Wiki/Email/Blog

AlwaysOn: What are the top five features or qualities that are driving people to use JotSpot?

Joe Kraus: There's a huge DIY content trend that's been going on for 10 years—actually, it's been going on for hundreds of years, but it's really accelerated in the last 10 to 12 years. The web has hugely accelerated the DIY content trend because now anybody with a web server and some knowledge of HTML can publish. Blogs further accelerate that trend because now anybody who can type into Word can publish on the web. Plain-vanilla wikis and these kinds of tools are also part of this trend, except that they're focused on not one-to-many, but many-to-many publishing.

So I don't think it's a particular feature that's driving their use but rather this trend toward DIY content publishing. This is a really powerful trend—one that's been playing out over the last dozen years and will continue to play out in the future.

JotSpot founder Joe Kraus explains how the web-enabled trend of do-it-yourself (DIY) content is driving new kinds of collaboration.

AlwaysOn: What are the top five features or qualities that are driving people to use JotSpot?

Joe Kraus: There's a huge DIY content trend that's been going on for 10 years—actually, it's been going on for hundreds of years, but it's really accelerated in the last 10 to 12 years. The web has hugely accelerated the DIY content trend because now anybody with a web server and some knowledge of HTML can publish. Blogs further accelerate that trend because now anybody who can type into Word can publish on the web. Plain-vanilla wikis and these kinds of tools are also part of this trend, except that they're focused on not one-to-many, but many-to-many publishing.

So I don't think it's a particular feature that's driving their use but rather this trend toward DIY content publishing. This is a really powerful trend—one that's been playing out over the last dozen years and will continue to play out in the future.
Kraus: DIY, do it yourself. Even though there's been a long-term trend toward DIY content publishing, there hasn't been any real advance in DIY application publishing. That's what's interesting and what we're trying to do at JotSpot: to extend that trend not only to publishing documents but publishing data and documents together.

People will be driven to use these types of tools by market need. The bottom line is that collaborating over e-mail with Word documents works; you can get by with it. But everybody seems to acknowledge that this form of collaboration is broken—that it's really hard to remain coordinated as more and more e-mail is passed around, and more and more Word docs get locked away in hard drives. It's really hard to bring people who are outside of the communication flow up to speed. How do they know what's happened, what people have talked about? People are experimenting with tons of different tools right now: They're experimenting with blogs; they're experimenting with wikis.

Another thing that drives people toward particular solutions is ease of use. One thing that people really like about JotSpot is that you've got a browser and Microsoft Word-style editor, and it all integrates into the flow of what you do with Word.

The web is about giving individuals control. I think that's what a lot of this is about. It's about removing the barriers to controlling content. When you look at the traditional internet, what's really a bummer is that 99% of it is outside of the user's control—that is, individuals need to get permission to make any changes.

But what's really interesting is that whenever systems stop asking for permission, participation increases. Take the web: It would never have taken off if you'd had to get permission from your internet service provider to put up a website—that's just too cumbersome. What's interesting about wikis, in particular, is that they enable people not to ask permission. As a result, you get a lot more participation—whether it's on an intranet or a team project or some kind of collaboration.

One interesting thing to come out of this is that you're getting a lot of pressure from the bottom of companies or groups (as opposed to pressure from the top) to use these tools. That's what I think makes this trend interesting: It's taking shape at the bottom and driving its way up, as opposed to being sold at the top and pushing its way down.

AlwaysOn: So more content is being created and more ideas are being generated?

Kraus: Yes. I often think back to my Excite days when I was managing a group of 450 people: I know a lot of them had great ideas, but how could I ever hear about them? The answer is, if you had an idea, you had to advocate. You had to fight to get someone to listen. So what do you do? You just don't even bother.

But here's what's really interesting about these DIY content tools: That engineer who was trying to get my attention could just put it on a page and say, 'Here's a list of things that I think we should be doing.' It's much more effective to gather ideas, gather information from the bottom because when you reduce the barrier to publishing, interesting things come out—ideas that had locked away in people's heads because it was too hard to get anything published.

AlwaysOn: This also creates new challenges of processing more data ...

Kraus: JotSpot has a blog application in addition to its wiki application, and what we've found internally is that the wiki has thousands and thousands of pages in it—which means that looking at what's been recently changed or added is not a useful exercise. There's too much information: Some of it is too granular (meaning I don't actually care about it), and some of it consists of minor changes to existing pages (which I don't need to know about).

What we've found is that you use the wiki as a giant content repository and safety net. Everything we record about the business lives there—the documents, the documentation, the marketing plans, the project plans, bugs, feature ideas. You need it; you know where to go to get it: a single, centralized repository of information that anybody can add to and change.

In contrast, we use the blog as a way to highlight five or six things we think people will find worthwhile. That's how I think these tools interoperate: You use the blog as a town crier—'Hey, I really need you all to notice this.' And the two tools work together in the sense that the blog references a wiki page, and that wiki page gets included right in the post.

Finally, in looking at the evolving style of communication inside of a little place like JotSpot, we use e-mail for urgency. So we use the blog for highlighting, the wiki as the repository, and e-mail to convey urgent messages.

AlwaysOn: But who's in control of the blog?

Kraus: It's a multi-author blog, so everybody can post to it. What we now put in the blog, we used to send out over e-mail. But that wasn't archived. New people would join, and they'd have no way to see what people had previously shared. The blog provided a much more effective vehicle for archiving that information. Plus, every post to the blog automatically becomes a wiki page, which means everything gets archived automatically.

It's interesting to note in one small environment the balance between e-mail, instant messaging, blogging, and wikis. What we're seeing is that each serves a purpose. We're one of those places where I'll instant-message someone sitting no farther away from me than you are now—to send a quick note in a point-to-point way that needs to be handled immediately.

I rarely IM programmers, though: The most important thing in programming is to let them stay in the flow state. This means that the most productive thing most people can do is actually turn off their IM client. So we're pretty judicious about IM use. I tend to e-mail questions to engineers because I don't want to interrupt them. How can you not use e-mail? We actually use very little e-mail internally, but you've got to use it for stuff externally.

Part 2 - Jotspot: do like excel

When you look at the web today, you'll see that web-based applications are still kind of hard: You don't do many of them, and you certainly don't do them casually. But when you take an Excel spreadsheet and make a database out of it—putting things in cells that aren't formulas such as vendor lists and partner lists—and then e-mail that spreadsheet to people, you're doing DIY application publishing and you don't even know it. You've built an app (the Excel spreadsheet); you've published it (in this case over e-mail)—and you're doing something that you could do much more efficiently if you had the right tools.

I really believe that JotSpot's mission over time is to transform web-based applications the way that Excel has transformed financial apps. We want to enable end-users to build these ad hoc, one-time-if-necessary web-based apps. And we want to enable a lighter-weight class of programmers to do a lot more than they ever could before.

As one example, we'd like an HTML person to be able to build a relatively complicated recruiting application—fully featured—that he or she could then resell on top of JotSpot. And this is just one of the many applications that could be built on top of JotSpot in a prepackaged way. Just as cells are Excel's main metaphor, JotSpot comes out of the box looking like a wiki with pages and links as its main metaphor.

AlwaysOn: Let's talk open web vs. closed web.

Kraus: What does that mean?

AlwaysOn: By allowing people to come in with spreadsheets and other applications that work with JotSpot but weren't built with JotSpot tools, you've made JotSpot an open-web product—which is in contrast to something like Google where you get what you get and you can't integrate other applications.

Kraus: There's been a lot of hype around open source, and there are people who believe that open web means open source and anything less than open source is not open web. I don't believe this. Open source is a really important movement—it's powerful in many ways, and we certainly benefit from it and contribute to it—but there are also a lot of interesting examples of late that take a different view of open web. These examples point to open data, not open source.

Take Flickr: Its code is closed, but there are enough APIs in and out of it that you have this tremendous ability to tinker, even if you don't control it. JotSpot, too, is very much an open-data-style service. Just as with Flickr, there are tons of APIs in and out. You can manipulate content inside of JotSpot even if you don't have source-code control.

To me, open web really refers to how tinkerable it is. It's really a spectrum of tinkerability. Some people believe the only way you can tinker is via open source, but I believe the new trend (especially in web-based services that are hosted by somebody else) is to facilitate tinkerability even without open source. That's why I think Google is not actually totally closed: They offer APIs into their service. Now, they could offer more ...

AlwaysOn: But do you think we're at crossroads where the Googles and Yahoos are the new AOLs?

Kraus: An example of something in search that's more open data would be A9, which has this nice RSS interface where you can get your results bundled in over an open standard. That to me is really interesting because it's taking the concept of search open data to another level.

AlwaysOn: With all the closed web data, it's like the Google guys bought tickets but don't want to go on stage, and Microsoft is the one being overwhelmed.

Kraus: Well, this is my theory: In the mid to early '80s, IBM was the feared Goliath. Then, Microsoft took that position in the late '80s and mid '90s, and IBM became the gentle giant. Now, I think Microsoft is becoming the gentle giant, and Google is becoming the new Microsoft. And I find myself strangely rooting for Microsoft in different places now—and I've always been suspicious.

Now, I'm not rooting against Google, but I do think there's a mantel passing. Google is the Microsoft of the mid '80s—incredibly powerful, incredibly capable, with a huge amount of engineering talent and capable of being in any business. Every entrepreneur knows (or should know) that there are five engineers at Google working on whatever he or she is doing.

But this is what I find really interesting: There's a transparency at Microsoft now that people understand that, yes, it has a monopoly on Windows and it's trying to maintain that monopoly and print a bunch of money. There's an industry shift going on, and I suspect is it means that Microsoft is going to do some more interesting things in this space. In the past, I would never have though Microsoft would do open-data stuff, but now I think it's going to move in that direction.

AlwaysOn: They've been pretty open about blogging; they want 20 bloggers to blog about Longhorn.

Kraus: Some habits die hard.

AlwaysOn: Yes, it's funny. But if you look at Bill Joy's 'six faces of the web,' Microsoft only owns one face—the desktop, if you will—so it's a wide-open game. I think people feel like there are a lot of things to do and still stay out of Microsoft's way. With Google, like you say, everyone's just worried that they're going to get into their business models. And the poor VCs: They all have Google envy, and they're suffering.

Kraus: They have tremendous Google envy. It's like OK, you didn't do the deal; you've got to move on: You're not going to find another one of those, sorry.

AlwaysOn: But if you're Tim Draper, you think they're just around the corner. It all depends on your attitude. These VCs are different kinds of people: They grew up to jump through all the right hoops, and those hoops were pretty identifiable. Prestige is very important to them. But now they've been given this new hoop, the Google hoop, and they feel like they're never going to be able to jump through it ...

Kraus: Right. And I'm just curious, did the same thing happen when Microsoft went public and it started becoming clear that those guys were worth a mint?

AlwaysOn: I think at the time (1986), Microsoft was only the fourth-largest software company, so it became the big early giant on a much more gradual trajectory.

Part 3 - Consumer Enterprise Model

The internet is radically blurring the distinctions between whole classes of software. You're not going to buy an ERP system in the same way you think about Google Maps, but I think for a big chunk of software, the internet is totally blurring the line between consumer and business applications.

This is an important trend to recognize—and why I think of Google as a threat. Even though our first application is more business oriented, our goal is to make the end-user, not the IT person, our customer. We're sold to end-users, and those are the people we need to addict.

AlwaysOn: When you look at the people who are using JotSpot now, are they creating collaboration environments for fun or for profit?

Kraus: They come thinking, 'I want to do this for my project team business,' and within a day they're thinking about how they can apply it to their families, homes, etcetera. So crossover products are happening, and I think anything that deals with groups collaborating is particularly ripe for crossover.

In general, though, I don't see a ton of application in the business world for social networking sites. One exception might be Linkedin, which I think has more of a business focus, but I don't see any of the other social networking sites really succeeding in the business environment.

AlwaysOn: Right now VCs are betting on the wrong idea: They're thinking of social networking like the search engine business—meaning there's going to be three brands, which collectively will attract hundreds of millions of people. In reality, though, I think there will be hundreds of millions of social networks consisting of 150 people each.

Kraus: I agree. I'm always a contrarian as an investor, though. Everybody is beating the living crap out of enterprise software right now—everyone's saying that enterprise software is dead—and from a JotSpot point of view, I certainly agree that there's a different way of doing things. But if I were an investor, I'd be looking for opportunities in the traditional enterprise software business.

I think the problem with investing in general—especially in the VC market—is that something gets hot and you get venture fratricide up the wazoo. The upshot is that all of these overfunded companies end up killing each other, even if the market opportunity is real. If I were in a venture capitalist's shoes, I would try to make the difficult decision and find some ugly, hardcore enterprise software deals—not because I think enterprise software is sexy but because I don't think that anybody is looking there right now. Everybody's declared enterprise software dead, and whenever everybody declares something dead, that's the time I want to go for it.

AlwaysOn: Yes. Oracle is still in business, and everything's consolidating. I'm coming from the point of view, though, that if you're going to do something, you've got to do it in an open-web environment because the era of building little silos is over.

Kraus: It's gone!

AlwaysOn: Right. But if that's the case, are we now living in the post-Google, post-Yahoo era? Are Yahoo and Google going to be like AOL was in '95 and suddenly realize that the internet is open?

Kraus: I think they're smart enough to transform themselves and adjust.

AlwaysOn: AOL did for a while ...

Kraus: Kind of. The problem is that companies addicted to a closed model have a hard time switching to an open model. But I think the advertising business actually benefits from an open model, especially if you look at what Google and Yahoo are searching for, which is more inventory. They're trying to open up. AdSense is a classic example of opening up the economic model, the economic engine.

What's funny to me is that while everyone was declaring AOL dead in 1995, they were right—just 10 years too early. People see the handwriting on the wall; they just underestimate how long it's going to take to come true. So I think you're right that traditional companies will need to adopt open-data models, but I think it will take a lot longer than people expect.

AlwaysOn: Recently, I attended a demo of Yahoo 360 because Jerry Yang asked me to check it out. As I sat in this room full of hardcore geeks, I could see that they were all completely dissatisfied because it didn't pass the do-it-yourself test. It was like déjà vu—but with Yahoo instead of AOL.

Kraus: I think 360 is an interesting example of a few mistakes. Yahoo builds great products; I just don't think 360 is one of them. I don't really have good comments on open vs. closed, but I remember launching products like this with Excite, where you can tell in 360 the primary goal was to integrate with other Yahoo properties. So it's beautifully integrated but kind of useless at the same time—as opposed to a number of other properties at Yahoo, which are beautifully vertical but not necessarily super well integrated. It's clear, however, that 360 was meant to be a hub, and as a hub it doesn't satisfy a constituency that wants more DIY capabilities or is used to more power in other tools.

It's pretty clear to me that web-based programming is actually becoming more and more about integration and less and less about building anything. What's interesting about the open-data trend is that it reinforces this notion that it's going to be faster and faster to build things. But the real heat of web services will happen in the consumer space. In fact, you can already see it happening in the number of businesses built on top of Flicker and the ecosystems built around Amazon and eBay: This is the new model.

The trend in web-based programming is integration, and it's enabled by open data—that's the feedback loop. The more people who want to do web-based programming, the more they're going to want to integrate. This in turn will force companies to open up their data, which will enable even more web-based programming. That's how the engine gets started.

Part 4 - Inovation and Profit

AlwaysOn: And that person is going to be paying by volume, by big bandwidth?

Kraus: There are primary buying features and secondary buying features: Pages represent a primary buying feature, and SSL and a bunch of other things represent secondary buying features. In addition to the page pricing structure, there are also bandwidth and storage limits—which again start out free and increase incrementally.

AlwaysOn: Do you have an e-commerce system in place to automatically calculate and charge for these things?

Kraus: Not yet; we're building it. What you have now in your JotSpot account is a thermometer that essentially shows how many pages you have now and (based on your current 30-day velocity) the number of pages we expect you to have at the end of the month. This allows you, the end-user, to make some decisions: Do you want to delete pages? Or do you just want to change levels, moving from the 200-page plan, say, to the 750-page plan?

AlwaysOn: That's cool. Obviously $49 a head wasn't ...

Kraus: We never charged $49 a head; we were $5 a head.

But I think the question is, which assumptions are correct? Do you believe you can actually get $5 per user per month and do a self-service business model and not have a sales force? Because the truth is, you can't afford a high-priced sales force at $5 a month; I don't think you can make your user projections. If you look at Socialtext—they're the ones who charge $40 a head—they can afford a sales guy. But it means you're not going to have many customers. I like high-volume, low-priced businesses that go after a mass market.

AlwaysOn: Is there a limit to the amount of people who can join a wiki?

Kraus: Not in this model.

AlwaysOn: So people will have multiple wikis within large organizations?

Kraus: Yes, I think so.

AlwaysOn: By groups and stuff like that?

Kraus: Yes.

AlwaysOn: And they'll eventually interoperate?

Kraus: Yes, you can link between them.

AlwaysOn: If you were to look at a pie chart three years from now, which slice do you think would pay the lion's share of your bills—consumers or corporate expense accounts?

Kraus: I don't have the slightest idea.

AlwaysOn: Doesn't that make you nervous?

Kraus: Not really. We think there's a huge opportunity here—Graham and I wouldn't have come out of the great and easy lives we had to bust our balls to do something small. It doesn't worry me terribly because I think the market will tell us. And the truth is, JotSpot isn't just a wiki: We've got a wiki application, but we've also got the blog application, project management, contract management, time and expense, and consumer applications—and we're opening it up so that developers can build a bunch of other applications and charge for them directly.

The idea is that JotSpot ends up getting marketed vertically. When you build a system as broad as JotSpot, it's difficult to predict what killer thing will be built on top of it. In the history of things like platforms, it's rare that the guy who creates a platform knows what the killer app will be.

AlwaysOn: So that's open web ...

Kraus: I think so. I think it's totally open web.

AlwaysOn: What other exciting things are going on at JotSpot?

Kraus: We recently did this thing called a hackathon, which is a really effective way of reintroducing innovation. Startups are supposed to have two advantages over large companies: They're supposed to be faster and more innovative. But I asked myself the question, is that really true? And I think the answer is that a lot of startups—especially as they start to get a few customers—stop innovating because the become so focused on customer-driven development. They get a year-round product roadmap, and suddenly every engineer is locked into it, and nobody is thinking creatively.

So we did this thing called a hackathon—a day-long event in which we told all of our engineers that we were going to start at 9 AM and go to 8 PM, and at the end we would gather to show off what we'd done. The criteria for the hackathon were that 1) you had to work on something that could be taken to full prototype in a day; 2) it couldn't be something you were supposed to be working on; and 3) it had to be useful to the company in some way. The work that resulted—the creativity—was unbelievable.

People came up with amazing things that would never have gotten built in the normal scheme of things, and there's at least one product to come out of it that we're going to be launching, that gets built into the service. The whole idea of the hackathon is related to the Google notion of taking 20% of your time to work on things that aren't what you're supposed to be working on but will still help Google. This is a great idea, but I don't believe it works on a steady-state basis in startups because you never really take that 20% of your time. That 20% quickly gets filled up with more pressing things. If, however, you sanction an entire day for that type of activity and do it, say, every six weeks like we plan to and get users involved, it's a great way to keep innovation and entrepreneurship alive.