The Economics of Evil Google | Hacker News Reader

The Economics of Evil Google

krugman.blogs.nytimes.com

129 pointsmkr-hn115 comments

This is getting ridiculous. It's been three weeks of highly public complaining/pseudo-analysis about Google Reader. Now it's 'critical public infrastructure', on par with public transit systems.

Can we have a reality check? It's an RSS reader. It was really good. Now it's shutting down. Nobody relies on it to get to work so they can feed their children. Building a replacement doesn't require digging tunnels, buying trains, hiring drivers. The economic argument is interesting, but I'm wondering if this wins for 'Most Hyperbolic Google Reader Shutdown Article'.

It's important to note that he's not actually suggesting the government take over Google Reader, just giving an economics-based viewpoint on Google's situation. Combined with the linked article, it makes for good reading.

>Now it's 'critical public infrastructure'

Don't get riled up, this was a piece by Paul Krugman. He thinks government should own /control / regulate everything.

Exactly. He's a Keynesian freak who lost it a long time ago.

Right. Because the steamboat ferries, railroads, interstate highway system, and the internet were created by the free market...

You seem to be responding emotionally to something you're tired of hearing about topically, not the substance of this piece which doesn't suggest any of the things you seem to feel it does.

It's about economics, particularly as they apply to something with fixed-to-increasing costs but with users for whom the value has a fixed upper limit and is declining based on the intensity of their usage. There's a lot more here than complaining and pseudo-analysis. It's a Nobel Prize winning economist talking about pricing models as they apply to usage/cost/scale!

He has an interesting point, but he assumes the outrage by the high intensity users is proportional to the perceived value. In fact, they're just making news; they need to generate content, and this is something that requires no research, just a vitriolic rant. I suspect if we had a good way to measure 'willingness to pay', many of the loud complainers would suddenly end up in the 'low intensity' bucket. Less high intensity users means less network externalities, which comes back to my original point that the whole phenomenon is all smoke and no fire - lots of people complaining about the loss of a service that isn't terribly unique or widely used.

He does describe interesting models/theories, but I don't actually think they're relevant. It seems cheap and link-baity.

> He does describe interesting models/theories, but I don't > actually think they're relevant.

I have to agree. Google is very efficient at profiting from casual users unwilling to pay, but clearly they just didn't exist.

> [...] he assumes the outrage by the high intensity users is proportional to the perceived value

Right, that's the point. This is a common style in economic blogs. They often follow the form: assuming some set of data, what are the economic implications?

> He does describe interesting models/theories, but I don't actually think they're relevant.

Right, they may or may not be relevant to this specific case (Reader), but I still found it interesting to learn about.

> It seems cheap and link-baity.

Hmmm, here we have a distinguished economist, who could no doubt live easily from his university job, instead spends time trying to bring scientific economic analysis to popular topics. Not sure how that is cheap and link-baity.

He does have a knack for catchy titles, though. If you aren't familiar with his blog style, "evil google" probably sounds bad (and emotionally charged), but I really doubt it was meant to be taken literally.

Hmmm, here we have a distinguished economist, who could no doubt live easily from his university job, instead spends time trying to bring scientific economic analysis to popular topics.

Meh... you could make a pretty compelling argument that economics isn't a science, and a pretty compelling argument that Krugman is a radical ideologue who is good at marketing himself. In any case, whether he's distinguished or not, whether or not he has a Nobel Prize, how else he could theoretically support himself, don't really enter into it... evaluating this piece of content (as well as any other) should be done on it's merits (or lack thereof).

It's kinda disappointing in this case, because that actually is a really interesting article from an economic perspective. But when he mixes it in with his psuedo-socialist, big government, leftist ideology, it immediately turns a lot of people off. I almost didn't click through just because it was Krugman. But he sucked me in with the catchy title. :-)

> his psuedo-socialist, big government, leftist ideology

What are you talking about? What part of the article made that impression on you and why?

That he has that ideology in the general sense is well known, so just the fact that it's Krugman automatically brings that into play. Per this particular article, the bit where he says:

So what’s the answer? As Avent says, historical examples with these characteristics — like urban transport networks — have been resolved through public provision. It seems hard at this point to envision search and related functions as public utilities, but that’s arguably where the logic will eventually lead us.

alludes to his ideology. But to be fair to Krugman, he is quoting somebody else, or referring to the works of others, for big chunks of this article, so I can't be too harsh on him.

It's not the RSS reader. It's the open publishing ecosystem.

Most clients point to Reader as the central feed aggregator. Most publishers point to Feedburner as the central publisher.

Google seized the commanding heights with Feedburner and Google Reader and captured all the publishers and the clients, and now they're killing the ecosystem.

http://www.zdnet.com/embrace-extend-extinguish-how-google-cr...

I don't see why they couldn't have integrated Reader into Plus without killing the ecosystem. Twitter, Facebook, LinkedIn are moving into news aggregation, and Google is killing a successful news aggregation system.

I don't understand their strategy, but it's seems sort of like, we want everyone on G+ and we don't care how heavy-handed we look or how early adopters feel, and we don't want an open ecosystem that people can use to pipe content into Twitter and Facebook.

People can pipe content into FB and Twitter using APIs. The only service you can't pipe content into is G+, which has no input API.

The NYT tends to have really good articles, but some of their columnists are so off the wall I can't figure out how they keep their jobs. Being a columnists gives the writer a self reinforcing celebrity factor then they just stick around way longer than they should.

FYI: Krugman is a Nobel Memorial Prize winning economist who is quite otherwise famous.

He's not just a columnist at the Times, he's the only columinst at the Times ;-)

http://krugmantimes.com/

Just about the only one worth reading. And the one[0] [in just about the whole of the news industry] with the best track record as far as making testable and correct predictions.

[0] http://www.poynter.org/latest-news/mediawire/130485/claim-kr...

Don't get me wrong, I'm a big Krugman fan. Even he had a good laugh at the Krugman Times and shared it on his blog.

quite otherwise famous = sadly enough by reinventing himself as a partisan talk show pundit.

Well, the people the shutdown affects are also the people most likely to write articles. Everyday people may not get much out of Google Reader, but neither do they follow hundreds of sites looking for a good story to write like journalists do.

Nor do hundreds or thousands or hundreds of thousands follow most everyday people.

>Nobody relies on it to get to work so they can feed their children

I don't think anyone uses that as the criteria of what should be made public. People rely on car companies and grocery stores to get home and feed their children, but economists aren't fighting to make those public. And federal parks are hardly critical infrastructure but I don't know many people who complain that they are public.

The article is written from a purely theoretical economic perspective, it's not a real proposal, so it's worth addressing it from that perspective. Krugman's argument is basically that there are certain products that add global value but cannot exist without people being forced to pay for it. Many economists - not all - agree this is a good criteria of what should be made public (infrastructure and research would fall in this category). It's hard to imagine that web services could fall under this, but I suspect a lot of people forget that the internet is still in its infancy and companies like google have been able to subsidize loss-creators through insanely high profits. That won't last, google is already scaling back, and the result may be a global destruction in value.

(To be clear I don't want to see things like google reader made public for idealogical reasons, and I think you and most people are the same way, but I think that's out of the scope of addressing Krugman's point)

The sore point for me was mostly that the article takes it for granted that Reader had this huge positive externality, because a lot of people are complaining. It's theoretically interesting to say that such a service might be worth publicly subsidizing. It's just annoying that he happened to use Reader as an example, because there has been a glut of news about Reader lately, and I think it's overblown.

Where does it do that?

Basically it says following: * some people I know are upset * we know from economics, but often forget, that some products can be not viable even if people are willing to pay for them * historical solution is to make them public service which also to Krugman seems hard to envison

Nowhere does he say anything about importance of Reader.

In the case of Reader, I believe a number of other services will likely fill the void, one or more which will be able to do it profitably. If Reader was made a public service, we would have lost the potential competitors, and we also would have lost the information on how it would be done profitably and exactly how profitable. I think it would have been nice if Krugman also at least mentioned the potential value of that.

Why? That was not the point of his article and it is too soon to tell if we would lost anything. Even if we would, he is not obliged to research and write about every possible aspect.

A good deal of useful thinking involves models, and I don't particularly understand the anger that results.

Google isn't a monopolist in this situation either, but it is useful to understand pricing dynamics where there is no competitor. It sheds light on price discrimination with coupons, for example.

Krugman mischaracterizes the article[1], since it does not claim that Reader was crucial infrastructure, but he doesn't argue that it is crucial infrastructure himself either. They both talk about the dynamics of a product that has network effects and becomes a public good.

We generally know from cognitive sciences how we economize in situations like having a reliably knowledgeable colleague. This has served as a counterpoint to the silly arguments about how Google is making us stupid, but nevertheless says that the specialization in our expertise that it facilitates has Google search's continued availability as a vital component.

The original article talks about whether discontinuations will cause users to be more cautious about adopting services that create dependencies, and whether it will inhibit the success of new ones that involve network effects.

>Google wants us to use its services in ways that provide it with interesting and valuable information, and eyeballs. If a particular Google experiment isn't cutting it in that category, then Google may feel justified in axing it.

>But that makes it increasingly difficult for Google to have success with new services. Why commit to using and coming to rely on something new if it might be yanked away at some future date? This is especially problematic for "social" apps that rely on network effects. Even a crummy social service may thrive if it obtains a critical mass.

Anyway, the important issues aren't perfectly typified by Reader, but it is a situation people are familiar with at this moment. Personally, I think it is worth thinking about why more of the shuttered projects haven't been sold off or at least open sourced like Wave. Combine that with profitable businesses that are acquired for the talent, then neglected to death, and it sounds like a lot of effort is being needlessly trashed.

It is not difficult to imagine that if books.google.com were able to give even more liberal access to its library, that struggling municipalities would be even less willing to fund their libraries. This hypothetical would be unambiguously good for the world if it lead to more access to information, but one can also imagine that this hypothetical could also lead to a dependency that was itself vulnerable.

There are a lot of things that won't lead to the end of civilization, but they can still be nevertheless worse than an alternate outcome.

[1] http://www.economist.com/blogs/freeexchange/2013/03/utilitie...

But that makes it increasingly difficult for Google to have success with new services. Why commit to using and coming to rely on something new if it might be yanked away at some future date?

This perspective just boggles my mind. If anyone decides not to use a Google product because it could "disappear just like Reader" or something, they are shooting themselves in the foot.

First, any product (especially web sites), made by Google or otherwise, can disappear. If it's not a profitable product/market, it will go away, whether it's run by Google or some guy in a basement.

Secondly, even if you suspect a product will eventually disappear, why not chose to use a superior product while it's around? This is especially true for something as inconsequential and fungible as an RSS reader. I mean, in retrospect, do hard-core Reader users wish they could go back in time and chose a different product instead of Reader for all these years? What would be the benefit of that? Whatever choice you made could have also disappeared. And if your answer is that you would have still used Reader, than that's all the evidence you need to know you shouldn't hesitate to use a new Google product, if it's the best option.

People still are unused to the idea of planned digital obsolescence.

I don't think you can call it "planned digital obsolescence" since "planned obsolescence" implies that the provider had a specific life span in mind when launching the product.

Google Reader is being shutdown but I doubt that when they launched they said to themselves "We will run this service until 2013 and then shut it down". Without that plan I don't think it can be planned obsolescence.

Perhaps the stumbling block here is that Reader is such a bad example, since it is mostly for consuming content non-interactively, and applications that use the Reader infrastructure won't have a difficult time replacing it. However, that neglects the importance of network effects for Google, developers and users.

Maybe you recall a chapter in Cannery Row where a gopher sets up in a perfect vacant lot full of everything it needs and without predators. (Or not, and I suppose the value of allusions rest on the network effects of a common canon) The gopher flourishes, but ultimately gives up because no female gophers ever appear, and it returns to a garden with gopher traps set out. The allegory is an accessible way to explain network effects, something that is often lost in the entrepreneurial rhetoric, where companies succeed or not regardless of the market environment or context of the community they foster or don't.

When choosing a large development platform, anyone reasonable doesn't only look at the tools and languages, but also considers the community and the market. Take for example Blackberry or the Windows Phone. I don't think developers believe that anyone owes them certain tools, or even a guarantee of a certain number of sales. However, these platforms will only get good talent if there is a credible marketability of anything they develop.

Hypothetically, lets assume that Visual Studio works really well for developing on Windows Phone and is better than any other set of tools for mobile development. Then let's assume that in 2015, Microsoft stops creating phone operating systems because the phones remain unpopular. A lot of developers have spent a couple years learning skills, much of which they can use elsewhere. If they'd been developing for Android or iOS, they would have even more skills that would be useful elsewhere. Now, someone shouts at you, "So you're saying that Microsoft shouldn't have released such a good set of tools?!" The question itself is silly.

The problem is not reducible to a set of tools, or evangelizing to developers, or confidence in the platform, or advertising to consumers. Assumptions about all of these together go into developers' decision making today, and it isn't even that deterministic, since there's an error component to how these perceptions that aren't based on accurate future-prediction, that will itself decide the future success of the platform.

The original Economist post didn't so much talk about Google owing anyone anything, but addressed the confidence problem. If people believe that a product won't be around very long, it won't get talented developers which won't attract users, which will lead to the product being canceled, or vice versa.

Let's say you want to put a small module on your webpage, and you are considering Google Gadgets. It might be more difficult to replicate the same functionality using HTML5. Are you worse off, because Google Gadgets is available? Google doesn't owe you any guarantee about the longevity of the system, but it would be more valuable right now if you could believe that Google Gadgets will still be around in a couple years, and your payoff matrix says that you would be worse off if you choose Google Gadgets over something else and it is later canceled.

Don't read everything in black and white, observe the shades of gray. More people will avoid Google products, because now they seem more fickle and prone to disappear. Google Keep is an excellent example: features wise it does not have a single killer feature that moves people off Evernote, and the fickleness is making it so that users prefer to keep paying Evernote instead of jumping onto the free Google equivalent.

They both talk about the dynamics of a product that has network effects and becomes a public good.

No, both Krugman and Avent are talking about private goods with fixed costs to the producer.

Network effects might increase the consumer surplus, but that's all. They aren't that important to the analysis. The important part here is the inability of the producer to engage in price discrimination.

If you read the Economist article again, I think you'll see that you may have come to a personal conclusion with regard to identifying the source of the problem. This is especially apparent, because neither of them even mentioned the possibility of variable pricing, as it was only brought up in my comment.

Both focus on the value of the community itself, the public good that emerges from the physical infrastructure that is privately built, that its value increases with size, that growth is influenced by size, and that early adopters are likely influenced by guesses about longevity.

The analysis could be wrong over the long term, but there is a growing consensus summed up in another front page post[1] right now, that describes how Google has caused some damage to its brand. Some of which is likely being felt right now in the chilled enthusiasm around the release of Google Keep—a product that relies on network effects more than Google Reader does.

[1]http://www.theatlantic.com/technology/archive/2013/03/finale...

"Nobody relies on it to get to work so they can feed their children."

That isn't true but why it isn't true is not immediately obvious. Google reader drives a ton of traffic, what that means in real terms is that there are a huge body of people who don't have time to regularly visit a blog, but they do have time to look quickly at what 10 or 15 blogs have recently posted. Thus they discover where there is an interesting article on "your" blog and visit. That is how it drives traffics to otherwise poorly visited blogs. It is more effective than G+, Organic Search, and Facebook in that regard.

Those bloggers that benefit from this feed do feed themselves and possibly their family on income they get from their writing either in ad-clicks or affiliate links.

When Reader goes away, a number of those blogs will be hurt, and those people will be scrambling to find other ways to drive traffic to their sites.

Now this will also hurt Google because there is a crap ton of AdSense for Content advertising on those blogs, and I expect that what use there had been of Blogger will drop below the 'minimum sustainable point' and so Blogger too will get tossed into the spring cleaning trash bin. Now its probably not noticeable (much) the last time we (Blekko search engined) looked through the blog crawls in depth Google was the dominant ad network provider on those blogs, but they do something like 3.7B$ of revenue in 'Partner sites' [1] per quarter and 1% of that being tossed out the window is only 37M$ per quarter (or about 150M$ for the year) isn't a huge loss for them but its going to hit a number of their "partners" quite sharply.

Unlike sites which are bogus flick-n-click ad reapers these folks are actually trying to use the system in the way it is supposed to work, they write quality content and Google helps them monetize it.

The thing that few people "get" about Google isn't that they are a search monopoly (they aren't) it is that they are an advertising monopoly (which they are). They are synonymous with "internet advertising" and having dealt with a lot of advertising partners, they all seem to end up re-selling Google's ads in one way or another. Its like having three different pizza restaurants but only one kitchen.

I'm hopeful that this will create opportunities for actually new advertising networks to appear> We'll have to wait and see though.

[1] http://investor.google.com/earnings/2012/Q4_google_earnings....

For perspective: "huge body" is ~millions monthly actives (but: many of them journalists), very likely less than ten million (http://www.quora.com/How-many-users-does-Google-Reader-have). YouTube just crossed 1 billion monthlies (http://www.billboard.com/biz/articles/news/digital-and-mobil...), 100x that. Which of the two would you push?

And what is the engineering cost for Reader vs YouTube?

No one owes those bloggers traffic. Certainly Paul Krugman is completely out of left field to suggest that the traffic is a public service.

I agree that Krugman over states the importance. My point was that Reader generates a lot of legitimate advertising revenue for Google, which they kill off, and in the process may kill off some of their smaller "partner sites." That refutes the equally over broad characterization that "nobody" uses Reader to put food on the table.

Where did he do that? Ryan Avent, whom Krugman links to, talks about Google providing public services but he's talking about their whole offering. Krugman is just playing with the economics of fixed-priced monopolies.

Are you really surprised that someone like Krugman would argue

"seems hard at this point to envision search and related functions as public utilities, but that’s arguably where the logic will eventually lead us."

That is the "logical conclusion" for people of his ilk.

And I'm not surprised by the first comment that I found on the article:

It is the height of capitalist folly to insist that a profit-seeking corporation is the right form of organization for operating facitilites that become essential global public utilities.

Why of course. This guy is just random article commenter, but you know your average, hardcore leftist thinks the same way. Facebook must come under UN or some other quasi-governmental control. It needs to be controlled. Well, duh...

Probably it is correct that Google dropping Reader will hurt some people's businesses. In the last week I went through a lot of hassle moving my blog from Blogger to Wordpress (which I have more control over). Would Google ever drop Blogger? Probably not but who knows. I also stopped using GMail as my primary business email service. I recently blogged about the 5 Google services that were terminated that affected my customers' and my businesses.

I like to use free services from Google, Twitter, and Facebook, but I certainly don't want to rely on them for business.

An analogy: my family likes to support small local businesses. For web services, hosting, etc., I prefer using companies that I pay fairly for services and who support me.

"It's been three weeks of highly public complaining/pseudo-analysis about Google Reader."

On March 13, Google announced it's killing Reader; that's 10 days a go. http://googleblog.blogspot.nl/2013/03/a-second-spring-of-cle...

I know I'll be complaining about this for quite some time to come.

Warning to people reading the comments before the article:

This article is not calling Google evil, it's examining the economic and social incentives of maintaining a public utility as a for-profit company and as a society. I tend to think that putting scare quotes around words in headlines is really condescending behavior from headline writers, but threads like this remind me that I'm not an authority of headline writing.

And for the people arguing that Krugman is an ignorant big government fraud for claiming that Google Reader is a public utility - Mr. Johnson's water pump on the corner that half of the people on the block use is a public utility. It's economic analysis, not a communist grab for your guns, or a socialist power play to keep capitalists from doing whatever they want to do, whenever they want to do it.

edit: hadn't refreshed in a while before replying. This comment is mischaracterizing the level of discourse in this thread now:)

You're completely wrong. Krugman hates capitalism and free markets, and is on a crusade to end them.

Claiming that X's property is a "public utility" _is_ a power grab.

No, it's claiming that it has utility for the public. There's a hint in the name.

That's a contradiction. I personally have utility for the public. So do private companies. You can't call these things public utilities without introducing a contradiction into the language.

People who try to introduce contradictions into the language only do so because they are _motivated_ to do so, because they want to get something dishonestly.

Another good example, besides the one we are discussing, is the idea that people now call whatever they want a "human right." That's a contradiction, because it's not necessarily the case that it's even _possible_ to _produce_ enough of whatever that thing is.

Mangling the language is a crime against thought.

This is the fundamental problem with modern liberal economic thinking:

>>First, it’s a well-understood though not often mentioned point that even in a plain-vanilla market, a monopolist with high fixed costs and limited ability to price-discriminate may not be able to make a profit supplying a good even when the potential consumer gains from that good exceed the costs of production. Basically, if the monopolist tries to charge a price corresponding to the value intense users place on the good, it won’t attract enough low-intensity users to cover its fixed costs; if it charges a low price to bring in the low-intensity user, it fails to capture enough of the surplus of high-intensity users, and again can’t cover its fixed costs.<<

Sure, it's well understood that certain people want others to provide a service that certain people find very valuable, but aren't willing to pay for. That doesn't mean that service must be provided by the government when private parties refuse to provide it.

In 2003, I wanted an online store where I could purchase all of my pet-related supplies. This would make me, and society, far more productive because we would not have to spend time going to the pet store and purchasing those products. We could all then spend that extra time being productive. Potential consumer gains exceeded production costs.

Unfortunately, actual consumer gains did not exceed pets.com's production costs. So it died. And guess what? It turns out that when a company learned how to properly structure, supply, and support such a service at the right price, people were willing to pay for it, and thus actual consumer gains exceeded actual production costs.

The government didn't have to step in and supply this vital service to its citizens. Some of the smarter citizens figured out how to do it, at assuredly a fraction of the cost of whatever government program might have been implemented.

Note: before you claim that pets.com failed for some other reason, please stop. I picked it as an obviously frivolous example. You can substitute most other recent government programs if you’d like.

His argument is not that it costs more to run Google Reader than there is benefit, so we need the government to run it. His argument is that the benefit to society of Google Reader is greater than the cost to society of Google Reader. Outside of a free market, it is better for society to have Google Reader. Unfortunately, there is no price structure that allows Google Reader to exist as a for-profit service. The market is inefficient, and the government should step in to fix this inefficiency. Therefore, it might be useful to run some services like Google Reader as government services.

I think that Google Reader should not be government-run, but he raises a larger point. There could be some internet services that would be better run by the government, and we don't really have any today. I don't think we can dismiss that possibility. I can't think of any sites like that right now, though.

Right, that's what he's saying. He believes that a certain good makes society better off. And yet, the pricing mechanism, the fundamental underpinning of capitalism, shows otherwise. (I'm speaking obliquely about Reader, but more broadly about nearly everything technocrats love that people aren't willing to pay the full cost of.)

You can argue that this reasoning is flawed (people aren't rational!), but the results are hard to argue with. The market does a fabulous job of allocating resources. Individuals, no matter how smart, don't.

Please read up on public goods, natural monopolies and externalities before you claim that the capitalistic market outcome always reflects what a good is worth to society.

Despite its problems, it is the best system we have. What is a better, asking krugman?

> people aren't rational!

As an aside, I'll point out that this is essentially meaningless as a critique of free-market economics, since people's actual behaviors are the only viable external indicators of their underlying preferences that we have, so the only reasonable conclusion to draw from the observation of allegedly "irrational" behavior is that the hierarchy of values motivating that behavior isn't what we expected it to be.

In other words, since all human behavior is a product of underlying motivations, the behavior itself can't contradict the actual existing motivations, but can only contradict others' erroneous assumptions about what those motivations were; all economic activity is therefore rational by definition.

With this in mind, the whole debate regarding the rationality of economic actors becomes nonsensical; the critique is an attempt to falsify a tautology.

> Ryan Avent, who argues that Google has been providing crucial public infrastructure...

Krugman proceeds assuming that were true, but it's not. There are plenty of free alternatives to Google Reader. All this proves that old saying; never pick a fight with someone who buys ink by the barrel.

Exactly. None of Google's free products amount to anything close to "crucial public infrastructure".

Google Apps could be critical for a lot of businesses, but they're charging for that now, and presumably making money.

Now, Internet ISPs, on the other hand, might fall under the category of "crucial public infrastructure" and could probably use some government oversight.

I found this from the FCC and it's very interesting: http://www.broadband.gov/plan/

In the 1860s, we connected Americans to a transcontinental railroad that brought cattle from Cheyenne to the stockyards of Chicago. In the 1930s, we connected Americans to an electric grid that improved agriculture and brought industry to the Smoky Mountains of Tennessee and the Great Plains of Nebraska. In the 1950s, we connected Americans to an interstate highway system that fueled jobs on the line in Detroit and in the warehouse in L.A.

Infrastructure networks unite us as a country, bringing together parents and children, buyers and sellers, and citizens and government in ways once unimaginable. Ubiquitous access to infrastructure networks has continually driven American innovation, progress, prosperity and global leadership.

In communications, the government stimulated the construction of radio and television facilities across the country by offering huge tracts of the public’s airwaves free of charge. It did the same with telephony through a Universal Service Fund, fulfilling the vision of the Communications Act of 1934 'to make available, so far as possible, to all the people of the United States, a rapid, efficient, Nation-wide, and world-wide wire and radio communication service with adequate facilities at reasonable charges.' ... Today, high-speed Internet is transforming the landscape of America more rapidly and more pervasively than earlier infrastructure networks. Like railroads and highways, broadband accelerates the velocity of commerce, reducing the costs of distance. Like electricity, it creates a platform for America’s creativity to lead in developing better ways to solve old problems. Like telephony and broadcasting, it expands our ability to communicate, inform and entertain.

Broadband is the great infrastructure challenge of the early 21st century.

Krugman is such a faux-expert, that is way overhyped. He clearly doesn't understand the technology very well, and he seems to be confusing RSS, the protocol, with Google Reader the app. And even then, RSS is not a "critical public infrastructure" that the government should maintain or whatever. It seems like he's just fishing for (more) fame/hits here, as there was no reason for him to write about this.

He writes "Google's decision to shut down Google Reader has upset a number of people I know" so I guess he felt it necessary to weigh in.

He's commenting as an expert in Economics, not internet protocols. In this case, he's looking at the economic effects of a service that is valuable to it's users, and that appears to be both slightly unprofitable and a nautral monopoly. He's not coming to any definite conclusions, he's exploring a set of ideas; not drafting policy. Your comment adds nothing to the discussion; so what are you getting out of making it?

> He's commenting as an expert in Economics, not internet protocols.

Unfortunately, he isn't either. He isn't an expert in internet protocols because he doesn't have a sufficient understanding of how the internet works. He isn't an expert in economics because there's no such thing as an expert in economics.

I mean this seriously: [macro]economics isn't an empirical science; no experimentation, no experience, therefore no expertise. Mastering untestable theoretical doctrine doesn't count as substantive expertise.

Not to get too deeply into the weeds of philosophy of science; but macroeconomics very is at least as much of a science as climate science. Observations are made, hypotheses are tested against reality, and predictions are derived that fail or succeed.

That said it also suffers from a similar pathology where academic investigation is systematically distorted, occasionally suppressed outright and researchers in the field are sometimes attacked ( literally ) for the views that are ascribed to them.

> hypotheses are tested against reality

How?

Retrodiction. http://en.wikipedia.org/wiki/Retrodiction

Avent's article is about Google in its entirety, not about Reader alone.

A relevant point of concern: Google Drive isn't actually downloading documents.

Maybe someone's already brought this up, but I just tried Google Drive's desktop application for the first time. It downloads what amounts to a shortcut that's a text document with a link to the file on Google Docs.

The benefit of syncing is lost--you don't have a copy of a usable file you can open in another program or save somewhere. Even Microsoft's SkyDrive downloads real documents made in the web version of Office.

The files on my Google drive are actual files, which I proved to myself by disconnecting from the internet and viewing them. Maybe there is a problem with a specific Google app?

I assumed that was for using Google's web apps offline. Did you verify the files on your hard drive contain something more than a link?

Yes, I have contents saved locally and in the cloud. But I'm not using Google Docs, just a my own files. Edit: yes, I tried a spread sheet and won't work offline since there is no offline spread sheet app. There may be a way to make it work offline as a packaged app.

I'm talking about files made in Google Docs: https://news.ycombinator.com/item?id=5428762

I can't figure out what you mean about the benefits of syncing being lost, because the data is clearly synced. However, I can explain the "shortcuts" you're referring to. Those are for native Google Docs, which don't have a universal representation in other applications (they're tied to remote or locally stored web state) unless you explicitly chose to export to another format. Whereas Microsoft's SkyDrive is just raw MS Office Documents.

The tradeoff between the two is that with Docs you get real-time editing and collaboration, but if you want a real local copy you'll need to target a specific export format. Whereas SkyDrive always keeps a local copy, but it requires you to explicitly save and resolve conflicts if you want to collaborate with another user.

Of course, if you want to you can just use Google Drive as a dumb file syncing mechanism without real-time collaboration or online editing. You just don't use Google Docs and tell Drive not to convert files to Docs on upload.

I opened a .odt file in Docs, but Docs cloned it into a new document and created a [filename].gdoc file in the local Drive folder. I don't see a way to work with the original file. It's nice that I can choose formats when exporting, and even export in bulk, but I want to decide which formats I work in. And it relies on this feature always being there.

It won't even let me just open a .odt. It opens in some kind of previewer when I click the file. Opening it takes clicking in a menu to select "Open in Docs." Then it clones to that magic format that's ethereal until exported.

Unfortunately, this is the same pain every application has with managing external and native formats. In the case of Docs, the state for online editing and collaboration is deeply integrated with the browser, and doesn't have an analog in external formats. So, those features just aren't supported if you're not using native Google Docs, because the underlying support doesn't exist in other formats. And if you are using native Google docs, then the only way to get the document in another format is to export it.

Accepting all that, I still don't understand the claim that the document isn't synced.

A .g[doc,sheet,...] file isn't a document. I want to open, edit, and save in .odt within Google Docs. There's no reason they can't make this possible. It was possible before Drive.

I get that the collaboration features need the native format. I don't need collaboration features most of the time.

> A .g[doc,sheet,...] file isn't a document. I want to open, edit, and save in .odt within Google Docs.

so, open it, edit, and then download as -> .odt and name it the same as the original file?

I don't really understand. You want its internal representation to be in odt? What does it matter as long as it's not observable except by an icon and a filename extension?

You don't understand why I don't want to depend on there always being an export option? I don't understand what you don't understand. I don't want to have to export to .odt every time I make an edit just to maintain portability.

I want Docs to work like its desktop predecessors work. Even Office's web applications do this. There's no reason Docs can't. I can make an edit in Word and instantly open to see the changes in LibreOffice, or run some kind of cross-cloud sync.

Maybe we're talking about different things? Here's the workflow I envision:

1: Create an outline for a book in Docs

2: Open it in LibreOffice from the GDrive folder

I can do this in 2/3 word processors.

This is what I keep trying to explain to you. Google Docs do have their own native format, it's absolutely not ODT, and if you want to edit files online they must be in the native format. It's as simple as that.

No, I get that. As I said, several times. I get that Google does that. But I don't understand why Google makes it only work that way. I need to be able to have it work the way all other office suites work because I want to be able to open it in them without exporting.

It can be an option off in the periphery. The interface can politely warn me with a little pop-out at the edge that I can't use collaboration features if I enable the "works like all other office suites" function for a document. Call it compatibility mode. Word's web application does this if I edit an ODT in it. LibreOffice does this if I edit a .docx in it. I don't mind.

edit: I just noticed your bio. Maybe you can have someone from the Docs team come over and look at this? They might understand what I'm saying.

I'll try to explain one last time. Docs is built from the ground up to support real-time collaborative editing, because that's how the Web works. I don't speak for anyone involved with Docs, but I wouldn't expect it to be rearchitected to support the use case you're asking for because it's a major change that introduces tremendous complexity such as: manual saves, document locking, whole document collisions, etc.

So, to sum all this up, "Docs fundamentally can't work the way you need it to work."

I'll stick to LibreOffice and Word.

I don't think it's fair to call Google "evil", but maybe I'm cynical because I spend a lot of time reading about banks, insurance companies, the mainstream media, walmart, lobbyists, drug companies, private prisons, garment factories, weapons manufacturers, gun dealers, patent trolls, the recording industry, and so on.

You're misinterpreting the title. Put "Evil Google" in scare quotes and it makes more sense.

It's remarkable how little economics has to say about Google, their motivations, and outcomes. There's nothing wrong with trying to fit new information into existing models, but think of it this way: if you were a fly on the wall for Goog's decision on Reader, would it bear any similarity to this model?

What is the alternative explanation? Reader evidently needed a very non-trivial amount of resources to maintain and didn't have anywhere near the amount of users to justify it. I lament that they couldn't keep it going, but even the guy who started the project inside Google said he and people working on it felt like its days were numbered from the outset.

After reading the article several times it seems to me that it has nothing to do with Google. The idea that a company that can't price discriminate and isn't able to make a profit to provide a useful good is interesting. Words like 'Evil Google' and 'Monopolist' are just added fluff for link bait.

From TFA:

"even in a plain-vanilla market, a monopolist with high fixed costs and limited ability to price-discriminate may not be able to make a profit supplying a good even when the potential consumer gains from that good exceed the costs of production"

The whole blog post is an analysis explaining why that's true. "Evil Google" may have set people off (but it's pretty clear from TFA that Krugman's not calling Google "Evil"), but "monopolist" is pretty central to the article.

I am not an economist but I don't see why being a monopoly or not has anything to do with it. Would a not-monopoly be able to make a profit if it had high fixed costs and no ability to price discriminate? That's without even delving into the question on if Google is actually a monopoly.

Avent's article (which Krugman's is discussing) discusses Google in the context of possibly being a monopoly. Krugman's presenting an analysis that shows that even a monopolist may not be able to make money in some circumstances where there's positive social value to a service being provided. One would naively expect that a monopolist would be able to make a profit and stay in business in those circumstances even if perfectly competitive firms couldn't, and they often can which is why we have regulated monopolies, but Krugman's showing that isn't always true.

If you want, you can take his post as a counterargument to the "why wouldn't Google charge for Reader" question: it's possible that they could not have charged a price that made it profitable. Without any data, who knows if it's true or not, he's just showing how the logic would fit together.

That was more or less my point. Krugman provides no evidence that the Google Reader follows that model. It is an opinion article entitled 'The Economics of Evil Google.' After the first paragraph there is no reference to Google only 'Monopolist' Compare that to the longer Economist article which is actually all about Google and somehow fails to include either the words 'Monopolist' or 'Evil'. I guess I just don't like opinion pieces.

What was more or less your point? It's a discussion of monopolies.

The article he's talking about is still worth reading, but this is a good summary if you don't have the time/interest to read the whole thing.

Here it is, and it's quite good:

http://www.economist.com/blogs/freeexchange/2013/03/utilitie...

"in the long run that's a problem for Google. Because we tend not to entrust this sort of critical public infrastructure to the private sector. Network externalities are all fine and good to ignore so long as they mainly apply to the sharing of news and pics from a weekend trip with college friends. Once they concern large swathes of economic output and the cognitive activity of millions of people, it is difficult to keep the government out. Maybe that deterrent will be sufficient to keep Google providing its most heavily used products. But maybe not.

I find myself thinking again of the brave new world of the industrial city, when new patterns of interaction led to enormous changes in economic activity, in culture and personal behaviour, and in the way we think. We upgraded ourselves, in terms of education, hygiene, and social norms, to maximise the return to urban life. And the history of modern urbanisation is littered with examples of privately provided goods and services that became the domain of the government once everyone realised that this new life and new us couldn't work without them"

See, e.g. the NYC subway system.

The "Dark Knight" quote of "You either die a hero or you live long enough to see yourself become the villain" seems apt here to be honest.

I'm no fan of Google. But what could be a thousand times worse than Google?

Oh I know. Search as a public utility.

That depends ... BBC get the job pretty well as a publicly financed utility. So a search that is focused on quality and not ads supported by levy on broadband with transparency in ranking and inclusion, with prohibition of censorship ... may not be such a bad thing after all. Search is important.

Although I would just create an IP law that forces the corporation to put the source code and assets in public domain on the abandonment of a service when it gathers above some usage and traction. Someone will pick it up if the interest is enough.

P.S. I am ready for the downvoting that will ensure.

Here's how Google should shut down a product:

Step 1: Over a period of say three months, slowly increase the average response time of the product.

Step 2: Listen to people complain that load time is getting unbearably slow

Step 3: "Oh, we'd love to fix this problem but because of $random_infrastructure_issue, and because we've got most of our developers working on Android, we think it's time to shut the product down. After all, it was getting old and slow, right?"

Step 4: Feel justified about their decision

Revoking Paul Krugman's Nobel Prize would be a public good.

Maybe this is a dumb question, but why wouldn't Google simply sell Reader to some consortium, public or private group to continue running it? In terms of money, shutting Reader down would get them less money than selling it, as long as the price is greater than the cost of transferring the infrastructure. However, maybe it will incur liabilities for Google?

http://www.change.org/petitions/google-please-sell-us-google...

If I were to hazard a guess, it is that that Reader is too closely tied to other Google technologies (caching, crawling, search) for it to really function outside of Google.

"Nobody relies on [mail delivery, a national highway system, telephones, wireless spectrum, police, libraries, etc] to get to work so they can feed their children."

Necessity for sustenance or transportation does not drive the need for something to be a public utility. I think we are really talking about the tragedy of the commons. This has been a political and legal debate for centuries. The private market is not able to maximize the average utility of citizens in every market and AND earn profit.

Couldn't read this article initially because it's hidden behind an Evil Paywall.

Solution: Evil Google. Googling the title in news and clicking that gets you in without paywall limits.

so now Google Reader is like the banks/aig/car makers? too big to fail?

Krugman lost it a long time ago. The final nail in the coffin was when he wrote that it actually was a good idea to mint a $1 trillion coin to dodge the legislation preventing to raise the U.S. debt ceiling.

If you're into economy at all you should know that disciples of Friedman did predict the precise situation the eurozone is in right now (including Spain unemployment, the Greek default, Italy and France struggling) before the first euro even circulated... Meanwhile Keynesians have never been able to predict anything more than two or three years before they happened.

Keynesians are ultra-short sighted but politicians loves them because they're basically telling them what they want to hear: more state intervention, create even more money (the $1 trillion coin was just ultra rubbish, really), etc.

His whole book "End this recession now" is a gigantic pile of shit.

The situation ain't good because of the Keynesian doctrine and yet these people want to do more of what does obviously not work: let's print more money.

I never had much sympathy for Keynesians but honestly since he wrote about that $1 trillion coin I've entirely lost interest.

He lost it.

Advocating for a $1 trillion coin is a recognition that, if interest rates are effectively 0%, there is no difference between short-term government debt and government-issued currency. Minting a coin is just a hack for the Executive to get around Congress's restriction on issuing debt.

I'm not sure what you're trying to get at with your Euro rant, bu Krugman has been calling the Euro a bad idea since at least 1998. http://www.pkarchive.org/global/tag.html

"Google... a monopolist with high fixed costs and limited ability to price-discriminate"

Google is not a monopoly and neither was Google Reader. Krugman is intentionally disingenuous here to further his argument.

Krugman knows Google isn't a monopoly. He doesn't care. He's always trying to push the same narrative with his blog: Big corp is evil so government should step in and save the day! For anyone familiar with his blog, his agenda is extremely transparent at this point.

Thank you for identifying that Krugman is both an economist and an ideologue. HN readers identify closely with the former so they often ignore or excuse the latter.

I think Google search is close to a monopoly, and so was Google Reader. Search is pretty much synonymous with Google. Apparently Google has 67% market share, but I'd guess that most people not using Google just use Bing or Yahoo! because that's the default in whatever browser they have. http://searchenginewatch.com/article/2244472/Google-Once-Aga...

A "monopoly" consists of 100% market share, not 67%.

Where does Krugman state that Big corp is evil? Are you talking about banks?

Krugman doesn't state that Google is a monopoly and he certainly isn't disingenuous here. He's just using the Google Reader situation and an article by Ryan Avent as a hook to make an interesting economic argument about monopoly markets with a certain cost and consumer demand structure.

BTW, building a competitive search engine is arguably associated with relatively high fixed costs, and any social site is hugely affected by network effects. These are both traits that can lead to natural monopolies.

Where does Krugman state that Big corp is evil?

The title of the post is, "The Economics of Evil Google." It's not written as a question, but a statement. Nowhere in the article does he give any indication that it is sarcastic.

Krugman doesn't state that Google is a monopoly and he certainly isn't disingenuous here.

He doesn't state it, he implies it [1]. Otherwise, what does being a monopoly have to do with Google? High fixed cost and limited ability to price discriminate are not characteristics exclusive to monopolies. I don't see the reason to refer to "a monopolist" unless he is implying that Google is a monopoly.

When I read the article, I concluded:

* Krugman believes Google had a monopoly on RSS. [1]

* Krugman believes that RSS is a crucial public infrastructure [2]

* Krugman thinks the solution is a government-provided RSS reader. [3]

Just curious, what did conclude from this article that is different?

[1] "First, it’s a well-understood though not often mentioned point that even in a plain-vanilla market, a monopolist with high fixed costs and limited ability to price-discriminate may not be able to make a profit supplying a good even when the potential consumer gains from that good exceed the costs of production."

[2] "It seems hard at this point to envision search and related functions as public utilities, but that’s arguably where the logic will eventually lead us."

[3] "So what’s the answer? As Avent says, historical examples with these characteristics — like urban transport networks — have been resolved through public provision."

I respect Krugman and agree with him often but this is obviously dumb. It's obvious to anything who knows anything about how the web works. Krugman should have consulted with such a person before posting this boneheaded entry.

Reply on news.ycombinator.com