Ministry of Intrigue

A few weeks ago I was having drinks with my good friend Greg, and the topic of social bookmarking came up. I was discussing some of my ideal features in a service, before we got back to the more serious business of catching up. A few days later, Greg sent me a link to Diigo, as it seemed to match a lot of what we were discussing at the time. I took a look at Diigo and was pretty impressed with what I saw, and I decided to give it a go. I’ve been using it full-time for about a week now, and feel comfortable giving you all a review of what I have found.

Diigo has a very attractive and subdued appearance, that is packed with features without being overwhelming.

To begin with, Diigo is an extremely powerful social bookmarking site. Obviously, Diigo does all the things you would expect of this type of service: you can save bookmarks, assign tags to them, and search the site for bookmarks that are also tagged with those terms or find people who have saved the same bookmark. Diigo also allows you to construct “Lists” of links. Lists are another way of structuring your data that you can use in conjunction with tags. Each List can be made up of any group of links that you can sort in whatever order you desire via a drag and drop interface. This is really nice to see a service that still understands that tags are not the end-all be-all of organizing content.

Diigo also allows you to import bookmarks from variety of sources, including Delicious, Magnolia, Simpy, Blinklist, Furl, Connotea, RawSugar and of course, your own browser. The import function worked well for me importing from Magnolia, although Diigo replaced the spaces in my multi-word tags with underscores. Diigo does allow multi-word tags if you encase them in quotation marks, so this was a quick fix, if a little annoying. When will people see the light and do away with space-seperated tags? Just let me use a comma-separated list. Diigo also exports all your bookmarks quite effectively in a variety of formats including RSS, CSV, Delicious format, as well as in formats for both Internet Explorer and Netscape bookmarks.

However, Diigo doesn’t just want to be a bookmarking service, they aim to be a flexible research tool, and allow you to highlight and annotate web pages to provide more directed commentary on what you are bookmarking. These notes can be private for your reference only, or publicly visible to any user. This immediately brings up comparisons to Clipmarks, except that this is very different. Whereas Clipmarks just takes your highlighted content and loads it into their service, Diigo also leaves those annotations in place in the form of highlights and sticky notes that are visible only to Diigo users. This allows you to not only share those annotations on Diigo itself, but also to visit the originating site and see those comments in context of the surrounding content.

An example showing a highlighted annotation with a private sticky note attached.

This annotation feature is particularly powerful when used in conjunction with Diigo’s social features. Diigo allows you to create groups which can be public, private or semi-private, allowing you to collaborate on research through the use of links and annotation. Diigo also allows you to attach notes and comments that are visible only to the group, which is an extremely useful feature when sharing the link both publicly, as well as in a group context.

In addition to collaboration, Diigo’s social side is excellent for content discovery. The service can provide recommended bookmarks from other members based off of the links you have saved in the past, as well as recommending other users whose bookmarking habits seem to match yours. Diigo takes the “social” in social bookmarking very seriously, and provides very effective tools for finding friends on the service, as well as finding new people who have interests similar to your own. Friending another user doesn’t mean just making them a contact, it enables you to generate buddy lists, allowing you to organize sharing of bookmarks with friends, as well as providing a messaging system. Whereas in many other bookmarking services the sharing and social features seem to occur more as a byproduct of the sharing process, Diigo puts those social networking features front and center. However, Diigo’s interface is very content focused as well, making it clear that this isn’t a social network as much as it is a social tool.

Saving content to Diigo is done primarily one of two ways: you can either install the toolbar application, which is available for Firefox, Flock and Internet Explorer, or you can use Diigolet, which is a bookmarklet they provide that should work with almost every major browser.

Examples of bookmarking. The first with the Diigo toolbar in Firefox, and the second using Diigolet in Opera.

The Diigolet is a surprisingly powerful bookmarklet, revealing sticky notes and annotations, as well as providing all the basic functionality a user needs. However, even with my hatred of adding additional rows to my browser window, the Diigo toolbar has won me over and become my tool of choice to interact with the service. Both tools will provide tag suggestions and assist with group functions, as well as the ability to send the link via email, however the toolbar goes even further. When using the toolbar, you also have the option of cross-posting your links to other bookmarking services, or even Twitter if you require. You can save simultaneously to Diigo, Delicious, Magnolia and Simpy, as well as to your own browser’s local bookmarks. Bookmarking to other services seems to work well, and saving to local bookmarks is a particularly awesome experience when using one of the latest betas of Firefox, which will attempt to auto-complete based on both history and bookmarks. It even correctly applies tags in the Firefox Places storage system, which is great but makes me wonder why the toolbar bothers to also build a hierarchal folder system inside Firefox as well, as the tags do that job already.

Another powerful feature that the toolbar adds is the Diigo sidebar:

The Diigo sidebar displaying my recent bookmarks.

As you can see in the above image, the Diigo sidebar allows me to search and browse both my bookmarks and the bookmarks my friends have posted. In addition it allows me to get current information about the page I am viewing via the “This URL” tab. I can access public bookmarks and annotations, and lists of Diigo users who like the site. Diigo also can provide quick metrics about a site that I am visiting via the main toolbar. Using the “About This URL” menu option will provide a overall popularity score for the site, including a breakdown of the number of links to the site from Diigo, as well as from Google, Delicious, Yahoo myweb, Bloglines, Technorati, and Digg. Diigo also provides a calculation of the site’s Google PageRank, which is a really awesome bonus feature that I just discovered today.

Diigo supports OpenID logins, which makes me smile. It would be nice if Diigo allowed you to sign up using just your OpenID, however you need to create a standard account and then associate an OpenID with it. This isn’t a big deal, but it would be a nice enhancement to see in the future.

My only real complaint about Diigo is the lack of an API for developers. I did send them an email regarding this, and was pleasantly surprised to receive email directly from the founders. They indicated that an API is in the works, and were receptive to some of my suggestions in that regard. As I have browsed through the user forums, this seems to be a common practice for the people behind Diigo to actively engage with their users for ideas, and respond constructively to critiques. Diigo does provide RSS for bookmarks, as well as blog widgets, but those weren’t sufficient for my needs. I’ve been able to work around the lack of API by using the toolbar for cross-posting to Magnolia and continuing to use that service for loading bookmarks into my site. However, I am eager for Diigo’s API because this workaround doesn’t allow me to take full advantage of the annotation features when loading the bookmarks into my link blog.

However, when it comes down to it I feel Diigo is really head and shoulders above the majority of competing social bookmarking services in terms of features, and the site itself is certainly more responsive than my beloved Magnolia, which is a wonderful service in itself, but runs slow as molasses. Based on this glowing review, it may come to surprise you I approached Diigo skeptically, even prepared to be antagonistic, but the service managed to win me over. I was unprepared for the seamless integration of Diigo’s social features, as well as the flexibility of the service, which strikes out like a fist of features into the collective nuts of their competition. If the forthcoming API meets the rest of the site’s high standards, you can expect me to be a happy Diigo user for quite some time to come.

I’m a little late on this one, but it’s still worth discussing. I also apologize for the length of this post, but I had more to say than I thought!

A few weeks ago, some of you may have read this post by Jeremiah Owyang regarding his predictions for the future of social networks. The article makes several assertions, such as social networks will create open APIs in response to market pressure from Facebook, which makes a lot of sense. However, the most interesting thing that he predicts is that Facebook will become a major player in the identity game.

Facebook will launch an Identity widget that I can embed on my blog. This allows only those who have registered to Facebook to leave a comment, many high profile blogs will do this, to avoid nasty anonymous comments, thus reducing the incident of Kathy Sierra type events.

He also suggests:

Facebook will have faster adoption that Open ID, as the consumer users will drive it. (Remember the mantra of consider joining before creating communities)

Now, Jeremiah Owyang is not assigning any value one way or the other, merely making predictions based off of the trend he is observing. Facebook certainly is aiming to be a single web platform for its users, and moving into identity management would be a logical step in that direction. The Facebook API does makes this possible, and in fact their are some sites that already support using FB login credentials, like Ma.gnolia. Given the boom in popularity of FB, it seems natural that users would find this an attractive option. After all, we already have many poor benighted souls that eschew email for messaging within FB or MySpace, so it is logical that those individuals would prefer to manage their identity through one of those networks. It is also certainly demonstrable that the lack of anonymity reduces the amount of active trolling in communities (although this is more an argument for identity verification in general as opposed to a FB-specific observation).

All that being said, isn’t this a step backwards?

We don’t need another walled garden, where another corporate entity controls the keys to our online identity, and eventually more users are going to realize this. Thus, I suspect any trend towards using the Facebook API to provide authentication elsewhere is ultimately only going to exist in the short term, only to be superseded by more open systems. Dave Winer gives an excellent explanation of this in this post on user lock-in when he says:

Facebook could easily be the place where the dam breaks. It’s attracting so many users, who may at some point realize that they want control of the data that’s locked up inside Facebook.

That is exactly the issue with entrusting our identities to any single entity, because ultimately it comes down to a matter of trust. As users grow more savvy, they will trust closed systems less and less. Or at the very least, users will become more aware of how incredibly inconvenient the walled garden approach to personal profile and identity data is, as Justin Baum explains quite well in his post on redundant relationships. The future success stories of the web will be the ones that can solve this problem.

I turn to Winer yet again, because he explains it very well:

A vendor will come along and they’ll store your identity but give you complete freedom to move it where ever you want when ever you want at no cost. They’ll make it easy to do so. And they’ll get rich doing it, if they want to.

Why?

[…]

It’s the basic trust proposition of the Internet. People will only trust a service that gives them complete freedom to come and go as they please. Further, they’ll want to come back if you send them to cool places. It’s why people like Facebook today, and why they’ll be tired of it tomorrow, if it only sends you to places within the Facebook silo.

The key to building these future successes is developing and using decentralized protocols like OpenID, and utilizing simple semantic markup such as microformats or even more complex syntax like FOAF. OpenID allows for a truly decentralized identity system, which allows you to switch identity providers on a whim, or even run your own. For more info on OpenID, I recommend you start by checking out Simon Willison’s excellent screencast on using OpenID. Microformats and FOAF provide methods by which you can describe people, relationships and other objects in such a way to be machine readable, but comprehensible to human beings, which opens the possibility of importing/exporting relationship and profile data between various services automatically. We need convenient and open ways of managing our online identities as well, and there are some very smart people out there who have already made some initial forays into that area.

In a later post, Jeremiah Owyang acknowledges:

I realize that Open ID is one solution, but let’s get realistic, it’s not being adopted because it’s too geeky, maybe they need a marketing evangelist, or a mass consumer tool will need to be birthed. Please note, I’m not opposed to the tool, I’m just looking at the market around me.

Owyang is right on the money here, in that these services need more simplistic tools for implementation and more effective evangelism. Currently, the focus of these projects has been marketing the idea to other developers, and they have been succeeding in getting quite a few technical folks pretty fired up about it. Now it’s time to really focus on bringing it to the attention of the masses. Firefox 3 will have built-in support for detecting and utilizing microformats, as well as supporting OpenID which will go a long way towards raising awareness and usage of those methods of describing information. Lightweight and interconnected (maybe even interchangeable) social networks are clearly the future of the web, and these technologies, or their descendants, are the keys to creating them.

None of this is to say that services like MySpace and Facebook are doomed. Quite the opposite, both of these networks are uniquely poised to write their brands into the history of this shift in web technology. If they can let go of the lock-in business model and focus on developing with these budding technologies, they can be situated as central points of influence in the future web, rather than eventually fading into obsolescence, as many a closed system has before them. Turn user profile pages into OpenID identity providers, and offer the ability to use already existing OpenIDs for either new accounts or to associate them with existing ones. Utilize microformats and open APIs to encourage user freedom, and as a result, build brand equity as well as increasing user loyalty.

Users don’t want a walled garden, they want an open field. It’s time to give it to them.