- From: Timothy Holborn <timothy.holborn@gmail.com>
- Date: Sun, 4 Jan 2015 19:14:35 +1100
- To: public-rww <public-rww@w3.org>, "public-webid@w3.org" <public-webid@w3.org>
- Message-ID: <CAM1Sok0ETK72mT0kzd4XS-V864vTCHqw2+cEuon06s57tvFqaA@mail.gmail.com>
Thinking about the domain / subdomain issue [1]. Problem essentially appears to be a financial issue, being the cost of a domain and wildcard SSL certificate. Therefore; The utility of SubDomains appears to be the cheapest method to establish unique storage. a problem therein becomes how to define these subdomains in a systematic manner, as to support the utility of these subdomains, on mass, without naming conflicts, etc. Why not use IPv6 IP Addresses as subdomains?. first issue that comes to mind, is the issue of subnets. IPv6 may offer opportunities to provide a static IP address for user-accounts, and in-turn, user-devices, but that means allocating a subnet not solely a specific address. When allocating an IPv6 Subnet, a /56 allocation would provide seems practical. My theory is that notwithstanding a user moving between one country and another, might this provide a means to supports portability of information from one server to another, should the end-user migrate between providers, and the creation by users of a multitude of 'subdomains' (data-spaces) within their allocation of address-space, without being required to pay for domains and wildcard certificates. perhaps the underlying assumption is that this may be a quick and effective method to mitigate some of the business models embedded in the web; that make it difficult to implement RWW / LDP, personal data storage, etc. Another consideration relates conceptually to whether the use of 'human names' (ie: domain names) is content or data? I can see a range of implications herein, and obviously the concept in-turn relates also to WebID-TLS; yet was interested in the thoughts of others. Tim.H. [1] https://www.youtube.com/watch?v=AzMmVr2VUFU#t=1930
Received on Sunday, 4 January 2015 08:15:15 UTC