Jeremy Miller has an excellent post the Anti Team. I thought I would add a few team members to the mix:

• The New Guy - not the newest guy in the team, but the one who is always two steps ahead of everyone else, keep insisting that we should re-write the entire system in XYZ technology of the day. Will go to an event and come back talking excitedly about all the new stuff that is Right Around The Corner.
  Memorable quote:
  
  The new guy; "If you used Ajax, it would be much better".
  Me: "Um... for the ETL processes??"
• The clinger - cling to the techniques and processes that were abandoned long ago, even though he knows they are bad. That is what he knows, and he doesn't like change. The clinger will listen carefully to any instruction about better approaches than the hundred line methods and triple nested case statements then politely ignore whatever you just said to go back to The Way We Used To Do It.
  Memorable quote:
  
  The clinger: "It works and I like it."
  Me: "Yes... but the CC anaylzer is throwing integer overflow errors"
• The blind - can program reasonably well, as long as there aren't any obstacles in his path. The moment that something is not in order, the blind is incapable of handling it and panics.
  Memorable quote:
  

  The blind: "It broke again"
  Me: Why?
  The blind: I don't know.

• That OSS Guy - Grumbles constantly when using closed source products, spends a considerable amount of time hacking various open source projects. Go to sulk in the corner when he cannot use his favoriate OSS library for a project.
  Hm... that reminds me of someone...

The last time that I saw something like this, it was over an operational map and I had to capture a hill.

Will you believe that the end result is even more complex?

This is a great read, recommended.

[Via Stefano] I reached this post about a thought experiment regarding open sourcing SQL Server under the GPL. The author is a former Microsoft employee, and he spends quite a bit of time trying to explain how GPLing SQL Server will not hurt Microsoft business model. The terms that he suggests are similar to the way MySQL works, a GPLed products for free and a commercial version + support avialable for $$$.

I just don't see the point here. SQL Server is one of those products that I feel that I can rely on with my eyes closed, and while having the source might be nice to DB geeks out there, I just don't see the big value there. (Having the source to the client UI would be nicer, but even this one has a good extention model for what I want.)

Frankly, as an OSS proponent, I just don't see the point. Microsoft doesn't stand to gain anything from this, except lose potential revenues. From my projects alone, I can count about ~100,000$ in licensing fees for SQL Server that wouldn't have made it if SQL Server was avialable under the GPL*.

SQL Server is one of the revenue streams of Microsoft, and I don't see this changing. Offhand, I can't recall any successfull, revenue generating, product that was open sourced. If I were to make that decision, I would have definately said NO.

The BCL, on the other hand ... :-)

* Choosing SQL Server is mostly influenced by the fast that Microsoft is behind it, that it has great tools and a lot of community behind it.

I package it without including the app.config, which means that the application fails when you start it up.

Please try downloading it again.

Part of my job is interviewing developers, and I get all sorts, some that sticks to mind:

• ASP.Net programmer - 2 years experiance - couldn't explain how clicking on a button on the client cause the event to fire on the server. "It is ASP.Net that does it." - he kept saying.
• An experiance programmer, that had a difficulity reversing a string.
• ASP.Net programmer, nearly broke down and cried when I showed him HTML - "give me back the designer, I don't know what this stuff is"

I had a discussion today about the value of using a non-Microsoft framework for a complex application. The point that came up repeatedly was that they want to jsut grab a programmer from the street and have then start fixing bugs from day zero. For any but the most trivial applications, that is not going to happen. They are seeing the same types of people (appearing with 3 years experiance with MPL* and 15 years of .Net experiance).

The argument against going with the non-Microsoft route is that they need to train developers for that, and that this is too hard. My thoughts on this subject are well known. I refuse to program to the level of someone that is not willing to make the effort. That is far too limiting, and usually lead to the Throw Code At The Problem syndrom, after which the Big Rewrite Comes To The Rescue.

*MPL == My Proprietary Langauge

Update: After writing this, I run into these links:

• Safety versus Power
• Protecting Developers from Powerful Languages

When you write a piece of SQL, and then spent 20 minutes apologizing in the docs for writing it, and another 30 minutes removing it afterward.

And accessible from http://wiki.ayende.com/, this time on ScrewTurnWiki rather than MediaWiki.

Huge thanks to Papo, which provided me with a nearly complete copy of the old wiki.

The only thing that I am still missing is the NHibernate Course page, now.

 

There is a new version out, with a couple of bug fixes (handling properties that throws exceptions, handling null results from queries). This version is also compiled against NHibernate 1.2 Beta 3, so hurry up and upgrade.

You can find it here.