2024: The year of desktop FreeBSD?

LOL 🤡

For anyone who's curious,

networking that's already weird enough (with traditional em(4) and iwm(4)).

– an em(4) wired networking example. Sleep the computer at home, un-dock, dock it at work: wrong DHCP address. Basic stuff.

Note, I'm not seeking help (after years of frustration, I arrived at a workaround that's reliable enough). Just laying it on the table.



Postscript​

 
May be /etc/rc.d/* scripts need suspend/resume sub-commands....

Up until now people just edited /etc/rc.{suspend,resume} at the indicated places to handle local conditions, unload and reload various subsystems etc.

But testing for existence and running of specific (perhaps) /etc/rc.d/{suspending,resuming} commands instead sounds a good idea, keeping the original rc.* files standard?
 
Ghostbsd.org/download "latest build" dd write to a USB flash drive. Insert to a PC and boot. You can test the Live Media version on your PC hardware with NO INSTALL. PC needs 4GB dram. There you can have a working "pseudo FreeBSD" Desktop and then get on with your real work tasks.

For those with Raspberry Pi 4B,3B,400 and an interest in a XFCE4.18 BSD Desktop O/S ; I could use some help testing GhostBSD-Arm64 on your Raspberry Pi.

https://ghostbsd-arm64.blogspot.com/2023/12/how-to-install-ghostbsd-arm64-into-usb.html
Download this one shell script file. or use from the ssd_test directory. http://ghostbsdarm64.hopto.org/packages/write_arm64_image_file_dd_to_da1.sh

1.) write_arm64_image_file_dd_to_da1.sh Start with this script to write an image into a USB flash disk drive stick http://ghostbsdarm64.hopto.org/packages/write_arm64_image_file_dd_to_da1.sh
scipt will Resize the 10G UFS parition image to a better larger size to fit in your USB flash drive stick.
next boot your Raspberry Pi 4B,3B,400 with this USB flash drive stick. You can edit -s80G to your preferred size for your USB Flash Drive.

2.) create_disk_da1_ssd_partitions.sh This will create 7 partitions on a USB SSD pointed to by ${DISK1} environment variable
 
(Above, I added a link to a related topic.)

That's a bit cryptic, Graham. I only found out what you meant because I'd already had that thread marked 'watch' or I'd be none the wiser; I certainly can't keep up with every link like this one, "comment 7" above:

https://forums.freebsd.org/threads/sleep-resume-caveats-and-gotchas.90831/post-636663

Nonetheless, thanks for info about the new suspend and resume KEYWORDs in rc.d/ scripts to accomplish what bakul suggested and I blithely speculated about above.
 
I disagree. We need a few desktop style #1 meta-port, desktop style #2 meta-port, and so on. Then simply pkg install one or the other. Someone with a good idea of what a simple desktop might look like should be able to cobble something together. The problem is that there are way too many options and way too many desktops available. Everyone will want to set it up their way and I doubt there will be any consensus of what the perfect one or two desktops might be. I'm hardly a good judge of that because I, an old fart who's been doing this way too long, am stuck in my old ways using CDE or other legacy desktops. And most people don't want to see or use that, though a few do.

I think something light such as xfce or lxde might be the best. The desktop installer port/package IMO tries to be all things to all people. It's too complicated for a brand new user with no previous computer experience. We need to keep it simple. The fewer the options the better, there's less to break or go wrong for new users.
I doubt a new user who is unable to read the handbook would be able download the install image to an installation media and begin the install in the first place.
 
"The year of desktop FreeBSD" was 2015 for me. 🤷‍♂️

It only happened because of increasing frustration with things breaking in my previous Debian setup. The final straw was somehow systemd-related, but I don't remember details. I then gave 11.0-CURRENT a shot (-RELEASE didn't fit the bill because it was lacking GPU support back then), found lots of stuff I liked and migrated my server to FreeBSD as well a while later. I like to start tinkering with something new on the desktop, IMHO you quickly get a better feeling for the system that way.
 
I doubt a new user who is unable to read the handbook would be able ...
The bigger problem is that a growing number of people seem to claim that they have a right on convenience. It's not that they cannot read or aren't able to understand. They just avoid any efforts.

When some BSD affiliated fans demand growth of the user base, those consumer minded people get on board. Make sure you you have a lifebelt usable then.
 
I doubt a new user who is unable to read the handbook would be able download the install image to an installation media and begin the install in the first place.

The thing is, people don’t have time for that; especially in production environments (ie. corporate offices, SMEs, production studios, etc.). They want something tangible and readily available to install and go. This is why macOS, and to lesser extent RHEL are sought after. They don’t to waste time and energy reading a handbook and mucking with a shell. They want to use applications to get sh*t done.

To piggyback on cy; just pick one.. just one desktop for people to use and for application developers to target; that’s it. This whole BuT mUh cHoIcE argument is bikeshedding nonsense. If someone isn’t satisfied they can grab the headless ISO and do their own thing.

Anyway, I vouch for a fork/fix of gtk2/gnome2.
 
Now what could be considered as a good advance in FreeBSD in the year 2024?

Hard to tell but not so hard to look at what failed in 2023:

 
"The year of desktop FreeBSD" was 2015 for me. 🤷‍♂️

2003 here, but this is my oldest screenshot from a quick scan of the archives. FreeBSD 4.5 on paqi, a Compaq 1500c (that ran 24/7 from 2001 to 2019).

It didn't seem like any big deal.

desktop1.png
 
… PC-BSD addressed the problem. …

I can't find any shots of my own from that era. Instead, here's a 2008 shot of PC-BSD on a 1998 computer:

 
I'm seeing a conflation between 'Desktop' as in 'Desktop Tower" and 'Desktop' as in 'Desktop Environment', like KDE... not to mention that KDE can run just fine on a laptop, and a laptop can sit just fine on the top of a desk.

What do you call a 17-inch, 6 Kg (12 lb) 'laptop' machine that is unwieldy to carry around (but has internal components that are difficult to replace with aftermarket stuff) ? 😩 'Desktop Replacement' 🤣 ?

I do try to use FreeBSD as a 'Daily Driver', but sometimes, Windows is just the tool that allows you to waste less time troubleshooting. Yeah, it's nice if you get your webcam and sound working properly for a Skype/Teams conversation on FreeBSD. But unless you're willing to spend the time it takes to get it to work on your personal hardware, it may be easier to just use Windows for the task.
 
The thing is, people don’t have time for that; especially in production environments (ie. corporate offices, SMEs, production studios, etc.). They want something tangible and readily available to install and go. This is why macOS, and to lesser extent RHEL are sought after. They don’t to waste time and energy reading a handbook and mucking with a shell. They want to use applications to get sh*t done.
Yes, these folks get paid for the sh*t. They make money.
So you say, somebody else should sit down and make the install more convenient, for nothing, so that these can make more money?? Are you kiddin' me?!?!?
 
Yes, these folks get paid for the sh*t. They make money.
So you say, somebody else should sit down and make the install more convenient, for nothing, so that these can make more money?? Are you kiddin' me?!?!?
I question if the entire premise is relevant. If the development is being used in a corporate use case these developments can be handled internally. Or the above named alternative systems can be used if they are a better fit for the desired work flow. I don't believe there is a real need here for this type of installer. If so it can certainly be developed as needed. Otherwise it cannot be considered a priority.
 
I question if the entire premise is relevant. If the development is being used in a corporate use case these developments can be handled internally. Or the above named alternative systems can be used if they are a better fit for the desired work flow. I don't believe there is a real need here for this type of installer. If so it can certainly be developed as needed. Otherwise it cannot be considered a priority.
No, it's actually an old stance, probably from the times of BOfH: "What do you mean? Do you think we have time to read what's written on the screen?"
 
No, it's actually an old stance, probably from the times of BOfH: "What do you mean? Do you think we have time to read what's written on the screen?"
It is possible for some users the prerequisite of literacy is too much to ask. In this case a simpler pre-installed solution should be used. I do find the idea that a user would download an image and burn or write that image to an installation media yet be unable to follow a written guide regarding additional system changes hard to believe.
 
Yes, these folks get paid for the sh*t. They make money.
So you say, somebody else should sit down and make the install more convenient, for nothing, so that these can make more money?? Are you kiddin' me?!?!?

The point is for people to use it and get work done. That's the entire point of FreeBSD existing. Whether they make money or not is irrelevant. It's a utility, not some uber-nerd-shell wrangling cult. You do realize BSD was birthed from a University? People need to get sh*t done. Shipping a default desktop is not as ridiculous as you may think.

No, it's actually an old stance, probably from the times of BOfH: "What do you mean? Do you think we have time to read what's written on the screen?"

Expecting mere mortals to waste time engaging in basic desktop administration is crazy and dismissive... much like a BOfH. It's obvious you've never administered desktops in these kinds of environments. In retrospect, macOS may be over-simplified but I've had a way easier time fixing crap and receiving less complaints in comparison to Windows. I can only imagine the stress/pain involved if you were to put someone in front of a shell and asked them to do post-mortem configuration to get a desktop up. They'd look at you sideways.

Even in the 80s people were using NeXT and SUN workstations with default desktops. The stance will always be legitimate. The point of a desktop is to use applications, not to feel superior with pointless mucking with a shell. It's not practical.
 
The point is for people to use it and get work done. That's the entire point of FreeBSD existing. Whether they make money or not is irrelevant. It's a utility, not some uber-nerd-shell wrangling cult. You do realize BSD was birthed from a University? People need to get sh*t done. Shipping a default desktop is not as ridiculous as you may think.



Expecting mere mortals to waste time engaging in basic desktop administration is crazy and dismissive... much like a BOfH. It's obvious you've never administered desktops in these kinds of environments. In retrospect, macOS may be over-simplified but I've had a way easier time fixing crap and receiving less complaints in comparison to Windows. I can only imagine the stress/pain involved if you were to put someone in front of a shell and asked them to do post-mortem configuration to get a desktop up. They'd look at you sideways.

Even in the 80s people were using NeXT and SUN workstations with default desktops. The stance will always be legitimate. The point of a desktop is to use applications, not to feel superior with pointless mucking with a shell. It's not practical.
Perhaps it's not practical that you use FreeBSD for your "desktop" tasks. I have found the setup of a FreeBSD system to take very little time even when using it as a "desktop" or workstation system. It does sound like you have a design in mind but you expect others to do the work for you. I would recommend you consider contributing your ideas to the GhostBSD efforts if possible. They do have a default desktop and a live system as well. You may have some input that could be valuable in their operating system.
 
You do realize BSD was birthed from a University?
Yes, exactly. Science. Increasing the realm of knowledge. The greater good for mankind. That's quite different from unpaid work for shops.

Expecting mere mortals to waste time engaging in basic desktop administration is crazy and dismissive... much like a BOfH. It's obvious you've never administered desktops in these kinds of environments. In retrospect, macOS may be over-simplified but I've had a way easier time fixing crap and receiving less complaints in comparison to Windows.
No, macOS is perfect for these usecases. They did absolutely the right thing: start with a qualified OS, put all the demanded convenience on top of it, and then attach an appropriate price tag.

I can only imagine the stress/pain involved if you were to put someone in front of a shell and asked them to do post-mortem configuration to get a desktop up. They'd look at you sideways.
As you say, people need to get sh*t done - but: why should I care?
Don't you see the difference between e.g. science, bringing things onwards for mankind as a whole, making new things possible, and individual people who may need whatever they may need out of their own decision (for which they are individually responsible)?
 
Back
Top