Google car sacrifice you for the sake of the many? (And Networked Road Neutrality)

Google self-driving cars are presumably programmed to protect their passengers. So, when a traffic situation gets nasty, the car you’re in will take all the defensive actions it can to keep you safe.

But what will robot cars be programmed to do when there’s lots of them on the roads, and they’re networked with one another?

We know what we as individuals would like. My car should take as its Prime Directive: “Prevent my passengers from coming to harm.” But when the cars are networked, their Prime Directive well might be: “Minimize the amount of harm to humans overall.” And such a directive can lead a particular car to sacrifice its humans in order to keep the total carnage down. Asimov’s Three Rules of Robotics don’t provide enough guidance when the robots are in constant and instantaneous contact and have fragile human beings inside of them.

It’s easy to imagine cases. For example, a human unexpectedly darts into a busy street. The self-driving cars around it rapidly communicate and algorithmically devise a plan that saves the pedestrian at the price of causing two cars to engage in a Force 1 fender-bender and three cars to endure Force 2 minor collisions…but only if the car I happen to be in intentionally drives itself into a concrete piling, with a 95% chance of killing me. All other plans result in worse outcomes, where “worse” refers to some scale that weighs monetary damages, human injuries, and human deaths.

Or, a broken run-off pipe creates a dangerous pool of water on the highway during a flash storm. The self-driving cars agree that unless my car accelerates and rams into a concrete piling, all other joint action results in a tractor trailing jack-knifing, causing lots of death and destruction. Not to mention The Angelic Children’s Choir school bus that would be in harm’s way. So, the swarm of robotic cars makes the right decision and intentionally kills me.

In short, the networking of robotic cars will change the basic moral principles that guide their behavior. Non-networked cars are presumably programmed to be morally-blind individualists trying to save their passengers without thinking about others, but networked cars will probably be programmed to support some form of utilitarianism that tries to minimize the collective damage. And that’s probably what we’d want. Isn’t it?

But one of the problems with utilitarianism is that there turns out to be little agreement about what counts as a value and how much it counts. Is saving a pedestrian more important than saving a passenger? Is it always right try to preserve human life, no matter how unlikely it is that the action will succeed and no matter how many other injuries it is likely to result in? Should the car act as if its passenger has seat-belted him/herself in because passengers should do so? Should the cars be more willing to sacrifice the geriatric than the young, on the grounds that the young have more of a lifespan to lose? And won’t someone please think about the kids m— those cute choir kids?

We’re not good at making these decisions, or even at having rational conversations about them. Usually we don’t have to, or so we tell ourselves. For example, many of the rules that apply to us in public spaces, including roads, optimize for fairness: everyone waits at the same stop lights, and you don’t get to speed unless something is relevantly different about your trip: you are chasing a bad guy or are driving someone who urgently needs medical care.

But when we are better able control the circumstances, fairness isn’t always the best rule, especially in times of distress. Unfortunately, we don’t have a lot of consensus around the values that would enable us to make joint decisions. We fall back to fairness, or pretend that we can have it all. Or we leave it to experts, as with the rules that determine who gets organ transplants. It turns out we don’t even agree about whether it’s morally right to risk soldiers’ lives to rescue a captured comrade.

Fortunately, we don’t have to make these hard moral decisions. The people programming our robot cars will do it for us.

 


Imagine a time when the roadways are full of self-driving cars and trucks. There are some good reasons to think that that time is coming, and coming way sooner than we’d imagined.

Imagine that Google remains in the lead, and the bulk of the cars carry their brand. And assume that these cars are in networked communication with one another.

Can we assume that Google will support Networked Road Neutrality, so that all cars are subject to the same rules, and there is no discrimination based on contents, origin, destination, or purpose of the trip?

Or would Google let you pay a premium to take the “fast lane”? (For reasons of network optimization the fast lane probably wouldn’t actually be a designated lane but well might look much more like how frequencies are dynamically assigned in an age of “smart radios.”) We presumably would be ok with letting emergency vehicles go faster than the rest of the swarm, but how about letting the rich go faster by programming the robot cars to give way when a car with its “Move aside!” bit is on?

Let’s say Google supports a strict version of Networked Road Neutrality. But let’s assume that Google won’t be the only player in this field. Suppose Comcast starts to make cars, and programs them to get ahead of the cars that choose to play by the rules. Would Google cars take action to block the Comcast cars from switching lanes to gain a speed advantage — perhaps forming a cordon around them? Would that be legal? Would selling a virtual fast lane on a public roadway be legal in the first place? And who gets to decide? The FCC?

One thing is sure: It’ll be a golden age for lobbyists.

5 Comments »

May 13, 2014

Full-text searching Harvard Library: a hacky mashup

Harvard Library has 13M items in its collection. Harvard is digitizing many of them, but as of now you cannot do a full text search of them.

Google Books had 30M books digitized as of a year ago. You can do full-text searches of them.

So, I wrote a little app [Note: I've corrected this url.] that lets you search Google Books for text, and then matches up the results with books in Harvard Library. It’s a proof of concept, and I’m counting the concept as proved, or at least as promising. On the other hand, my API key for Google Books only allows 2,000 queries a day, so it’s not practical on the licensing front.

This project runs on top of LibraryCloud, an open source library metadata server created by the Harvard Library Innovation Lab that I co-direct (until Sept.). LibraryCloud provides an API to Harvard’s open library metadata and more. (We’re building a new, more scalable version now. It is, well, super-cool.)

But please note that this HOLLIS full-text search thingy is NOT a project done by our highly innovative and highly skilled developers. I did it, which means if you look at the code (github) you will have a good laugh. Also, this service will fail in dull and interesting ways. I am a horrible programmer. (But I enjoy it.)

Some details below the clickable screenshot…


Click on the image to expand it.
googleHollis screen capture

Click here to go to the app.

The Google Books results are on the left (only ten for now), and HOLLIS on the right.

If a Google result is yellow, there’s a match with a book in HOLLIS. Gray means no match. HOLLIS book titles are prefaced by a number that refers to the Google results number. Clicking on the Google results number (in the circle) hides or shows those works in the stack on the right; this is because some Google books match lots of items in HOLLIS. (Harvard has a lot of copies of King Lear, for example.)

There are two types of matches. If an item matched on a firm identifier (ISBN,OCLC, LCCN), then there’s a checkmark before the title in the HOLLIS stack, and there’s a “Stacklife” button in the Google list. Clicking on the Stacklife button displays the book in Harvard StackLife, a very cool — and prize winning! — library browser created by our Lab. The StackLife stack colorizes items based on how much they’re used by the Harvard community. The thickness of the book indicates its page count and its length indicates its actual physical height.

If there’s no match on the identifiers, then the page looks for a keyword match on the title and an exact match on the author’s last name. This can result in multiple results, not all of which may be right. So, on the Google result there’s a “Feeling lucky” button that will take you to the first match’s entry in StackLife.

The “Google” button takes you to that item’s page at Google Books, filtered by your search terms for your full-texting convenience.

The “View” button pops up the Google Books viewer for that book, if it’s available.

The “Clear stack” button deselects all the items in the Google results, hiding all the items in the HOLLIS stack.

Let me know how this breaks or sucks, but don’t expect it ever to be a robust piece of software. Remember its source.

Be the first to comment »

November 7, 2013

What I want those Google barges to be

We now know that the Google barges are “interactive learning spaces.” That narrows the field. They’re not off-shore data centers or Google Glass stores. They’re also not where Google keeps the porn (as Seth Meyers reported) and they’re not filled with bubblewrap for people to step on, although that would be awesome.

So here’s my hope for what “interactive learning spaces” means: In your face, Apple Store!

Apple Stores manifest Apple’s leave-no-fingerprints consumerist ideal. Pure white, squeaky clean, and please do come try out the tools we’ve decided are appropriate for you inferior Earth creatures.

Google from the beginning has manifested itself as comfortable with the messy bustle of the Net, especially when the bustlers are hyper-geeky middle class Americans.

So, I’m hoping that the “interactive learning spaces” are places where you can not only get your email on a Chromebook keyboard, play a game on an Android tablet, and take a class in how to use Google Glass, but is a place where you can actually build stuff, learn from other “customers,” and hang out because the environment itself — not just the scheduled courses — is so stimulating and educational. Have hackathons there, let the community schedule classes and talks, make sure that Google engineers hang out there and maybe even some work there. Open bench everything!

That’s what I hope. I look forward to being disappointed.

1 Comment »

September 2, 2013

Merging my Google Plus accounts

I made the mistake many years ago of creating a Google Accounts email address in addition to my existing Gmail account. Thus I have been plagued (granted, it’s an excellent example of a First World Problem plague) with two out of sync accounts.

Gmail works fine because “self@evident.com” is my public-facing email address and has been since about 1994 when first I took the evident.com domain. (Yes, children, there was a time when you could register an existing word with all its vowels just by being the first to claim it.) When you send mail to that address, it shows up in my evident@evident.com Google Account. It also shows up at my dweinberger@gmail.com account, which now has 12,722 unread messages in it. Nevertheless, the “system” works for me.

But it does not work for me at YouTube.com, where I have two accounts that cannot be merged. I’ve tried.

And I thought it didn’t work at Google Plus. But recently I’ve been getting friend requests (or Circle requests, I guess) at G+ for evident@evident, whereas my social network (such as it is) is at dweinberger@gmail. Since I do very little with G+ anyway, it only bothers me because I hate rejecting friends’ requests, even though they’re trying to join a G+ that I don’t ever check and that currently has a total of 7 people in it. So, I googled for info, and found that Google Takeout promises to move my dweinberger Circles over to evident. Google seems quite serious about it: access is limited during the first 48 hours, the transfer takes up to 7 days, and you can only request one transfer every six months.

We’ll see how it works. In any case, I do appreciate the Google Data Liberation Front commitment.

And perhaps now my Circles will be unbroken.

3 Comments »

September 4, 2012

Who forces Google to remove search results because of copyright claims?

According to a post at TechDirt by Riaz K. Tayob, Google has released data on which organizations request certain search results be suppressed because of copyright issues.

From TechDirt:

It may be a bit surprising, but at the top of the list? Microsoft, who has apparently taken down over 2.5 million URLs from Google’s search results. Most of the the others in the top 10 aren’t too surprising. There’s NBC Universal at number two. The RIAA at number three (representing all its member companies). BPI at number five. Universal Music at number seven. Sony Music at number eight. Warner Music doesn’t clock in until number 12.

The velocity is increasing:

As it stands now, Google is processing over 250,000 such requests per week — which is more than they got in the entire year of 2009. For all of 2011, Google receive 3.3 million copyright takedowns for search… and here we are in just May of 2012, and they’re already processing over 1.2 million per month.

The requests and Google’s responses must both be generated automatically. This raises once again the problem with having robots enforcing the law: They don’t know about leeway, which means they (a) lack common sense, (b) have no way of balancing against greater goods, and (c) can’t tell when Fair Use should provide an exception. (Here’s an op-ed I wrote in 2003 about this.)

We saw this this weekend as robots blocked the use of perfectly legitimate film clips at the Hugo Awards. Ridiculous. And scary.

3 Comments »

August 27, 2012

Big Data on broadband

Google commissioned the compiling of

an international dataset of retail broadband Internet connectivity prices. The result was an international dataset of 3,655 fixed and mobile broadband retail price observations, with fixed broadband pricing data for 93 countries and mobile broadband pricing data for 106 countries. The dataset can be used to make international comparisons and evaluate the efficacy of particular public policies—e.g., direct regulation and oversight of Internet peering and termination charges—on consumer prices.

The links are here. WARNING: a knowledgeable friend of mine says that he has already found numerous errors in the data, so use them with caution.

Be the first to comment »

July 17, 2012

Yahoo’s patents

It has been bruited about that maybe Yahoo has hired Marissa Mayer, employee #20 at Google, in order to get acquired by Google. I cannot see the sense in that as an acquisition tactic, but it has led to further speculation that Google is interested in Yahoo for its patents.

Now that makes sense! In fact, here are just four of the many valuable patents Google would acquire from Yahoo:

  • US PATENT 893749039 Improving the rapidity of the embarrassment of a corporate board through non-vetting techniques

  • US PATENT 989209374 Significantly depressing corporate value by the refusal of no-brainer acquisition offers through the innovative application of self-importance

  • US PATENT 463874738 A new calculus of corporate value that rewards the acquisition, mishandling, and abrupt closure of genuinely innovative services with loyal user bases.

  • US PATENT 784789909 Techniques for the alienation of a company user base by re-imagining customers as consumers and services as Big C Content.

(The truth is that I have a soft spot for Yahoo as one of the original engineer-led sites, and I hope Marissa can lead it back from the brink.)

 


Hanan Cohen points to DearMarissaMayer.com. I’m more ambitious than that; I’d substitute “Yahoo” for “Flickr” on that site.

Be the first to comment »

July 13, 2012

How Google may turn its Kansas City broadband project into a business

As you likely know, Google is in the midst of providing ‘ultra high speed fiber’ access to the residents of Kansas City (MO and KS). (‘Ultra high speed‘ means at least 1gb, which is 50100x faster than your 10mgb connection.) This has been positioned as an experiment, and as a poke in the eye to the incumbents to “show ‘em how it’s done.” And it has apparently made the incumbents nervous enough to offer residents a bounty for tips about the deployment.

Now Bill St. Arnaud speculates about how Google is going to turn this into a business. I have zero idea if he’s right, simply because I don’t know enough to have an opinion, but it sure is some interesting speculation.

Bill’s post is very readable, so I suggest you not rely on my summary, but here goes. First, Bill wonders how Google could hope to make back its investment in the physical infrastructure, since providers need about 40% of the market to subscribe to drop the per-user cost sufficiently. But (Bill figures), the incumbents will never let Google take 40% of their market. So, Bill figures:

Google will offer a basic free high speed Internet to each and every home, perhaps bundled with Google TV using their new set top box. A variety of premium services will also be offered for additional fees. I would not be surprised that Google decided to offer a basic 1 Gbps service to every home. This would clearly differentiate Google from the cableco or telco and make it almost impossible for them to compete without undertaking a massive investment themselves.

But, Bill guesses that the premium services will still not make the venture profitable. So, he speculates that Google…

…could offer to peak manage the customer’s power usage, by briefly turning off air conditioners and hot water tanks. They could also install smart thermostats and other devices to further reduce energy consumption. The money in the energy savings would be used to pay for the fiber or premium services, rather than being returned to the customer as piffling amount of energy savings.

So, the deal to users would be: We’ll give you incredibly high speed connectivity (or we’ll give you some great premium services) if you’ll let your energy company install a smart thermostat and manage your peak energy consumption in ways you won’t much notice. The user’s energy bills don’t go down (or don’t go down proportional to their energy consumption decrease), and the energy company shares the money with Google.

I’m not convinced that users would take the deal positioned that way. Maybe I’m positioning it wrong, but it seems like a pretty complex offer. I think I’d rather take a deal with my energy company to lower my usage and my costs, and then decide if I want to pay Google for fiber access or for premium fiber access. I already resent the cablecos for making their “triple play” (telephone, tv, Internet) pragmatically a requirement to get any one of the three. A double play of Internet and energy savings would be even weirder.

But, Bill knows approximately 50x what my own poor brain fiber does. The key is, I believe, in the energy company making the claim that the decrease in energy consumption will be minor, the noticeable impact on the user will be negligible, and the monetary savings would be “piffling.” If he’s right, it’ll be fascinating to watch.

 


This isn’t right, is it?

Seemingly wrong WolframAlpha result

9 Comments »

June 29, 2012

[aspen] Eric Schmidt on the Net and Democracy

Eric Schmidt is being interviewed by Jeff Goldberg about the Net and Democracy. I’ll do some intermittent, incomplete liveblogging…

NOTE: Live-blogging. Getting things wrong. Missing points. Omitting key information. Introducing artificial choppiness. Over-emphasizing small matters. Paraphrasing badly. Not running a spellpchecker. Mangling other people’s ideas and words. You are warned, people.

NOTE: Posted without having even been re-read. Note note (a few hours later): I’ve done some basic cleanup.

After some amusing banter, Jeff asks Eric about how responsible he felt Google was for Arab Spring. Jeff in passing uses the phrase “Internet revolution.”

ES: Arab Spring was enabled by a failure to censure the Internet. Google enabled people to organize themselves. Especially in Libya, five different militias were able to organize their armed revolt by using the Net. It’s unfair to the people who died to call it an “Internet revolution.” But there were fewer people who died, in part because of the incessant media coverage. And we’ve seen that it’s very easy to start what some call an Internet revolution, but very hard to finish it.

JG: These were leaderless revolutions, crowdsourced revolution. But in Egypt the crowd’s leaders were easily pushed aside after Mubarek fell.

ES: True leaders are very hard to find. In Libya, there are 80 militias, armed to the teeth. In most of the countries there were repressed Muslim groups that have emerged as leaders because they organized while repressed. Whoever takes over inherits financial and social problems, and will be thrown out if they fail.

JG: Talk about Google’s tumultuous relationship with China…

ES: There are lots of reasons to think that China works because its citizens like its hierarchical structure. But I think you can’t build a knowledge society without freedom. China wants to be a knowledge society. It’s unclear if China’s current model gets them past a middle income GDP. Google thought that if we gave them free access to info, the Chinese people would revolt. We were wrong, and we moved Google to Hong Kong, on the open side of the Great Firewall. (We had to because that’s the Chinese law.) Now when you enter a forbidden query, we tell the user that it’s likely to be blocked. We are forbidden from announcing what the forbidden terms are because we don’t want employees put in jail.

JG: Could Arab Spring happen in China? Could students organize Tianamen Square now?

ES: They could use the Chinese equivalent of Twitter. But if someone organizes a protest, two people show up, plus 30 media, and 50 police.

JG: Google’s always argued that democratization of info erodes authoritarian control. Do you still believe that?

ES: The biggest thing I’ve learned is how hard it is to learn about the differences among people in and within countries. I continue to believe that this device [mobile phone] will change the world. The way to solve most of the world’s problems is by educating people. Because these devices will become ubiquitous, it’ll be possible to see how far we humans can get. With access to the Net, you can sue for justice. In the worst case you can actually shame people.

JG: And these devices can be used to track people.

ES: Get people to understand they have choices, and they will eventually organize. Mobiles tend to record info just by their nature. The phone company knows where you are right now. You’re not worried about that because a law says the phone company can’t come harass you where you’re sitting. In a culture where there isn’t agreement about basic rights…

JG: Is there evidence that our democracy is better off for having the Internet?

ES: When we built the Net, that wasn’t the problem we were solving. But more speech is better. There’s a lack of deliberative time in our political process. Our leaders will learn that they’ll make better decisions if they take a week to think about things. Things will get bad enough that eventually reason will prevail. We complain about our democracy, but we’re doing quite well. The US is the beacon of innovation, not just in tech, but in energy. “In God we trust … all others have to bring data.” Politicians should just start with some facts.

JG: It’s easier to be crazy and wrong on the Net.

ES: 0.5% of Americans are literally crazy. Two years ago, their moms got them broadband connections. And they have a lot of free time. Google is going to learn how to rank them. Google should enable us to hear all these voices, including the crazy people, and if we’re not doing that, we’re not doing our job.

JG: I googled “Syria massacre” this morning, and the first story was from Russia Today that spun it…

ES: It’s good that you have a choice. We have to educate ourselves and our children. Not everything written is true, and very powerful forces want to convince you of lies. The Net allows that, and we rank against it, but you have to do your own investigation.

JG: Google is hitting PR problems. Talk about privacy…

ES: There’s no delete button on the Net. When you’re a baby, no one knows anything about you. As you move through life, inevitably more people know more about you. We’re going to have to learn about that. The wifi info gathering by StreetView was an error, a mistake, and we’ve apologized for it.

JG: The future of journalism?

ES: A number of institutions are figuring out workable models. The Atlantic [our host]. Politico. HuffingtonPost. Clever entrepreneurs are figuring out how to make money. The traditional incumbents have been reduced in scale, but there are plenty of new voices. BTW, we just announced a tablet with interactive, dynamic magazines. To really worry about: We grew up with the bargain that newspapers had enough cash flow to fund long term investigative research. That’s a loss to democracy. The problem hasn’t been fully solved. Google has debated how to solve it, but we don’t want to cross the content line because then we’d be accused of bias in our rankings.

JG: Will search engines search for accuracy rather than popularity?

ES: Google’s algorithms are not about popularity. They’re about link structures, and we start from well-known sources. So we’re already there. We just have to get better.

JG: In 5 yrs what will the tech landscape look like?

ES: Moore’s Law says that in 5 yrs there will be more power for less money. We forget how much better our hw is now than even 5 years. And it’s faster than Moore’s Law for disks and fiber optic connections. Google is doing a testbed optical installation. At that bandwidth all media are just bits. We anticipate a lot of specialty devices.

JG: How do you expect an ordinary, competent politician to manage the info flow? Are we inventing tech that is past our ability to process info?

ES: The evidence is that the tech is bringing more human contact. The tech lets us express our humanity. We need a way of sorting politicians better. I’d suggest looking for leaders who work from facts.

JG: Why are you supporting Obama?

ES: I like having a smart president.

JG: Is Romney not smart?

ES: I know him. He’s a good man. I like Obama’s policies better.

Q&A

Q: Our connectivity is 3rd world. Why haven’t we been able to upgrade?

A: The wireless networks are running out of bandwidth. The prediction is they’ll be saturated in 2016. Maybe 2017. That’s understandable: Before, we were just typing online and now we’re watching movies. The White House in a few weeks is releasing a report that says that we can share bandwidth to get almost infinite bandwidth. Rather than allocating a whole chunk that leaves most of it unused, using interference databases we think we can fix this problem. [I think but please correct me: A database of frequency usages so that unused frequencies in particular geographic areas can be used for new signals.]

A: The digital can enhance our physical connections. E.g., a grandmother skyping with a grandchild.

JG: You said you can use the Net to shame govts. But there are plenty of videos of Syria doing horrible things, but it’s done no good.

ES: There are always particularly evil people. Syria is the exception. Most countries, even autocratic ones, are susceptible to public embarrassment.

Q: Saying “phones by their nature collect data” evades responsibility.

ES: I meant that in order to their work, they collect info. What we allow to be done with that info is a legal, cultural issue.

Q: Are we inherently critical thinkers? If not, putting info out there may not lead to good decisions.

ES: There’s evidence that we’re born to react quickly. Our brains can be taught reasoning. But it requires strong family and education.

Q: Should there be a bill of rights to simplify the legalese that express your privacy rules?

ES: It’s a fight between your reasonable point of view, and the lawyers and govt that regulate us. Let me reassure you: If you follow the goal of Google to have you as a customer, the quickest way to lose you is to misuse your information. We are one click away from competitors who are well run and smart. [unless there was money in it, or unless they could get away with it, or...]

Q: Could we get rid of representative democracy?

ES: It’ll become even more important to have democratic processes because it’s all getting more complicated. For direct democracy we’d have to spend all day learning about the issues and couldn’t do our jobs.

JG: David Brooks, could you comment? Eric is an enormous optimist…

ES: …The evidence is on my side!

JG: David, are you as sanguine that our politicians will learn to slow their thinking down, and that Americans have the skills to discern the crap from the true.

David Brooks: It’s not Google’s job to discern what’s true. There are aggregators to do this, including the NYT and TheBrowser. I think there’s been a flight to quality. I’m less sanguine about attention span. I’m less sanguine about confirmation bias, which the Web makes easier.

ES: I generally agree with that. There’s evidence that we tend to believe the first thing we hear, and we judge plus and minus against that. The answer is always for me culture, education.

Q: Will there be a breakthrough in education?

ES: Education changes much more slowly than the world does. Sometimes it seems to me that education is run for the benefit of the teachers. They should do measurable outcomes, A-B testing. There’s evidence that physics can be taught better by setting a problem and then do a collaborative effort, then another problem…

1 Comment »

March 1, 2012

[2b2k] Moi moi

Google has posted my authors@google talk. Thank you, Google!

And Steve Hargadon has posted the hour interview he did last night as part of his Future of Education series, in which we talked about knowledge and education. Thank you, Steve Hargadon!

1 Comment »

Next Page »