ext4 supports a feature called 'journalling' which prevents filesystem corruption in the event of a power failure or system crash. When you are certain you don't need it, you can choose ext2. But for the vast majority of users, ext4 is the best option.
The only time you should really specify ext2 is if you are either using a device that you need to minimize the number of writes to (such as a USB Flash drive), or you are creating a read-only file system, such as for an initial RAM disk.
You specify an ext2 file system instead of ext3 when you don't want to use the extended journaling feature that ext3 offers.
Only when sharing the filesystem with another Linux system that uses an older filesystem such as ext2.
UFS is the main filesystem for FreeBSD, but ZFS is also a popular choice. ext2, msdos, ntfs (read-only), and smb are also supported.
EXT2 is a filesystem. Like FAT and NTFS for Windows, It doesn't have a boot record. It all comes down if you use MBR or GPT partition tables on the disk. The former will have what you are looking for, whereas GPT only has a "protective" one.
No, ext2 does not have journaling support. This wasn't added to ext until ext3.
Quite simply, because Windows hasn't been programmed to recognise ext-type filesystems. Out of the box, Windows won't be able to identify ext3 and ext2 filesystems, and will probably consider them corrupted. Fortunately, you can install a driver for ext2 and ext3 filesystems into most versions of Windows. It's called ext2 IFS, and is linked below. With this driver installed, you can mount most ext-type partitions as ext2. (NOTE: ext3 is technically not supported. This means that using ext2 IFS will disable journaling, as that's the main difference between ext2 and ext3)
Part of an answer: Every *nix has its own filesystem. Here's some examples. An arrow "->" means "was replaced by". Linux: ext->ext2->ext3Sun Solaris: FFS->UFSBSD: FFSIBM AIX: JFSHP HP-UX: HFSSGI IRIX: EFS->XFSLinux can read most or all of these.
Ext3 is a tiny bit slower than ext2 is, but it holds tremendous advantages. There is really only one difference between ext2 and ext3, and that is that ext3 uses a journal to prevent filesystem corruption in the case of an unclean shutdown (ie. before the filesystem is synced to disk). That makes ext3 a bit slower than ext2 since all metadata changes are written to the journal, and then flushed to disk, but on the other hand you don't risk having the entire filesystem destroyed at power failure or if an unwitted person turns the computer off uncleanly. You don't have to check the filesystem after an unclean shutdown either. Ext3 has three levels of journalling. Metadata (ie. internal filesystem structures) are always journalled, so that the filesystem itself is never corrupted. How ordinary data is written to the file system is controllable, though. The default option is the "ordered" mode, which causes file contents to be written to the filesystem before metadata is even committed to the journal. The highest reliable mode is called the "journal" mode, which causes file data to be committed to the journal before it is flushed to its final place, like the metadata. The least reliable mode, but rumoured to be the fastest, is called the "writeback" mode, which makes no promises at all regarding the consistency of file data. Only metadata is output reliably in writeback mode. So as for anything else, it's mainly a matter of priority. If you don't want ultimate speed, go with ext3. If you need the highest speed that is theoratically aquirable though, then go with ext2. For that to be effective you'll probably need a really advanced hard drive controller, though.
Typically one uses an older file system to be compatible with older systems that do not support a newer (safer) version.
Block is a generic OS concept. Whenever OS wants to read data from the hard disk, file system tries to read a block of data instead of one character at a time. This improves the performance. (disk is a mechanical device) Size of the block varies for each file system, user can specify the block size when a file system is created on the device. Creating a file system on Linux is equivalent to formating a device on Windows. When you format a device on windows you can specify the block size. To create a file system on Linux mkfs command is used. The default block size is 4K but this can be modified with -b option. For eg. $ mkfs -t ext2 -b 2048 /dev/sdb1 this command will create ext2 file system on the device with block size 2048. There is a limitation imposed on maximum size of block by ex2/3/4 file system. Maximum block size should be 4K and minimum is 512 bytes.
The normal common file system is ext2 or 3.
Most memory cards are formatted with any denomination of the FAT file system (the most commonplace nowadays being FAT32, with a select few opting for exFAT - but not everything supports exFAT). However, you can format the volume with whatever filesystem you desire and what your requirements need (FAT, NTFS, EXT2/3/4, ReiserFS, HFS/HFS+, and so on)
The ext2 file system was a common file system for Linux systems. It has been supplanted for the most part by ext3 and ext4, which are backwards-compatible with it. It is still used on USB drives since it is not journaled, reducing the number of writes made to the drive (but thus increasing corruptibility).