The .NET Sweatshop (v2) : CodePlex

An Ode To GotDotNet

SandyK — Thu, 22 Feb 2007 00:31:00 GMT

About two years ago, I stepped into this role with a primary responsibility being to help get GotDotNet back on its feet. As the first Microsoft community for .NET, GDN held a dear place in the hearts of many. However, it had clearly fallen on hard times and there was a strong proposal to end the site. Personally, I had used GDN way too much to see it go down without a fight. When Betsy Aoki left the team, I gave a little insight to the story of how we stepped up when no one else would. It's hard to believe all of that was two years ago.

It was a battle where we had some pretty nasty things said about us while we tried to right the ship. Our first move was "Project Tourniquet", which was literally to stop the bleeding. We then opened CodeGalleries and re-did the site design to usher in a new age of GDN. Outside of a tough deployment in November 2005 (when we rolled out the new chrome/site design AND moved from 1.1 to 2.0) and a pair of weird hiccups in 2006 that each lasted about a day, we've reached a smooth operating machine. I can remember when the tide truly turned. After the November 2005 deployment (at which time, I was so frustrated that I was ready to shut down the site then and there), I wrote a service to ping our eight most popular pages once every 15 minutes and ensure that the site was not only responding, but doing so with something other than the infamous "Troubleshooting in Progress" screen that so many people. As the months went on, I went from agonizing about those statistics to getting excited as the numbers started getting really good. Meanwhile, the general mood of customers completely changed. I can't tell you how many glowing mails I've gotten from people practically apologizing for their previous harsh words (although they were admittedly warranted). It's been a fun ride and we had some great uptime statistics in the last twelve months. I look at Betsy Aoki, Jana Carter, and George Bullock (the three PMs for GDN during my tenure) and see three people that helped turn site once called an "embarassment" into a legitimate example of a .NET web application. In the process, I learned a ton about running a web-site and dealing with customer requirements on the fly. You can say I owe a lot of my career sensibilities to GDN.

However (you knew there'd be a 'however'), the project we were doing in parallel with the rebirth of GDN was the birth of CodePlex. As proud as I am of the revival of GDN, I am more proud of the soup-to-nuts story of CodePlex. Jim Newkirk and I fought hard to build a site that would treat sensisbilities of a community developer as a priority and to do it from the ground up. With CodePlex, I think we really nailed it and continue to do so. As CodePlex approaches its one year anniversary and continually grows its traffic, the need for two communities from the same team addressing the same customers is nearing its end. GotDotNet hit an all-time high in traffic the month before CodePlex came out, but has been declining since then. It's clear that this isn't a conincidence. So, this summer, we will be closing the doors of GotDotNet. It was a hard decision, like having to put your dog to sleep (well, I've never had a dog, but I'd imagine it's similar). Over the next few months, keep an eye on the front page of the site to keep abreast of how we'll handle the migration. To those of you who supported GotDotNet and put such compelling content on the site, I want to say thank you. And if you have any great new ideas, I suggest you give CodePlex a try.

Gravediggers

SandyK — Wed, 11 Oct 2006 06:46:15 GMT

I was taking a look at the latest news around CodePlex recently and I started noticing a fascinating trend. Many people have described us as "Microsoft's SourceForge Killer", suggesting we built this site to eliminate SourceForge. In fact, that's a trend with everything Microsoft does. Think about it. Everyone calls Zune the iPod killer. The Motorola Q Smartphone is the Blackberry killer. MSN Soapbox is the YouTube killer. Meanwhile, the same onus is put on products that are targeted at Microsoft, as Linux is the Windows killer, Open Office is the Office killer, Firefix is the IE killer, and Google is the uber-Microsoft killer. The one thing in common with these products is no one ever really achieves the kill. What's fascinating is that everyone assumes that the only reason to join a market is to completely destroy its competitors. Now, I won't deny that the people from Zune would love to bring the iPod to zero market share and have every man, woman, and child with a Zune. But let's be honest--you can't get rid of an incumbent that easliy and thinking anything could kill the any of these great products is borderline ridiculous. At best, you may trigger an inflection point that may slow momentum (and I'd imagine that's what Zune and Q are all about). In cases where dominant products were signiificantly reduced of market share, it took several versions before the tide really shifted. You need to get in the game and THEN you can start laying claim to some market share. IE 1.0 was NOT the Netscape killer. But IE 3.0 and IE 4.0...

For CodePlex, we've never once suggested we were trying to go after SourceForge or kill SourceForge. Even three versions down the line, I expect SourceForge will do just fine. CodePlex is very different than SourceForge. Our goal isn't 100,000 projects to people comfortable with CVS or SVN, many of whom are the lone developer and using the site more for code storage than community development. Our hope is that CodePlex appeals to a new audience that may have stayed out of the community collaborative game because the bar for SourceForge wasn't a site with which they felt comfortable (like me :-) ). We've made CodePlex work well with Visual Studio, we've tried to focus on features that allow distributed teams collaborate more easily, and we're putting in features that enable the community to get involved--even if they are not developers on the project. We're trying to streamline participation and that's our niche. There's still lots of room for SourceForge, CollabNet, Codehaus, and even Google Code Projects. Don't get me wrong--I do hope some SourceForge users decide they want to switch, but I am not naive enough to think we are going to put a dent in their business. I believe the difference is in the intent of our sites: they are a huge code repository and we are a community collaborative development site. Building another SourceForge does nothing to help Microsoft's business. But spurring collaborative behavior in the Microsoft developer community that creates useful applications and utilities for Windows? Sign me up...

That's why I'm really excited about the ideas being circulated around the team about upcoming features. As you'll notice with Jim Newkirk's blog, we are doing three week deployments (fulfilling our desire to "release often") and continually evolving the site based on customer feedback as well as a series of backlog items that we've been thinking about for quite sometime. Plus, we have another great developer who'll be joining the team soon. I won't spill the beans on who it is as I'm sure he'll be announcing it shortly, but if you are a fan of Enterprise Library, you'll be very pleased. I know I am.

So, put away the shovel as SourceForge isn't going anywhere, but stay tuned for some great stuff that'll make CodePlex even more useful to our developers. Of course, if you want to come over from SourceForge, we don't have a problem with that either. ;-)

{Velvet Revolver - Contraband}

CodePlex: The XBox for Community Developers

SandyK — Sat, 01 Jul 2006 02:07:00 GMT

I once blogged about "Glory and Danger" and the idea of taking on GotDotNet. One of the things I hinted at was, while we would do a revamp of the site, we did want to consider something from scratch that would be a step above. Then, I patiently stayed quiet about what we were doing. Well, the cat is clearly out of the bag now. A year or so ago, I was asking people how we should support community development going forward. After a year where we primarily stayed under the radar, CodePlex is now live and ready for the world. The team has even started their own blog. Having watched the thing in development over the past year, I can't help but smile when I cruise around the site. So many of these features were once little more than slideware. To those who think the "little guys" at Microsoft can't come up with a pretty good idea and have it supported, we are proof that anything is possible if you fight hard enough for it. I can still remember sitting in my old office, talking to Jim Newkirk about how we could affect the greatest amount of change in the company. We both believed that community software development was (and still is) vital to the health of the developer ecosystem and finding a place where people could do it easily would be so important to Microsoft. In our case, we both ended up leaving the patterns & practices team to join the Microsoft.com Communities Team to make it happen. As time went along, we were fortunate to add some killer talent to the team--guys that could've worked anywhere, but joined us because they saw the possibilities of what this team was doing. I couldn't help but think that this was what Microsoft is still all about--get a bunch of smart guys in a room and let them work on what makes them passionate. Then, open it up to the world and continually iterate to make it work better for them. Of course, it's never that easy. Jim and I have had countless late-hour IM conversations, trying the navigate the project through the many challenges that any software project (especially at a large company) goes through. I honestly believe no one else I know could've pulled off the project better than Jim did. The guy is top-notch both technically and personally and I am five times happier for him than for me.

As the team got closer to the finish line, I was reading a book about the development of the XBox and it was amazing to see the parallels. The leaders of that project knew that the secret to its success was creating a platform where developers could do their magic. While it's specs were awe-inspiring for that time period, the standalone XBox was useless, It's really about the games and what the game developers could do with the power of the XBox. My hope is that we do for community development what XBox has been doing for game developers--make the process easier and more enjoyable.

While this release is something to be extremely excited about, I am even more excited about what I think we can do in the future. We already have tons of ideas and we are hearing from many customers who have their own input that is helping us guide our roadmap. Comment on my blog or on the CodePlex blog and let us know what you think. Plus, there is the discussion board on the site itself. We really want to hear what you think, so let the team know. And keep poking around the site--I think there's gonna be some killer games, er, projects that will debut in the coming months.

What has Jim Newkirk been up to?

SandyK — Tue, 16 May 2006 20:09:00 GMT

I remember reading someone posting a comment to a blog saying that Microsoft hired people like Jim Newkirk only to keep them from doing stuff for the community. Not so. If you want to know what Jim has been doing, check out his blog entry today: http://blogs.msdn.com/jamesnewkirk/archive/2006/05/16/598935.aspx

Oh, and he's got some impressive help as well: http://www.agileprogrammer.com/dotnetguy/archive/2006/05/16/14807.aspx