Sloppy Reporting on the Self Driving Car | Hacker News Reader

Sloppy Reporting on the Self Driving Car

informationdiet.com

158 pointscjoh55 comments

I really like the ideas behind this "infoveganism", but lord do I hate the name. I think the author chose it because in his mind veganism has to do with health. Nothing wrong with that association (i don't think there's anything wrong with veganism, either), but in most people's minds veganism = restrictions mixed with a sense of superiority. And the thing is, there's nothing about demanding that actual content be present in your media that places a restriction on what content enters your brain. As the author points out, the content you are presently getting is what's restricted by virtue of being so barren of fact.

So if you were to stick with the food metaphor (not that I think that's a good idea), veganism doesn't real work. It's more like adding 13 essential vitamins and minerals to your otherwise toxic fruity pebbles.

Actually no, I think it has to do with both health and ethics, as sticking to a plant based diet involves both. Our information diets have as much social ethical consequence as our food ones do.

But for those of us who are more than happy to enjoy meat, veganism often doesn't mean either health or ethics, it means people who go without some of the best food and often feel superior about it.

Not saying that this view of veganism is correct or fair, just that that's the immediate connection.

Sure. I mean hey-- the most dangerous place in Washington is between me and a chicken wing.

The good news is, the term ticks of as many vegans as it does non-vegans.

I don't think its an unfair commentary, since even the poster you're responding to suggested veganism has much to do with ethics.

Presumably those who feel they are behaving ethically by committing an action, also feel that those who do not follow their standards are less ethical. That's the feeling of superiority that wafts off vegans.

I'm not entirely sure it's a requirement, personally I feel it's more ethical to cycle than to drive a car - but when I do cycle, I've never felt superior to anyone at all.

Then again, I regularly use taxis, get lifts... so maybe I am to anti-car-cyclists as people who occasionally go a day without meat are to vegans.

This is a contradiction. Either you believe the action is ethically superior or you don't. It makes no sense to believe it's ethically superior except when you do it. I would very much hope you believe you're doing the right thing.

I think you're confusing the belief that your actions are superior with the action of being a dick. They're not directly related — it's just that the latter is likely to reveal the former.

Is it impossible to think that something you do is a more ethical thing to do than something someone else does without a.) judging the person who isn't doing it or b.) at least thinking that their decision not to do it is wrong?

I'm not sure I know the answer..

Sure. It's more ethical for an self-professed alcoholic to not drink, as their alcohol consumption likely affects others. That said, it's easy to not drink alcohol and not judge others for drinking it.

That said, do that too often, and you're just around a bunch of annoying drunk people...

Exactly! Just like people who aren't infovegans think that infoveganism just means eschewing mainstream media and feeling superior about it! It's a perfect analogy!

I couldn't even figure out what the word was til about 20 seconds of staring at it. It would be better as info-veganism, I was trying to parse it as in-fov-egan-ism, or in-fove-gan-ism and such.

Yeah, I hate the name.

If you want to stick to food, I'd go with "whole story movement" or "slow story movement" or something. This seems to be about putting personal effort into getting a better result for yourself, vs. making sacrifices for the good of others (largely the point of veganism, as far as I can tell).

That does indeed seem to be the sense in which infoveganism is meant: You consume information according to a strict set of ethical standards (which you feel is superior) rather than what is most pleasant or convenient. Just like with veganism, it is mostly defined by what you choose not to consume, but it is also important to replace the "bad" things you discard with enough good things. It isn't just about "actual content being present in your media" any more than veganism is about veggies being present in your meals. It's about only taking in good content and eschewing content which (by the philosophy's standards) is bad.

When I first heard it, I didn't like it because I thought it was based on the popular conception that vegans practically starve themselves. But having read more about it, I think the concepts are indeed similar.

Vegans consume food according to what is pleasant within their ethical standards. Yes they refuse more food due to ethical boundaries, but that doesn't mean they've fundamentally changed how they consume food.

Meat-eaters for example mostly refuse to eat human meat, but choose to eat any other kind of meat that is pleasant--cannibalism excluded.

This standard is strict as well, there isn't any leniency allowed for cannibalism in the common day.

I think you're using a different definition of "pleasant" than I was. The idea of something being ethically pleasant is odd to me. I basically meant "tasty." Some vegans find, say, fried chicken to be quite tasty, but they refrain from eating it nonetheless. The same cannot be said of human flesh for most people.

As for the suggestion that vegans are no more dedicated to their diet than anyone who does not eat human meat, I find that frankly bizarre. Refraining from cannibalism in a society where cannibalism is illegal is extremely different from avoiding all animal products in a society where animal products are nearly ubiquitous. People who don't eat other people are essentially going with the flow. (This is not to say that they would be cannibals in a different situation, just that few are even presented with the choice, and choice is an important distinction here.)

Most Americans eschew insects, dogs, and cats. That seems like a better analogy.

This is a great article detailing a manifestation of several problems:

- "News" as entertainment. The primary purpose of many articles isn't to inform, but to entertain. Or otherwise provide utility other than a deep, informed, and/or unbiased/NPoV perspective.

- "News" as marketing. This is so prevalent in tech (and has been for decades) it's passed from "not funny" to trope. Company launches (itself, product), hires PR team, shops story to various media outlets, breathless hagiography ensues. The real market is usually 1) investors and 2) market shaping (for consumer products), though influencing legislative and regulatory environments (as in the Google car story) may also be a goal.

- "News" as linkbait / attention generator. Though we think of this as a modern phenomenon, and Web-based concepts of eyeballs and social networking don't help, it's also old as dirt. So you get crap like PC World, Engadget, and even the New York Times with recursive, internal references, and lacking (as Clay Johnson points out) the truly useful references to original sources. HN is a great antidote to this -- the "Ron was wrong, Whit is right" direct link (and discussion of) PKI RNG weaknesses.

And yet ... to a large approximation of "it works", it works -- as a business plan, these are fairly successful market strategies.

Hard information markets are hard.

They are only hard to support from advertising.

For any service which requires payment for access to the information, said information tends to be hard. Blomberg news and the Wallstreet Journal, are examples for this in terms of financial news.

WSJ is largely a mouthpiece / propaganda arm of Fox and the Murdoch empire these days. Some would say it's always had a strong bias toward, shock, surprise, Wall Street interests. There's a great quote somewhere by Warren Buffet if you can find it about the WSJ's editorial staff believing its own writing.

There are a number of information venues that are member/subscriber based, and which do relatively well: NPR, Mother Jones, the Christian Science Monitor, Consumer's Union. The Economist Newspaper (as it styles itself) has a rather interesting revenue model, roughly one-third each subscriptions, ads, and EIU. The last is the Economist Intelligence Unit, a bespoke research arm, though I suspect much of its work ends up in the public edition at least in the form of background and depth.

There's a great deal which could be written on markets and funding streams for information goods -- not just news, but arts, invention, data, and other works. Over the past several millenia, we've seen patronage, sponsorship, performance, publication, subscription, advertising (both direct and sponsorship), data mining, music sales (sheet, player roll, analogue recording, digital downloads), mechanical royalties, and other methods. Each has a profound influence on the nature, quality, and form of work produced.

This obsession with self-promoting links is what makes reading sites like BoyGeniusReport and everything in the Engadget family so frustrating.

Editorial rules seem to be that any in-story link MUST be a self-reference to some keyword match inside the site and the only external link (singular) allowed is the one at the end of the article using text like "Source" or "via XYZ.com".

God I hate this on About.com. They're almost always the first result when I search for something, and every link on their page goes to another About.com page. And none of them have the information I'm searching for. I've added them (along with all of gawker media) to my hosts file.

One of the neater features I've seen Google implement is that if you click a link and then hit back within a short period, a little link shows up next to that search result that says "hide everything from about.com?"

Mashable now link all companies and products to their seo spam pages too. These sites are just getting ridiculous.

I posted a lengthy complaint about the ridiculous state of news blogs the other day but it didn't quite make it:

http://news.ycombinator.com/item?id=3564257

It's a quick thought, but what we could do is the following: Have a wiki-like site, where the community update source information about news items. On the items pages, they also make links to news articles on the web. Along with this there is an analysis tool, so if you see an article on a news site somewhere, you pass the URL to this site, and it will point you to the relevant wiki page based on the URLs that are set on the wiki page or by guessing the best one for you based on content and similarity to other news reports.

You mean, like, say, WikiNews?

Specifically: citing your references: http://en.wikinews.org/wiki/Wikinews:Style_guide#Citing_your...

I so wish so called 'journalists' would take this to heart.

Although sometimes they would 'harsh the buzz' as it were. I recall a series that ABC news did on unemployment and they profiled a gal who, in their story, seemed like a sad case of working hard and not getting anywhere. Unfortunately for them there was enough PII in the story that one could do a bit of Googling and figure out who this person was, only to find they were a convicted felon (drug possesion with intent to sell), in and out of rehab, and recently 'laid off' when their Nova job term timed out and the employer elected not to hire full time. That is also a sad story but it would not have 'worked' for what the news producer was going for 'Regular folks, working hard, and unable to find work.'

I agree its quite different to omit facts to make your story more credible, buts another form of this 'info-veganism' which. Generally I've been happy with the Economist (they seem less tempted by this sort of thing than others) and it suggests that there is once again an opportunity to create a news organization. (If you are old enough to remember, Fox News was created on the back of the first Iraq War with the tagline 'we report, you decide' trying to take the editorial spin out of reporting which was so evident in CNN's coverage, of course once it was a going concern that changed.) I believe there is a really disruptive opportunity here which I'm happy to share if anyone is interested.

Yeah! What's the disruptive opportunity?

So the opportunity is to create a video news network based on a streaming / store model rather than the 24hr model. Lets say you put together such a network, as you cover stories you put them up on your server. Your customers click 'the news' and they get a newscast of all the current news, but once they've seen a story they don't see it again unless there is actually additional information.

Things like sports highlights, weather, and finance have a regular 30 minute update cycle. So push the button and you can watch a 'show' which is probably 30 - 40 minutes long of the current news. If you watched the news this morning and its a slow news day the icon indicates no new updates.

You click news you are interested in and it tailors your broadcast to you're interest. Don't care about entertainment? not there. Don't want to hear about the latest hijinks on the debate circuit, not there. Interested in updates on NASA's budget? there.

You can surface potential stories of interest in the 'crawl', click/press it and that story gets added to your newscast.

Same basic organization as any other news organization, but the transmission is network based.

News transition Nightly -> Morning + Noon + Night -> 24hr

Differentiation Liberal leaning / Conservative / more local / less local / more weather / more sports / etc.

Monetization - advertising.

Disruption - remove transmission costs, pull cable networks out of the middle, move to NetFlix/AmazonPrime/Hulu with a higher revenue share. Product hook is "like a Tivo that watches the 24hr news channel and only records the new/interesting/current bits." Value add - saves time for you, no need to wait around for the top story to repeat, press go and its there. Partnership opportunities with business hotels for additional monetization revenue (they pay for your feed (probably generic version, not customized) and insert hotel specific ads into the video (rev share). The customer value proposition is news now, no repeats.

When it happens it will disrupt the crap out of Fox/Cnn and will add a missing service to the NetFlix/Roku/Amazon/Hulu's of the world.

Sounds like a good idea. It'd be a lot cheaper to do in text (and wouldn't require any technology that didn't exist in 2000); has anyone done it? If not, why not? If so, what happened? Does making it video greatly change the value proposition?

The infrastructure cost of it is a barrier. One has to understand the 'news' is "the highest value video production out there next to porn" [1], you take a producer, a camera person, and a talking head, and you record what you see.

And to the text only question, efforts using text have not been successful, people seem to prefer to consume it as video or audio when doing something else. The goal is to disrupt Fox, not the Chronicle, you need to do it in a way that the average person has very little friction between wanting to know 'what's happening' and getting the news. They have been voting 'video' by their consumption habits to date.

Most of the efforts I've seen that have started along here have chickened out when it came to being the network vs being a repackager. They spend a weekend writing a server that repackages the AP feed and maybe reads it aloud. But the editorial voice is pretty important for this endeavor. You really do need to be the news organization.

The other barrier is that the 'value chain', which is to say the way in which news is monetized is very well understood and so you cannot succeed by being another 'channel' to get to the same stuff someone else has, if you are successful they cut you off from the content. Or if you leverage someone else's network they cut you off from the bandwidth. So you either have to 'go big' here, or not go at all. That means a studio for putting the final product together, a mobile reporting force which can get to places, and partnerships in place for places where you cannot afford to get to yet.

The change which has been occuring which will make this possible today when it wasn't possible before is that enough consumers have the requisite network bandwidth to their home or office to support it, and they are likely to have a device which can stream it (Roku, PS3, Samsung 'smart' TV, etc). The sweet spot of this market is the 22 - 42 year olds who want a better news experience than they are getting and have high comfort with technological solutions. As a 'network' you also already have a business model that can slot you into a relationship with folks like Comcast, DirectTV, or Time Warner. Partner with them and their 'smart set top' on a rev share basis and you give added value add for their customers, and pricing leverage against those networks you are disrupting.

[1] This was a quote from the guy who was the GM or EVP of the KRON TV news group at a talk he gave back in the 90's in Palo Alto.

> Two people are required to be present in an autonomous vehicle, both holding a valid driver's license.

Why?

It's a pilot program for legalizing on road research projects. Presumably, someone felt that two people paying attention to the autonomous actions was better than one.

One paying attention to the research and one paying attention to the traffic situation sounds like the right division of labor.

In my experience you need the driver to fully pay attention to the road with one hand on the kill switch and the other on the wheel while the passenger works the computers doing the driving. This is just legislating common sense.

These are Research Vehicles, not final products.

A million dollar bond from the state seems kind of excessive

A pile-up could easily cause a million dollars worth of damage. A measure that makes sure that the responsible businesses are serious about safety doesn't sound that unreasonable to me.

I guess it makes sense when only large businesses are the ones to own these automobiles. It puts individual ownership out of reach.

>Businesses wishing to test their self-driving cars have to buy a million dollar bond from the state.

It's not for individual owners.

Well its not about individual ownership its about businesses testing commercial self driving cars right?

Seems to me that a company is going to need to sink many millions in to design, build, and test a truly road safe self driving car no? Even considering it as a hardware and software mod to an existing model.

Individual ownership of driverless cars is out of reach, and always has been. This pilot program brings it a step closer, despite clearly just a test (e.g. requires two drivers in the car)

It isn't even that big a number. People with lots of assets carry that much or more liability insurance and they aren't paying enormous amounts for it.

How is a robot caused pile-up any different from a human caused pile-up? Do human Nevada drivers have to carry a million dollar bond?

Human drivers are not a new technology being field tested in a new and challenging environment.

Drivers in most U.S. states (I can only speak for VT, PA, and MA) are required to carry liability insurance in order to drive legally. The amount varies.

Robot drivers are a new and unproved technology. After not too many years I'd expect that they will cause way fewer accidents per mile driven than human drivers.

Add that there is not a whole lot of case law on robot driven vehicles to understand how far the accident will go legally (unknown expense for an unknown number of parties)

Funny. I was thinking it sounded low.

You were not alone, that I can say.

The bond does not represent Max liability. The coproration and probably even the individuals in the vehicle would be liable as well.

But I agree that it is low.

Not excessive as long as Nevada has a complete monopoly ;)

One crash survivor with a broken neck could eat up that million quick.

That would restrict it to big business.

But the requirement that two drivers need to be in the car will restrict it to no one who wants to invest in it to be more profitable.

Right now this is simply for experimentation and I wonder who'd drop one million dollars just to experiment?

Let's see now: CMU, Stanford, MIT, UPenn, USC, LAAS, USyd, Oxford, Caltech, just to name a few. All of these universities have already dropped at least that much building technology for the DARPA Grand Challenges or related research demonstrations.

Right. But I want my car to drive me to work and back while I sleep or read. But to get there I think first business must commercialize it through their simple desire to increase profits, then it may become cheap and accessible enough for average Joes like me.

When you require two people with a driver's license in the car at all times you kill any profitable commercial reason to use self-driving cars. And that to me is a huge mistake.

Commuting to work (if you drive > 2 0minutes at highway speeds) is probably the most dangerous thing by far that all of us do on any given day.

Distracted drivers, drunk drivers, sleepy drivers (like me!) this is why we need self-driving cars.

But I think the path to mass use of self-driving cars goes through commercial use of self-driving cars.

I guess it wasn't clear enough that this is the regulation for testing autonomous cars on public roads, not for deployment of autonomous cars.

I don't like the fact that comments require connecting to facebook. Which might explain the lack of comments on the page.

Reply on news.ycombinator.com