A Service That Needs to Exist

There are lots of fun and even useful new services springing up as part of Web 2.0.  But every time someone points me to one, I have to go through all of the rigmarole of signing up, feeling nervous that I didn’t read the EULA (End User License Agreement) thoroughly, typing in hard-to-read phrases to prove I’m a human and not a bot, and responding to a registration email.  Michael Parekh blogged an hilarious account of this annoying process here and points out that the annoyance factor  is a detriment to the growth of Web 2.0 services.

Some services don’t require signup, of course.  Many services allow you to read content freely and anonymously but require registration in order to create content of any kind.  For example, you can look at del.icio.us tags or digg ratings without an account but you can’t tag or digg anything unless you register. 

If services like these eliminate the registration requirement, they face the very real danger that they will be overwhelmed by bots designed to tag or digg whatever web page the bot’s master wants to promote or defame.  There really is a reason for making us prove we’re not bots when we sign up.  Similarly, the requirement that we provide a valid email address and authenticate it by responding to email is necessary to prevent promoters of this or that from creating hundreds of accounts from which to influence ratings.

So far wikipedia and wiktionary have been the exceptions among web services and have allowed unrestricted posting and editing of posts without registration.  These services have little to fear from bots since so far bots aren’t interested in writing articles but the wikis do have a problem with vandalism.  Wikipedia recently had a spate of bad publicity from a defamatory article posted as a “joke” and has introduced but not yet used the concept of a “semi-protected” page which can only be modified by those who have registered – and not even the most recent registrants.

The need for registration is unlikely to go away.  But why is it necessary to register separately for every service we use?  I think there’s an opportunity for some entrepreneur to create a one-registration-for-many-services business.  If this doesn’t happen, most services will have to be gobbled up sooner rather than later by companies like Yahoo on which hordes of people already have accounts.  When Yahoo bought photo service Flickr, it allowed those with Yahoo accounts to use Flickr without creating new accounts.  Presumably the same thing will happen with newly acquired del.icio.us which will then grow even faster than it was before.

We have been trying to create a game for readers of hackoff.com using Flickr and del.icio.us but are stymied that the subset of hackoff.com readers who are also registered to use both del.icio.us and Flickr is probably so small that almost none of our readers would be able to play. But, if everyone who has a Yahoo registration can use both of these services without further hassle, then we’ll be able to build our game.

The point is that are a huge number of new applications (both permanent and transient) which could be built from all the pieces and services available on the web but the need to sign up for each service separately stymies that.  A service that we register with once, convince that we are humans, and which then credentials us for many other interesting services would be a huge benefit. Assuming it was well run and reliable, I think that new services might come to rely on it as the PayPal for registration and not each build their own annoying process.

If I were still starting companies (I’m not) then I’d be interested in this opportunity. Might try to start with a standard EULA and getting new services to agree to adopt this. The advantage of a standard EULA is you really might read it once and then know that every service which subscribed to it wasn’t asking any more of you then you already agreed to. Services that feel they need a non-standard EULA would have to justify why they are asking for something different than what most other services need.

Anyway, I’m giving this idea away for the holidays. Really, my motives are not quite pure:  I’m sick of signing up for each service separately and really would like to be able to build interesting combinations of services without asking people to register separately for each component service.