I’ve decided to use Diigo as my key bookmarking app. Here’s why.
There’s a lot of useful stuff on the web, but it’s usually hidden in an awful lot of crap. As a professor and design researcher, it’s important for me to have access to a wide assortment of good information. The web can deliver that information only as fast as I can identify it. So when I do find something particularly useful, I want to make sure I can remember (a) that it exists and (b) where it is.
Enter bookmarking apps. These are web-based applications that let you bookmark other sites, and then search your bookmarks quickly. Because the bookmarks are stored on servers elsewhere on the web, there are two immediate and important benefits. First, you can get at your bookmarks from anywhere (i.e. you don’t have to carry your laptop around with you just for the sake of accessing your bookmarks). Second, you can share bookmarks with others (this is an interesting and under-utilized method of collaboration these days).
Years ago, I’d pondered the few existent bookmarking services and found them all lacking on one particular front: you can tag a website, but the service won’t index the site too. I think indexing would be good because the only “best” set of tags is an exhaustive set of descriptors and keywords that describe the item, which no one really wants to type in. Most keywords are already in the item’s content, so why should one have to type them in again? By indexing items, one can use fewer tags, which also makes tagging easier and more consistent.
So, since I couldn’t find what I needed, I decided to roll my own. It’s quite buggy still, but SERF does work, sort of. I call it “anti-social bookmarking” because I’m the only user who can add/edit items. That’s mostly because (a) the software is still, as they say, pre-alpha, and (b) it’s only for storing the links that I care about for my teaching and research. I’d love to develop it further, but I can’t get the funding I need to develop it as a research project, and I can’t find programmers I trust to work on it for me a-la FOSS. (Though I’m always open to opportunities.)
But the web waits for no man, and while SERF is good for some things, there are more items that I want to track than I can possibly add to SERF – partly because many of them don’t have that much to do with my work.
So I needed something else, something I could use to pitch items into as temporary storage, but still functional enough that I can use it to quickly look things up when I need them – and make them available to my students and colleagues.
I’d had a Delicious account for ages, but never really used it. By the time I realized that I really needed a bookmarking app, there were competitors. So for the past couple of months, I’ve been playing with Delicious, StumbleUpon, Diigo, and Twine. And I’ve recently come to the conclusion that Diigo is what I need. I decided this based on five criteria: bookmarking functionality, look and feel, extra functionality, shareability, and performance.
Delicious, Diigo, and Twine all have comparable basic bookmarking functionality, including appropriate bookmarklets and toolbars. Basically, you navigate to the page you want to bookmark, click the right icon and you get some sort of popup that lets you tag the item, add a description, and, depending on the service, do various other things. Twine has the most web-2.0-ish facility, but that doesn’t necessary impress me. All but StumbleUpon recommend tags based on others who tagged the same resource. But since I have my own tagging scheme that I find is significantly different from those of most other users, this doesn’t really matter to me.
As far as basic bookmarking goes, all four systems are pretty evenly matched.
Look and feel
Look and feel is, for me, extremely important, partly because I do appreciate an aesthetically pleasing web site, but also because I’ve found that look and feel ties very closely to usability.
Here, Delicious really shines. It has an absolutely brilliant look and feel. The screen is laid out in very carefully chunked areas that have consistent structure and content no matter what part of the site you access. Every little design feature of the page serves a very clear purpose, and everything is there only once – there’s no redundancy. Everything is visible: no strange icons or bits of text magically appear when you mouse over something. (That really annoys me.) I’m sure people who specialize more in web usability could go into much more detail, but suffice it to say that the Delicious look and feel is, for me, like a breath of fresh air.
Diigo comes in second. It has a good layout with just a little redundancy. Everything is visible. I don’t like Diigo’s rendering of the tag cloud; it’s not as elegant as that in Delicious. Indeed, the entire site is just not as elegant as Delicious – it looks slightly rough and seems to not use screen space quite as well as does Delicious. (This might be because Delicious has been around for so much longer that its developers have had the time to carefully tweak the look and feel much more.)
StumbleUpon is next. It’s interface is interesting, but not flexible enough for me. I wish there were other ways of arranging things; I don’t mean just skins or themes here, I mean the arrangement of the actual items. It has an ugly tag cloud. I wish there was an easy way to get rid of the screenshot thumbnails for each entry; I can understand that some people might like them, but I find them useless. And I was never able to get rid of them. I don’t like how they string tags and the URL on one line, because if you have more than a few tags, the URL is truncated. Indeed, showing the URL is of limited use anyways. The other systems expect you to mouse over the item’s title and look at the address bar of your browser. Put all these things together, and you get an app that I just don’t feel comfortable using, compared to the others.
Twine, unfortunately, comes up last here. I tried real hard to understand the Twinerly way of doing things, but I just couldn’t get it. I keep loosing track of which part of the screen I should be looking at, and what all the boxes were for. The problem isn’t exactly the layout, but it’s the content. For instance, under My Items, you can filter the list of your bookmarked items in lots of different ways – too many ways, if you ask me – including by “related people,” the meaning of which I’ve not yet deciphered. Lists of items also seem to include all manner of items including individual bookmarks, twines (socially-constructed feeds of items on particular topics), comments you may have posted on individual items, and other things too. I find this confusing, and I wish they were somehow compartmentalized.
Long story short, every time I use Twine is like the first time. Which is not good.
Each of these services provides extra functionality of one type or another. Caveat lector: this category is very subjective, because the functionality I value is the functionality I need – which might not be what you need.
All four services offer alternative ways to establish groups around specific topics or subjects. Each site has it’s own particular way of doing it, but they all really amount to the same thing. They also all offer RSS feeds for virtually any list of items you can generate.
StumbleUpon has a very interesting service: you specify what kinds of websites you like (about, say, science fiction, or AJAX programming, or Labrador Retrievers) , and then when you hit the Stumble button, you are taken to a random site of that kind. You then rate it with a simple thumbs up or down. StumbleUpon’s recommendation engine uses your ratings continually fine tunes it’s recommendations. I really like this approach – it’s clean and simple – but it works for people who aren’t looking for anything in particular, which is a category that doesn’t include me most of the time.
Twine’s unique contribution is it’s AI-ish tagging engine that is supposed to help you tag things and organize them into feeds (twines). The problem for me is that my personal tagging style is informed by years of study on classification systems and taxonomies, which means I’m not your average tagger. So Twine’s recommended tags rarely line up with what works for me.
Delicious offers little in the way of extra functionality except for being able to search the entire tag cloud of all entries, which can be amusing, but no where near as much fun as StumbleUpon. But Delicious is the purist’s bookmarking site and extra functionality is just not what it’s about. Which is fine by me.
Then there’s Diigo. Diigo allows you to highlight and annotate sections of a web page, and share those annotations with other users. Installing the Diigo toolbar gets you a collapsible sidebar in which all these annotations can be read side-by-side with the corresponding source material. This means that groups of people can collaboratively analyze documents online and asynchronously. It also means I can note specific passages that are important to me, so that when I come back to them later, when I’ve forgotten why I bothered to bookmark the site, I can just read my own annotations and remember what all the fuss was about. Absolutely brilliant, especially for scatter-brained eggheads like me.
In terms of extra functionality, Diigo runs away with the prize.
By shareability, I mean the ease with which bookmarks can be shared with others. Here, again, Diigo pulls out in front – just by a tad – by providing synchronization of Diigo bookmarks with some other services, including Delicious. I haven’t seen comparable services from the others.
All four systems support RSS feeds, and sharing via other mechanisms like email, or the formation of groups or other structures like them.
In terms of access speed, page load speed, and general zippiness, the only site that I found quite slow indeed was Twine. Not only did the pages take long to load, but there are often quite dramatic pauses between clicking on a link and getting any response. I know it’s not my ISP, because everything is behaving properly. Perhaps Twine’s AI features combined with its recent and sudden increase in popularity is putting a bit of pressure on their servers. I hope that’s all it is.
Still, speed is everything these days. Web 2.0 apps were originally conceived especially to improve (perceived) performance on the client side. So a slow Web 2.0 app, such as Twine, is really a contradiction in terms.
StumbleUpon has the fun factor, and Delicious has the brilliant look and feel. These are important. But to me, nothing can beat the raw functionality of Diigo’s annotation system. Twine, unfortunately, just didn’t make the grade.
So, while your mileage may vary, I’m convinced Diigo is the cream of the crop for me. And since I sync my Diigo bookmarks to Delicious, I also have an excellent “Plan B.”
I also continue to check updates on both Twine and StumbleUpon, because they’re different from what Diigo offers, and I occasionally find some really cool sites that way.
Sometimes, strange things happen on the Internet. One of these things is Compete.com’s graph of unique visitors to Delicious, Twine, and Diigo. The sudden spike in activity for Delicious relates, I think, to their offering connectivity with Twitter. Twine’s increased activity, however, is different; it’s more gradual and shows all the marks of classic exponential growth (i.e. “going viral”) that mark word-of-mouth transmission. The dip in the past month could be an anomaly – after all, some people just get caught up in things when they go viral, and eventually come the other side saying: What was I thinking? – or it could be something else; we’ll just have to wait and see.
It’s the exponential growth that gets me. What in the world got into people? I’ve been experimenting with Twine since long before October 2008 when the growth started, and I noticed nothing. Sure, there were some changes to Twine’s user interface a few months ago, but they seemed pretty minor to me.
And then there’s the Alexa rank data for all four bookmarking sites, which you can see at the top left of my Computers tab on Pageflakes. Since it’s rank that’s displayed, the low number is best. Notice that Twine is the lowest rank, and StumbleUpon is first. Also notice that something happened just before April. This tells a different story than does Compete.com. And I can’t find any explanations anywhere.
So much data; so little information.
Lesson: I didn’t know what I really needed till I played with these systems. You can rarely solve a real problem just by thinking about it – no matter what they taught you in high school. To solve real-world problems, you have to act, to do things that poke at the problem. This sets up a feedback loop between the problem and your brain that let’s your brain work both consciously and unconsciously on the problem. This means you’ll solve the problem faster.
So when you’re looking for any kind of system, take the time to play with the alternatives and take the time to reflect on the good and bad points of each. Make notes. Then weigh the pros and cons of each and make an informed decision. It’s harder and takes longer than just choosing something, but the upfront cost will more than pay off in increased productivity (and fun too!) later on.
Also see my last post about my web presence: A First Attempt.