At the University of Illinois, Ray Ozzie worked on
PLATO Notes, one of the earliest collaboration applications.
Later he wanted to develop collaboration
software of his own, but couldn't find funding. After
he led the development of Lotus Symphony, Mitch
Kapor and Jonathan Sachs decided to invest in
Ozzie's idea, which would become Lotus Notes. Instead
of working as an employee, Ozzie founded Iris
Associates in 1984 to develop the product for Lotus.
It was an unusual form of startup, but it worked.
Lotus Notes was the first widely used collaboration
software. The first release shipped in 1989, and
Iris was acquired by Lotus in 1994.
In 1997, Ozzie founded Groove Networks, which built Internet-based workgroup
collaboration software. Microsoft acquired Groove in 2005 and named
Ozzie chief technical officer. In June 2006, he took over as chief software architect
from Bill Gates.
Livingston: When you started Groove, where were you and who was there?
What was the first piece of code anyone wrote for Groove? What did it do?
Ozzie: When we first started Groove in the fall of '97, we worked out of my
house. Initially, it was my brother Jack, Eric Patey, and Brian Lambert. A few
weeks later, we moved to an office space at the Cummings Center in Beverly,
Massachusetts. A couple months into the project, another former Iris engineer,
Ken Moore, joined our team. The first thing we coded was a primitive version
of our synchronization algorithm.
Livingston: How did you come up with your ideas?
Ozzie: The common theme to both Iris and Groove was the fact that the ideas
were not based on technology, but on a need I saw for users or potential customers
for the product. I'm an engineer by training and I tend to be one of
103
Ray Ozzie
Founder, Iris Associates,
Groove Networks
7
CHAPTER
these people who believes he can accomplish basically anything in software--
it's just a big toolbox. So if you know that you can accomplish anything you set
your mind to, what's worth accomplishing?
I've never taken the perspective of "build a cool piece of technology and see
where it goes." It's more or less been based on an intuition about a hole in the
market--or, more accurately, a future hole in the market. At any given time,
you've got to have a technology roadmap in your mind and a market roadmap as
to where things are headed--broadband is getting increasingly pervasive or
wireless is getting increasingly pervasive, or something is going on--and trying
to project out several years, because it will take you several years to build anything
that's worth building. So you don't want to fill today's needs, but try to
capture some window that will happen in the future.
In Notes, it was (and this is hard to imagine because it was a different time)
the concept that we'd all be using computers on our desktops and therefore we
might want to use them as communication tools. This was a time when PCs
were just emerging as spreadsheet tools and word processing replacements,
still available only on a subset of desks, and definitely no networks. It was '82
when I wrote the specs for it. It had been based on a system called PLATO that
I'd been exposed to at college, which was a large-scale interactive system that
people did learning and interactive gaming on, and things like that. It gave us a
little bit of a peek at the future--what it would be like if we all had access to
interactive systems and technology.
With Groove, it was an observation that the nature of work was changing.
Technology at that point had largely been applied to helping people work
together within corporate boundaries. People were increasingly going to be
challenged trying to apply that same technology across boundaries, because you
can't control the technology chosen by your business partners. I might choose
Notes, you might choose Exchange, the other person might choose someone
else.
We saw a lot of frustration when our customers tried to deploy systems
across enterprises. So we came to the conclusion that what we really needed
was to build a system that just worked instantly, right after download, for the
end users.
Livingston: Were you trying to basically build a "better Lotus Notes" for the
Web?
Ozzie: Lotus Notes ended up being a multifaceted piece of software; it had
email, it was used for collaborative workspaces for people to do dynamic work
together. It was used as a content management system, as an application server.
Groove was really meant to fulfill just the collaborative workspaces piece.
We were laser-focused on the notion of people needing to dynamically assemble
in a virtual environment, share documents and their thoughts in order to get
work done very quickly, and then disassemble. In the work environment now,
increasingly you have to work with partners or customers directly, and this concept
of rapidly forming virtual workgroups would be an increasing challenge
and opportunity moving forward. The Web itself on the open Internet is an
alternative way of doing this, but we were really targeting people who needed
104 Founders at Work
to work in a highly mobile fashion, behind the firewall, outside the firewall, and
in a secure manner. So we went for a desktop architecture.
Livingston: So this is a big problem that you were approaching. How did you
start?
Ozzie: Before I start a company, I typically write a couple of founding documents.
One of them is very outside-in: it's a scenario-based document, describing
the high-level challenge that I'm trying to address and the end user
scenarios that we are trying to solve. This attempts to explain what we're trying
to accomplish to anyone who joins the company or we might need to get financing
from.
Then I create a second, bottom-up document describing the different technologies
that will have to be assembled to accomplish that vision.
The first thing we did in both Iris and Groove was get a big open office and
recruit a core team of people. Generally these were people I'd worked with
before, so I wouldn't have to get past the trust issue involved in understanding
what they are good at and what they're not.
And we'd just sit down with the whiteboards and just try to work through
some of the more difficult algorithms, make key tooling decisions. Early on in
Groove, we had a very big decision to make: do we do it in C++ or do we do it
in Java? These types of decisions are important because you can never go back
on them once you've started down that path.
In Groove's case, there was a very risky piece of technology--a certain algorithm
for synchronization that we didn't even know if we could do. And we didn't
want to hire more people and really get going until we knew we could
accomplish it. It took about 3 to 4 months before we were confident that we'd
be able to actually build what we wanted to.
Architecturally, Groove was a real contrarian play at the time. This was in
'97, an era where most people were saying, "Things will move from other architectures
to the Web." We were basically saying, "The Web will hit its limits at
some point for certain applications, and we want to go to a peer-to-peer architecture
that would complement the Web, not replace it." For a certain class of
applications it would be very effective. It's a masterless synchronization where
people could do things like work independently on all these different peer
nodes and the algorithms would get everyone in sync.
It can get very complex when you have a dozen people and they're in different
subnets. Eventually these people come together and it's complicated to
make sure all the changes get applied in a uniform fashion for everyone. So we
worked through that on the whiteboard and then in a prototype. Once we were
sure we could build it, we decided to hire the first 15 to 20 people and just
embarked on the project.
Livingston: Masterless synchronization was a novel technology that you guys
really had to work through?
Ozzie: It had been done for years in a variety of settings--especially in an academic
setting. But the commercial PC environment is a very harsh one. People
Ray Ozzie 105
reboot PCs, they restore them from backups, they lose them. It has to be very
resilient. We wanted to make sure the algorithms we were using would scale to
what we needed.
All those early technology choices were like that. Initially we thought we'd
be using Java, but we ended up not using it because we concluded that there
would never be a stable runtime environment that we could count on on all
desktop PCs. It didn't seem like Sun, with all due respect, really was on a path
to having a stable client-side environment. And we needed the thing to work
within several clicks on random PCs worldwide without anybody supporting
them. So we ended up having to do a lot of extra work using C++.
Livingston: Was there an initial customer who was so happy with the product
that you just knew Groove was going to fly?
Ozzie: We launched Groove in beta in October 2000, 3 years to the month
when we first formed the company. We didn't ship the first commercially available
version of Groove until April 2001. When we did, we announced a 10,000
seat deal with GlaxoSmithKline, the major pharmaceutical company. They are a
big Notes customer, but saw the opportunity for Groove to address some of the
cross-boundary collaboration needs they have in bringing new products to market.
In hindsight, that initial sale may have hurt us more than it helped. We
deluded ourselves into thinking we could sell Groove into enterprises like
GlaxoSmithKline far more quickly and systematically than turned out to be the
case. We really hadn't paid our dues yet in terms of making Groove "enterprise
ready." We did that in subsequent releases of the product, but still struggled to
develop a successful, repeatable sales model for the enterprise. It was
extremely difficult to sell new technology like Groove into enterprises at a time
when their sole focus was on reducing costs and increasing security.
Livingston: What else was hard in those 3 "stealth" years?
Ozzie: The thing that's not really characteristic, that doesn't really translate
from both of my startups to what other entrepreneurs do, is that I think of the
challenges I take on as 10-year challenges, not filling a quick market niche.
There tends to be some time where I'm building up a level of technological
advantage for when we get to market. With technology, there's no such thing as
a sustainable advantage, but you can get a good running start if you concentrate
on doing something hard really well.
In Notes, it was the database and replication environment and the security
aspects. In Groove, it was the security aspects again and this transaction synchronization
and the peer-to-peer XML-based communications. Most people
find risk and uncertainty very daunting. In both Notes and Groove, there was
both technological uncertainty and market uncertainty. We knew we were
embarking on something that was technologically very difficult and would take
several years. But you know that the market is going to change during those
years, so virtually everything you do, you have to late-bind the decisions. You
can't completely predetermine all the user interface or integration decisions.
106 Founders at Work
You cannot early-bind marketing and positioning decisions because the market
and competitive environment will be different.
Some people cope with uncertainty by being really comfortable in their own
little box. Some developers, for example, will divide the problem and divide the
problem until they only have to work on this little piece of the database or this
little piece of the communications, and they just don't worry about the stuff
above that. They leave that to people like me to deal with, in terms of the risk
and continuing to be on the right path. To be on that long of a time frame, you
have to be able to change as the market changes.
So there were a number of things over the course of the years at Groove
that changed dramatically. At one point early on, we were giving an equal focus
to the media/entertainment and productivity applications of our technology.
When we started Groove in '97, it was the Bubble, and because you can apply
technology in many ways, we thought that we'd bring it to market to serve a
number of different things. By the time we brought it to market in late 2000,
things were starting to get a little serious, and we decided to concentrate on the
productivity realm instead of consumer applications.
Then once we really doubled down that path, it meant that we had to take a
lot more enterprise manageability things seriously than we had early on, which
brought with it a lot of burden and a lot of changes within the company.
Livingston: If you do have this long time frame, are you extra nervous about
competitors? And do you have to manage people's expectations differently?
Ozzie: In a startup, you're on this mission together. Everyone has to feel that,
and you have to hire people who are willing to believe in something they are
trying to accomplish. And in that era, it was very challenging in two dimensions.
Hiring in the dot-com era, when a lot of these people's friends were getting
rich, was hard. But the other thing was that the type of software we were building
had many systems software elements to it. A lot of it was lower-level communications,
storage, application framework-type code, and hiring was more
difficult for that type of talent at that time. In an earlier era when DEC was big,
it was easier to hire systems software talent.
But what held people together was the belief that you're really going to
change the world. I think that's the nature of many startups. You believe that
what you are doing is going to have a dramatic impact. You might not exactly
know how, but you really have a belief. That keeps you going and going through
many changes and a lot of uncertainty.
Livingston: What about managing your investors' expectations?
Ozzie: That's a difficult subject. There are pros and cons to taking money. The
best kind of company is one where you don't have to take any money.
Livingston: Did you use your own money for Groove?
Ozzie: Yes, I funded the first few years myself. But eventually I took some
money from Mitch Kapor and then others. Not so much because I needed it at
that point, but because I knew that, ultimately, you cannot accomplish something
completely on your own. You really need to develop a network of people
Ray Ozzie 107
who win when you win. Being on the East Coast, I believed that it was very
important to establish a good network in Silicon Valley, where I didn't have a
presence.
I'd worked with Mitch for many years, and I felt that he could make the
right introductions. So I first took money from Mitch, then he made some
introductions to VCs. One of them was Accel, and I took money from them. I
ended up spending quite a bit of time in the Bay Area, meeting a lot of people,
and ultimately that network helped a lot.
Iris was a corporate partnership with Lotus. I was 27 years old and didn't
have the money to fund it then. Getting the product built was an amazingly positive
experience. We had structured a great contract that funded the product--
it was a unique partnership, a corporate startup kind of R&D partnership. But
that brings its own challenges. When you have an alliance with a major corporation
from an early stage, what you build really has to relate to the other, larger
goals of that corporation. You may not be completely tied up, you still can
accomplish your vision, but it would make no sense to be funded by a company
and be completely aligning yourself with their competitors' offerings.
In a startup environment, it's much rougher in terms of making your numbers.
There's much less patience. Once you start down the treadmill of taking
venture capital, it's "how many rounds before people give up on you or you have
a positive exit event?" So you're really setting yourself up. The best by far is to
structure it such that you don't have to take money.
Livingston: You also took money from Microsoft. I know they thought very
highly of you, but do you think they also invested to keep an eye on what you
were doing?
Ozzie: That's exactly why they did. They were a straight investor, meaning
there was no technology sharing or anything like that as part of the investment.
I think Notes probably got a little bit out of control from Microsoft's perspective.
They didn't really track its market very closely while it was emerging, and,
had they watched what was going on, perhaps they might have been able to
respond a bit more quickly.
So I think with Groove, it was essentially buying a look at what kinds of customers
found this technology attractive. More than anything else it was market
tracking. They knew enough about the technology, because once we came out
of our stealth phase we were very open with everyone about the kind of technology
that it was built on. And we were very confident about that because we
knew how hard it was to build.
At both Iris and Groove, we believed Microsoft was our prime competitor.
Livingston: Even at Groove? But Microsoft seems so ambivalent about the
Internet...
Ozzie: If there's going to be a trend that's largely horizontal, Microsoft cares.
Because Microsoft's bread and butter is serving the masses--whether it's consumers
or enterprises--with low-cost technology that solves many problems.
And other people layer upon it more vertical solutions.
108 Founders at Work
We were pitching Groove as a fairly horizontal technology. We were applying
it to productivity challenges, but to the extent that it had the potential to
catch on broadly, they would certainly have been the biggest competitor.
Livingston: Looking back, was the Microsoft threat real?
Ozzie: Oh yeah, they are brilliant technologically and from a business strategy
perspective. If you believe that Microsoft is your competitor, it's better to keep
stealth and then embrace them at the right time, when you believe it can be to
your advantage to embrace them. In the case of Groove, we were having distribution
challenges, we needed money, we were raising a round. One of the
biggest questions we were encountering with our enterprise customers was
"Why isn't Microsoft just going to crush you tomorrow?"
And although I brought some credibility to the table because of my background
at IBM, having Microsoft as a backer only helped us within those enterprise
accounts.
Livingston: Back to Lotus Notes--were you already working on an application
when Lotus discovered and then funded you? What was the history there?
Ozzie: As I mentioned earlier, I first wrote the spec for Groove in 1982. But I
couldn't find funding for the idea. So in 1983 I was hired by Mitch Kapor and
Jonathan Sachs at Lotus Development, just after Lotus 1-2-3 release 1 had
shipped. I did a small amount of work on 1-2-3 1A, then led a small team to create
Lotus Symphony, one of the first "suite" products. I agreed to do Symphony,
if Mitch would help make introductions to VCs and help get Notes off the
ground. The day Symphony shipped, Mitch made good on his word. But
because Lotus was in a good cash position, rather than introduce me to VCs,
Mitch suggested Lotus supply the capital. I then formed Iris Associates in
Westford, Mass., with three other programmers in December 1984.
Livingston: What surprised you the most?
Ozzie: How difficult the go-to-market challenges are. I suppose it shouldn't
have surprised me, but in both the cases of Notes and Groove, building a market
in something that's new can be as, if not more, challenging than building the
technology. We were building some very complex technology, and I thought,
since we were developing to what seemed to be a fairly straightforward customer
value proposition, going to market would be a lot easier.
Changing people's habits is extremely difficult. Notes came out at a time
when things were kind of booming from a tech perspective. But Groove came
out at a very difficult time. It was just post-Bubble and IT spending was really
down. If you are serving the consumer, everyone expects not to have to pay for
anything. In business, if you're talking with IT, it's just very difficult to justify
any incremental spend.
I guess as a tech entrepreneur I would nurture relationships with people
who are outside your skill set on the marketing and sales side or business development
side. Relationships you know you can trust. As a technologist, it's very
difficult to hire someone on the marketing and sales side because they're so different
than technologists and you don't know who to trust. It takes about a year
Ray Ozzie 109
to really understand whether the people who you are partnering with trust you
and know they will rely upon you just as much as you know you will rely upon
them.
That's where I think working for another company and building those relationships
is extremely valuable. Frequently, people think just running from
school out into doing a startup is the best thing to do. But I think that getting
some experience within a number of companies is really positive because you
meet people and you start to develop patterns in your mind of the types of people
that you need, and the types of people that you can trust, and the types of
people you never want to work with.
Livingston: What advice would give someone who was thinking of starting or
joining a startup?
Ozzie: For someone who's joining a startup, just learn about leadership from
the people at the top of the company. Watch how they talk to people, watch
how they present to people. Companies take their shape based on the personality
characteristics and human interaction characteristics of the founders. This is
true in every company. Learn about the kind of culture that you want to create
in your own company based on the positive and negative aspects that you witness
in the people that are your leaders.
Learn to respect and appreciate other people's skill sets, because you are
going to need other people if you do start a company and you are a technologist.
Understand that it's a rare, rare case when a tech entrepreneur is the right one
to lead a startup for a long period of time. You have to feel comfortable in your
own skin in terms of what you're good at and what other people are good at.
Know when the shift to chief technologist is the right thing for the company.
You have to be comfortable with the fact that you are separate from the
thing that you're building, and that the team and the people financing you will
have joint custody over the asset that you create. You have to respect that and
not associate your own success and failure with the success and failure of your
"child."
Livingston: Is there anything that you learned from Iris that you applied to
Groove?
Ozzie: In terms of the culture, there were some really strong positive things.
People doing things for the right reason. Never say to people that you are doing
it for the money. Don't do it for the money.
Everyone knows that one reason you go to work and do what you do is the
hope that ultimately you'll be compensated. But you don't have to say it, and it
doesn't have to come through. It should be about the mission. It should be
about changing the world. It should be about how you can impact the lives of
users, partners, and the employees themselves. It's not just about this big payday.
The more you focus on the things that matter when you are talking to people
who want to believe in you, the more they will believe in you and the more
it will be a sustainable entity.