[Scuttles]

Where I come from, it is considered bad engineering to build a product that can only run on a single platform, can only run on one particular OS, and the product is at the mercy of future decisions by a 3rd party. We go out of our way to ensure we are not dependent on any single thing.

Even if it was the best choice at the time to go the Microsoft route, it may not be in the future... however -- SE has zero choice now otherwise they'd have to re-write their entire product... that sucks as a business because you have little choice over your own product now.

[GeorgeBeech]

>Where I come from, it is considered bad engineering to build a product that can only run on a single platform, can only run on one particular OS, and the product is at the mercy of future decisions by a 3rd party.

This seems like a bit of a waste of resources to try and run an internal bespoke application on every platform imaginable. Now, if you are developing a product that you are then going to sell to other people to run on their own hardware this make more sense.

There are some very big benefits to writing to a specific platform, especially in the performance space.

In the end of the day it is a trade off between trying to eliminate every 3rd party dependency (next to impossible) or picking a solution or company you think will be around for a long time, and forging strong relationships with them.

We regularly talk to people at Microsoft (and all of the people who create and build to tools we use), give them feedback, and get bugs fixed. There is very little that comes down the pipe from them that we are not away of ahead of time and in some cases have helped shape through early access programs.

> SE has zero choice now otherwise they'd have to re-write their entire product...

This is not true at all, we have choices if MS decided to blow everything up. Not great choices, but we have them. Choosing between two or three crap options does not mean that options do not exist.

-George (SE Sysadmin)