Legitimate gripes, but …
#4 MS SQL Server 2008 supports multiple inserts
#6 In .NET development, paging isn’t done in the SQL. It’s done through the data adapter.
#8 Sure MySQL will run on Windows, but you lose stability and performance … and not because of Windows :]
#9 MS SQL is just as free as MySQL. They both have “Community”/”Express” editions, but if you want to get to the enterprise level, they are both very expensive.

MySQL has it’s flaws too, but I won’t get into that.

#3, #4, #6, #7 those imho concern the software that u use and not the core of the DBMS.
For example : You could write a plugin for Management Studio that translates USING to the one you want and the execute the query. Of cource it’s better to have out of the box functionality.
So when you re building a small/medium scale application it’s is not that important what you will choose. All perform pretty well. But when you need something more advaced? Let’s say replication, security at column level etc, MS SQL (or perhaps Oracle) is a better solution from MySQL.
Also keep in mind the Guid feature provided by MS and supported by SQL Server!

Another cool thing in MySQL that it took MS SQL until version 2008 to get an equivalent of:

INSERT … ON DUPLICATE KEY UPDATE … lets you handle the situation whereby some data may or may not already exist, and if it does, you want to update it, but if it doesn’t, you want to insert it. This is reasonably commonly needed functionality.

MS SQL 2008 implements the newest SQL standard for this functionality, namely, the (truly cool) REPLACE statement. But prior to that, even in MS SQL 2005 (let alone 2000, 7.0, 6.5, 6.0, and all prior versions), you’d need either an IF EXISTS (…) BEGIN UPDATE … SET … END / IF NOT EXISTS (…) BEGIN INSERT (…) VALUES (…) END _or_ IF … IN (…) BEGIN UPDATE … / IF … NOT IN (…) BEGIN INSERT … _or_ an INSTEAD OF … Trigger to do that in pure SQL, or else have your client logic handle it.

MySQL has much more powerful data formatting capabilities out-of-the-box. For instance, you could easily format a datetime column as, for instance, “Monday, June 22nd, 2009 @ 12:34 pm” (or even, if you know you’ll be outputting to HTML, “Monday, June 22nd, 2009 @ 12:34 pm”) using a single SQL function call. With MS SQL 2005 and later, you could, if you have sufficient permissions, write a CLR (Common Language Runtime) function in any supported .NET language that leverages .NET’s quite powerful formatting API, and get powerful formatting functionality as an SQL function, but there is none built-in with even remotely that level of functionality.

Some more from my personal list:

1. For one project I found it very useful that MySQL 5 natively handles bitwise operators on 64-bit integers (BIGINT). I was disappointed to discover that SQL Server 2005 only allows bitwise operators on 32-bit integers. If I remember correctly, SQL Server 2005 silently returns bad results for such comparisons.

I believe MySQL developed such support in order to handle ENUM and SET datatypes faster and with more flexibility.

2. As was mentioned above, date manipulation and formatting functions are far superior in functionality and ease of use in MySQL compared to SQL Server.