[ale] Maddog’s take on recent Red Hat source distribution changes

Jon "maddog" Hall jon.maddog.hall at gmail.com
Sun Aug 20 17:47:31 EDT 2023


I would also like to say something about the logs being in ASCII as they
were in the early days of Unix.

When I was at DEC we had a friend of mine put forth a proposal for binary
log files instead of the ASCII files that were being generated.   His
proposal was a mechanism that detected if there was any new information
coming from the error routines.  If there was not he would simply increment
the count of the error message and put a time stamp on it.   This reduced
the amount of data that was generated from the error routines.

He then wrote an "error formatter" program which could take the errors
being generated and reconstruct them into what would have been generated
from the ASCII strings.  The program could also let you form windows of
when you are interested in seeing the error messages formatted.

The benefit of all this was that the straight ASCII error messages would
often be generated so fast on the servers of that day (much less today)
that they would fill up the file system and create even more problems
(including lost error messages).

md

On Sun, Aug 20, 2023 at 5:20 PM Jon "maddog" Hall <jon.maddog.hall at gmail.com>
wrote:

> Hi Steve,
>
> You, like so many other people in this argument, appear to be assuming
> that everything in "Linux" is GPL V2.x.   Mostly it is only the kernel that
> is GPL 2.x.   Many other packages in the distribution are other types of
> licenses, produced by many other entities.   I have not looked at RHEL for
> a long time, but I am willing to bet there are packages that are solely
> written by Red Hat, copyrighted by Red Hat and licensed by Red Hat perhaps
> under the GPL, but perhaps under some other license.  There are other
> packages that are under some "permissive" license like BSD, and have no
> requirement to send along the source code to the end user or anyone else.
> Even if there were only packages in RHEL that GPL, the entire distribution,
> the set of bits required to install and qualify as RHEL could have a
> different license.   In the past the only thing that was required was the
> removal of the Trademark information.
>
> If I was "defending" Red Hat, it was only to point out that an "Enterprise
> Linux" is more than just the bits in the ISO.   It is the QA, the channel
> partners, the support people, the development of training and certification
> and many more things.  If all of this was easy to do then certainly some of
> the other distributions that are looking for the enterprise customers would
> be alternatives to RHEL  SUSE has been around as long as Red Hat (maybe
> even longer).   They should be glorifying that Red Hat is changing its
> licensing.   SUSE should simply say "Buy SUSE, it too is Enterprise
> Linux".   Instead SUSE wants to clone RHEL.
>
> My point in writing the article was to bring back to peoples minds the
> original reasons for Free Software.   The fact that people would get binary
> code and not be able to do what they needed with it.  The fact that
> companies would release binary code and then either force you to have a
> maintenance contract with them to get the bug fixes or enhancements you
> need.   The fact that over time the companies would walk away from the
> release you were counting on and never patch it again.  Can you say
> "Windows XP", with close to 12,000,000 systems still running it?  The fact
> that you could never have a third party support organization have as much
> knowledge about the OS as the closed-source engineers, so you were stuck in
> buying your support from the supplier?
>
> Sure, some of the distributions also make it a point of freely
> distributing their code to everyone, whether they are a "business partner"
> or not.   It is one of the reasons I use those distributions.
>
> I also wrote the article because people were saying how bad IBM was to
> Free and Open Source Software.   I wanted to remind people about some of
> the history of IBM and FOSS compared to some of the companies who are now
> beating their breasts and saying how good they are.   The same companies
> that do not release their source code on hundreds of products they ship.
>
> As to systemd vs init files, I have no bones in that argument, just as I
> stay away from vim vs emacs discussions.   However, I do remember the
> discussion about package managers versus distributions via tar files.
> Digital's Ultrix had its own "package manager' named setld(8) and our
> engineers complained about having to put their software into setld(8)
> packages.   "What was wrong with tar" they would say.
>
> I pointed out that tar did not check for dependencies, and there was no
> facility in tar to remove files and not disturb the dependencies of other
> programs while you were doing it.
>
> "Oh yeah" was the response.
>
> Finally, at the bottom of my blog article, was the real meat of what I had
> to say.   In the aftermath of Red Hat's announcement four different groups
> came forward and said that they were going to create a clone of RHEL.   If
> these four groups were going to spend the time and money to each create a
> clone, that means we would have four more "RHEL"s.   I pointed out that
> this was, at best, three clones too many.   I also pointed out that what we
> really needed was a competitor to RHEL, a better RHEL, not just a clone.
>
> But that is hard work, and from my experience would take much more money
> than the 10 million dollars that SUSE was putting forth, even if they only
> managed to copy the RHEL release source code (all of it, GPL or not) and
> rebuild it.   It is the rest of what is needed that would cost a lot more
> to bring to the table.   It probably could be done.  I would encourage it
> to be done.
>
> Sorry if I gave you any other idea of what I was saying.
>
> maddog
>
>
>
>
>
> On Sun, Aug 20, 2023 at 2:52 AM Steve Litt via Ale <ale at ale.org> wrote:
>
>> Scott McBrien via Ale said on Mon, 31 Jul 2023 17:15:00 -0400
>>
>> >IBM, Red Hat and Free Software: An old maddog’s view
>> >lpi.org
>> >
>> >
>> >Maddog, I know you’re sometimes around.  As a long time Red Hatter
>> >(since 2001), not only did I learn some history from your article, I
>> >appreciate its opinion and thoroughness.
>>
>> I liked the history too, but I'm skeptical about the parts about
>> Redhat. Below is section 6 of GPLv2:
>>
>> =======================
>> 6. Each time you redistribute the Program (or any work based on the
>> Program), the recipient automatically receives a license from the
>> original licensor to copy, distribute or modify the Program subject to
>> these terms and conditions. You may not impose any further restrictions
>> on the recipients' exercise of the rights granted herein. You are not
>> responsible for enforcing compliance by third parties to this License.
>> =======================
>>
>> Seems pretty clear to me. If you let somebody have the program, whether
>> for free, for a million dollars, alone or bundled, if the program was
>> GPLv2 you have to give them the source, and more importantly, you have
>> the right to redistribute the source.
>>
>> I know, I know, we can argue about first, second, third, fourth and
>> fifths parties, but section 6 is clear that if you come into possession
>> of the software you have the right to redistribute it.
>>
>> And yes, I know, you can't buy Red Hat without agreeing not to
>> redistribute the source. Once again, Red Hat requires the breakage of
>> section 6. They get away with this only because their army of lawyers
>> provide a chilling effect on a paid-for recipient asserting his or her
>> rights under section 6.
>>
>> Of course, I'm not surprised a bit at Red Hat's Microsoft style action.
>> Within a few years of the spectacular Bob Young's departure, Red Hat
>> became an anathema to Linux, culminating in their promotion of systemd
>> to all distros. They could have kept quiet about systemd, which would
>> have given them a competitive advantage had systemd been beneficial.
>> But they knew that with Debian and Ubuntu as competitors, their sales
>> would suffer, because their distro contained the inferior systemd, if
>> they didn't promote Debian and Ubuntu to incorporate systemd. The
>> purpose of systemd was to complexify Linux so they could sell their
>> services and training.
>>
>> Yeah, I can't prove the final two sentences of the previous paragraph,
>> but here's a smoking gun:
>>
>> http://asay.blogspot.com/2006/10/interview-with-red-hat-cto-brian.html
>>
>> In the preceding link, search for the word "complexity". The only way
>> they make money is if software is complex.
>>
>> In the article, I resent the word "freeloaders". If Red Hat doesn't
>> want other distros to redistribute them, they should switch to Linux or
>> Mac. Once again, I'm pretty darn sure that if there's a deep pockets
>> lawsuit from a Red Hat customer who chooses to follow section 6 of
>> GPLv2, Red Hat will lose, although our present disfunctional Supreme
>> Court might side with them because they're a business. Their license
>> provisions not to redistribute are just there for chilling effect.
>> Trouble is, FSF doesn't have the money to fight a legal battle with Red
>> Hat.
>>
>> The article mentions "right to make a profit." Nobody ever suggested
>> they should give away their training and consulting for free. Just don't
>> violate GPLv2.
>>
>> The article also mentions, and I quote, "if you are not maximizing
>> your revenue with the resources you have, you are not paying fiscal
>> responsibility to your stockholders." Not true. You can't murder to
>> maximize revenue. Not even if you won't get caught. You can't steal to
>> maximize revenue. And I'm pretty darn sure you can't violate a license
>> to maximize profit. The rest of this email concerns my personal opinion
>> of Red Hat...
>>
>> I haven't used Red Hat since 2003, and I wouldn't use Red Hat if they
>> were the last distro on earth. Matter of fact, I wouldn't use any RPM
>> based distro just to make sure Red Hat doesn't somehow mess me up.
>>
>> About systemd: If I and let's say six of my friends were paid half of
>> what Red Hat paid Poettering and his crew, we could have incorporated
>> every feature of systemd without creating a massively entangled mess.
>> If Red Hat didn't enable Poettering, systemd never would have gone
>> anywhere, and Linux would be better for it. Any systemd features that
>> were really necessary would have long ago been incorporated outside of
>> the init system. Systemd was never about improving Linux, it was about
>> complexifying Linux.
>>
>> http://asay.blogspot.com/2006/10/interview-with-red-hat-cto-brian.html
>>
>> I very much liked the history, but I just don't buy the article's
>> defense of Red Hat.
>>
>> SteveT
>>
>> Steve Litt
>> Autumn 2022 featured book: Thriving in Tough Times
>> http://www.troubleshooters.com/bookstore/thrive.htm
>> _______________________________________________
>> Ale mailing list
>> Ale at ale.org
>> https://mail.ale.org/mailman/listinfo/ale
>> See JOBS, ANNOUNCE and SCHOOLS lists at
>> http://mail.ale.org/mailman/listinfo
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.ale.org/pipermail/ale/attachments/20230820/2f273cac/attachment.htm>


More information about the Ale mailing list