Ritalin shipping….

SELECT METHODS OUT OF NAMESPACES “Nm2″ WHERE
DepthOfCreateA “Nm1.Foo” == 1

In a broader sense, though, why am I supporting following the guidance? Because without experience, changing the recipe has a much higher risk of creating a bad dish. That’s all.

Yes, most Scrum documents talk about 30-day sprints. It’s in Ken Schwaber’s books, it’s on his web site, etc.

http://www.controlchaos.com/old-site/rules.htm

So you see, you’ve just proved my point…

Many people think it’s better not to do 4-weeks sprints but 2-week sprints. Why should we start with the rules of the book, when we know that doing something else is better in our situation?

A team of one? That isn’t a team and as such has no need for a team management model like Scrum, XP, Chrystal, or anything else IMO. I also find myself working on individual projects and without team mates quite often. In those times, I do not use any formal process other than my own to-do list.

I would also argue that even in this situation, the focus can and should remain on value delivery. I can be Lean all by myself.

A team of one? That isn’t a team and as such has no need for a team management model like Scrum, XP, Chrystal, or anything else IMO. I also find myself working on individual projects and without team mates quite often. In those times, I do not use any formal process other than my own to-do list.

Now ask this, “How can I truly excel in a vacuum?”

I don’t agree. This is often just not possible.

How am I supposed to apply the standard 4-week iterations of Scrum when all of my projects only last for 3 weeks on average?

How am I supposed to apply the 5-9 team members rule when I have a team of just one person?

I have blogged about this in “Copy-Paste Reasoning (or Adopt, Skip Adopt)”
http://www.noop.nl/2008/06/copy-paste-reasoning.html

That was my favorite, sounds like the project manager mantra. But hey, they probably stayed at Holiday Inn last night.

I’ve seen this happen with Microsoft Dynamics. Same story there. Instead of ditching the technology, source control got thrown away. If this is what we get from Redmond after 30 years of IT evolution, then it’s simply unacceptable.

Then for me the EF practically excludes source control this way. Why is it then that numerous other ORM solutions out there can get this one right from the first time. As I mentioned in my post, it’s simply a matter of getting the priorities straight. Supporting OO principles and source control are one of those priorities, otherwise it is a useless piece of technology.

I can only recommend to step outside the narrow Microsoft vision, and have a look at other solutions like NHibernate, IBatis or another non-intrusive data solution. When I mean that the EF team are real professionals, then I also mean that the NHibernate contributors are gods!

I want to say kudos the people who started this whole “vote of no confidence” initiave. They are the ones who aknowledge that something has to change in the .NET space, especially regarding some Microsoft products. The ALT.NET community are the only ones I know of. So instead of bashing this initiave, lets improve the way we want to write software for a change.

EF does not exclude the use of source control. It makes it difficult because of more frequent collisions than desirable.

The list you quoted from is my summary of the vote of not confidence petition, stating my opinion that none of these things make EF a faiure.

Data-centric designs in the sense of EF and VB6 are quite different animals as I’m sure you’re aware. Mapping is a non-negotiable. EF provides it, although it clearly has issues.

Metadata is a point of source control collisions. That’s not an easy problem to solve.

Kathleen

You know, it might be fun to keep an archive on a project wiki or something - it’s so easy to forget how far we’ve come sometimes when we’re dealing with the day-in/day-out. It could be cool to be able to occasionally pull out “look where we were just last quarter”.

And just the fact that you were able to put it together so quickly is perfect (I’d be worried people would obsess over getting it just right and so spend a bunch of time on it).

(oh, and BTW, NOT cheating… check out scott berkun’s post on the irony of creative change - I think what you did can fall neatly into that category http://www.scottberkun.com/blog/2008/the-irony-of-creative-change/)

I now realize that the design needs to be changed to be testable to isolate the statement into a separate method.

Bottom line is: “If you can’t test it, your design is wrong.”

Na… I also vote for NOT Cheater.

“Art of the possible” and if I’d have to show the same demo/feature walkthough 4 times right after the other it would be possible that I’d have to kill someone… :p

I think the important point is the intent. The intent was to not waste time. To provide the same experience for all four shows and it wasn’t to hide anything. And given that the demo was based on working, shippable code… (and if challenged you could drop into the app and just do it)

I think you were being smart and efficient.

But yes, as you and Paul suggested, I think the shorter video overview and an ah-hoc demo of the app based on feedback/questions from the current people in the group.

But then again, since I’m SUCH a Scrum noob I could just be talking out my a……

As Paul said, the real value is in having a prepared script so that you can efficiently give a demonstration without forgetting anything.

I think your idea of limiting it to a 5 min demo is good, and would make doing it 4 times less tedious and more interactive as it leaves more room for questions (less is more is this case). If you don’t want questions until the end then you could ask people to save them until you are done (writing them down if necessary), since they will not be waiting more than 5 mins to get an answer.

If you are worried about not showing everything that you want to show then why not follow a basic hand-written script?

Paul.

First Open Spaces meeting I’d ever seen; it’d be interesting to try having brownbags at work in a scheduled meeting room, but with no scheduled/planned topic, using that format. I’m wondering what people in this building want to talk about but never get the chance to talk about, because no one wants to make the PowerPoint slides or get the code samples ready…

Not all projects get this much attention and indeed most of them have just a couple of maintainers. You can always step up to the plate and participate or even donate. If there is a lack of documentation, then by donating the developers can start looking at that.

For me the strongest point in using OS projects like Castle and NHibernate is their strong communities. There are several user groups with a lot of interesting stuff to learn:

http://groups.google.be/group/nhusers?hl=nl
http://groups.google.be/group/castle-project-users?hl=nl

I hope you enjoy the book as well as I have

I would check with Rick Strahl to see if he knows.
http://www.west-wind.com/weblog/

I guess what concerns me the most about using software libraries like nHibernate and Castle Project is that it seems like it’s one or two guys working on stuff in their spare time and updates and bug fixes are done… “when they are done” (there is no set release schedule, and the author’s primary job obviously takes precedence). Also the documentation doesn’t compare to the wealth of documentation and coverage that the Microsoft framework libraries and add-ons (LINQ to SQL, EF, ASP.NET MVC, etc) get. Microsoft just has a ton of internal people writing stuff, and a lot of people outside of Microsoft are also writing about these new technologies.

I guess those are problems that OS has always had, and are not going anywhere soon.

Oh well, I’m excited and interested in finishing the book. I also am participating in the ASP.NET MVC, Javascript for Ninjas (always wanted to be a ninja as a kid!), and ASP.NET Brownfield Development MEAP books. I’m a big fan of Manning (I have at least 10 or more softcover books and a few PDF books) and a long time ago I did some work as a technical editor on Fergal Grime’s .NET book for Manning.

This the guy that translates between the developer and the customer, right?

Wasn’t he the first to go once the “consultants” came in?

/

I find the Tortoise WIndows shell integration very useful for managing those kinds of artifacts.