Observations on Twitter, Noise, Filters, Spam, and Value

This video contains a lot of thoughts on Twitter and the value of Twitter… lately things have seemed really noisy, with a lot of content that I may not want to see, some spam thrown in for good measure, and some otherwise tech-savvy folks who engage in what I feel are questionable Twitter practices. I talk about the value in one’s Twitter stream along with particular behaviors which degrade the value, wrapping things up with some solutions (both technical and behavioral) that I feel will increase the value of Twitter for individuals. The video is a bit long (a little over 13 minutes) but I’m hoping folks will watch and comment. Let’s increase the value of microblogging for everyone!

Facebook Terms of Service Change: Content is now Licensed Forever

A few weeks ago I posted about deleting my Facebook account due to their terms of service being an overly broad rights grab for any content posted on the site.

There was a lively discussion with over 30 comments left on the post. Commenter Matt Behrens noted that Facebook’s terms indicated that a user could remove his or her content and that Facebook’s license would expire at that time.

Not anymore.

While researching for my presentation, I discovered that Facebook updated their terms of use on February 4th. The section about a user being able to revoke the content license by removing the content from Facebook is now gone. There is no verbiage that indicates a user may remove Facebook’s license. Facebook now claims a perpetual license to any content posted on their service, with no way for a user to terminate that license.

Your Grandmother’s OpenID: URL vs. Email

One usability issue that comes up in OpenID discussions is that of URL as identifier, noting that perhaps e-mail would be a better option, because not everyone has a “personal” URL. Two quick thoughts:

  • There was a time when most folks didn’t have an email address, either.
  • Most of the folks who seem to be the “target audience” for email as identifier probably don’t even have a custom/personal email, but rather they’re using something like bobsmith47@hotmail.com. That’s hardly any more personal or identifiable than a URL tied to one of the big social networks.

ReadWriteWeb, Movable Type, and Vidoop OpenID: Broken

Warning: user-focused “I just want this damn thing to work” rant ahead.

It’s no secret that I’m a big proponent of OpenID. Last year I took a stand that I would avoid participation on tech-focused blogs that didn’t support OpenID. Shortly thereafter, I discovered that I couldn’t sign into ReadWriteWeb using my Vidoop OpenID.

I contacted Vidoop, and was told:

I’ve done some more troubleshooting with our friends at ReadWriteWeb, the issue is that they aren’t configured to handle “https” (SSL) connections, and we don’t allow OpenID associations with websites who don’t use SSL to verify with us.

This makes sense to me. The main reason I use Vidoop is because of their excellent security and the fact they require a SSL connection is a good thing. I then contacted Richard at ReadWriteWeb; he bounced the issue to his Movable Type consultant, who responded with this:

I have forwarded this on to Byrne at Six Apart. This seems like something that should be supported natively by MT’s OpenID auth module. I will share any feedback that comes back from 6A.

That was in March 2008. It’s now February 2009, and I’m still receiving “An error occurred: The sign-in attempt was not successful; please try again.” when attempting to log in to what should be an easy authentication system for end users.

Why is this so hard? What’s going on?