[olug] CentOS 6.0

George De Bruin sndchaser at gmail.com
Mon Jul 11 20:38:41 UTC 2011


I'm actually really trying to understand how hard the whole process is.

On the one hand, I can see where combing through each RPM, and replacing
strings, and images, etc. is a really difficult job.  OTOH, I wonder if
there would at least be some way to automate an identification system that
could at least fingerprint the known-good files, and produce some kind of
three stage report on the files (ie, green light is fingerprint matched,
known good or "clear" files, red light is no-fingerprint, definite copyright
identification, yellow light could be non-fingerprint matching - but signs
of GPL or CC license identified, or clear signs of non-Copyright / Patent).

I'm about to embark on building my own cheapie "de-duping" script for home
use.  I know there are plenty of products that can do it already, I also
know that I want to have some of the fun of trying to work out a sound
methodology for myself.  (It's really the first step to a set of utilities
that I want to build...)  So, I would think that if there is a way to easily
define a de-duping methodology, then it shouldn't be too difficult to work
out some kind of fingerprinting mechanism for identifying known good vs
known not-good files for CentOS should be possible.

George

P.S. How does Scientific Linux manage to do it?  Seems to me there could be
some cooperation between those two groups.

On Mon, Jul 11, 2011 at 10:42 AM, Dan Linder <dan at linder.org> wrote:

> On Mon, Jul 11, 2011 at 09:50, Jon Larsen <jon at jonlarsen.us> wrote:
>
> > I wonder if adding delta rpms contributed to the complexity of the build
> > system.
> >
>
> I can't tell, nor do I believe anyone outside of their development team
> knows...and that's a root problem from what I've seen.
>
> I haven't asked to help directly (I just don't have the free time to give
> back more than my ramp-up time would take), but from what I've read on many
> blogs is that the few outsiders that do get into the development team were
> completely baffled by the lack structure to the "system".
>
> From the outside, this anarchy seems to have been most visible during the
> accidental hostage situation a year ago when a lead contributor went dark
> and stopped working with the community (
> http://news.cnet.com/8301-1001_3-10300222-92.html).
>
> I hope they can keep it up and automate the heck out of the build process.
>  Since they don't have any major financial backers (that I know of), they
> are always going to be cash strapped.  If they had an automated build
> system, I'd donate many spare CPU cycles to do test builds if it were
> possible.
>
> Dan
>
> --
> ***************** ************* *********** ******* ***** *** **
> "Quis custodiet ipsos custodes?"
>    (Who can watch the watchmen?)
>    -- from the Satires of Juvenal
> "I do not fear computers, I fear the lack of them."
>    -- Isaac Asimov (Author)
> ** *** ***** ******* *********** ************* *****************
> _______________________________________________
> OLUG mailing list
> OLUG at olug.org
> https://lists.olug.org/mailman/listinfo/olug
>



-- 
---
Faster moments spent spread tales of change within the sound...



More information about the OLUG mailing list