There are serious concerns about the IPv4 address space
limitation and the allocation requests of several GSM operators for the rollout of
general packet radio services (GPRS). Numbers between 32 and 160 million addresses
I think the real question is : Is IPv4 required for all mobile
The biggest 'boom' will be the introduction of Internet-enabled mobile
phones. But do each of these devices require one (or several) IPv4 addresses ? I
don't think so. IPv6 should be a much better solution. See below my arguments.
The use of IPv6 allows for a phone to have a fixed, unique part of the address, perhaps
some form of coding of the phone numer. Another part of the address may be assigned
by the network where the GPRS terminal is currently connected in order to facilitate
- The few Internet enabled mobile phones do not access the current Internet
web servers directly. The low storage and processing capabilities of such devices
are incompatible with the growing complexity of current Internet services. Instead,
these mobile phones have to rely on gateways that convert the data to a simplified
and more compact format. Communication between the mobile phone and the gateway can
use other protocols than IPv4.
- Other (non-WWW) application level protocols require
heavy modifications to be able to talk to GSM phones, either by changing the server
software or by implementing application level gateways. The new software can be made
compatible with IPv6.
- It is extremely unlikely that a direct connection of GSM
phones to PC class computers will be required, at least considering the current applications
that use peer-to-peer connections. Voice-over-IP is a different matter and the existing
standards are not followed anyway.
- There are fair chances that IPv6 will become
widespread until the GPRS rollout will actually take place. Several *NIX-like operating
systems already support IPv6 (including the popular FreeBSD and Linux) and Microsoft
is currently providing a beta version of the IPv6 protocol for Windows 2000.
Last but not least, this is an opportunity to avoid the limitations and drawbacks
of older technologies by implementing better protocols rather than trying to make
a quick-and-dirty patch for a 20 years old protocol.