Jason Fried says that the most innovative software in the next 10 years will come from companies like his, those who build web-based applications for very small businesses.

Though much of Jason’s mantra seems to be “do it our way”, I think that most of what he’s saying dovetails nicely with the notion I’ve been playing around with: tryability.

Jason says: “What they (independent freelancers) crave are low/no-learning curve, simple focused tools that let them get their work done quickly and then get out of their way.”

Note the low/no-learning curve part, I think that’s key. This is exactly the feature that people need when they can’t find an immediate answer to the question: “how is this application better than the one I’m currently using?”.

It’s not that people aren’t smart enough to figure this stuff out on their own. Given enough time, anybody can do anything, or close to it. It’s that we lack the window of attention to do all the things that we want to do. How long, do you think, would it take to evaluate all desktop email applications right now? One day, a week, a month? The point is that nobody is going to take the time to find out!

So, web apps get rid of one major hurdle that steals time: downloading, installing and upgrading. This is a main point that any web-based app provider makes, and it really makes a difference.

What makes even more of a difference, however, is having No Learning Curve. If I can start using an application immediately, and find it useful, then I’m a customer. The fact that I don’t have to install anything is great, but it doesn’t show the value of the application. Having No Learning Curve does.

Ideal Tryability = No Learning Curve

Sometimes we spend a lot of time and effort doing various things that we don’t want to be doing because they just have to be done:

1. Between yesterday and today I’ve spent 5 hours trying to get my wireless network working with the VoIP box
2. Between yesterday and today I’ve spent 3 hours doing various things with my car, taking it to the shop, calling about it, explaining how bad it is driving recently, etc.
3. Recently I spent 8 hours trying to get rid of viruses on an old windows box at home. So frustrated, I gave up and asked Jason from work to do it, promising him a share of the money I make on it when I sell it on eBay

Sometimes we refuse to spend a lot of time because they just don’t have to be done:

1. Trying out new things when we have a decent one to begin with
2. Taking the time to install new software that we’re not sure is useful yet
3. Taking the time to learn something new when doing it the old way takes x amount of time
4. Doing something the old way after we’ve learned the new way
5. Seriously considering a new idea when it potentially conflicts with our established thoughts

Tryability is a new term I made up to represent the pain of trying something new. In a similar way that “usability” is a measurement of how usable something is, “tryability” is a measurement of the effort involved of trying something new. It is made up of several factors:

• learnability: how easy is it to use for the first time? (or learn new features)
• usability: how easy is it to use over time (includes learnability)
• effort: the time and energy involved in trying it out
• attention: the attention it takes to find and learn about it (as opposed to learning to use it)

These things all overlap, of course. The reason why I’m using the new term “tryability” is that it involves an attention factor…things get much harder when there are other things to attend to. In other words, we have to make an effort to divert our attention to try something. We could probably house this under “usability”, but usability testing is often conducted in controlled situations…which don’t include fractured attention.

For web applications tryability is crucial because tools are making it so easy to post information online. How many ways are there to make a web page? A million?

As a sort of follow-up to my post on the interface elements for subscribing to feeds, I’ve provided two feed URLs below in order to see what various platforms do with them. One is in the regular http:// format, utilizing the http protocol, while the other one uses the feed:// pseudo-protocol.

• feed://bokardo.com/feed/
• http://bokardo.com/feed/

Specifically, it would be nice to know what platforms support feed://, so we can migrate away from those instances where people click on an http:// feed and get a bunch of XML that doesn’t mean anything to them.

Also, I know you’re already subscribed to the feed, but test it out on your platform anyway, so we can see what works and what doesn’t. Note, there may be different effects depending on where you’re clicking on them. For example, the action taken when clicking within a web-based screen reader might be different than a desktop reader.

Let me know how it goes…

Steve Rubel over at Micropersuasion (a site I pay attention to) has written a letter to Microsoft: Dear Microsoft, I am Dumping You. It’s an interesting read simply because Steve took time out to say this in such a public way. (he’s got a big audience)

His reason for switching is that he’s finding more value in Web 2.0 applications than he is in desktop applications that aren’t part of the Web platform. He mentions Gmail, Flickr, and Backpack. These applications, of course, are web-based and therefore can be run on any machine with a modern browser. And Steve’s browser of choice will be whatever he’s running on his new Powerbook.

Cross-reference this sort of activity with the recent gains by Apple in overall marketshare. They’re up to 4.5% of new computers shipped, which is higher than they’ve been in years. Note, however, that computers shipped only means computers sold, so that the install base is potentially much higher. In other words, percentage of computers used doesn’t depend on computers shipped, it depends on where people actually spend their time. I’ve heard claims that Apple’s installed base might be as high as 16/17%. I mention this because many folks I know that use macs tend to wear them out, using them for 4 or 5 years. My Dad had an original iMac for 5 years before he upgraded…

Regardless of what side of the OS fence you’re on, Steve’s post is really about the Web as Platform. Now that we can do most of our work using web-accessible stores of information, we don’t need to be bound to Outlook, Mail, or Eudora, for example. We simply use whatever interface we desire on top of a web service like Gmail, or even the Gmail web interface itself.

From a pure strategic point of view, this benefits Apple big time. The question becomes, if it doesn’t matter what OS you get, why not get the most stable one that runs on top of the best hardware? Since the OS lock-in is going the way of the Dodo, we’ll all have more choice about which hardware we use. That’s another great benefit of the Web as Platform.

Rubel says, in a final shot to Microsoft: “start embracing the Web as OS”.

And I say: “Amen to that”.

You’ve known about it for a while, but now it’s beginning to go mainstream: Marrying Maps to Data for a New Web Service. (may need to login)

Not really new information, but a nice summary. Mr. Markoff obviously sees the potential in these mapping services. So why is mapping so popular? It’s one of the most basic human condition that we want to know where we are, and we want to know where we can find things.

Also, I was trying to add all the new services to my google maps page for a while, but it became overwhelming in a short period of time. Luckily, another enterprising person has created: Google Maps Mania. Just look at the right hand navigation bar to see the innovation going on…

What do you want to map?

My latest article is up: Using Ajax for Creating Web Applications.

Most ajax articles have been about the technology itself, how it isn’t anything new or how to get up to speed using it or how it actually works. This is great information, of course, but it has been written about everywhere. So I went a slightly different route.

For UIE’s audience, made up more of usability professionals than techies, I wrote it from an angle of a design team choosing what web application approach to use. Should they go with the sophisticated Java/Flash approach or the web-standards, paged-based approach? With Ajax, we might have a middle-of-the-road approach that combines the best of both.

This is interesting…two days after I wonder whether there will be/are API-only services, Joseph Scott asks the same thing. He thinks “yes”, and offers an interesting example: spell-check.

Obviously, there are things that we should only need to implement once. APIs let us do that, by implementing the functionality on our servers and then offering it to others. If someone were to create a spell-check API, then conceivably everyone in the world could simply hook into that. Gives a lot of power to the spell-check service provider, but that’s what companies are looking for, right?

That’s why Google and Yahoo are fighting like mad to give you services, even if they are free right now. They want be the service of choice. I doubt they will be free forever…

In general, there are two kinds of interfaces in Web 2.0

• API (Application Programming Interface)
  This is the “web as platform” interface. Companies with data stores (eBay, Amazøn, Google) create APIs for developers to hook into. In doing so they provide public access to information that would have been private previously, completely changing the way that people interact with the site. The Google and Yahoo Maps APIs are two of the latest to be released. These are the building blocks of web services.
• AII (Application Interaction Interface)
  This is the interaction interface that is built on top of the APIs, and is most often called a web application. Developers build Web 2.0 interaction interfaces to provide value above and beyond what the solitary sites can add themselves.

Andale.com is a great example of an interface built on top of an API, the eBay API. On Andale you can track the selling activities of products previously sold on eBay. This allows a much clearer picture of what items actually sold for, and it is functionality that eBay doesn’t provide itself.

Interestingly, most Web 2.0 companies start out with both interfaces. Amazon, for example, had the Interaction Interface years before the Programming Interface. I wonder if there are any companies out there building just the API, without providing an Interaction interface. We’ve got lots of sites with just the Interaction Interface, but few with only the Programming Interface. (Andale is just the Interaction Interface, of course…but I could imagine them adding even more value with their own API.). Also, it could be argued that an RSS feed is an simple API, of sorts.

Over time, companies with valuable content might just have APIs to that content, without providing their own interface. After all, interfaces are hard, why not let other people create them? You’ll still be able to license that content assuming it is providing value.

Know of any API-only sites? | Bokardo Interface (AII)

In order to remix data into a new interface on Web 2.0, you need access to the data as well as access to functions that can be done on that data. Current web standards provide us with the tools for both: XML for the data and Javascript for the functions. Thankfully, these two technologies are what web developers have been using for years, in various flavors.

Google Maps is the current Remix King (e.g. Gmaps Pedometer), and it is built this way. In the newly released Google Maps API, we see these two things at work. Consider the following code:

var map = new GMap(document.getElementById("map"));
map.centerAndZoom(new GPoint(-122.141944, 37.441944), 4);

This code creates a new map and centers it on Palo Alto, California. This is the function part being done in Javascript. The element in which the map is placed looks like this:

<div id="map" style="width: 500px; height: 300px"></div>

This is the data part being done in XML (XHTML in this case).

For hardcore coders this is a breeze. For most folks, however, (like web developers beginning to dip their toes into the remix waters of Web 2.0) this might be a refreshing sign. We’ve already got the requisite skills to remix Google Maps, or any web service that plays nicely. It’s plain old tag-based markup and Javascript.

If you didn’t know it already, go check out the Google Maps API. It is the first official API for Google Maps. Also note that Google has both an API for its desktop search as well as an API for its Adwords campaign.

This is a huge Web 2.0 move. Provide the tools for remixing and lets folks do it themselves. APIs become the “services” of the Semantic Web by way of their openness. When developers can latch onto parts of the maps service via code, they’re doing so because the API has provided the necessary semantics. If the API didn’t provide the semantics (if the API didn’t explain what everything was), then the developers would have to hack the thing. The problem with hacking, of course, is that the company may change the API at any minute. With an “official” API, changing is less likely.