The .NET Sweatshop (v2) : business

The Death of the Smart Client is Greatly Exaggerated

SandyK — Fri, 30 Mar 2007 10:24:00 GMT

I was watching a great presentation recently from a Berkeley professor who came to speak to a bunch of senior people in my division. He threw out a lot of interesting sayings, but my favorite one may have been "the most dangerous three words in the investment community is 'This time, it's different!'" When I see a lot of what is going on in the industry, I get the same sense of deja vu. Advertising business models. The death of the PC. The idea that XML will change the world. And, of course, the network is the computer--where the offshoot is that the browser is more important than the OS. People, we've been here before. Now, don't get me wrong--I do think things are different this time and I do believe changes are afoot and I am bullish about those changes. However, the MVP Summit that took place a couple of weeks ago was a great reminder of the fact that the world is not 100% sold on browser apps and there's plenty of power in the "Smart Client".

It started when I met up with Billy Hollis for the first time in a couple of years. Billy and I spent a lot of time together during my days in patterns & practices, but hadn't seen one another lately. Billy, the King of the Smart Clients, still felt pretty good about the importance of client apps and the fact that ASP.NET did not need to be the lone choice for developers out there. It was nice to get the gospel (and those who know Billy realize that with that southern accent, gospel just seems like the right word) that didn't imply that everything needs to be intermediated through a browser--which, in my moments of frustration on the topic, make me think of being told all food needs to be fed through a straw. The next day, I met up with a host of MVPs for a breakfast that was organized by Alan Griver (aka Yag) from our team. We shared some of the social computing stuff that we're working on and integrating into the upcoming releases of Forums and Blogs. While the response on the new ideas was good, there were still many who loved NNTP and wanted it supported. It mostly wasn't about the technology, but rather about some of the things you just can't do with a browser app, such as successful off-line operation. I guess AJAX isn't that asynchronous.

No matter how wonderful AJAX becomes, it will never provide a richer customer experience than a smart client. I know--there are those out there who will disagree. Hey, I know I prefer poking around an AJAX site instead of downloading an EXE file, installing it, and then finding out it stinks. In many cases, I admit that I don't bother getting that far. But to rely on a browser as the sole means for access? I don't know if that's an ideal long-term strategy. Personally, I think it's the difference between the person you want to date and the person you want to marry. A browser app is easy, quick, no fuss, no deployment hassles, install files, etc. It is one URL away and, if you don't like it, you move on. No commitments, no strings attached. It's dinner and a movie. A smart client requires deployment, installation, OS compatibility (though browser apps have to now deal with browser compatibility), DLL installation (notice I didn't say "hell"--we don't have that any more, right?). etc. It's a much more committed relationship, in a sense.

Many of my single friends love the former, but (and I suppose I say this as a happily married guy) the browser app is fleeting in the long-term when compared to the benefits of the smart client once you know you're ready to commit. Once you are sold on the service, you have to be frustrated by the things you can & can't do in the browser. All this computing power and I am at the mercy of my network bandwidth. The sandbox nature of a browser is great when you don't trust a site, but when you do (as I do with Amazon, eBay, and several other established brands that have given me great customer service and earned my trust), it's incredibly limiting. As one example of someone who saw this and wouldn't compromise on the experience is Apple with iTunes. You would think Apple would rather create a browser-based application with a plug-in than have to support writing an application for (gasp!) Windows. But they continue to write and update it. Why? Because I don't think you can build that rich an experience through a browser. Will you ever? Perhaps, but by then, client capabilities will have increased as well and Apple will have done something new that the next versions of IE, Firefox, and Safari won't allow.

So what happens next? I have a belief that technological progress is often affected more by trends than by need or opportunity (not always, but often). If people believe AJAX-enabled browser applications are the future, the Smart Client becomes an after-thought and investors/large companies/etc don't want to go in that direction. I am hoping that is not the case here. I think the rise of APIs, which is thriving as a mash-up technology, could give way to an increased number of smart client applications. An example that has worked for me in Flickr. I like the web API, but I've started using some of the smart client apps to handle the heavy workload tasks (like batch uploads). The Flickr Uploadr is a great utility, but doesn't get the "ink" that any Flickr mashup gets. Personally, I think the ideal would be to do what Microsoft gives you in their messaging offering. To give Outlook & Exchange their due, you have two ways to access your e-mail. There is Outlook for richer processing of data, off-line usage, and access to the hard-drive for archives. Then, there is Outlook Web Access when you are on a kiosk or another person's PC or if you don't want to start up Outlook and just need a quick & dirty way to check if you have mail. Personally, I use both very liberally. I think you could make a really good Amazon application. Or eBay. Maybe they could happen. Maybe I could beg Billy to write them...

Peanut Butter Jelly Time? (Part II)

SandyK — Wed, 27 Dec 2006 10:59:23 GMT

Following up on the Yahoo memo, I found it interesting to read the section regarding the effort level at Yahoo. I believe the larger your company grows, the harder it is to get everybody driving hard--especially at the levels of a startup. For as much as people love to lobby for the "startup mentality" at large companies, I think that's rare to get that to happen. I would love to tell my team to operate like a startup, but if my VP (or for that matter anyone up to Steve Ballmer & Bill Gates) hands down a mandate that we use a certain product, restrict a branding decision, or align with a corporate direction, I am obligated to push that direction down on my team. And by accepting these uber-decisions, the team knows that it answers to a higher authority. Startups don't have that. With the exception of the venture capitalists, the highest authority is usually sitting just down the hall or, in some cases, in the same room. No "blaming it on the man"--you are the man. That's empowering. Yahoo is a mature company and a random developer in Yahoo Mail eventually has to make sure he is aligned with Terry Semel and his band of merry executives, who, in turn, need to keep Wall Street happy. Sorry, but that's different than a couple of guys in a garage with a dream--no matter how much you encourage your team otherwise. Go ahead and try to convince him to work like it's a startup.

So as a manager for the world's largest software company, does that mean I'd accept anything less than a 100% effort from people on my team? Of course not. Occasionally, you spot people that are dogging it and I expect the managers on the team to call them out and suggest they look for more interesting employment if we can't keep them entertained (and that has happened in my career). Complacency is contagious and it's nice to nip that in the bud.

But what of those people ready to work, but lacking the motivation that is built-in to to startup's "do-or-die" culture? In other words, how do you maximize the effort at a large company? I think it's twofold. First is to "accentuate the positive": I think you need to focus on a separate set of incentives at play with a larger company, not the least of which revolve around the magnitude of the impact they can make. While the startup is exciting, 9 out of 10 startups fail (heck, I've failed once myself). Doing something at Microsoft is like doing it into a megaphone and the thrill of knowing your effort will be enjoyed by millions of people (which is what I can boast to my team) is pretty powerful as well. It's the difference between the effort an indie band makes trying to score a record deal and the effort that U2 makes during a sold out set at Madison Square Garden. Just because you've made it doesn't mean you can rest on your laurels--in fact, the expectations are higher. Nothing in the Yahoo memo suggested that the success they have had should hold them to a higher standard. It's as if they'd be better as the two Stanford PhD's who create a web directory. Embrace your success! Second is to eliminate the negative: while I can hope that they feel enough autonomy to let operate at a startup pace (complete with insane hours and the sense of being on a mission that makes the effort worth it) and feel that level of empowerment, I realize that it's foolish to set it as an expectation when I'm at a 70,000 person company. That's said, minimizing the impact of the rest of the company on the individual contributor, shielding them from the unnecessary garbage that often comes in. Only then do you get the empowerment and autonomy that creative people crave. That spurs effort. The delta between what comes from the team themselves and what reaches the customers should be minimal. That's empowering. Getting caught in the exec spin cycle of "tweak this, change that" cuts down on the empowerment. Of course, what, then, is the role of the "boss"?

The Yahoo memo did make mention of the bureaucracy facing their company and how ownership was split around the company in different disciplines (Engineering, Finance, Marketing, etc). With no clear ownership, execs were being forced to make decisions. My question is "which decisions?" In most cases, execs don't need to make decisions, but rather they need to assign who should make which decisions and stand behind those decision-makers (unless they are in strong disagreement, in which case, a discussion should occur). They should also ensure proper collaboration between the appropraite parties to come to compromises when necessary and arbitrate when no other solution can be reached. I like what Eric Schmidt at Google said about his role as CEO (I believe this came from the SF Chronicle, but I lost the link):

(Schmidt's) primary job is to run Google's numerous staff meetings. To get the most out of them, he relies on two philosophies. One is getting opinions from as many people as possible, particularly dissidents, under the theory that decisions are better when based on many viewpoints. Another is to set deadlines because, unless there's a reason to end an argument, people will continue to argue. He described a time when founders Sergey Brin and Larry Page were arguing violently about Internet browsers and software that had pitted them against one another in front of other managers during a meeting last year. After a while, Schmidt, who was also in the fight, put his foot down. He emptied the room and told the co-founders they were to reach an agreement by 6 p.m. because the dispute was confusing the staff. They resolved it by the next morning.

That's a powerful way to lead: ensure all avenues are reviewed, get the objections out of the way, and force the elimination of dangerous stalemates. Life is all about dissenting opinions, tradeoffs, and compromises and your ability to navigate effectively is the difference between success and failure. The exec should be a part of the debate and, when necessary, step up and make the decision. Yet it shouldn't be an in a vacuum, but rather the exact opposite. Empowerment isn't absolute authority, but rather an voice that will be heard loud and clear--particuarly as it relates to one's expertise. It's not consensus. It's not even necessarily a democracy where majority rules. It's a collaboration where the collective establish mutual respect and can generate something greater than a solo entity and the best ideas win. Imagine that...

Peanut Butter Jelly Time? (Part I)

SandyK — Thu, 21 Dec 2006 10:12:00 GMT

The shakeup at Yahoo is a fascinating reflection of the life and times of a tech company. Now many consider Yahoo a competitor to Microsoft and I have no interest in disparaging their company, but this is an interesting business case study and there is a lot of learn here from a strictly academic perspective. So ignore my MS bias and assume my statements are made from the perspective of a innocent bystander.

Seems like even before the shakeup, there'd been a lot of dissension over at Yahoo, which included an e-mail entitled "The Peanut Butter Manifesto" that came out last month. In it, a Yahoo senior exec called the company to task for (among other things) spreading themselves too thin across too many projects (losing focus) and pretty much giving a half-baked effort on their projects. He seemed to think the company lost its drive and was wallowing in their bloat. Interesting. There are two parts of the document that are interesting: the diversification of interests and the operational efficiency of a disenfranchised workforce. I've got a lot of thoughts on both, so I'll split them into two blog entries.

Judging the "Peanut Butter" effect as it relates to an overdiversification of business interests, it makes for an interesting debate. According to rumors that I recall floating around, Google is undergoing the same sort of internal scrutiny and recently shut down their Q&A service as a possible harbinger of things to come. While their success has been far greater than Yahoo, they are being critical about how they are execute on their charter. They also believe that they need to focus on the businesses that have the highest chance of success instead of the spaghetti approach of throwing a bunch of stuff against a wall and seeing what sticks. When Jack Welch was at GE, one was his most popular approaches to optimizing its success was to tell businesses "either you're #1, #2, or you will be #1 or #2 in an market within a couple of years. If not, out you go." Well, those might not have been his exact words, but you get the point. Welch was focused on markets that he could successfully compete in. For students of the BCG 2x2 matrix, he wanted to be a Star or a Cash Cow, but had no interest in the question mark or the dog. For those who have read Jim Collins' Good to Great, this sounds like the "Hedgehog" concept (fwiw, I really didn't like that book) in determining what you do best. With the Yahoo memo, there was a suggestion that they were complacent in being the question mark (high growth, low market share) and that was impeding their ability to succeed in their other businesses and hurting their overall corporate health. Some might suggest that Microsoft is in danger of doing the same thing.

I agree that simply getting into businesses for the sake of getting into the business is a bad idea. I also agree that you need to learn when to say no. But I also believe that there are some businesses that may take a while to incubate and you need to know where to preach patience and where to come quick. I look at other companies and wonder how their decisions are being perceived internally. eBay's decision to buy Skype shows a plan to diversify in a way that (in my mind) still doesn't quite align with their strategy to be the world's greatest flea market. Amazon's S3 play is a bold move that has nothing to do with retail. Are these good or bad? My instinct is to say "bad", but I've been too quick to judge in the past. I do recall chuckling when Apple came out with the iPod, thinking "what does the iMac company have to offer the gadget world? Here's another Newton..." Well, I was obviously wrong about that. Now, in retrospect, that was a great move by Apple. But wasn't that spreading the Peanut Butter a little bit? I have to admit that I like a lot of Yahoo's offerings. I think their foray into being a content company wasn't as flawed as people are trying to make it out to be in retrospect. The problem with bold, gutsy decisions is that you are a hero when they succeed (Steve Jobs and the iPod) and an idiot when they fail (John Sculley and the Newton)--and oftentimes, the result has as much to do with luck as with any specific aspect of the decision. For Microsoft, diversifying into enteprise computing has given them a growth engine that has been pivotal in the last several years. Building a home and retail presence with XBox and Zune is meant to give us something to hang their hat on for the next decade. To base Microsoft entirely on operating systems and productivity applications (the historical primary money makers) and cut all else would completely ignore the fact that you need to prepare for life after your core competency. Thank goodness Microsoft didn't give up on SQL and hopefully we'll be saying the same about Zune in a couple of years.

Again, I'm not suggesting haphazard business launches for the sake of letting workers do whatever the heck they please. But when I see things like Live Labs, I am excited, rather than pertubed, by what to bodes for Microsoft's future. Cutting businesses provides a short-term fix and does quench the Wall Street thirst, but be careful where that leads you in the long run...