Yep, names like p4p1 are soooo much better than eth0 :-(

[Robert Myers]

On Sat, Oct 15, 2011 at 9:30 PM, inode0 <inode0 at gmail.com> wrote:
> On Sat, Oct 15, 2011 at 8:08 PM, Robert Myers <rbmyersusa at gmail.com> wrote:
>> On Sat, Oct 15, 2011 at 8:59 PM, Rahul Sundaram <metherid at gmail.com> wrote:
>>> On 10/16/2011 04:24 AM, John Aldrich wrote:
>>>> On Thu October 13 2011, Darryl L. Pierce wrote:
>>>>>> The adults also realize that Fedora already has a process pretty much
>>>>>> exactly as Thomas described, and participate in it if they want to.
>>>>>
>>>> Really? What is it? How do we access this wondrous special forum on
>>>> changing device names on a whim of a dev?
>>>
>>> Nothing major like this gets changed on whims.  This change was done to
>>> solve real world problems.
>>>
>>> https://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming
>>>
>>> The special forum is just linux kernel mailing list and fedora devel
>>> list where this was broadly announced and discussed.  All major features
>>> follow the feature process
>>> (http://fedoraproject.org/wiki/Features/Policy) and gets voted upon by
>>> the Fedora Engineering Steering Committee which itself is a body elected
>>> by the Fedora contributors.   If you subscribe and follow fedora devel
>>> list, you can follow all such discussions in the future.
>>>
>> A foolish consistency is the hobgoblin of little minds.
>>
>> Your description of the process betrays exactly what is wrong with it.
>>  It's a hobby shop for geeks with a very narrow perspective on
>> computers and how they are actually used.
>>
>> Even Microsoft tests changes on *users*.  Of course, Fedora is sort of
>> a beta distribution for RedHat.
>
> Everyone is welcome to help test new features.
>
> http://fedoraproject.org/wiki/Test_Day:2011-01-27_Network_Device_Naming_With_Biosdevname
>
> Users can with very little effort become contributors in this project
> and helping with testing is one of the places where people with just
> about any skill set can make valuable contributions.
>
So, let's run this process in the imagination.  Everyone involved in
Test_Day:2011-01-27_Network_Device_Naming_With_Biosdevname has been
fully briefed on the changes and has their minds focused on them.
They have not just upgraded or something and are trying to get on with
business.

Let's suppose that one of these lowly testers is insufficiently awed
by the company he is keeping to say, "You know, this is really stupid.
 Things that I have done from memory and flawlessly for years are
totally confusing and screwed up because of your need for
consistency."

What chance is there that such feedback would have any influence
whatsoever on a "major change" already decreed from on high?  This
discussion is pointless.  If you're going to argue endlessly about why
changing "eth0" to "p4p1" would be a major inconvenience to almost all
end users, there is very little point in discussing the matter.

Robert.