I’m a bit confused by all the kerfuffle around the Google toolbar rewriting links on people’s sites. I mean, for the most part it looks like it’s an opt-in thing and I don’t see that this is a particular problem. Cory posted his vehement defence of the proposal:
Plenty of Cowbell asks whether I like the sight of an ISBN corresponding to one of my books being rewritten. My answer: Hell ya! This shows how an authors’ association like the Science Fiction Writers of America could collect its members’ ISBNs and affiliate IDs for their favorite web-stores and provide plugins that would rewrite every single instance of my ISBNs on pages viewed through the plugin with a link to my affiliate account on Amazon, making me some serious coin. Wanna support an author? Install her plugin and help her feed her kids. Wanna support a charity? Install its plugin and have all the affiliate links rewritten to its benefit. Wanna support youself? Install the plugin that rewrites every ISBN with your own affiliate ID.
So here’s the bit that worries me. We’re talking about plugins at the moment, right? Places where there is no reason why anyone should feel forced to use the device concerned. Now move back to the browser market and lets posit a world where browser market share for Internet Explorer has fallen back to some semi-reasonable level – let’s say 40%. Now in Cory’s model, I can see no reason why Microsoft shouldn’t decide one day to replace every single Google Adword (or similar advertising structure) with its own advertising on each and every web page that you visit. I mean – this could be the way that you finance a new browser, you remove advertising from the page and replace it with advertising that makes you money instead of the content creator. You could do the same with RSS readers.
Now given that there are other browsers in the market, there’s still choice in this picture – you could switch to any of them and not have to experience the world the Microsoft way (or the Google way if they created a browser). Unfortunately, given that it would make apparently no difference to the actual content or design of the page in any way, then I’d doubt that many normal members of the public would really care that much – that is until the sites that they liked to visit started to shut down one-by-one. And that’s before the other browser manufacturers realise they cannot compete with the concept someone milking advertising revenue from every single site on the internet and decide to follow them down the path to mutually assured destruction.
Genuinely and honestly, I would like someone to explain to me why such a situation could not emerge as an evolution of the stuff that Cory’s talking about. Reassure me, if you will. Why won’t this happen??
10 replies on “On the potential for browsers to replace all local advertising…”
That is a scary thought. I don’t think it could happen, but my reasoning isn’t as convincing as I’d like.
The first thing to bear in mind is that sites reliant on advertising could simply refuse requests from the browser in question, so it would have to willfully misidentify itself. That probably rules out any remotely respectable company from even contemplating it.
I doubt that a new browser that did this would get off the ground, because new browsers are initially adopted primarily by geeks, who are going to be more aware of, and concerned by, the implications than the average member of the public. I think if even Firefox suddenly started doing it now, the bulk of its users are still sufficiently passionate about such things that they’d abandon it en masse (I’m ignoring the fact that it’s open source, which pretty much means it couldn’t happen anyway).
That limits potential culprits to mainstream browsers with a relatively computer illiterate userbase, which basically means IE. Microsoft can be pretty arrogant at times, but I don’t think they’d want to make quite that many enemies, especially with the supreme court watching over them for anticompetitive practices.
In fact the only people I can imagining contemplating such a thing are AOL. As their market share declines it seems like just the sort of desperate measure they might take, but by that point I shouldn’t think it would have much impact on anyone.
Hmm… none of those arguments are as persuasive as I hoped they would be 🙂
First of all, are you talking about something which MSIE does automatically, without the user ever requesting it, on each page the user visits? If so, then that works a very different way to Google AutoLink, so blaming Google for this would be deeply unfair. They’re no more responsible for such a feature than Tim Berners-Lee. Microsoft could have done this at any time – why should they start now? The answer to your question in this case, btw, is that it would be commercial and legal suicide.
This is one of the things deeply pissing me off about the arguments raging around – people are positioning Google as having suddenly broken some kind of taboo that makes all kinds of evil content-modification okay. They have done nothing of the sort.
But, okay, let’s take the feature the other way. Let’s suppose that MS come up with a feature that has to be manually activated in isolation, just like AutoLink, and it switches ads. As far as I’m concerned, that feature is a completely valid thing to build. Now what you have to do is persuade me why anyone would ever turn it on.
wait; first you postulate that IE’s market share has fallen to 40%. Meaning that (oh joyful day) that most of the public has woken up and started using another browser.
Then MS would do something draconic like this; geek sites will be up in arms, lots of noise, and more ‘regular’ people will switch away from IE. Even smaller market share.
Will the remaining people locked in with the ‘link rewriters’ then seriously affect the internet economy? (such as it is) I think the effect would be far less than you fear.
I think the answer to your concern is the main thing we’ve gained in this whole discussion regarding AutoLinking. The Google initiative, in contrast with the earlier Microsoft initiative (i.e., Smart Tags), is a transformation that appears for a single user at his or her request only, on a page-by-page basis. The earlier implementation was turned on in the browser and then from that point on transformed all pages as they were displayed. I feel that between those two points is a bright line that divides the acceptable from the unacceptable, and furthermore that it’s a difference that most users can understand.
There has to be a balance between the publisher, the user, and the maker of the medium through which published material is displayed. I think those that suggest that no modification is permissible *by the user* are making an inconsistent distinction between tools they like and tools they don’t. Is there any difference between adding links to a page within the viewport and asking a service to translate a page into a different language? Maybe, but is the difference logically consistent? I doubt that. To me, by long-standing tradition on the web, once a page is in a browser, users do have the right to perform all kinds of transformations. Just as when you buy a CD you have the right to rip the CD and perform transformations on the audio file (you just can’t necessarily distribute them).
For the browser maker to automatically and systematically perform transformations without specific user intervention is a completely different matter, though, and I don’t see how Google’s initiative has any bearing on whether we’re on a slippery slope towards that or not. Google is very specifically NOT doing that, so to suggest that AutoLink opens the door to that is akin to suggesting that F1 Racing should be banned because individuals might speed in their Honda on a residential street.
To Yoz – I’m not getting angry with Google but I’m not mischaracterising the debate either. As I understand it, the argument that people are using is that as long as people have a choice about whether to use the functionality concerned then it’s okay to run any kind of proxy like this. I’m suggesting that people have a choice at a higher level in terms of what browser they use, and that a good way of funding a browser might be to do this kind of thing for adverts.
What I’m specifically asking is whether there is any qualitative difference between the two actions rather than an issue of scale. Both a plug-in or a browser are things that you are not forced to use, there are alternatives that you could utilise, and one of them is draconian and damaging where the Google implementation is not. But that doesn’t answer the question of whether the concept is good or bad, or how we distinguish between implementations.
To Mikel: The part that I have trouble with is this idea that using a certain browser is in itself not a choice.
Let me change the frame of the debate from Microsoft and say that Omniweb developed a new browser that did really cool things that no other browser did and they were able to give it away for free because it changed the advertising on people’s sites. No imperative to use it. None to switch. User gets total choice as to which browsers they use and how they use em. Even if the option to turn this off is buried somewhere in the preferences, is this really something that we think is acceptable?
I think my issue with this debate is not that what Google have done is unacceptable, because it does not appear to be – nor that they have intent to do other stuff that is bad, but that there isn’t a clear dividing line between what kinds of things people could do that are acceptable or are unacceptable and what the logically consistent reasons might be for that.
Yoz nails it. If a company — any company — rewrites a web-page without telling me that it’s doing so, without letting me switch it off or switch away, then it is doing wrong. The wrong thing it is doing is tricking me and locking me in. The wrong thing ISN’T rewriting webpages.
So if MSIE did what you propose, we’d fight it. If MSIE said, “here’s a browser that lets you choose to do this,” I’d support it.
Tom: One of the things I specifically included in the set of rules I wrote for this topic (and I hate to bring this across as some kind of Ultimate Law, but it neatly defines where I stand on the issue) is that such content-modifying features should ideally be specifically enabled in isolation by the user. So, for your OmniWeb example, I’d say that if the ad-swapping feature was off by default and had to be specifically enabled by the user (who had also been fully made aware of its function) then it would be completely fine by my rules, no question.
But now let’s assume that OmniWeb has this feature enabled by default, all other things (full disclosure, ability to disable) being equal. By my standards, we’re already in dodgy territory and need to look on a case-by-case basis, because you don’t know whether the user has downloaded OmniWeb specifically for the ad-munging feature or one of the others.
I should specify that what I say is acceptable, by my rules, is acceptable on a kind of “free speech for software” basis. If someone makes a plugin that takes my site and scrawls a bunch of vicious hate across it, I say that they have every right to distribute it, but they’re probably a bastard in most other ways.
I don’t know if, ultimately, I can answer what you’re looking for here, though it’s an interesting question. I’ve drawn a line that describes what is always acceptable, and you’re looking for a line that does the opposite. I think such a line is much harder to describe and much more subjective.
simonG already mentioned this, but didn’t emphasize it enough, in my opinion. any site that depends on advertising to sustain itself will quickly refuse requests from browsers that remove/replace those ads. users will find they can’t access all sites, and switch browsers. end of problem. of course, the browser could conceal it’s identity, but that’s a technical issue that could be dealt with in ways similar to all the CSS rendering hacks used today.
There was (or perhaps still is) a Supreme Court case in the US somewhat related to this. If I remember rightly, it concerns a video rental shop in Utah, in which the manager edited out all the naughty bits and rented out the bowdlerised versions. The film studios complained that this was a copyright infringement by virtue of creating a derivative work for commercial purposes.
So, here’s the thing: is AutoLink a facilitator like the bloke making the cuts, or is it more like a copy of Avid with a sophisticated plugin to detect smut? I think the latter.