Pick your headline: Google steals content… Google declares war on publishers… Google violates copyright…. Google is evil…. None of the above… All of the above….

Pick your headline: Google steals content… Google declares war on publishers… Google violates copyright…. Google is evil…. None of the above… All of the above….

: Correct me if I’m wrong, but Google is repeating one of the great sins of AOL, promising to “accelerate” the web by caching — that is, copying and storing — web pages to serve to you.

It’s one matter when the search engine caches a page you can’t get anymore; that’s a copyright violation but an all-in-all benign one in the sense that it’s only giving you content you could not otherwise see (no different from, say, the web archive).

But it’s quite another matter for Google to get in the way of serving current content. This means that the page is served from Google rather than from a publisher’s server, which means that the publisher cannot count the traffic and serve targeted and dynamic advertising.

It also means that Google is copying content on its servers and serving it from there and thus is violating copyright.

And it means that Google is in a position to snoop on data on consumers’ usage of sites that Google does not own: That is, Google will know what the consumers on my site are doing better than I will for these “accelerated” pages.

I’m also reading that Google shortstops cookies, which affects not only publishers but also users. Cory Doctorow points out: “…but of course you’re not anonymous to Google, which knows about your search history (if you’ve got cookies on), your email address (if you register for Groups, etc), your friends (if you use Orkut), your email (if you use gmail), and even has your credit card (if you use AdWords or Answers).”

And all this is done in the name of accelerating the web in an era when most of us have broadband and it doesn’t need much acceleration.

Again, if I’m getting that wrong, please tell me. But I’ll just bet we’ll see publishers telling their lawyers go to fetch.

Here’s PaidContent’s writeup; here’s Steve Yelvington’s writeup; here’s John Battelle’s.

Search Engine Journal reports that the accelerator is screwing up, serving the wrong pages to the wrong people and even signing people into forums under the wrong names.

Fantomas gives instructions on how to block the Google accelerator.

: See earlier cautionary notes on Google here and here.

Bottom line: I see no reason why I should be expected to trust these guys anymore.

33 responses to “Pick your headline: Google steals content… Google declares war on publishers… Google violates copyright…. Google is evil…. None of the above… All of the above….”

  1. tony Avatar

    but in one day of browsing theyve saved me 5.6 minutes!

  2. Narc Avatar

    Reasons like this are why copyright law needs to be overhauled. Not that it will be, at least not to any benefit for the consumer. Whether or not this is a significant copyright infringment is debatable (not that I’m a lawyer), and probably will have to be settled in court. I figure that any content producer that objects to this is welcome to refuse to serve thier content to Google referrers.
    Some of the major ISPs do this sort of thing. AOL, as you pointed out, Earthlink, and probably others. I see no big reason why Google’s transgression — if there is one — is significantly worse than theirs. If a user is concerned with the potential privacy implications, then he is welcome to not use such an accelerator.
    By reading your blog, Jeff, I count no fewer than three copies of your copyrighted work that I’ve had to make. I don’t even use an accelerator. And yet, I have no license from you to do so. I could argue that there’s an implied one based on the fact that your blog has been made public on a publically-accessable website, but I’m not sure that would hold up in court.
    Furthermore, you’re replicating my copyrighted work (this comment) by allowing it to be copied by many people that view your blog.
    Your claim that “most of us use broadband” is also incorrect. Roughly half of Americans with Internet access at home have broadband. Roughly half don’t. I suspect most of those that don’t are lower-income, simply because DSL and cable are more expensive. So things like this are valuable, and work in the public interest.
    Frankly, I’m of the opinion that this is a tempest in a teapot.

  3. Jeremy Tunnell Avatar
    Jeremy Tunnell

    Jeff,
    (IANAL)
    I can’t remember what case it was, but simple caching has been determined to NOT be a violation of copyright. I believe it is called an intermediary exemption.

  4. Skate Avatar

    All web pages are cached at some point. In order to view a web page, it has to be copied. Putting up a web page includes de-facto permission to cache it–that is how the internet works. If people don’t have permission to copy your web page at all then they don’t have permission to view it at all.
    AOL does massive amounts of cashing, that is also part of what a proxy server does. I really don’t have a lot of sympathy for those who would complain about Google’s web accelerator based on copyright paranoia.
    More important to me is that Google is collecting reams of data about users and their is no way for users to control how that massive data file is used. Every search you have ever performed on Google is logged and now they can actually watch your web browsing directly. Scary. I don’t care if they do say “Don’t be evil.” That is just a saying, not a contractual obligation they have.

  5. Jeff Jarvis Avatar

    Skate: It prevents publishers from measuring traffic effectively — which matters to revenue — and from serving dynamic and targeted ads — which also matters to revenue. Yes, AOL does it too, and publishers fought against AOL… and AOL doesn’t matter much now.
    As for the contractual obligation: It is a brand obiligation. And when they violate that, they violate the relationship with their public.

  6. Eduardo Avatar

    Google says they don’t prefetch ads. So I guess they get them as in a normal page request. They should be aware of the potential problems here, because they would be affected as well (think about AdSense).

  7. Eduardo Avatar

    On the other hand, I would hardly need any Web accelerator, not because of broadband but because I do most of my “browsing” through RSS feeds. (I use Bloglines, where another form of “cacheing” content is arguably taking place.)

  8. Derik Avatar

    Good point Eduardo. I also use Bloglines pretty extensively and that’s the first thought that popped into my head. It retrieves the RSS feed once for many people (in the case of this blog 1,424 people at last count).
    What of it Jeff? Are you going to start howling at them too (I think they’re owned by Ask Jeeves)?

  9. billg Avatar

    I’m always amazed when anyone trusts an intermediary like Google in the first place. The only thing standing in the way of any such site’s altering/hiding/defacing content is the intent of the people running it.
    There’s never been any more to trust or mot trust Google than there is to trust or not trust a blogger who splashes an unlinked quote from somewher else.

  10. Michael Zimmer Avatar

    Bloglines and RSS is slightly different that Google’s act of caching. I choose to allow RSS feeds on my site, enabling Bloglines to come and grab the posts. Google just does it. (I could use the robots.txt protocol to block their crawlers, but few people know of that option)
    But this issue is more than just one of copyright. As I noted here, Google’s use of cached pages is tantamount to redefining the very nature of the Web: instead of it being a distributed network of independent computers, the Web now becomes merely pages of content housed on Google’s own private servers.

  11. Derik Avatar

    “I choose to allow RSS feeds on my site, enabling Bloglines to come and grab the posts.”
    You choose to allow HTML pages on your site, enabling Google to come and grab the pages.
    Explain the difference please.

  12. Undertoad Avatar

    A technology is bad because it doesn’t serve the ADVERTISER correctly?
    I’m not sure I’m reading the right blog this morning. I thought the Internet was going to return control to me, the reader/consumer…

  13. Michael Zimmer Avatar

    Derik – The difference is that by providing RSS feeds I consent to the sharing of my content. It is through my action that my content is to be scraped and delivered by a 3rd party. With Google’s cache, they come and take it without asking and without my express permission.
    The purpose of RSS is expressly to share my content. The purpose of my HTML code is so browsers can present my page to a viewer. Just becuase HTML can be grabbed doesn’t mean that’s what its there for.

  14. Derik Avatar

    Still not seeing a difference. A format change is not an explicit agreement to all reproduction (see the AP’s RSS feeds, which are obtainable in Bloglines). An RSS feed is not automatic consent.
    Don’t fall into the trap that because RSS stands for “Really Simple Syndication” that is necessarily implies that the data contained therein is expressly for syndication. It’s a standardized format, nothing more, nothing less.
    And what about corporate/organization proxies? A good portion of the Internet using public gets their content served through large-scale proxies (which is EXACTLY what Google is doing) without even knowing it. That’s been true for just about as long as the Internet has been around. Take a look at any Networking textbook. It’ll tell you how to do it in almost excruciating detail.

  15. Armando Avatar
    Armando

    While I have not looked into the details, it seems to me that you are correct in your analysis.

  16. Skate Avatar

    Jeff wrote:
    “It prevents publishers from measuring traffic effectively”
    I can see your point on this issue in terms of stats. I also see an issue when pages are served up improperly because of mistakes by Google’s cache system. My only point was that caching is not an issue in terms of copyright, or shouldn’t be.
    Of course, I’m betting Google’s accelerator has no ill effects on Google ads ;-)

  17. Skate Avatar

    BTW,
    AOL’s caching was rather insidious in at least one way: They actually created derivative works by re-compressing graphics in their own proprietary and lower quality format. I assume they have finally stopped this practice which made my graphics look awful to AOL users, but I don’t know for certain.

  18. Michael Zimmer Avatar

    Derik –
    1) I don’t mean to suggest than any use of RSS is “automatic consent.” Just that my decision to add a “subscribe to this feed” link on my site is my own act of consenting to distribution of the content. (Along with my Creative Commons license, of course).
    2)This is more than just a proxy server, right? Not only are your browers requests going through Google’s proxy servers, but instead of Google routing those requests to the Internet itself, they are using their cached versions of pages. While the use of a proxy has its own issues in terms of Google being able to track my activities, the issue here is about cached, copywritten content.

  19. billg Avatar

    Undertoad:
    What does advertising have to do with your being in control of anything? For that matter, why should you be in control of something someone else is doing?
    Not be smarmy. We see that claim a lot, and I’ve never understood it.

  20. rivlax Avatar
    rivlax

    I can feel the backlash to Google coming. Just last night my wife, while blogging, remarked that if it’s proven that Google is lefty biased then she’ll never use it again. Plus, she said. The two guys who started it are Montessori products and she hates Montessori and everything that comes with it.

  21. Tony Gentile Avatar

    I drew a picture to express my sentiments. ;-)
    http://www.buzzhit.com/2005/05/google-meets-fockers.html

  22. Anonymous Avatar
    Anonymous

    Why are most bloggers hysterical paranoids?

  23. Bill Quick Avatar

    Why are most bloggers hysterical paranoids?

    Why are most drive-by trolls anonymous cowards riding on the back of somebody else’s time, talent, and money?

  24. CHARLES BEX Avatar

    Is it a surprise that google would pull such an act, to usurp the internet?

  25. Reid Avatar

    Even if we want to put aside the ethics of this, it’s a totally boneheaded move.
    Imagine you are running Web Accelerator, and you log in to your favorite blogging or webmail application. You’re now in a place that is not normally spidered, or cached by a third party, because it’s behind an authentication scheme. And yet, Google’s little app begins pre-fetching links. Like the ones that say “Delete.”
    Web Accelerator also ignore Javascript pop-ups, like the one that says “Are you really really sure you want to delete this entire blog, because it can’t be undone.”
    It’s already happening. Others in forums like SomethingAwful.com report that with GWA turned on, each new page the view in the forum shows them logged in as some other user … with access to that user’s account and details.
    Lots of fun ahead here, to be sure.

  26. Matt Avatar

    Jeff,
    Why not ask some questions and do some research before running around in a hyperbole-fueled panic?
    First you say that it will affect dynamic ads. Is this true? Let’s see what Google has to say about that.
    From Google’s site describing the Web Accelerator:

    6. Will Google Web Accelerator affect traffic for ads on my site?
    No. We do not prefetch ads.


    You can even still get statistics:

    5. Will Google Web Accelerator affect my server load or usage statistics?
    It depends on whether your pages are cacheable. You can identify page requests prefetched by Google Web Accelerator through the HTTP header X-moz: prefetch. You can learn more about this header on the Mozilla website.
    It might end up that Google isn’t really being truthful about these things, or it might end up that they are being honest. The bigger point is that you and others commenting on this are speaking in rather hyperbolic terms about something which you do not have a technical background to be discussing. I’d excuse that, but your journalism skills are failing you: why not write an email to Google:
    7. What if I have a question that you haven’t answered here?
    We’d love to hear from you. Contact us at labs+webaccelerator@google.com.

    It took me exactly 5 minutes to look this stuff up. It seems like you’d rather shoot accusations first and ask questions later. If none of this turns out to be untrustworthy behaviour on Google’s part, you’re going to look rather foolish.
    Just take the 5 minutes to look things up and send emails to Google asking questions. Then I’ll take what you write seriously.

  27. Matt Avatar

    Jeff,
    None of the people you link to in any way support your hyperbolic Google Is Evil headlines.
    Lest it seems like I’m defending Google: I’m not. They seem to have botched the roll-out of this (people seeing other people’s pages, erroneously deleted pages, etc.). That’s really no different than the roll-out of any other application — there will always be bugs. It’s just that this application hooks itself deeply into the web browsing experience, so the bugs will hook themselves deeply into your web browsing experience too.
    Also remember: Google isn’t pushing the accelerator towards normal users. It’s on Google Labs, which is their preview/beta-test site. On that page (I looked it up!) it says that the web accelerator is one “of our favorite ideas that aren’t quite ready for prime time.” That’s clear indication to me that there might be bugs!
    But let’s not say Google Is Evil And The Internet Is Dying just because of a few bugs. If they fix them and people find the web accelerator useful, great. If they don’t fix them, or if people don’t find it useful, the accelerator will fade away. (which is also mentioned on the Google Labs FAQ: “What will happen to these experiments over time? That depends in part on you. Your comments and the feedback from other users could elevate a project to consideration for integration into Google.com. Other demos may disappear and never be spoken of again.”
    None of this is cause to get worked up about and start saying Sue! Sue! Sue! Snoop! Evil!
    Look, I’m glad you’re commenting on things, and it makes me happy to see you fostering discussion of some really important issues. But is it too much to ask to just do a little bit of research and fact-finding first? To just make a fire-and-brimstone-style post without having done so makes you seem arrogant and beyond the need to be informed and circumspect.

  28. Jeff Jarvis Avatar

    Matt: Thanks. This is why I asked people to correct me. I cannot find the page you reference; your link is broken and so is the one on the site. So your quotes are helpful. I’m not saying the precache ads but if they do get in the way of cookies, then our ability as publishers to serve targeted, dynamic ads under rules about frequency and such are possibly impaired. Also, counting a precache of a page is not the same as counting a read of a page and advertisers will probably not allow precaches to be counted unless they know, under iab and abc rules, that a page was specificaly requested and displayed. We went through this with AOL and I see it happening again with Google. As for whether it’s just for geeks… if it works from their busines stance, then it won’t be, and now is the time to raise concerns, which I have done.

  29. Matt Avatar

    Jeff,
    Thanks for the reply. Here are the URLs in non-link form. They’re working here for me. [Perhaps you’re still suffering from yesterday’s Google DNS hack? Maybe more reason why Google (or anyone) caching everyone’s web surfing might be a bad idea? ;)]
    Accelerator Webmaster FAQ:
    http://webaccelerator.google.com/webmasterhelp.html
    Google Labs:
    http://labs.google.com/
    Google Labs FAQ:
    http://labs.google.com/faq.html
    I agree that now is the time to raise concerns — which is why I’m hoping that maybe you’d email them and see if you can get answers to your questions about what happens with cookies and hit counts. From the FAQs, it seems like they’re attempting to address at least some of the problems, and I would hope they’d come up with a way to address the rest of them too.
    If they don’t address the concerns then that’s the time to start saying that there’s no reason to trust Google. [and also, that’s why I think it’s better to save the ‘we don’t trust you’ comments until later — if people get used to hearing them, they’ll dismiss them as paranoia.]

  30. Derik Avatar

    “1) I don’t mean to suggest than any use of RSS is “automatic consent.” Just that my decision to add a “subscribe to this feed” link on my site is my own act of consenting to distribution of the content. (Along with my Creative Commons license, of course).”
    “Subscribe to this feed” is now automatic consent to any and all reproduction? Not really.
    “2)This is more than just a proxy server, right? Not only are your browers requests going through Google’s proxy servers, but instead of Google routing those requests to the Internet itself, they are using their cached versions of pages. While the use of a proxy has its own issues in terms of Google being able to track my activities, the issue here is about cached, copywritten content.”
    THAT’S EXACTLY HOW A PROXY SERVER WORKS. IT CACHES. PICK UP A BOOK.

  31. socioeconomics Avatar

    I am on Google’s side on this.
    I believe the fact that it has those pages on its cache benefits my sites.
    A lot of users first find that page and then follow to the home page of the site.
    Webmasters reserve the right to use their robots.txt files accordingly if they don’t want google to cache some of their pages in the first place.
    If this post is about just another way of squeezing money from google, that’s something else of course.

  32. Chris Avatar

    Hey Jeff,
    I’ve been doing a lot of thinking since this post (which is why it has taken me days to reply), and I think maybe there is some middle ground where google users could use the accelerator, and content publishers could regain some control.
    What if the web accelerator worked like most blog pinging services? If you own a website and would like to participate in google’s accelerator program, you sign up and then whenever your content is updated, you ping google and google goes back out and gets a fresh copy of the site.
    That way everyone all around opts in to the program.