TFS Version Control and more ..

A change in our ETM Architecture Plans

mrod — Thu, 22 May 2008 15:58:23 GMT

As you guys already know we are working on a set of architectural investments that will allow us to scale out the TFS deployment. These scale out capabilities allows administrators to easily manage the load put by their teams on Team Foundation Server while simultaneously reducing the TCO of our offering. Our vision is Manage Complexity, Achieve Agility. We published the details of our plans in a spec located here: Enterprise Team Foundation Server Management

Today I wanted to let you know of a change: Supporting Multiple TFS Instances on one Application Tier (physical machine) has been CUT.

Why you might ask?

Before answering the question, let us look at what Multiple TFS Instances per physical machine (AT node) provides you. In short, it gives you the ability to have multiple TFS Applications or top level websites on one machine given further isolation for the different teams that use the system. For example, I can recycle, make an update, change settings to TeamA without affecting TeamB because they are on different processes. There is also a rolling upgrade scenario that becomes easy once you allow these multiple instances.

Perfect! Now the problem is that you can accomplish this today by using virtualization. Have this machine split into two machines and now you can install TFS in it and have even a greater control on the settings, memory, etc. Essentially we are enabling an already existing and soon to become very hot/popular technology. Given this, we felt that our time and resources were better spent doing other features and leave those scenarios that were enabled to the virtualization technologies.

Any thoughts or complaints please add a comment or just email me directly.

- mario

Team Foundation Server Properties Spec

mrod — Fri, 09 May 2008 17:48:33 GMT

Yesterday we released our first specification discussing Team Foundation Serve properties and our thoughts on the feature. If you get a chance please go through it and provide us with feedback as it is always important for us to validate our assumptions with customers.

Here are the links:

Directly to the spec: Team Foundation Server Properties

Site that hosts the spec: http://msdn.microsoft.com/en-us/teamsystem/bb936702.aspx

Please email me or post to the forum your feedback.

- mario

Version Control Areas of Innovation

mrod — Thu, 08 May 2008 19:19:11 GMT

If you look at Version Control over the 1990’s you are not going to find a lot of innovative and revolutionary features. I believe that there were certain industry factors that were responsible for this "dormant" period, but we won’t talk about those since they are not all that relevant.

On the other hand since the early 2000's or so we are seeing a lot more activity in this space:

· Accurev introduced their TimeSafe® technology and their streams

· We introduced a unified server allowing easy integration between version control, work item, build and user data, and don’t forget shelvesets

· IBM has now followed with an interesting collaborative centered experience in their Jazz product line

· Perforce continues to explore new visual representations of the data stored in version control

· Git is making quite a noise after winning the Ruby on Rails team with their distributed model.

As you can see the next five years are going to be interesting as the ALM space becomes more competitive and a lot more innovative features hit the market. In version control there are three key areas to innovate on (this of course is my humble opinion):

1. Branch and Merge. Because today is too complicated end-to-end

2. Collaboration and Integration. People don’t develop software in isolation; successful teams communicate early and often without fear, and data is accessible and linked no matter where it resides

3. Performance. Because no one likes to wait for information

At the end, whoever can deliver on these successfully will be the “desired” tool of choice. I think we are making heavy investments in each of those areas so again I am very excited about our present and future as a product unit.

Since we are talking about innovation I would encourage you to see the following presentation titled “The Myths of Innovation”, link: http://www.youtube.com/watch?v=amt3ag2BaKc . It gives you a very good perspective on innovation and how much failure and human drive, not book or classes are an integral part of a successful idea. Check it out you will not be disappointed!

- mario

What does TFS merge uses: local or latest?

mrod — Tue, 06 May 2008 22:35:51 GMT

Recently our team got this question asked through the forums, and since I remember when we were making this decision, I thought it would be best to put it in paper and do a quick blog.

First let’s talk about the scenario: I have a workspace and I have not done a GET in a week or so but I want to merge a parent branch, which I have mapped, to my private branch. The question then is: would merge use my local workspace contents for the source or would it use the latest on the server? In addition, what does it use for the target?

When discussing this model we need to look at the intent of the customer. Do they want to merge the changes they have in their workspace to the target location? After careful analysis the answer to this question comes back as NO. They want to merge the contents of the branch. Since they want to merge the contents in the branch then it makes sense that for Source we always default to use the latest committed version and never pay attention to the contents the user has in their local. Effectively we assume that in a merge operation it never makes sense (unless the user specifies a workspace version range) to utilize the local source versions (which might not be even committed) and merge them to a target.

Summary: Merge uses Latest committed server version for the source of the merge

Now lets look at the target. In this case we need to pend the changes so this requires that you have the target mapped in your local. Going back to the scenario we ask the same question: what does the user wants to accomplish? We get two high level scenarios:

1. The user wants to bring certain changes (cherry pick or a range) from a branch into his local stabilize those changes and then integrate with whatever is latest

2. The user wants to do a catch up merge and is not interested in doing any integration with their local changes

On both of these cases we need to pend the change and have it reside on the user's local workspace. Due to this "requirement", looking at the server version is not ideal because we can easily put the users target workspace in an inconsistent state since some files will have new server content (participating in the merge) and other ones will not. Updating content for the user without an explicit Get action is always a recipe for disaster if the operation has a tendency to be partial on the items being updated. Keeping people from getting into this dependency mess is the main driver of our decision.

Summary: Merge uses local (workspace version) for the target of the merge

Regardless of this decision we always advice our customers to perform a Get in the target if they are going to be doing a catch up or big merge so they minimize the conflicts at check-in time.

There you have it ... hopefully you can use this information to make better decisions or to train your team in some of the TFS concepts.

- mario

Interesting HBR IdeaCast article

mrod — Thu, 01 May 2008 18:20:32 GMT

Today I came across this article and I thought to share with you because I think it holds a lot of true for software development teams. I think the notion of always providing value because we are smart or have this certain level of expertise at times is detrimental to the overall success of that idea.

Podcast: http://blogs.bnet.com/intercom/?p=1648

Column: http://discussionleader.hbsp.com/goldsmith/2008/01/tips_for_managing_smart_people.html

Also read the comments as some of them are very insightful

Here is my high level take on this:

Speak up if you think the idea is not great. Meaning that if you think the value add of your comment is more than 30-50% of the current idea at that point I think is well worth it
Don't comment or provide extra features just for the sake of doing it. Sometime a "That is great !" is all that is necessary to get that idea to completion and into the customers hand
Foster a collaborative team environment were people feel free to share and collaborate on ideas so you avoid these type of problems (this I think is the most important and also the most difficult to achieve)

External Team Foundation Server Tools

mrod — Mon, 28 Apr 2008 17:47:40 GMT

The other day I was in a customer call and I was asked for a list of TFS External tools, I am defining those as any TFS feature or tool that is not included in the packaged product. Although I still don't have a compiled list the best advice I could think of was to give links to the places where this information is available. Here are those links and a bit of information about them:

1. TFS Development Center (http://msdn2.microsoft.com/en-us/tfs2008/default.aspx)

This is the main Team Foundation Server site and hence it provides a lot of information about our product, current news and released external tools. You can check out the CodePlex and Downloads sections for the latest releases (e.g. TFS MSSCCI Provider)

2. TFS Power Tools (http://msdn2.microsoft.com/en-us/tfs2008/bb980963.aspx)

Our Team Foundation Server Power Tools are a set of enhancements, tools and command-line utilities that improve the Team Foundation Server user experience. We try to release a new version every three to four months in order to keep with the growing customer demand

3. CodePlex (http://www.codeplex.com/Project/ProjectDirectory.aspx?ProjectSearchText=TFS)

In here just run a query with the keyword TFS or VSTS and you will get a significant number of tools developed by our community.

4. Third Party Tools

Attrice SideKicks (http://www.attrice.info/cm/tfs/index.htm) :Team Foundation Sidekicks is a suite of tools for Microsoft Team Foundation Server administrators and advanced users providing Graphic User Interface for administrative and advanced version control tasks in multi-user TFS environments.

TeamPrise (http://www.teamprise.com/): Teamprise offers a suite of client applications for accessing Team Foundation Server from outside the Visual Studio IDE.

Team System Rocks (http://teamsystemrocks.com/): Community site maintained by some of our MVP's focusing on news and tools. Their files tab contains a list of available tools that customers can utilize.

Personify Design (http://www.personifydesign.com/products/default.aspx): A Team System Partner with two products: TeamLook and TeamSpec

If you know of more links which I have not included here please let me know and I will add them.

- mario

Enterprise Team Foundation Server Management Spec

mrod — Wed, 23 Apr 2008 23:44:36 GMT

As you know we are constantly looking at ways to improve our product in order to meet the growing market demands of our customers. Our enterprise customers today are looking to consolidate and usually decide to purchase one server for their teams and create a significant number of team projects in it.

The problem with this approach is that they lose some isolation capabilities (e.g. work item fields showing for all teams, changesets are interleaved across non-related projects) , and hit product limits (i.e. 500 team projects) relatively fast. In addition, management features like backup and restore of a "single team project" together with load balancing capabilities are a challenge in today's architecture.

Team Foundation Server started a series of investments in the TFS 2008 release and we are now planning the next major revision of the architecture by focusing on a set of server framework services/components, encapsulation concepts and deployment/management tools that addressed the core market problems.

Our vision aligns with the Microsoft’s Promises to IT Professionals and Development Teams and it defines the direction for our current and future investments in the space.

Vision: Manage Complexity, Achieve Agility
Reduce the TCO of deploying, managing, and customizing TFS as an enterprise-wide service so IT departments can focus on delivering new business value

Today we are releasing the first of a series of specifications that have to do with this work. If you get a chance please spent some time reading it and tell us what you think:

Enterprise Team Foundation Server Management

You can find other specs located here: http://msdn2.microsoft.com/en-us/teamsystem/bb936702.aspx

If you do not post to our forum please feel free to drop me a line or send your feedback to mariorod@microsoft.com

Feature Design and Simplicity- Part 3

mrod — Tue, 22 Apr 2008 00:46:43 GMT

I have gotten some emails by readers about this series so I thought to continue the momentum and talk about the next law: Time.

It is no secret that "Thin is in" with customers and that there is nothing more frustrating and at times upsetting than waiting for an application to do some task. TFS has worked a lot on server performance in order to minimize this wait time for our customers and we continually get better. Just recently, we discover a way to improve unshelve and we think we can get a pretty big boost on performance which has us really excited.

On the other hand, sometimes doing back end work is not enough to improve the perception of time and you really have to spend time on the design to make the experience better for our users. Take for example the branch visualization work that we are doing (specs will be publish soon enough for you guys to enjoy).

In Branch Visualization we try our hardest to remove all of the information we don't need and also minimize the experience to excel at the scenario the user is leveraging the visualization for. This allows us to remain very thin and reduce any wait time the user will experience. In the first phase this is not implemented but our second phase will bring those design elements in. The solution in this case was simple, take the navigation of the branches and marry it with the target selection so the user discovers the answer one at a time, to the best of approximations. That is a lot better than sending the server 20 targets, which you may not need, and waiting for the server to come back with the answers.

Here is a screenshot of what it might look like:

So always remember to optimize for time because at the end "savings in time feel like simplicity" for the end user.

- mario

Slick Design

mrod — Tue, 15 Apr 2008 15:55:23 GMT

We Program Managers are always looking for the latest technology, trends and designs that will be enable us to deliver our vision of the feature to our users in its simplest and most powerful form.

I was doing some research on tilt UI and physics UI and I came across this site, which in my opinion is one of the best implemented:

http://www.whitevoid.com/application.html

Now clearly this cannot be used to navigate your source tree, unless it is very small, but you can extract a set of design ideas from it. Let me give you some of the examples that I think make this implementation a good one:

Simplicity - there is no application junk hanging around the site, the content is the UI and nothing distracts you

Color - the background color plus highlight/ illumination centers you on the UI and since it is not static (moving elements) the navigation feel immersing. In regards to colors, I think the color and mood of a UI is one key element of a great looking application

Annotations - The orange callouts (Hot, 20) are a great simple way to get your attention while not distracting from the content. The help content that appears (zoom in, open) are also a great way to tell you what to expect from the click action.

So there you have it, a slick UI with lots of good elements that we can apply to our applications.

Team Foundation Sidekicks 2.1 Release

mrod — Mon, 14 Apr 2008 15:20:58 GMT

Hi,

Attrice has released a new version on their Sidekicks which you can now get here: (http://www.attrice.info/downloads).

In addition to bug fixes, there are two big features of note (more details at http://blogs.microsoft.co.il/blogs/tfsidekicks/archive/2008/04/08/get-team-foundation-sidekicks-2-1.aspx):

· Code review by work item selection

· Integration of Sidekicks into Visual Studio 2005/2008

· View changeset details wherever relevant

Lastly they now provide an undocumented ability to plug in your own Sidekick into VS/application. Send them an email if you want to learn more about it.

Eugene does a great job with these Sidekicks and the majority of the improvements are based on your feedback so if you get a chance check it out and see if there is something in there which your team might get use of.

- mario