December 14, 2013

Twitter’s Rapid Reversal -- and the Rising Dilemma of “Public” vs. “Publicized”

By now you probably are well aware of what must have been one of the fastest major policy reversals in the history of social networking, when a few evenings ago, Twitter changed their definition of user blocking, and in the face of an enormous outcry that became an international news story within a matter of hours, announced that they were for now reverting back to their original methodology.

There are a couple of rather obvious lessons here, and at least one not so obvious lesson that is likely at least as important.

First, at Internet speeds, it's possible to lose years worth of user love in a heartbeat, if you're seen as suddenly changing the rules in a manner that your customers (users are customers, right?) feel in the large to be antithetical to their interests.

Secondly, when you screw up, don't prolong the agony -- get in front of the issue as fast as possible. When a mea culpa is in order, get the sword out, make your proclamation, and correct the situation as quickly as you can. Twitter's rapid response to a sudden crisis (albeit of their own making) was indeed both wise and prudent.

But this leaves us with a gaping question -- how did Twitter so grossly miscalculate the likely reactions to their policy change, and so massively underestimate their impact?

I suspect that part of the answer involves understanding and appreciating the issues of "public" vs. "publicized" (allow me to coin the term "PvP" for short) in social networks -- a category of concerns only now really coming into focus and discourse.'

"Public is public" -- you've likely seen me say this many times. It is foolhardy to assume that a public posting will be seen only in the context in which it was originally made, or to pretend that a public statement can somehow be effectively erased after the fact. The so-called "right to be forgotten" -- that suggests trying to censor information that has already been extant on the Web, either from websites, search engines, or both, is entirely impractical and potentially vastly dangerous to fundamental free speech rights and more. Such anti-speech laws must be vigorously opposed.

Yet there is a difference between sending a posting out to the members of a mailing list, or your followers on a social network -- vs. having the same message blaring out of loudspeakers for all to hear on every street corner of the planet.

The difference relates fundamentally to "discoverability" -- how likely it is that any given posting will be seen or found beyond the context in which it was originally made. In other words, simply because a posting is publicly available to find in a search engine and to read via a public URL, doesn't equate to purposely "publicizing" that material beyond its original posting context.

There are many facets to this dilemma, and various aspects of them apply to most social networks today, including Twitter, Facebook, Google+, and others.

The aspect perhaps most relevant to the recent Twitter reversal relates to an increasingly popular concept in social networking, amounting to the idea that it's acceptable -- even desirable -- to present different logged-in users, and non-logged-in observers, completely disparate views of the same underlying discussion threads, streams, or user status states.

So, for example, logged-in user A may see a different "public" stream of discussion than that observed by logged-in user B, and non-logged-in C may see something different in other ways -- in some cases actually even more complete than what A or B might be seeing, depending on possible user blocking relationships between B and A.

This takes us to the heart of the Twitter controversy. Under their now aborted changes, blocking a user would have only prevented the blocking party from seeing what the blocked party was saying -- the latter would be able to continue publicly harassing the blocker -- and having such remarks seen by everyone who would have seen them previously -- except the party who triggered the ersatz block itself. This situation was quite reasonably described by critics as being roughly equivalent to being offered a blindfold and earplugs to deal with someone covering your home with graffiti, or, to characterize it a different way, "Lie back and think of England [go ahead, look it up!] ...

Supporters of Twitter's changes argued that blocking a user never really stopped them from seeing your public postings anyway, and that not informing someone that they were blocked would help prevent possible reprisals from the blocked party (on the theory that they wouldn't catch on to the fact that they were blocked).

But most any battered woman is likely to tell you that assuming harassers are stupid, or that ignoring them is a solution -- is utterly and perhaps lethally wrong.

It is true however that those public postings would still be visible if you went out looking for them. But the inability to directly associate with them, within the context of the specific streams and threads themselves, is still highly significant once blocking has been enabled.

There's a slippery slope aspect to all this as well in the broader social networking context.

Once you accept the proposition that it's OK to not inform someone that they've been blocked, and to present them with a version of a stream or thread that is actually more limited than that seen by other users or even the public, it becomes much more acceptable to spread this mindset into other areas.

User A may not realize that the comments they posted on page B are only visible to A, and not to anyone else (unless A logs out and inspects the page from that vantage point) -- or that comments written by A and queued for moderation, or rejected by moderators, may still appear to logged-in A as if they are publicly viewable, even though they are not yet (or never may be).

The practicality of such approaches when attempting to manage large social networks seem clear enough, but the resulting dilemmas are arguably ethically dubious at best.

These conflicts become even more noteworthy as the concept of "public" spreads into third-party contexts beyond the scope of original postings, a situation I alluded to above.

When a posting made publicly to a set of followers becomes routinely visible and highlighted to affinity and interest groups who are not largely congruent with that original posting audience, the impact of the posting itself can change in a fundamentally qualitative manner, both unexpected and unwelcome from the standpoint of the posting party.

In essence, the public posting has now been publicized in a place and manner not in accordance with the poster's original intentions.

By analogy, if you went looking for a job by posting your information publicly on a tech job website, you probably wouldn't want that information crossposted to a publicly available sex magazine (or perhaps you would, but the point is that you likely don't want the job website to perform that crossposting without your explicit permission.)

As you can see, the entire "public vs. publicized" arena is nontrivial to grasp in its scope, and I can really only scrape the surface here today.

But the next time you hear discussions about public information on the Web, particularly in relation to social networks, the next time you hear someone say "public is public" (including me!), I urge you to consider the fascinatingly complex maze of twisty passages that reside between public and publicized, and how best we may find our way through them without the use of magic wands or magic words [insert "XYZZY" joke here? Naw ...]

Be seeing you.

Disclaimer: I'm a consultant to Google. My postings are speaking only for myself, not for them.

Posted by Lauren at December 14, 2013 11:45 AM | Permalink
Twitter: @laurenweinstein
Google+: Lauren Weinstein