The moving target of OS support

[Daniel van Vugt]

We did. C++14 was completely unnecessary and the reason why you can't 
build Mir on our latest LTS today (without PPAs etc). The change to the 
code was mostly cosmetic. Yes, that did make our lives slightly easier 
in parts, but to what cost? If the cost is that some great developers 
using trusty never get involved in Mir then I think it's too high.


On 12/08/15 10:09, Christopher James Halse Rogers wrote:
> On Wed, Aug 12, 2015 at 12:03 PM, Daniel van Vugt
> <daniel.van.vugt at canonical.com> wrote:
>> I feel that is just making excuses to not aim higher. The whole
>> platform changes every six months and yes Linux developers are used to
>> the pain that comes with that. But would it hurt us to try and make
>> Mir one of the more stable parts of that platform?
>
> Yes! We don't¹ arbitrarily decide to use new libraries or features
> because they just happen to be there. We use new libraries or features
> because they make our jobs easier. Not using them does indeed hurt us.
>
> ¹: Or, to be pedantic, shouldn't. I don't think we've arbitrarily
> introduced new dependencies in the past, though.
>