When will be created a great desktop experience for OpenBSD?

classic Classic list List threaded Threaded
103 messages Options
123456
Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Christopher Turkel
I'd like to see an updated FVWM as the default WM.

On Wed, May 8, 2019 at 7:18 AM Mohamed Fouad <
[hidden email]> wrote:

> if you are suggesting updating the openbsd installer to include dwm as an
> option. Even that it adds one more click to the installation process, it
> would work as a sharm for some people :P
>
> On Tue, 7 May 2019, 2:04 am Clark Block <[hidden email] wrote:
>
> > In 2019 still there is not a great desktop experience for NetBSD.
> However,
> > the new "OS108" is seeking to improve this with a NetBSD operating system
> > paired with the MATE desktop environment.
> > So, OS108, a derivative of NetBSD, has just been released:
> > https://os108.org/?ez_cid=CLIENT_ID(AMP_ECID_EZOIC)
> >
> > When will be created a great desktop experience for OpenBSD?
> >
>
Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Peter Nicolai Mathias Hansteen
In reply to this post by Clark Block
> When will be created a great desktop experience for OpenBSD?

I think it is important to keep in mind that in order to achieve
*anything* in the OpenBSD project (or other open source projects for
that matter) the way forward is to work *with*, not against, the
developers and their code.

The short version is, please present your ideas of what you want to do
with sound reasoning and if at all possible supplement with patches
posted to tech@.

The patches stand a better chance of being accepted (perhaps along
with their developer) if the submitter can take comments and valid
criticisms from competent people (again mainly the developers) in
stride and seems willing to stay around as maintainer in the longer
haul (ie not slink back to the shadows after a release or two).

For anyone considering taking up the theme of this thread, please
consider whether this could somehow be made into the package with only
minimal impact on the base system.

Such a package could for example leverage all the tools already in the
base systems to generate something like bsd.graphic.{rd,is,fs} and
offer a skeleton for a site.tgz for the generated install medium.

If this sounds a lot like what is very achievable with the tools
already in the OpenBSD base system and seasoned OpenBSD admins would
do comfortably with a relatively simple autoinstall, it's because that
is exactly what it is.

But if there is an actual use case spot we're missing, this would be
the way to filling it with the least amount of extra work for everyone
involved.

--
Peter N. M. Hansteen, member of the first RFC 1149 implementation team
http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/
"Remember to set the evil bit on all malicious network traffic"
delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Fish Kungfu
I love the default minimalism, simplicity and freedom of OpenBSD to make it
how I want it.
My "Perfect OpenBSD": spectrwm, dmenu, urxvt (with perl tabbing), tmux, etc.

On Wed, May 8, 2019 at 7:40 AM Peter N. M. Hansteen <[hidden email]> wrote:

> > When will be created a great desktop experience for OpenBSD?
>
> I think it is important to keep in mind that in order to achieve
> *anything* in the OpenBSD project (or other open source projects for
> that matter) the way forward is to work *with*, not against, the
> developers and their code.
>
> The short version is, please present your ideas of what you want to do
> with sound reasoning and if at all possible supplement with patches
> posted to tech@.
>
> The patches stand a better chance of being accepted (perhaps along
> with their developer) if the submitter can take comments and valid
> criticisms from competent people (again mainly the developers) in
> stride and seems willing to stay around as maintainer in the longer
> haul (ie not slink back to the shadows after a release or two).
>
> For anyone considering taking up the theme of this thread, please
> consider whether this could somehow be made into the package with only
> minimal impact on the base system.
>
> Such a package could for example leverage all the tools already in the
> base systems to generate something like bsd.graphic.{rd,is,fs} and
> offer a skeleton for a site.tgz for the generated install medium.
>
> If this sounds a lot like what is very achievable with the tools
> already in the OpenBSD base system and seasoned OpenBSD admins would
> do comfortably with a relatively simple autoinstall, it's because that
> is exactly what it is.
>
> But if there is an actual use case spot we're missing, this would be
> the way to filling it with the least amount of extra work for everyone
> involved.
>
> --
> Peter N. M. Hansteen, member of the first RFC 1149 implementation team
> http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/
> "Remember to set the evil bit on all malicious network traffic"
> delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
>
>
Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Roderick
In reply to this post by ropers


On Wed, 8 May 2019, ropers wrote:

> Tangentially related: Does anyone here routinely use the default fvwm?

Yes, routinely default fvwm, also twm, and nothing else. Not only
with OpenBSD.

Rodrigo

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Eric Furman-3
In reply to this post by Peter Nicolai Mathias Hansteen
On Wed, May 8, 2019, at 7:38 AM, Peter N. M. Hansteen wrote:

> > When will be created a great desktop experience for OpenBSD?
>
> I think it is important to keep in mind that in order to achieve
> *anything* in the OpenBSD project (or other open source projects for
> that matter) the way forward is to work *with*, not against, the
> developers and their code.
>
> The short version is, please present your ideas of what you want to do
> with sound reasoning and if at all possible supplement with patches
> posted to tech@.
>
> The patches stand a better chance of being accepted (perhaps along
> with their developer) if the submitter can take comments and valid
> criticisms from competent people (again mainly the developers) in
> stride and seems willing to stay around as maintainer in the longer
> haul (ie not slink back to the shadows after a release or two).
>
> For anyone considering taking up the theme of this thread, please
> consider whether this could somehow be made into the package with only
> minimal impact on the base system.
>
> Such a package could for example leverage all the tools already in the
> base systems to generate something like bsd.graphic.{rd,is,fs} and
> offer a skeleton for a site.tgz for the generated install medium.
>
> If this sounds a lot like what is very achievable with the tools
> already in the OpenBSD base system and seasoned OpenBSD admins would
> do comfortably with a relatively simple autoinstall, it's because that
> is exactly what it is.
>
> But if there is an actual use case spot we're missing, this would be
> the way to filling it with the least amount of extra work for everyone
> involved.

Peter, it's not going to happen because it would require someone to do work.
The whole point is to try to get others to do it for you.

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Steve Litt
In reply to this post by Christopher Turkel
If you do that, you'd better crank waaaaay up on its fonts. Fvwm fonts
are so small that if you have bad vision, you can't read the screen
well enough to increase the font size.

It's easy for a well-sighted person to reduce fonts, but for the poorly
sighted person who can't read the screen in the first place, it's a
long, difficult process.

SteveT


On Wed, 8 May 2019 07:20:29 -0400
Christopher Turkel <[hidden email]> wrote:

> I'd like to see an updated FVWM as the default WM.
>
> On Wed, May 8, 2019 at 7:18 AM Mohamed Fouad <
> [hidden email]> wrote:  
>
> > if you are suggesting updating the openbsd installer to include dwm
> > as an option. Even that it adds one more click to the installation
> > process, it would work as a sharm for some people :P
> >
> > On Tue, 7 May 2019, 2:04 am Clark Block <[hidden email]
> > wrote:
> > > In 2019 still there is not a great desktop experience for
> > > NetBSD.  
> > However,  
> > > the new "OS108" is seeking to improve this with a NetBSD
> > > operating system paired with the MATE desktop environment.
> > > So, OS108, a derivative of NetBSD, has just been released:
> > > https://os108.org/?ez_cid=CLIENT_ID(AMP_ECID_EZOIC)
> > >
> > > When will be created a great desktop experience for OpenBSD?
> > >  
> >  

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

noah pugsley-3
In reply to this post by Christopher Turkel
Updated FVWM or a different default config?

Sent from mobile.
  Original Message  
From: Christopher Turkel
Sent: Wednesday, May 8, 2019 04:22
Cc: OpenBSD Misc
Subject: Re: When will be created a great desktop experience for OpenBSD?

I'd like to see an updated FVWM as the default WM.

On Wed, May 8, 2019 at 7:18 AM Mohamed Fouad <
[hidden email]> wrote:

> if you are suggesting updating the openbsd installer to include dwm as an
> option. Even that it adds one more click to the installation process, it
> would work as a sharm for some people :P
>
> On Tue, 7 May 2019, 2:04 am Clark Block <[hidden email] wrote:
>
> > In 2019 still there is not a great desktop experience for NetBSD.
> However,
> > the new "OS108" is seeking to improve this with a NetBSD operating system
> > paired with the MATE desktop environment.
> > So, OS108, a derivative of NetBSD, has just been released:
> > https://os108.org/?ez_cid=CLIENT_ID(AMP_ECID_EZOIC)
> >
> > When will be created a great desktop experience for OpenBSD?
> >
>

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Roderick


On Wed, 8 May 2019, noah pugsley wrote:

> Updated FVWM or a different default config?

I hope, no one comes to the idea to change the configuration.

In FreeBSD fvwm is awful configurated and I do not want to waste my time
configurating it. That is why I use there twm (easy to configure).

Rodrigo

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Matthew Graybosch-2
On Wed, May 8, 2019, at 5:18 PM, Roderick wrote:
>
> On Wed, 8 May 2019, noah pugsley wrote:
>
> > Updated FVWM or a different default config?
>
> I hope, no one comes to the idea to change the configuration.

I'm not really fussed about the default FVWM config; it's easy enough to pull a copy into $HOME and start tinkering with it once you've read fvwmrc(5), or to switch to cwm (1) (my favorite). Hell, it wasn't that hard to install OpenBSD with MATE and LibreOffice on a neighbor's old Intel NUC so her kids had a machine to do homework on without getting distracted by Fortnite (but wait until they discover hack(6)).

I'm happy with OpenBSD as a desktop machine, even though the OS wasn't made with me in mind, and I appreciate all of the work the developers have put into it for their own purposes. Thanks!

--
Matthew Graybosch
"'Out of order'?! Even in the future nothing works."

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Christopher Turkel
I'm in favor of an updated FVWM with a simple config.

On Wed, May 8, 2019 at 5:43 PM Matthew Graybosch <[hidden email]>
wrote:

> On Wed, May 8, 2019, at 5:18 PM, Roderick wrote:
> >
> > On Wed, 8 May 2019, noah pugsley wrote:
> >
> > > Updated FVWM or a different default config?
> >
> > I hope, no one comes to the idea to change the configuration.
>
> I'm not really fussed about the default FVWM config; it's easy enough to
> pull a copy into $HOME and start tinkering with it once you've read
> fvwmrc(5), or to switch to cwm (1) (my favorite). Hell, it wasn't that hard
> to install OpenBSD with MATE and LibreOffice on a neighbor's old Intel NUC
> so her kids had a machine to do homework on without getting distracted by
> Fortnite (but wait until they discover hack(6)).
>
> I'm happy with OpenBSD as a desktop machine, even though the OS wasn't
> made with me in mind, and I appreciate all of the work the developers have
> put into it for their own purposes. Thanks!
>
> --
> Matthew Graybosch
> "'Out of order'?! Even in the future nothing works."
>
>
Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

chohag
In reply to this post by noah pugsley-3
noah pugsley writes:

> Updated FVWM or a different default config?
>
> Sent from mobile.
>   Original Message  
> From: Christopher Turkel
> Sent: Wednesday, May 8, 2019 04:22
> Cc: OpenBSD Misc
> Subject: Re: When will be created a great desktop experience for OpenBSD?
>
> I'd like to see an updated FVWM as the default WM.

Then you'll need this: https://www.openbsd.org/faq/faq5.html

Why is this thread still here?

Matthew

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Charles A Daniels
In reply to this post by Clark Block
I'd like to chime in here, on a slightly different subject.

I think the OP (Clark) raises a point, but I suggest he's coming it
from the wrong angle. I think there's something here to discuss that I
have not seen mentioned in this thread thus far.

TL;DR: the OpenBSD (and friends) way of thinking is falling further and
further out of fashion with respect to mainstream computing  -- I justify
this statement, posit on the need for action, and propose a starting
point.

Disclaimer 1: I use OpenBSD at various points to refer to the piece of
code, to the development philosophy, to the development team, and to the
community of users. I try to make clear which I am referencing; sorry if
it's confusing.

Disclaimer 2: I am not an OpenBSD developer. I have contributed only in
very minor ways. I don't speak on behalf of anyone other than myself as
a user of OpenBSD. If it seems at points that I am speaking on behalf of
"OpenBSD" (by any of the previous definitions), I intend that as an
appeal to my perception of what the community of users and developers
feels and thinks, based on my interactions with them here and elsewhere.
If I am wrong in this respect, I invite corrections.

It certainly seems that there is a great disconnect from the canonical
(small c) definition of "great desktop experience", and the OpenBSD (and
friends) definition. I feel that the broader notion of what a "great
desktop experience" means within the context of the 2019 zeitgeist has
trended towards pandering to the user, in my view to the point of being
patronizing.

"The users cannot be trusted to manage their own files, it's too hard and
will confuse them."

"The users cannot be trusted to install their own programs, it's too hard
and will confuse them."

...

"The users cannot be trusted to make decisions, it's too hard and will
confuse them."

You get the picture (hopefully).

Part of this is perhaps because the users are "bad at using computers".
I think most anyone who has helped a computer-illiterate family member
or friend with any technology related problem for more than 5 minutes
will see the truth in this statement. But I think it's not the users
fault; many might argue "but if the users would only learn X, Y, an Z
DE/WM/OS/app/etc". I feel that many _do_ argue this, with all the talk
these days of "pushing the envelope", "modern UX", "innovation", and so
on in big blinking neon letters. Ultimately, what this means is telling
the users "yup, you learned $paradigm, but now we have $paradigm++
because it's the new big thing". If you're a corporate user on a box you
don't control, or you just don't have the experience to do systems
administration on your own, you have to suck it up and deal with it.

That probably won't be a relateable sentiment to nearly anyone likely to
ever read this document. But as a thought experiment, let's imagine if
vi got a fresh now UX paradigm every year or so, and let's pretend for
the sake of argument that we can't patch it or revert. I think all of us
would not want to use such a program very much. vi takes a while to
learn, and while I (as a diehard vim user) would argue against the
notion of the vi paradigm as the One True Way to edit text, it is
certainly a very powerful tool... because of the time put in to build
muscle memory and intuition about it, knowing that that knowledge will
be applicable to vi implementations for decades to come. Without the
ability to trust that time spent front-loading learning will not be
wasted when $paradigm goes to $paradigm++ in a year, who would ever
invest effort into learning more than the bare minimum?

Remember that the typical computer user sees their box the way most of
us probably see our cars. It doesn't matter how it works, as long as it
does, but nobody wants a car where the gas and brake pedals switch every
second Tuesday and you wake up one morning to discover the head unit is
now entirely in Sumerian.

It would seem that this creates a self-perpetuating feedback loop. The
users have a difficult time using the software because they don't learn
it, so the software changes to accommodate the users better, which
further puts folks off of ever learning any of it very well (by
punishing the ones who try). I suggest that this trend has become so
prolific that it has seeped into the general human population's
consciousness around how interacting with computers works.

Think about it. How many software packages do you know of where a user
could learn how to use it well once and have that knowledge be
applicable for years or decades thereafter? This is something we expect
(as technical folk) of shells and editors, scripting languages, and so
on, but it is not something that the layperson using a GUI can now or at
any point in the past reasonably expect.

Remember also, that every developer was one a user at some point. It
sure seems that the wall you have to climb over to go from user to
developer keeps getting higher and higher every year.

There are several problems that I perceive to face us (OpenBSD users
and our friends in similar projects). It ultimately boils down to "we
don't exist in a vacuum", but to be more specific:

1. Most people want to use their computers to get work done, and that
often requires inter-operating with software and file formats that are
used by operating systems other than OpenBSD. This generally amounts to
having people to Do The Work to maintain that compatibility, and I think
we're doing alright here.

2. More importantly, we need contributors. Developers yes, but also
people to report bugs, people to test new features, people to hang
around on misc@ and help new (and old) users troubleshoot. For a project
to last (and I do wish for OpenBSD to last, and I believe most people on
this list do to), it must have sufficient "new blood" coming in to
sustain (1), and also to push it's loftier technical goals forward. It
seems from the outside looking in that there aren't so many new people
coming into OpenBSD (maybe someone working more closely on development
can comment on this trend or lack threof?), and I would argue that a
large part of this may be because it doesn't conform to people's
expectations of how computers should work "these days".

Certainly, OpenBSD seems to have little in the way of technical or
philosophical mindshare in the broader computing community, let alone
among laypeople. Perhaps the latter is not so important. Indeed the
former is also not a good goal in of itself, but is a necessary evil up
to a certain point.

I think the way that much of the Linux (et. al.) world has solved this
(2) is "make it more like what Apple/MS are doing so people think it's
easier". The theory is perhaps that new users are drawn in by the
"friendly" (read: similar to what they are used to) "desktop
experience", and these folks may become community members, then
contributors, and eventually developers. Maybe it works.
Ubuntu/Fedora/GNOME/KDE et. al. certainly have a lot of people working
on them (weather or not the work being done is any good is a discussion
for another thread/flame war). Detractors of those projects might call
it "selling out"... others might just call it "selling".

I would propose that "just do what everyone else does / what people
expect" isn't the right approach for OpenBSD. Maybe the core team (the
people really Doing The Work, i.e. not me) feels differently, but I feel
that I am echoing sentiments I have observed on this list and in other
OpenBSD related communities: OpenBSD is about technical excellence, only
solving problems can be solved right, and not expending precious
dev-hours on implementing unnecessary (or worse: broken) features. It's
not about working the way some arbitrary group of people think it
should, but rather about working the Right Way, and being Excellent.

So what should we do?

I posit that we (the community of OpenBSD) community should be trying
to... perhaps "push" is the wrong word... but "make clear the value of"
the OpenBSD Way (see previous paragraph). Make that way of thinking, and
solving problems more desirable for others to use. I would suggest that
just "make it easy" isn't the right answer, since I think we want to
attract people that can do hard work and have some perseverance, but
those people need to get "hooked" in the first place, and have a general
path forward. I think we're on to something here with respect to how
things are done, in a way that the broader technical community doesn't
appreciate.

What does that look like?

I don't think that's as clear. Damn it Jim, I'm a programmer, not a
sociologist. I think a first step is to have some consensus that there
is an issue to be addressed, and at least vaguely what it is. However, I
will propose that this isn't a matter which has a technical solution,
which is unfortunate since the OpenBSD team is clearly very good at
building great technical solutions. I am willing to help to the extent
that my abilities and limited time will permit.

In closing -- OpenBSD might not have a great desktop experience per se,
but I would argue that it's better than any of the other desktop
experiences that are out there right now. I think we can probably keep
making it better, and that we should. However I don't think shipping a
different WM/DE is going to help.

RFC.

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Steve Litt
On Wed, 8 May 2019 22:43:00 -0400
Charles <[hidden email]> wrote:

> I'd like to chime in here, on a slightly different subject.
>
> I think the OP (Clark) raises a point, but I suggest he's coming it
> from the wrong angle. I think there's something here to discuss that I
> have not seen mentioned in this thread thus far.
>
> TL;DR: the OpenBSD (and friends) way of thinking is falling further
> and further out of fashion with respect to mainstream computing  -- I
> justify this statement, posit on the need for action, and propose a
> starting point.

Ex-actly! Certain OSes and distros, and OpenBSD is one of them, cater
to the hands-on, DIY type of person. Turning OpenBSD into just another
Ubuntu would be a disservice to such people.

[ snip a bunch of true and insightful writing ]

> However I don't think shipping a
> different WM/DE is going to help.

Back in 2010 or thereabouts,  when I used OpenBSD on a laptop
regularly, OpenBSD offered a bunch of WM/DE's in its package manager.
That's a wonderful thing, because different people have different
workflow techniques. I assume OpenBSD still has several different
WM/DE's.

I don't know whether OpenBSD has KDE or Gnome, and don't really care. I
kicked KDE off all my boxes in 2012 because it's it's a massively
entangled monolith. As far as Gnome, even if it *could* be used in the
absense of systemd, I view Gnome as a gateway drug to the
Freedesktop.org worldview of having every software strongly linked to
every other software, and I want no part of that noise.

One point I didn't see in RFC's post is stability. When I used OpenBSD
back in 2010, subjectively it seemed more stable, more consistent, and
less surprising than any Linux I'd ever used (and of course than any
Windows I'd ever used). If my computer were just for web browsing,
social networking, email, and storing photos and videos, Ubuntu or Mint
would be stable enough. But the way I work, I often have over 50
windows open. I can't afford the massive instability bestowed by "we do
it all for you" user interfaces.

SteveT

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Theo de Raadt-2
In reply to this post by Charles A Daniels
Interjecting.

About 80 people are building & maintaining an operating system, and
following their whims to fix the problems that they believe to be most
relevant.

What is this email exchange about?  Let me lay it out.

1. We are being told what to do.
2. We are being told what is important.
3. We are being told our choices are wrong.

There are no diffs, no source code proposals, nothing.

There are only wishes.

Can you guys please figure it out?

Charles <[hidden email]> wrote:

> I'd like to chime in here, on a slightly different subject.
>
> I think the OP (Clark) raises a point, but I suggest he's coming it
> from the wrong angle. I think there's something here to discuss that I
> have not seen mentioned in this thread thus far.
>
> TL;DR: the OpenBSD (and friends) way of thinking is falling further and
> further out of fashion with respect to mainstream computing  -- I justify
> this statement, posit on the need for action, and propose a starting
> point.
>
> Disclaimer 1: I use OpenBSD at various points to refer to the piece of
> code, to the development philosophy, to the development team, and to the
> community of users. I try to make clear which I am referencing; sorry if
> it's confusing.
>
> Disclaimer 2: I am not an OpenBSD developer. I have contributed only in
> very minor ways. I don't speak on behalf of anyone other than myself as
> a user of OpenBSD. If it seems at points that I am speaking on behalf of
> "OpenBSD" (by any of the previous definitions), I intend that as an
> appeal to my perception of what the community of users and developers
> feels and thinks, based on my interactions with them here and elsewhere.
> If I am wrong in this respect, I invite corrections.
>
> It certainly seems that there is a great disconnect from the canonical
> (small c) definition of "great desktop experience", and the OpenBSD (and
> friends) definition. I feel that the broader notion of what a "great
> desktop experience" means within the context of the 2019 zeitgeist has
> trended towards pandering to the user, in my view to the point of being
> patronizing.
>
> "The users cannot be trusted to manage their own files, it's too hard and
> will confuse them."
>
> "The users cannot be trusted to install their own programs, it's too hard
> and will confuse them."
>
> ...
>
> "The users cannot be trusted to make decisions, it's too hard and will
> confuse them."
>
> You get the picture (hopefully).
>
> Part of this is perhaps because the users are "bad at using computers".
> I think most anyone who has helped a computer-illiterate family member
> or friend with any technology related problem for more than 5 minutes
> will see the truth in this statement. But I think it's not the users
> fault; many might argue "but if the users would only learn X, Y, an Z
> DE/WM/OS/app/etc". I feel that many _do_ argue this, with all the talk
> these days of "pushing the envelope", "modern UX", "innovation", and so
> on in big blinking neon letters. Ultimately, what this means is telling
> the users "yup, you learned $paradigm, but now we have $paradigm++
> because it's the new big thing". If you're a corporate user on a box you
> don't control, or you just don't have the experience to do systems
> administration on your own, you have to suck it up and deal with it.
>
> That probably won't be a relateable sentiment to nearly anyone likely to
> ever read this document. But as a thought experiment, let's imagine if
> vi got a fresh now UX paradigm every year or so, and let's pretend for
> the sake of argument that we can't patch it or revert. I think all of us
> would not want to use such a program very much. vi takes a while to
> learn, and while I (as a diehard vim user) would argue against the
> notion of the vi paradigm as the One True Way to edit text, it is
> certainly a very powerful tool... because of the time put in to build
> muscle memory and intuition about it, knowing that that knowledge will
> be applicable to vi implementations for decades to come. Without the
> ability to trust that time spent front-loading learning will not be
> wasted when $paradigm goes to $paradigm++ in a year, who would ever
> invest effort into learning more than the bare minimum?
>
> Remember that the typical computer user sees their box the way most of
> us probably see our cars. It doesn't matter how it works, as long as it
> does, but nobody wants a car where the gas and brake pedals switch every
> second Tuesday and you wake up one morning to discover the head unit is
> now entirely in Sumerian.
>
> It would seem that this creates a self-perpetuating feedback loop. The
> users have a difficult time using the software because they don't learn
> it, so the software changes to accommodate the users better, which
> further puts folks off of ever learning any of it very well (by
> punishing the ones who try). I suggest that this trend has become so
> prolific that it has seeped into the general human population's
> consciousness around how interacting with computers works.
>
> Think about it. How many software packages do you know of where a user
> could learn how to use it well once and have that knowledge be
> applicable for years or decades thereafter? This is something we expect
> (as technical folk) of shells and editors, scripting languages, and so
> on, but it is not something that the layperson using a GUI can now or at
> any point in the past reasonably expect.
>
> Remember also, that every developer was one a user at some point. It
> sure seems that the wall you have to climb over to go from user to
> developer keeps getting higher and higher every year.
>
> There are several problems that I perceive to face us (OpenBSD users
> and our friends in similar projects). It ultimately boils down to "we
> don't exist in a vacuum", but to be more specific:
>
> 1. Most people want to use their computers to get work done, and that
> often requires inter-operating with software and file formats that are
> used by operating systems other than OpenBSD. This generally amounts to
> having people to Do The Work to maintain that compatibility, and I think
> we're doing alright here.
>
> 2. More importantly, we need contributors. Developers yes, but also
> people to report bugs, people to test new features, people to hang
> around on misc@ and help new (and old) users troubleshoot. For a project
> to last (and I do wish for OpenBSD to last, and I believe most people on
> this list do to), it must have sufficient "new blood" coming in to
> sustain (1), and also to push it's loftier technical goals forward. It
> seems from the outside looking in that there aren't so many new people
> coming into OpenBSD (maybe someone working more closely on development
> can comment on this trend or lack threof?), and I would argue that a
> large part of this may be because it doesn't conform to people's
> expectations of how computers should work "these days".
>
> Certainly, OpenBSD seems to have little in the way of technical or
> philosophical mindshare in the broader computing community, let alone
> among laypeople. Perhaps the latter is not so important. Indeed the
> former is also not a good goal in of itself, but is a necessary evil up
> to a certain point.
>
> I think the way that much of the Linux (et. al.) world has solved this
> (2) is "make it more like what Apple/MS are doing so people think it's
> easier". The theory is perhaps that new users are drawn in by the
> "friendly" (read: similar to what they are used to) "desktop
> experience", and these folks may become community members, then
> contributors, and eventually developers. Maybe it works.
> Ubuntu/Fedora/GNOME/KDE et. al. certainly have a lot of people working
> on them (weather or not the work being done is any good is a discussion
> for another thread/flame war). Detractors of those projects might call
> it "selling out"... others might just call it "selling".
>
> I would propose that "just do what everyone else does / what people
> expect" isn't the right approach for OpenBSD. Maybe the core team (the
> people really Doing The Work, i.e. not me) feels differently, but I feel
> that I am echoing sentiments I have observed on this list and in other
> OpenBSD related communities: OpenBSD is about technical excellence, only
> solving problems can be solved right, and not expending precious
> dev-hours on implementing unnecessary (or worse: broken) features. It's
> not about working the way some arbitrary group of people think it
> should, but rather about working the Right Way, and being Excellent.
>
> So what should we do?
>
> I posit that we (the community of OpenBSD) community should be trying
> to... perhaps "push" is the wrong word... but "make clear the value of"
> the OpenBSD Way (see previous paragraph). Make that way of thinking, and
> solving problems more desirable for others to use. I would suggest that
> just "make it easy" isn't the right answer, since I think we want to
> attract people that can do hard work and have some perseverance, but
> those people need to get "hooked" in the first place, and have a general
> path forward. I think we're on to something here with respect to how
> things are done, in a way that the broader technical community doesn't
> appreciate.
>
> What does that look like?
>
> I don't think that's as clear. Damn it Jim, I'm a programmer, not a
> sociologist. I think a first step is to have some consensus that there
> is an issue to be addressed, and at least vaguely what it is. However, I
> will propose that this isn't a matter which has a technical solution,
> which is unfortunate since the OpenBSD team is clearly very good at
> building great technical solutions. I am willing to help to the extent
> that my abilities and limited time will permit.
>
> In closing -- OpenBSD might not have a great desktop experience per se,
> but I would argue that it's better than any of the other desktop
> experiences that are out there right now. I think we can probably keep
> making it better, and that we should. However I don't think shipping a
> different WM/DE is going to help.
>
> RFC.
>

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Theo de Raadt-2
In reply to this post by Steve Litt
Steve Litt <[hidden email]> wrote:

> > However I don't think shipping a
> > different WM/DE is going to help.
>
> Back in 2010 or thereabouts,  when I used OpenBSD on a laptop
> regularly, OpenBSD offered a bunch of WM/DE's in its package manager.
> That's a wonderful thing, because different people have different
> workflow techniques. I assume OpenBSD still has several different
> WM/DE's.
>
> I don't know whether OpenBSD has KDE or Gnome, and don't really care. I
> kicked KDE off all my boxes in 2012 because it's it's a massively
> entangled monolith. As far as Gnome, even if it *could* be used in the
> absense of systemd, I view Gnome as a gateway drug to the
> Freedesktop.org worldview of having every software strongly linked to
> every other software, and I want no part of that noise.
>
> One point I didn't see in RFC's post is stability. When I used OpenBSD
> back in 2010, subjectively it seemed more stable, more consistent, and
> less surprising than any Linux I'd ever used (and of course than any
> Windows I'd ever used). If my computer were just for web browsing,
> social networking, email, and storing photos and videos, Ubuntu or Mint
> would be stable enough. But the way I work, I often have over 50
> windows open. I can't afford the massive instability bestowed by "we do
> it all for you" user interfaces.

Very good points being made here.

I'm going to match those points -- and similar to everyone else -- misread
what is being said, and agree we should choose *ONE* window manager and
delete all the others, and force people into a single working model.

Look, it is clear that is what everyone wants.  I apologize for the
group -- we got distracted by the bogus model of "lots of choice is good
choice".

We'll get started on that community requested goal immediately.

And I can assure, we will succeed: over many years we have assembled an
accompolished team of code deleters.  Internally we'll make a quick
decision about which window manager satisfies you all, delete the rest,
and prepare for the mission of convincing you all that we are correct
and you should all adapt to that choice or run Linux instead.

Personally I am a twm fan, but similar to others in this conversation
I'll be humble and limit my viewpoint to 80% validity, as long as kde or
anything fancy like that doesn't stand a chance I'm open to any point of
view.

Thank you all for your input.


Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Craig Skinner-3
In reply to this post by ropers
On Wed, 8 May 2019 00:23:09 +0200 ropers wrote:
> Tangentially related: Does anyone here routinely use the default fvwm?

Yep - daily.

Cheers,
--
Craig Skinner | http://linkd.in/yGqkv7

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Mihai Popescu-3
In reply to this post by Clark Block
> ... we should choose *ONE* window manager and delete all the others, and force people into a > single working model

There is no need to be so cruel. You can release OpenBSD like hosting
plans: free, basic, advanced and pro. You can rename them differently,
like Home, Enterprise, Business, Ultimate. Free will have that *ONE*
wm, then you add something for each level of upgrade. Needless to say
that starting from basic one has to pay :-{.

Still wondering what "great desktop experience" means ...

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Charles A Daniels
In reply to this post by Clark Block
@ Steve

>  One point I didn't see in RFC's post is stability. When I used OpenBSD
>  back in 2010, subjectively it seemed more stable, more consistent, and
>  less surprising than any Linux I'd ever used (and of course than any
>  Windows I'd ever used). If my computer were just for web browsing,
>  social networking, email, and storing photos and videos, Ubuntu or Mint
>  would be stable enough. But the way I work, I often have over 50
>  windows open. I can't afford the massive instability bestowed by "we do
>  it all for you" user interfaces.

This is also true. In my experience with Gnome, KDE, et. al., these
fancy configuration menus and wizards generally wind up being leaky
abstractions. Writing a simple format into /etc/hostname.if has in my
experience had far fewer caveats than NetworkManager or nm-applet. I had
mostly addressed stability in terms of UI/UX design, but in the broader
software quality meaning of the word it's a good point.

I would be fine with using a fancy tool to configure everything... if it
worked and was consistent. So far the only such tool I've found to
deliver on that (actually functioning and being consistent) is OpenBSD's
/etc/.

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Patrick Harper
In reply to this post by Clark Block
Aren't new users forced to use fvwm already since that's the default? Removing the options won't stop those inclined from finding one they like in packages (unless alt. DEs and WMs are set to be culled from ports as well O_O). However, there is an opportunity for all GUI ports to include a mechanism to add a shortcut to executables in a given WM's menu, that would be made more practicable with a forced default.

None of the Xlib window managers are resolution independent from what I can tell. Making fvwm comfortable to use on my 4K screen involved doubling the sizes of fonts, titlebars, borders, icons and the viewport window in .fvwmrc. I didn't bother trying to do the same with all the other fvwm modules and I don't think anyone should be expected to. Until the day comes when all of this, plus the X11 utilities (xterm et. al) can be configured automatically based on the Xserver dpi setting, cwm is the easiest to set up as only the border and font sizes need changing.
--
  Patrick Harper
  [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: When will be created a great desktop experience for OpenBSD?

Wolfgang Pfeiffer
In reply to this post by Mihai Popescu-3
On Thu, May 09, 2019 at 12:52:18PM +0300, Mihai Popescu wrote:

>Still wondering what "great desktop experience" means ...

Bullshit, that's what it means.

In nearly 20 years on mailing lists and on the internet, I don't
remember to have ever seen anyone asking such a stupid question like
the one in the subject line. Ever.

The OP doesn't seem to have the slightest grip on the fact that it's
everyone's own job to configure and set up their "Desktops", just the
way they want to have it. Or, alternatively, he expects the OBSD
developers to fulfill his lazy wishes so he just can sit and wait see
his "great desktop experience" roll onto his computer, freshly made -
and at no cost, for sure - by the OBSD coders.  Mindbogglingly stupid,
an impudent, this idea ..

Wolfgang

--
OpenBSD: "software we primarily develop for ourselves -- in the
hope that other people are like us and need similar things."
  -- Theo de Raadt

123456