[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
EL6 RPMs for Fedora 14 Issue
- Date: Thu, 22 Dec 2011 08:16:02 -0600
- From: joel.sherrill at OARcorp.com (Joel Sherrill)
- Subject: EL6 RPMs for Fedora 14 Issue
On 12/22/2011 08:07 AM, Ralf Corsepius wrote:
> On 12/22/2011 02:48 PM, Joel Sherrill wrote:
>> On 12/22/2011 06:50 AM, Ralf Corsepius wrote:
>>> On 12/22/2011 01:09 PM, Peter Dufault wrote:
>>>> On Dec 22, 2011, at 6:49 , Ralf Corsepius wrote:
>>>>> CentOS is a server distro, which is aiming at "long term API
>>>>> In general, it's a good choice to run a web-server or a data base
>>>>> on it, but is a bad choice for development purposes.
>>>> I would have agreed until I found the EPEL (Extra Packages for
>>>> Enterprise Linux) packages. Centos/Epel may not be good for working
>>>> on the head, but it is good for working on products, and with the
>>>> EPEL packages I've only had to build a few things myself.
>>> Well, EPEL doesn't change much about the key weakness of CentOS: The API
>>> and ABI freeze CentOS is based on.
>>> I.e. one sooner or later one gets stuck in CentOS being tied to ancient
>>> ABI/APIs which will prevent packages on it to take advantage of new
>>> features or from being added to CentOS at all.
>> And that's what is good about it for "products" as Peter referred to
> Correct ... vs. development == here: rtems-4.11.
>> When the products/hardware is updated, The RTEMS version will be
>> updated to the latest with matching tools and development OS.
> That's not what I am talking about.
> I am referring to people who are wanting to work on/develope on new SW
> on 5 year old designs (CentOS5) or older CentOS4.
> This kind of undertaking is non-reasonable and non-sensical.
> Spear-head development/the bleeding edge needs to take place in a
> rolling release model, until "the development is finished".
> When it's "finished", it's reasonable to switch to freeze the
> environment on what's current then.
>> It may
>> be foreign to your experience but it is the way products with over an
>> 18 month lifespan are supported.
> Don't try to generalize your limited views.
And you have over 25 years of experience in development, deployment,
and support of embedded applications in multiple domains?
At this point, I don't think my experience is limited.