Porting X11Libre to FreeBSD.

I have been following but remained quiet so far because I think my bias is quite obvious. xlibre-server, xorg-apps, xorg-fonts, xlibre-drivers, xorg-utils should just be one monolithic package.

But.... I appreciate that is not how the port is going to be best received by the project ;)

I think your choice is fair. I guess if people do want the full Xlibre distribution then an xlibre meta-port could then pull them all in (servers and auxiliary) just like the existing xorg port.

Have there been any groans from the maintainers of the existing Xorg port? Perhaps this is an opportunity to simplify and sidestep any issues that would have caused breaking changes with xorg.
we currently have "xorg" and "xorg-minimal". I'm not sure of the exact difference but I believe they pull in different. So both are "meta-ports" but minimal pulls in a subset of the other.

To me, the question becomes "does having separate ports for apps, fonts, drivers, etc make it easier to maintain"? As in how often are the components updated vs the overall? Kind of like "pkg-base" :) yes poking the hornets nest there.

So xorg/xlibreI personally don't care if monolithic or meta ports, I just want it to work.
 
I knew that OpenBSD was doing it and I think that's because they have their own fork of Xorg, Xenocara. In my opinion, X11 shouldn't be in the base at all because it isn't a software that everyone needs and should stay as just a port.
Not in NetBSD and OpenBSD ;) .

I think there is a difference between "OpenBSD installing it by default" and "It's part of base". A long long time ago, FreeBSD installer roughly asked "are you installing a server or workstation" Picking that meta option caused a predefined set of packages to be installed. Choose Workstation, you got X installed. Server, X not installed. So not part of base but automatically installed based on specific choices.
OpenBSD/NetBSD installer may have made the decision to "always install X" which is different from "X is part of base". I don't believe that OpenBSD/NetBSD considers X to be part of base but rather something to always install. I could be wrong, but I've got a 50/50 or 6's/3's chance of being right (don't ask my wife)
 
we currently have "xorg" and "xorg-minimal". I'm not sure of the exact difference but I believe they pull in different. So both are "meta-ports" but minimal pulls in a subset of the other.
I think xorg-minimal lacks drivers (i.e for people who don't want to pull in the 1mb intel X11 driver when they use the X11 modesetting driver). It also lacks things like xeyes because they are seen as redundant for most people.
To me, the question becomes "does having separate ports for apps, fonts, drivers, etc make it easier to maintain"? As in how often are the components updated vs the overall?
Indeed. Splitting it apart was probably one of the factors that Linux has stagnated as a desktop platform for a decade.

The distributed / bazaar approach to development is often fast and flexible but for big projects with lots of interconnecting things, the monolithic approach is better. Potentially people will see this if PkgBase ever happens too.
So xorg/xlibreI personally don't care if monolithic or meta ports, I just want it to work.
I suppose a main concern is that it is easy for developers to access and work on. One of the nicest things about CDE is that it is monolithic and all-in-one. So its quite easy to add / fix stuff. Xorg/Xlibre/Xenocara I feel could work well like this.
 
I think there is a difference between "OpenBSD installing it by default" and "It's part of base". A long long time ago, FreeBSD installer roughly asked "are you installing a server or workstation" Picking that meta option caused a predefined set of packages to be installed. Choose Workstation, you got X installed. Server, X not installed. So not part of base but automatically installed based on specific choices.
OpenBSD/NetBSD installer may have made the decision to "always install X" which is different from "X is part of base". I don't believe that OpenBSD/NetBSD considers X to be part of base but rather something to always install. I could be wrong, but I've got a 50/50 or 6's/3's chance of being right (don't ask my wife)
Yes, I understood the difference now but still, I prefer to have Xorg installed onto /usr/local but not onto base system like /etc. I wouldn't like FreeBSD to make an offer to install X. I like building my own system on my own piece by piece and I think most of FreeBSD users would prefer that too. I mean FreeBSD installer shouldn't prompt for anything desktop related but let you alone on your own. I like the separation between base system things and packages/ports.
 
Xorg should be kept out of base, along with upstream Linux graphics stuff. Having “meta releases” should probably be maintained within a separate project or SIG. The way Ubuntu does it IMO.
 
OpenBSD and NetBSD provide Xorg/Xenocara as a "set".
Its not quite "base" but is actually what FreeBSD used to do when Xorg was monolithic, clean and not tangled in with other random ports as dependencies.
FreeBSD has very few sets now. I think just kernel, base and (arguably) ports.

The Linux graphics stuff is so tightly bound to a specific kernel version that it isn't a perfect candidate for packages. But I agree, it isn't perfect for base either. Included in the future(?) Xlibre "set", specific to a FreeBSD release could be a good compromise though.
 
OK, a few clarifications. I think I have used some inaccurate wording that leaded to a misunderstanding.
  • What I meant by "a base set" is not the X server being included in the base.txz set, what I meant by it is the X11 server and et al. being packaged into their own optional sets that then can optionally get installed during the installation process to the base folders (/usr/X11R* and /etc/X11) like in the other BSDs.
  • I fully support the Idea of the X11 sets being optional, not every one needs them.
  • I don't care that much about the exact location of the installation, be it /usr/X11R*/ or /usr/local/X11R*/, What I like is all of it being centralized to a single location.
  • OpenBSD and NetBSD technically don't have their own full forks of X.Org, they have their own frameworks around X.Org that provides them with their needs this includes build system modifications, patches and ... , But at the end they still track changes from X.Org. Xenocara.org and X.Org on NetBSD good reads on the topic.
  • In OpenBSD and NetBSD the X11 sets are optional you can select them during the install time if you want to, these are some pictures of the installers first two are from NetBSD, The last is OpenBSD:

inst-install-type.png

inst-sets.png

VirtualBox_openbsd61_18_08_2017_13_28_53.png
 
Cart before the horse way of thinking. Code is a prerequisite to all of this.
Yes, code is important, but also engagement.

And call it what you want, it doesn't change the fact that Xlibre is still in its infancy and therefore not a well maintained project. It sooner or later will have to figure out in some way the points I've raised.

And only if it is able to do so, and to release in a reliable manner, it can be rightfully named a well maintained project. At the moment it is not.

It's a fork with potential by a developer with questionable reputation among the community. Nothing more, nothing less.
 
OK, a few clarifications. I think I have used some inaccurate wording that leaded to a misunderstanding.
  • What I meant by "a base set" is not the X server being included in the base.txz set, what I meant by it is the X11 server and et al. being packaged into their own optional sets that then can optionally get installed during the installation process to the base folders (/usr/X11R* and /etc/X11) like in the other BSDs.
  • I fully support the Idea of the X11 sets being optional, not every one needs them.
  • I don't care that much about the exact location of the installation, be it /usr/X11R*/ or /usr/local/X11R*/, What I like is all of it being centralized to a single location.
  • OpenBSD and NetBSD technically don't have their own full forks of X.Org, they have their own frameworks around X.Org that provides them with their needs this includes build system modifications, patches and ... , But at the end they still track changes from X.Org. Xenocara.org and X.Org on NetBSD good reads on the topic.
  • In OpenBSD and NetBSD the X11 sets are optional you can select them during the install time if you want to, these are some pictures of the installers first two are from NetBSD, The last is OpenBSD:

inst-install-type.png

inst-sets.png

VirtualBox_openbsd61_18_08_2017_13_28_53.png

That should work. I believe there’s current work on graphical installer by the committers; I wonder if they'd include an install option like mentioned above. I think it’d be a good compromise between keeping vanilla clean and ease of configuration for people.

it doesn't change the fact that Xlibre is still in its infancy and therefore not a well maintained project.

You know it’s cold outside, when you go outside and it’s cold.

Rome wasn't built in a day.
 
Interesting.

It seems the X.org maintainers themselves have been outright gatekeeping patches and improvements from contributors, for the past four years. Insane. Things make more sense now.

I’m really looking forward to this port.
Dang, didn't knew about it...

After reading that, my views had been changed against XLibre.

Thanks.
 
To get philosphical for moment, the guy says that they are keeping him out, with the implication that his stuff is good, and only not getting in becaue of prejudice. I don't know enough about coding to judge, but Musashi, the swordsman of the 1600's (and some 1500's) once said, As you do one thing, so you do 10,000. Actually, I assume he said it in Japanese, but, whatever. 一事万事
(Ichi ji, ban ji)

This guy is an anti vaxer, which indicates that he tends to ignore common scientific knowledge for unproven pop and rather stupid theories--those of you in the US, where measles is making a comeback because of similar ignorance know how dangerous this can be. .
So, I wonder if it's a matter of him having the same attitude of ignoring what some experienced people say to go his own way--in other words, going about it as he does his vaccination beliefs. And I wonder if *that's* what his problem is, that the xorg people are ignoring him the same way legitimate scientists would ignore his anti vax beliefs.
 
To get philosphical for moment, the guy says that they are keeping him out, with the implication that his stuff is good, and only not getting in becaue of prejudice. I don't know enough about coding to judge, but Musashi, the swordsman of the 1600's (and some 1500's) once said, As you do one thing, so you do 10,000. Actually, I assume he said it in Japanese, but, whatever. 一事万事
(Ichi ji, ban ji)

This guy is an anti vaxer, which indicates that he tends to ignore common scientific knowledge for unproven pop and rather stupid theories--those of you in the US, where measles is making a comeback because of similar ignorance know how dangerous this can be. .
So, I wonder if it's a matter of him having the same attitude of ignoring what some experienced people say to go his own way--in other words, going about it as he does his vaccination beliefs. And I wonder if *that's* what his problem is, that the xorg people are ignoring him the same way legitimate scientists would ignore his anti vax beliefs.

You’re comparing Apples to Airplanes. Let’s leave our socio-political opinions and/or biases out of this.
 
This guy is an anti vaxer, which indicates that he tends to ignore common scientific knowledge for unproven pop and rather stupid theories--those of you in the US, where measles is making a comeback because of similar ignorance know how dangerous this can be. .
He could believe that a $DEITY made the Earth in 7 days for all I care. If his technical skills are up to scratch, he can still produce great results with a computer (just don't position him as head of the NHS or equivalent).
 
Well should the personal believes and behaviour of a software developer matter for the project? Normally not so much, if he's not committing criminal offences.

But we are living in a world today, where Code of Conducts are widespread and adopted in many projects (which could be the topic of another discussion if this is a good or bad thing).

So people are today more sensitive to such things. Also coding always contains a big social component as well.

So in short: yes, it matters to many, because coding is not just a technical task but involves many human interactions as well.
 
Hmm... May have found a bug in bsd.port.mk and makesum.sh.
The makesum target errors out when relative paths are set for the DISTINFO_FILE, because makesum.sh is not cd'ed into the current directory.
 
To get philosphical for moment, the guy says that they are keeping him out, with the implication that his stuff is good, and only not getting in becaue of prejudice.

I'm the author of the article being referenced. I just want to clarify that it's not just "one guy", there's dozens of contributors whose patches have not been released for four years.

The political views of that one guy is just a red herring.
 
So people are today more sensitive to such things. Also coding always contains a big social component as well.

So in short: yes, it matters to many, because coding is not just a technical task but involves many human interactions as well.

If people are manipulated to a particular view that's not beneficial for technical code, that seems an issue.

If X11Libre was faster than Xorg, any OS that doesn't have it for reasons other than someone not packaging it yet, it'll be an arms race for me to see what OS is serious enough to use (FreeBSD's the first I've seen semi-interested). I like performance :p

I'm certain every contributor to open-source has something in their past that could be blown-up today to make them seem bad, which makes little sense considering the code done was usable and probably good before the media frenzy. I'm still not convinced code can be discriminatory, so what would other people have against contributing to code? Or, if Wayland was as good as people claimed it to be for at least 8 years, why would this be a discussion now?
 
Back
Top