I came across an interesting concept the other day of a developer of an open source project who was developing the project “behind the scenes” with out any feedback to the community, now this stuck me as odd while first reading there comments about what they were doing.

What had happened in this case was the developer wasn’t happy with being pushed in any particular direction on there project so they had stopped openly advertising what they were doing on the project, with out stopping other people from using there code, this is where i came up with the concept of Open Source Vs Open development.

When most people think of open source the link these two concepts  of not only having the source code available but also being directly involved in the development cycle of the project through bug reports, submitting code patches etc, and most project this is how they are run, however in this case it was no longer occurring as the developer had no wish to implement new features and was more focused on maintaining it and fixing bugs at there own pace.

my question to you is have you come across a case of open source but not open development? do you think there is a place where this makes sense for a project to be developed with this methodology? or should all open source projects also employ open development?

Well recently I got IPv6 enabled on the VPS that is currently running my mail server, now this I thought would be a great opportunity to start to move into the new generation of IP on my hosting by getting this server enabled. when this was assigned I was interested to note my humble VPS with just 128mb of ran was not allocated a single IP address but a whole subnet that I could use. Now knowing that IPv6 has some 340 trillion trillion trillion unique addresses I wasn’t so surprised by getting more then one address, but my surprise came when I found out how much I was given.

My VPS was assigned 2607:f128:42:d1::2/64 as its IPv6 allocation, now that may not seem to imposing at first glance but a small amount of research and you will see that this allocation contains more IP address space then the ENTIRE ipv4 address space, this makes me wonder why this is the case, and why the MINIMUM allocation any of the regional address providers will give out to an organization is a /32 block, this seems wasteful and unnecessary.

Now I understand that IPv6 has been designed so there are far more addresses then we will ever use but is that a reason to waste them like this so early in its adoption. as was the case with ipv4 where early in its life cycle large blocks were handed out to the point where we now see that things like NAT are required to keep even a portion of the address space available for use.

Would it not be more prudent to give a single server a much smaller allocation of ip addresses (perhaps just 1 even?) rather then taking such a huge chunk out of the address pool we have seen the effect this has had on IPV4 with some organizations being given entire /8 blocks of address space early in its life cycle (although some have since been returned for reallocation).

Now this may never become an issue with the huge address pool available to us through IPv6 but I believe that this should at least be considered before this becomes so wide spread that every IPS and hosting provider in the world is using a /32 and they are running out.