FreeBSD discussion on HN

I don't doubt that the project will likely not turn around and become less and less relevant. I work as a developer and most of my peers have no idea what a Free BSD is or what it does.

I was actually pretty shocked to see that FreeBSD does not have an official Twitter or Facebook page. Why not? Isn't there always talk about how we need to get FreeBSD more "out there" and "visible"? Isn't social networking one of the top ways of doing that right now?

Out-of-box usability is terrible, probably thanks to the licensing conflicts (things like sudo are not there by default), and the fact that even simple things like having a good, preconfigured shell for root and the users. The fact that we still have csh as the default, which NO ONE knows how to script with, is actually a big deal.

I think the problem with FreeBSD as a project is that they have always "kept it real". Instead of paying attention to what would make it more usable and approachable, they have stuck to their guns on stupid things like using csh while Linux can give a full, working desktop or server with a single (graphical) installer button click.

It does not make sense to make things difficult for beginners for the sake of customizability or for the benefit of skilled users. It should be the other way around: everything should work well enough on its own, and then the elite can do what they want with it from there.
 
With all due respect to the foundation I think that their funding approach should become a bit more aggressive.
There are many large corporations who could sponsor the project with millions of dollars.
 
Pushrod said:
I don't doubt that the project will likely not turn around and become less and less relevant. I work as a developer and most of my peers have no idea what a Free BSD is or what it does.

I was actually pretty shocked to see that FreeBSD does not have an official Twitter or Facebook page. Why not? Isn't there always talk about how we need to get FreeBSD more "out there" and "visible"? Isn't social networking one of the top ways of doing that right now?

Out-of-box usability is terrible, probably thanks to the licensing conflicts (things like sudo are not there by default), and the fact that even simple things like having a good, preconfigured shell for root and the users. The fact that we still have csh as the default, which NO ONE knows how to script with, is actually a big deal.

I think the problem with FreeBSD as a project is that they have always "kept it real". Instead of paying attention to what would make it more usable and approachable, they have stuck to their guns on stupid things like using csh while Linux can give a full, working desktop or server with a single (graphical) installer button click.

It does not make sense to make things difficult for beginners for the sake of customizability or for the benefit of skilled users. It should be the other way around: everything should work well enough on its own, and then the elite can do what they want with it from there.
Click to expand...

I share only part of this view, in a general sense[1], but more from a hardware standpoint. Yearly I kludge through disk setups ( yesterday, a boot0cfg wiped all the data on a backup disk somehow; a guided gpart setup failed with a somewhat terse error, but poking around in /dev
I found usable labels with which to continue the re-pre-fstab setup. (IMO maybe partially
solved were each EXAMPLE in the gpart (etc) man pages preceeded by a paragraph verbosely explaining
what the situation is with the disk, then an explanation after the example explaining what the new
situation is (again, verbosely) and explaining the meaning of errors one might encounter. However, it
sounds like something for a full-time paid FreeBSD developer. Similarly, a PDF flowchart to print
(1990 shareware style) for configurations (cups, cron, disklabel, gpt, mbr), but again, much more
time to implement than I suspect anyone has time for.

[1] Something lacking overall: a lack of time; what with new hardware protocols arriving from
elsewhere; etc etc.
....
Relevant? I don't see it as ever being less relevant than it is. Yesterday reviewing a microsoft
directory structure; and daily browsing a cutting-edge linux forum, I see those as far worse off than
FreeBSD (from a user's standpoint) (given a few years of experience in FreeBSD in the latter), if
one wishes to keep the same installation over a matter of many years and not just a few.
 
It does not make sense to make things difficult for beginners for the sake of customizability or for the benefit of skilled users.
Click to expand...

Yes, yes it does; you are beginner only once.
And I don't think it's that big of a deal as it sounds.

It should be the other way around: everything should work well enough on its own, and then the elite can do what they want with it from there.
Click to expand...

Not possible since it goes against #1.
 
Pushrod said:
I work as a developer and most of my peers have no idea what a Free BSD is or what it does.
Click to expand...
I know plenty of developers who have, or rather had when I last spoke to them, not heard of concepts such as iSCSI. People knowing about your product does not necessarily make it any less useful.

I was actually pretty shocked to see that FreeBSD does not have an official Twitter or Facebook page. Why not? Isn't there always talk about how we need to get FreeBSD more "out there" and "visible"? Isn't social networking one of the top ways of doing that right now?
Click to expand...
Why should they have a Facebook page? Should they also run TV-adds? I don't care how many people use the same OS that I do, or wear the same brand of shoe, or eat the same type of bread. I use what I think is appropriate. I like FreeBSD, because it does things the right way (mostly, and IMO), not because it is the popular choice.

Out-of-box usability is terrible, probably thanks to the licensing conflicts (things like sudo are not there by default), and the fact that even simple things like having a good, preconfigured shell for root and the users. The fact that we still have csh as the default, which NO ONE knows how to script with, is actually a big deal.
Click to expand...
_I_ like csh. Also I prefer having shell scripts in a simple /bin/sh-style which can be moved between computers without relying on additional shells, in this case bash, which all the linux-distros seem to rely on. Cleanly splitting and labelling is a good thing.
As for sudo - most uses of it that I have seen were about giving _all users_ root rights. I have not seen many how-tos which discussed delagating rights or reducing the number of users that had access to root-accounts. It is better to do the right thing, than to do the easy thing. Especially in the long run. Users in the wheel group can 'su root', which is what sudo is generally uses for anyway. Sudo used properly - good, sudo abused - not good.

I think the problem with FreeBSD as a project is that they have always "kept it real". Instead of paying attention to what would make it more usable and approachable, they have stuck to their guns on stupid things like using csh while Linux can give a full, working desktop or server with a single (graphical) installer button click.
Click to expand...
The OpenBSD installer is great in my mind. It is a simple text-installer which asks questions, and then configures. Much more light-weight than anything, I don't need a mouse to use it, I can install over serial console, etc etc. Point-and-click has much bigger requirements.
If you compare how long the Windows-installer (point-and-click-deluxe) takes to run through compare to the OpenBSD one, I know which one I prefer.

It does not make sense to make things difficult for beginners for the sake of customizability or for the benefit of skilled users. It should be the other way around: everything should work well enough on its own, and then the elite can do what they want with it from there.
Click to expand...
I never struggled with FreeBSD when I was new. For one reason only - the documentation. When I had a problem I could read manuals, man-pages or something else. The answers were there, not always immediately, but I found them. Often I found information that I was not looking for, which helped me solve other problems down the line.
The question you are asking is similar to: 'how much can we dumb down the product so that Tom,Richard, and Harry can all use it without having to read the manual.
 
mix_room said:
People knowing about your product does not necessarily make it any less useful.
Click to expand...
No, but it sure helps when you're trying to pitch it to the decision makers.

mix_room said:
Why should they have a Facebook page? Should they also run TV-adds? I don't care how many people use the same OS that I do, or wear the same brand of shoe, or eat the same type of bread. I use what I think is appropriate. I like FreeBSD, because it does things the right way (mostly, and IMO), not because it is the popular choice.
Click to expand...

People know what Linux is. People learn it because it gets them jobs. People use it because things work on it and are supported. People don't know what FreeBSD is, and they don't donate their time or their money to it.

mix_room said:
_I_ like csh. Also I prefer having shell scripts in a simple /bin/sh-style which can be moved between computers without relying on additional shells, in this case bash, which all the linux-distros seem to rely on. Cleanly splitting and labelling is a good thing.
As for sudo - most uses of it that I have seen were about giving _all users_ root rights. I have not seen many how-tos which discussed delagating rights or reducing the number of users that had access to root-accounts. It is better to do the right thing, than to do the easy thing. Especially in the long run. Users in the wheel group can 'su root', which is what sudo is generally uses for anyway. Sudo used properly - good, sudo abused - not good.
Click to expand...

sudo is a half-decent band-aid for the brain-dead UNIX "light switch" security model. If people don't want to learn how to configure it, what makes anyone think those same people will want to learn how to configure anything else on the system?

mix_room said:
The OpenBSD installer is great in my mind. It is a simple text-installer which asks questions, and then configures. Much more light-weight than anything, I don't need a mouse to use it, I can install over serial console, etc etc. Point-and-click has much bigger requirements.
If you compare how long the Windows-installer (point-and-click-deluxe) takes to run through compare to the OpenBSD one, I know which one I prefer.
Click to expand...

I know which I prefer as well. I prefer the point-and-click-and-do-something-else-while-it-finishes installer. I don't want to be asked questions about stuff where I am almost certainly going to want defaults for nearly everything. Non-defaults can be changed later quite easily anyways.

mix_room said:
I never struggled with FreeBSD when I was new. For one reason only - the documentation.
Click to expand...
Good for you.

mix_room said:
When I had a problem I could read manuals, man-pages or something else. The answers were there, not always immediately, but I found them. Often I found information that I was not looking for, which helped me solve other problems down the line.
Click to expand...
I'm not going to argue against the merits of good documentation, but I think it takes a back seat to usability. Millions of iPhone users agree. Documentation shouldn't substitute for poor UI design or needlessly complicated configuration.

mix_room said:
The question you are asking is similar to: 'how much can we dumb down the product so that Tom,Richard, and Harry can all use it without having to read the manual.
Click to expand...
The issue is not "dumbing things down"; the issue is "does anyone want to use this, and if not, what changes do we need to make?"

I think the slumped donations are more significant than people might want to admit. If the project is not being supported, it will die.

I have been using FreeBSD for ten years, and I owe a ton of my knowledge to it, but in honesty, I could have stuck with any OS, even Windows, and have learned just as much about computer hardware and software. It was good at the time (got in at version 4.2) but at this point, I am only using it because I am too lazy to find a way of embracing a Linux distro to the depth that I have with FreeBSD, and because I don't have a practical way of reformatting 7TB worth of disks without losing data.

FreeBSD will become an OS for a small sub-subculture, and perhaps for academic purposes, and that's about it. It lacks a dictator, so it will not have a direction; it lacks any kind of corporate backing so it will never have the man power to stay modern, and the refusal to improve out-of-box usability and performance will sentence it to the aforementioned niche.
 
Pushrod said:
sudo is a half-decent band-aid for the brain-dead UNIX "light switch" security model. If people don't want to learn how to configure it, what makes anyone think those same people will want to learn how to configure anything else on the system?
Click to expand...
How many linux how-tos with broken configs ("allow all") have you seen, how many with proper ones? The question is not if the tool is good or not, but rather having useful defaults. In FreeBSD the default is useful: if you are in the wheel-group you can su to root, if not then you can't. This is already a stronger security model than the "allow all to sudo". It is also not much harder.

I know which I prefer as well. I prefer the point-and-click-and-do-something-else-while-it-finishes installer. I don't want to be asked questions about stuff where I am almost certainly going to want defaults for nearly everything. Non-defaults can be changed later quite easily anyways.
Click to expand...
The OpenBSD installer in particular is very quick, it asks its questions in less time than it takes the Windows installer to ask its. The OpenBSD installer might not be very attractive, but it is a well thought out solution. Adding a graphical interface on top of it shouldn't be hard, but it wouldn't add any benefit - only make it look prettier.

I'm not going to argue against the merits of good documentation, but I think it takes a back seat to usability. Millions of iPhone users agree. Documentation shouldn't substitute for poor UI design or needlessly complicated configuration.
Click to expand...
I personally think that FreeBSD has superior usability to the common Linux-distros, to not talk about Windows. All the things are in the right place, I know where to look for installed programs, where I can expect the config files to be etc. Other extreme - Windows, where I have no idea where a particular piece of software puts its stuff. Linux is somewhere in the middle.
Usability is much more than a novice user being able to use a tool quickly.

I think the slumped donations are more significant than people might want to admit. If the project is not being supported, it will die.
Click to expand...
In the end the donations are not bad at all, almost 100 kUSD in 10 days is impressive. (http://www.itwire.com/business-it-n...teran-confident-of-reaching-fund-raising-goal - 220 kUSD on 2nd December, 305 kUSD today)

and because I don't have a practical way of reformatting 7TB worth of disks without losing data.
Click to expand...
On a side note - you should have backups.

FreeBSD will become an OS for a small sub-subculture, and perhaps for academic purposes, and that's about it. It lacks a dictator, so it will not have a direction; it lacks any kind of corporate backing so it will never have the man power to stay modern, and the refusal to improve out-of-box usability and performance will sentence it to the aforementioned niche.
Click to expand...
It seems to me that quite a lot of modern things are already in FreeBSD - ZFS,DTrace, etc etc. Computing is not the desktop.
Linux has much worse problems - stemming specifically from a fragemented community, just look at the sounds card problems that they have, and their new wierd form of system initialization. FreeBSD does a much better job of having a useable system here.

EDIT: Changed amounts of USD to be correct.
 
jackp said:
Also -- one poster mentioned the donation mechanism on the FreeBSD Foundation website is triggering the spam detectors on the email address linked to the Foundation's PayPal account: https://news.ycombinator.com/item?id=4895015

Can someone look into it? Or contact someone who can?
Click to expand...

The problem was that paypal payments come in "from" the donor rather than paypal, and paypal was exempted but not all e-mails that were from paypal but listed as from the donor. This has now been solved.
 
emaste@ said:
The Foundation raised over $40,000 from over 650 donors in three days: http://freebsdfoundation.blogspot.com/2012/12/stunning-news-website-fundraising.html

There's also a followup on HN: http://news.ycombinator.com/item?id=4912361
Click to expand...

It looks like it hit /. as well: http://news.slashdot.org/story/12/1...3200-for-the-freebsd-foundation-in-three-days

Amazing how much vitriol in every one of those threads. Nice to know that ad hominem style headlines work best to push SEO linkbombs.
 
Pushrod said:
I was actually pretty shocked to see that FreeBSD does not have an official Twitter or Facebook page.
Click to expand...

Like these?

Twitter: https://twitter.com/freebsd
Facebook: http://www.facebook.com/pages/FreeBSD-Foundation/102507223153209

eadler@ said:
The problem was that paypal payments come in "from" the donor rather than paypal, and paypal was exempted but not all e-mails that were from paypal but listed as from the donor. This has now been solved.
Click to expand...

Thanks for taking care of it!

emaste@ said:
There's also a followup on HN: http://news.ycombinator.com/item?id=4912361
Click to expand...

Another followup on HN this morning (a resubmission of that news article); it's currently on the front page: https://news.ycombinator.com/item?id=4920891
 
There is not reason to change FreeBSD. The Handbook and man-pages are more than capable of walking a novice through. If a beginner cannot read, it is not a reason to waste coding time on making a MS version of FreeBSD. Isn't PC-BSD enough? FreeBSD is quite simple to configure, when compared to re-configuring an overly script managed system. Or, maintains higher performance than a system with a G.U.I. to configure everything.
 
You must log in or register to reply here.
Back
Top