Categories
- Benefits (5)
- Company News (40)
- Enterprise 2.0 (107)
- Information Management (23)
- Keep It Together (8)
- Product Announcements (36)
- Productivity (15)
- Software Development (31)
I think it was roughly a year ago that Gordon and I realized we needed to integrate with Microsoft Outlook. We looked up from our laptops and stared each other in the face. We both shared the same terrible, sinking feeling.
It’s hard to describe to non-programmers, or even to non-Microsoft programmers, what integrating with Outlook is like. The best analogy I can think of is trying to read Chaucer’s Canterbury Tales in its original Middle English. The innards of Outlook are archaic. Much of the dos and don’ts of modern application design were learned — the hard way — by the Microsoft Outlook developers over the course of decades. And to my knowledge, Microsoft has never attempted a modern translation of the classic. They’ve put new covers on the old book — improved the look and feel, swapped the icons, etc. — but they’ve apparently left the core exactly as it was written, apart from bug and security fixes here and there.
A page from The Canterbury Tales
It’s a difficult job. The only other company I know of that’s doing something similar is Xobni, and they have a much larger programming team — Investors have poured several million dollars into their pockets. And that’s allowed Xobni to give away their Outlook integration for free.
What Gordon and I realized was that not only did we have to integrate with Outlook, we had to do it with a smaller team, on a tighter budget, against a competitor that had foolishly already set the price to zero.
Pure madness, then. But we had to do it anyway.
For an entire generation of knowledge workers, email is collaboration. Regardless of its merits as a collaboration tool, it’s the one they’ve been using for decades now, and it mostly gets the job done at small scales.
So if you want to get these knowledge workers to try something different for collaboration, you’ve either got to get them to switch away from email or you’ve got to integrate with their desktop email client.
There are many, many folks trying to get people to abandon email. There are numerous instant messaging applications, wiki solutions, social media and networking sites, blogging and microblogging tools — on some level, Web 2.0 could be viewed as nothing more than alternatives to email. And thanks to services like Facebook and Twitter, home users are becoming more willing to move away from email as their primary means for communication and collaboration.
But email is the way things are done in the enterprise, and Microsoft owns the enterprise desktop, and Outlook is their email client.
If we want to help people work together, Infovark has to integrate with Outlook. Even if the engineering challenges are large, the competition has more resources, and nobody is likely to pay us a dime for our effort.
Welcome to the sad economics of software start-ups.
If there’s a core architectural principle behind what we do at Infovark, it’s this: Go where the information lives.
At a time where web applications and cloud computing are all the rage, we’ve gone to the desktop instead. Why? Because that’s where the information is. And most of the information on the desktop is locked away in the email client, the largest information silo in modern business.
Infovark digs out email, contacts, tasks, appointments and meeting information from Outlook, relates these items to each other and to the files on your computer, and shares what it finds with you and your peers. Our primary means to do this is through our web interface, but we knew we had to show this information in Outlook as well.
Here’s some early screenshots of Infovark at work in Outlook. These are works in progress.
Image 1: Infovark in Outlook
We’ve also got a tool bar that will allow you to add or remove email from Infovark’s brain. You can also apply ratings to the email or share it with coworkers if you like. (The icons shown here are temporary.)
Image 2: The Infovark Command Bar
And finally, here’s a closeup of the Infovark task pane. You can add tags, search the items in your Infovark library, and view related email and attachments. Again, the icons are temporary.
Image 3: The Infovark Outlook Task Pane
Eventally, we hope to add and expand on what you can do within Outlook. We’d also like to extend these features to the rest of the Micorsoft Office suite. But our primary focus will remain on the Infovark web interface. We’ll be posting screenshots of that in upcoming posts. So stay tuned, and send us your comments!
A few items I found this week on the noisy Internet started me thinking a bit about our industry, and why content management is actually interesting. No really, stay with me on this one….
When I was a kid, I always figured that one day there would be a super awesome computer that I could ask to do anything and it would instantly comply. Factual things, like “How far is it from here to the moon?” tasks like “Design for me my own awesome personal car based on things I like” or “Do my homework” and other, more esoteric things like “Does that girl in Math class who keeps poking me with her compass really like me?”
Now that I’m something of a grownup, we’re a lot closer to that kind of a solution actually existing. In fact, the Internet could potentially do pretty well at all of these questions. And yet, things aren’t exactly as I imagined them.
For starters, when I imagined my computer, I figured it would, you know compute things. Like, it would somehow measure the distance between here and the moon. And that it would build me blueprints for my new car by computing complex algorithms and lots of, you know — computery figuring out of stuff. The magic of the Internet as we know it is not really powered by computation at all. It’s people powered.
At its heart, the most useful things on the Internet are all just content management. They’re reordering, re-indexing and re-presenting existing information. For its almost limitless usefulness, Google itself doesn’t “know” much of anything. The only thing Google actually computes is how to relate existing content in the form of webpages, and how to present it at the appropriate time. (That’s no easy task, either, by the sounds of it. Just look at the amount of power they consume to bring those results to us.)
Which is why I was particularly intrigued to hear of Stephen Wolfram’s new project, Wolfram Alpha. It looks like this is an effort to produce a computation-based system, more in line with the computer of my juvenile fantasy. Unlike Google, Wolfram Alpha plans to actually compute answers, not just find them. It’s taking a seriously hardcore computer science approach to the problem of knowledge. I would dearly love to see this thing succeed, but I suspect that it won’t live up to pre-launch expectations.
At the other end of town, the other thing that got me thinking was Amy’s new project for South by Southwest (SXSW), the Pepsicozeitgeist. It’s a near-realtime, twitter-powered look at interactions between people at the conference. (It’s largely inspired by her original Twistori site, which does much the same thing for Twitter.) As well as being a fascinating time-waster, it’s a classic case of remixing people and their content, slicing and dicing information from the twitter data cloud. The computation going on in these solutions is all in looking for patterns in the content — finding the best ways to relate things — in divining commonality and relationships, and counting results.
In fact, when I look at all the amazing products and innovations that have arrived since web 2.0′clock, they all share this common element. Reasonably lightweight, simple computations being quickly processed against a collectively valuable, up-to date content repository. That’s what lets you find things, and learn new things, make new associations. And underneath it all is Content Management. Regardless of whatever it used to be about, it is now about preserving, maintaining and making these data repositories current, available, and interoperable so that value can be derived from them. Which is actually pretty interesting and important.
These new data libraries of information we are building for enterprises need to be constructed more like Twitter, and less like virtual paper archives if they are going to be useful to us in these ways.
At least, until some giant Hal 9000 type computer can do all our thinking for us. Meanwhile, I guess we should all just keep typing.
Do you want to create a must-have software application, a compelling new business model, and the next Internet darling? (I sure do!) Then Daniel Tunkelang offers some sage advice: when in doubt, make it public and permanent. He cites Blogger, Twitter, Flickr and YouTube as examples of previously private communication brought into the public sphere.
The lesson is that while security features remove objections in the customer’s mind, they don’t help you sell the product.
The first peril of social software is that nobody uses the application, and you’ve wasted time and money deploying it. The second peril is that everyone uses the application, and you’ve got to spend all your time scaling the system and filtering out all the noise.
Successfully navigating the dangerous waters of social software takes boldness and skill. Expect that you will need to help row the boat. And be prepared to lose a few sailors on the way.
Epic heroes don’t come through these tests unscathed. But they don’t let the wrecks of unsuccessful voyages scare them away, either.
Since you can’t make a case for Enterprise 2.0 using ROI calculations, it’s tempting to rely on anecdotal evidence instead. Once the storytelling begins, it usually doesn’t take long before social networking sites like Wikipedia, LinkedIn or Facebook come up. But there’s great danger in using these sites as justification for Enterprise 2.0 investment.
Consider Wikipedia. Looking at Wikipedia’s statistics and Alexa’s Internet traffic data, we find that out of the entire online population less than 10 percent visit Wikipedia during the course of a year. And that includes all visitors — the rates for people that contribute content to Wikipedia is a mere fraction of that.
The numbers for other large websites are equally depressing. Using social network statistics from Jeremiah Owyang of Forester from earlier this year, MySpace and Facebook have 8 percent and 4 percent of the online audience respectively. What gives? Aren’t these the leading lights of Web 2.0?
What makes these sites work is massive scale. You can take a “build it and they will come” approach if the population of potential users is large enough. In the case of the world wide web, we’re talking about 1.4 billion people according to Internet World Stats.
But in an organization of 100 people, you’d be fortunate to get 10 people using the system and extremely lucky to have 3 people contributing information. Would a user adoption rate of less than 10 percent qualify as success in your business? I don’t think so.
If you want your Enterprise 2.0 deployment to work, you have to focus on user adoption.
The hard truth is that if you want an Enterprise 2.0 application to succeed, you must be prepared to sacrifice everything to delight your customers. And these sacrifices will include treasured organizational imperatives. You’ll have to do things like:
You’ll find yourself in the midst of turf wars and political infighting. There will be squabbles over budget and resources. Change management and training will consume months, if not years, of your time.
Sure, you can duck these issues, and take a “build it and they will come” attitude. And if you are fantastically lucky, you might just achieve the same adoption rate as LinkedIn: 3 percent.