My Opinion of Smart Tags

Authored June 2001

Recently, as part of an updated beta of WindowsXP, Microsoft released "Smart Tags", a new feature within Internet Explorer. In a nut shell, if enabled, the feature will search the text of a web page and highlight certain words or phrases. Selecting a highlighted phrase will present the user with an inline box offering them related (MS owned) web links. Here is an image of Smart Tags in action from Scripting.com.

After several articles (most notably a negative feature in the Wall Street Journal) promoted the existence of this feature, it has become yet another rallying cry against Microsoft. Detractors claim MS is committing copyright infringement and modifying their pages behind their backs.

I disagree with this opinion and, of course, I'll you why.

[Update: On June 28th it was widely reported that MS had decided to indefinately delay the release of Smart Tag in Internet Explorer. The feature is still available as part of the OfficeXP suite however.]

The Nature of the Beast

One of the most commonly heard complaints is that "MS shouldn't be able to add links to my content". Now I would agree with this totally if MS were adding links, but they're not. The highlights used by the Smart Tags feature are a zig-zag underline (think misspelling in Word). They do not resemble standard hyperlinks and as such should not be confused with them by even inexperienced users. They are obviously "extra".

The next point belabored by critics is that Smart Tag references will point to MS online properties. I do not see how this is a bad thing in and of itself. For those of you unaware, MS has some damn fine properties out there. Even if Smart Tags were only coupled with TerraServer and Encarta.com the benefit could be enormous, especially to online homework help and education. Many Microsoft properties are best of class: Carpoint, Encarta, Money Central, MSNBC, etc. Yes, of course MS is using the feature to drive advertising revenue, but the content under those banner ads is generally going to be very good.

Also be clear that Smart Tags are disabled by default (although this may change by release). Even if MS promotes them in some way only a percentage of users will enable the feature. Those users that do enable the feature will expect value. Should MS Smart Tags end up as nothing more than advertising traps the simple result is that people will ignore or disable them. MS simply must provide value to gain any increase in advertising revenue.

Smart Tags are configurable by third party companies, although the default content will of course be Microsoft's. The complexity of the task makes it uncertain whether other companies will go to the effort to customize the feature. I predict however that at least one, AOL, will. The future of Internet Explorer as the engine for the AOL client software is in contention as I write this. If it does make it into AOL 7, Smart Tags, if enabled at all, will NOT point to MSN properties. Although few companies sport the resources available to AOL, the likes of ZDNet, Amazon.com, Excite, and Google among others all stand to gain from such customization.

Which brings me to another point: this feature, via a special "meta" tag, can be disabled by any web author that does not wish to have it active on their site. Yes, it can be argued that creates extra work to disable a feature that (so the argument goes) should never have existed, but I find this a weak argument. Web authors should worry about engaging their readers, not about inhibiting functionality of the browser.

Some critics have compared Smart Tags to traditional content theft which has a thief copying content and hosting it elsewhere to place advertising within it. This is a politically driven, incorrect characterization. Smart Tags work only as an overlay. Content, changes, and site traffic are not affected by them.

Who Holds the Reigns?

I find this whole debate amusing since in my view those authors arguing against Smart Tags are, in fact, arguing to take control away from the user. It's the dirty-little-secret of the Web that most Web Designers (if given the chance) would gladly take most, if not all, control from the users. Search various online web design/programming discussions and you will find hundreds of requests asking how to disable the "back" button, "view source", or "save" features.

Authors want the freedom and lack of expense that Web publishing offers but still want to maintain rigid control of their content, presentation, and access. I was once victimized by content thieves (see this HotWired Article or this Berst Alert for the story) and I agree that authors' rights need to be upheld in the online world. However a line needs to be drawn between the author's right of ownership and the users fair use.

I feel that anybody has the right to modify my content for their own personal use. I see this as no different than a college student highlighting a text book or a housewife clipping a recipe from a magazine. The highlight doesn't change the content or smack of copyright violation. When that book is sold back to the university and distributed to another student there is still no problem. The highlights are clearly not "original content" and will not be confused as such.

Smart Tags blur this line a bit: the "highlighting" is provided by a third party, but I see little difference. Enabling them is a user decision that doesn't affect the page's author or other users.

A web page user may choose to apply a different style sheet, thus utterly destroying my meticulously designed layout. A user may choose not to download images or sounds thus ruining the feeling and atmosphere I was building. A user may choose to disable JavaScript, Cookies, Java, etc thus reducing the utility that I sought to offer them. A user may even choose to run my content through an auto-translation service like Babblefish thus ruining the meaning of my content more that Smart Tags ever could hope to.

Conclusion

Future versions of HTML and XML seek to make the web understanble to machines, not to humans. Tim Berners-Lee has written and spoken at length about the "semantic" web, a technology vision where automated agents scour online data and pluck out items of interest. The content, in this view, adopts the context of the user not the author. Smart Tags exist only in the context of the user and as such should not be considered an issue by web authors.

As the Web matures authors will find many new technologies like Smart Tags taking control from them and placing it in the hands of their audience. Will web authors be able to handle that fundamental shift?