Most European IXPs grew from non-commercial ventures, such as research organisations (e.g. CIXP, which developed at CERN), academic institutions (e.g. VIX, which was started at the University of Vienna), or collaboration between ISPs (e.g. LINX, which was formed by 4 commercial ISPs and the UK academic network). A number of European IXPs have expanded their operations into multiple sites, usually using commercial co-location space where the original sites can no longer cater for the number of connected ISPs, but their management has remained non-commercial. By comparison, in the US the majority of IXPs are commercial, and some commercially run IXPs have emerged in Europe (e.g. XchangePoint in London).
Whether a new IXP should be commercial or not is a complex question, and a full discussion is beyond the scope of this document, however it would appear that most of the emerging IXPs have opted for a non- commercial approach. There are several possible reasons that this is the case. Firstly, most existing IXPs have evolved from a common requirement of their founders to improve Internet connectivity rather than being built as a company. Secondly, given the involvement of non- commercial academic, research and governmental institutions, it is very likely that it is much easier to start an IXP as a co-operative effort without having to establish all the various elements that need to be in place in a commercial company. Thirdly, a non-commercial entity is possibly better placed to maintain neutrality.
