This process works well; however, it requires you to configure an IPv4 address on the Galileo platform and potentially set a static IPv4 Address on the connecting computer. If you use IPv6 instead, you can avoid this configuration process all together.
Both Windows and Linux (including the yocto on the Galileo) are IPv6 enabled. As part of the IPv6 protocol, in the absence of a static or dynamic address from DCHP, it will create its own link local address with the fe80 prefix. So technically, you should only have to plug in a patch cable (most Ethernet adapters perform auto crossover) between the Galileo and the laptop Ethernet adapters allowing the TCP/IP network stacks to automatically assign a link local IPv6 address once the adapter link occurs. After which, just connect to the Galileo over the IPv6 address via a SSH terminal. The LINUX_IMAGE_FOR_SD_Intel_Galileo_v0.7.5.7z Boot Image has all the component already enabled to make this process work. If you are booting from Galileo SPI flash and want to use IPv6, you will need to load the IPv6 Module; this can be done with the following sketch command... system("modprobe ipv6");.
Galileo IPv6 Address
Windows IPv6 Address
To get the IPv6 Address from Galileo, run the following Sketch system command to dump the ifconfig to the serial monitor...
system("ifconfig eth0 > /dev/ttyGS0");
Then just use PuTTY or another terminal program to connect to the IPv6 address via SSH.
This process avoids the whole creating a static IPv4 address on both the Windows and Galileo side or setting up bridging.
--Matt Royer