Privacy is a serious thing. That’s what I thought when last week I heard about the subpoena on Google, Yahoo, MSN and AOL. Last week I was also concerned about how all these players except Google had accepted the DOJs request to deliver what was previously announced (in the press) personally identifiable information. Yesterday at Search Champs (the event I’m attending in Redmond) all of us attending learned otherwise.

Getting to the facts

Yesterday we heard (initially under NDA, released after a request by some of us in the Web 2.0 Workgroup) about what was actually released by MSN - and personally, I must say I was quite surprised to hear the facts in a non-press manipulated way. So to cut it to facts only:

• No personally identifiable information or IP addresses were shared
• 1 Million random results from the index, and the related search queries were released
• There was no connection to child pornography, but to children searching for porn on the web

Now, there’s a couple of funny things here. Firstly, anyone can go and request the search query information from the search engines (particularly if you’re doing it for academic research purposes); Second, how the press manipulated the story to lead everyone to believe that this was somehow related to child pornography, when it was actually related to children looking for porn on the internet - and to quote Chris Pirillo, who didn’t see a Playboy as a kid?

Should we be concerned?

The problem here isn’t what was shared, because we’ve now learned that it was nothing identifiable. The problem here is the precedent. This case proves that the government (whether it’s the United States government, or any other) can request information about what you’re looking for on the web. Now, everybody knows that this information is being collected everyday, but what do we - as users - expect companies to do with it?

Personally, I’d like companies to be transparent about this sort of situation - it is easy to assert that MSN’s biggest mistake in this story was not telling its users about the DOJ request. But I don’t want search companies to go at this alone. There needs to be a conscious effort to protect the user’s data, from all these players - together, not individually. Until we see that, this sort of story will continue to surface.

Discussion

Some of us at the Web 2.0 Workgroup decided to do a podcast with some great guests to discuss the situation. We had myself, Joshua Porter, Alex Barnett from Microsoft, Chris Pirillo, Thomas Vander Wal, Dion Hinchcliffe and Brady Forrest from the MSN Search Team.

Download the podcast - 42 minutes, 7mb

Alex and Joshua are talking about this as well. Update: Dion and Robert Scoble also have their takes on this.

So next week I pack bags (well, a couple of bags because I fly light) and fly out to Seattle for Microsoft Search Champs v4. I was invited (mind you, for the first time), and I must say I’m pretty excited and honored to be a part of it - I must admit I’ve never been the biggest fan of Microsoft’s previous online efforts (as I’ve discussed with Robert Scoble when I was in california), but I am interested in figuring out both what they’re up to, as well as how I can help making it better.

I’m also taking the opportunity to talk to some people I know from Seattle, and those that are not from around but that happen to be there next week. So if you’re going to be in Seattle, drop me an email and lets have dinner or something (my lunches are set to be in Redmond, I believe).

I plan on doing some shopping, and I definitely want to go to the Seattle Public Library - which is one of my favorite architectural projects by Rem Koolhaas. There may also be sort of an impromptu dinner with some well known Web 2.0 people (and members of the workgroup), but I’ll keep you all posted on that. What else is there to do in Seattle next week? Got tips for someone who’s never been? Drop me an email, or leave a comment. Thanks!

Even though I’ve talked about this in the past, I feel like it’s so important that it needs to be said again and memorized (in fact, get sticky notes and prepare to tag this to your walls). If you’re using public APIs, rely on the functionality they give you, but always have a backup plan.

Think about it, almost 50% of the current web applications rely on public APIs, put out there by Google, or Yahoo!, Amazon, or even one of the smaller players. The problem is when the APIs fail, when the terms and conditions for usage change and a big chunk of your application’s functionality is lost overnight, no questions asked and no warnings issued. What do you do then? If you have no backup plan to save the functionality provided by your API, you’ll have downtime and risk failure altogether.

So, when planning your web-application, keep in mind that the more in control of functionality you are, the less likely it is for it to fail. There is no set solution for substituting functionality provided by public APIs like, for example, Google Maps. And there’s no guarantee that anything will ever fail. But if you’re building a business, you better be ready for everything - good or bad.

Further reading:

Service dependency on the new web, Published October 27
Public APIs - Arghh, Published January 11th on Forest and the Trees
Web 2.0 and the Drive-by upgrade, Published October 24 on O’Reilly Weblogs

The Beta tag - seems like every single application needs one nowadays. While I disagree with launching as a Beta (or with the Beta label) as an excuse for incomplete or inconsistent functionality, there’s a lot of people doing it. Mike at Techcrunch has a very good post with many of the issues he sees (and I see as well) with the web applications being launched every day.

A day doesn’t go by where in my email I get a new invitation for an upcoming web application or service that, “while in beta” is “too great to miss”. What if it fails? Should I be disappointed, or should I just expect it to be fixed because it’s beta software? In fact, why should I see it now instead of waiting for the fully-featured, 100% working version of it sometime soon?

Sometimes, it’s better to wait and release properly, than to launch early, under pressure, and fail miserably. Make sure you read Mike’s post, it’s worth it.