Wireless GUI: "Killer App" for FreeBSD?

Much as I agree that it shouldn't be more Linux-y, I do wonder....

There are certain advantages to becoming more mainstream. Whatever folks might think about Ubuntu, it seems that it went a long way in its heyday at least, in getting both hardware and software vendors to give Linux more serious consideration.

While this might just be an old guy's vision of younger folks, I think that part of what happened was that as Ubuntu made it easier to use on a laptop, kids could be cooler than their friends by having something different. Had it been too hard to get working with most laptops, though, many of these kids might have given up. Then they go to college, come out, go work in IT, and they have this fondness for Ubuntu and Debian based systems.

Now, FreeBSD doesn't have anything like that. While we are all thinking, Great, who needs it, there is some value in having numbers, and having a growing influx of younger admins and developers liking it. PCBSD, while it lowers the barrier a bit, still has various issues with, for example, finding some Linux distributions if it installs its own grub, or being found by other Linux distributions, at least on ZFS, if using said distro's grub. (Note that this was in one quick test, while trying to decide whether to recommend it to a less technical friend, and while I'm sure it can be overcome, the entire experience seemed as if it would require more support than I was willing to give.)

I can easily find myself in the mood to support either viewpoint. If only a few people in the next generation of admins and developers have no interest in it, it will slowly fade away.
Yet, Linux becomes more like Windows with each iteration, even allowing for my old curmudgeon prejudices.

I do think there are advantages to lowering the bar, at least to some point. Then, I look at what the younger developers who seem to only be able to comprehend using a system on a single user laptop that goes to different places, and how they make it harder for sysadmins with the things that get put into the latest versions of Linux, and my view goes back the other way--let's leave the bar high and let the young, smartphone oriented developers go and continue to ruin Linux.

Big Disclaimer. I'm old and grouchy and I'm sure that I resent some changes just because they are changes.
 
scottro said:
Whatever folks might think about Ubuntu, it seems that it went a long way in its heyday at least, in getting both hardware and software vendors to give Linux more serious consideration.
Click to expand...

I wouldn't call Unity "serious"... :e

scottro said:
Yet, Linux becomes more like Windows with each iteration
Click to expand...

Yes, and that's because Windows users use Linux because it's shiny and free (and some of their friends obviously ignored that Snowden said that Linux can be opened easily).

If FreeBSD was more popular, it would end like Ubuntu, given that FreeBSD is the "user-friendly desktop BSD" compared to the other "plain" BSDs. I, for one, wouldn't want that.
 
Cthulhux said:
...given that FreeBSD is the "user-friendly desktop BSD" compared to the other "plain" BSDs...
Click to expand...
According to whom? Do you mean PC-BSD? FreeBSD comes without even Xorg, whereas other BSDs include it in the base. I think there is some context to the thread that I'm missing.
 
According to my personal experience. I gave PC-BSD a short try during my first weeks with "the BSDs" just to find out that I can achieve quite the same results (without the PBIs) within not much more than an hour with a "real" FreeBSD. Setting up FreeBSD is amazingly easy compared to OpenBSD and NetBSD if you haven't read the manual first but already used Linux in the past.

Of course that's relative. PC-BSD, for example, is hard to use compared to Windows.
 
scottro said:
Yet, Linux becomes more like Windows with each iteration
Click to expand...

If that means they are removing the bloat from the desktops, it is a good thing. Granted, I gave up on Linux little less than 10 years ago, so I don't know what it is today. You made a very good post in my opinion.

Cthulhux said:
Yes, and that's because Windows users use Linux because it's shiny and free
Click to expand...

I don't think being shiny has anything to do with it. There are lot of shiny programs that don't get user acceptance. Take a look at Windows 8 for comparison. Free is another thing, but it must be accompanied with a value to the user. A useless but free program is still a useless program. I think that is the most significant reason why many Linux projects are not gaining significant advances in their competition against proprietary software. This includes thinking that a good interface for a casual user is a shiny interface, rather then an empowering interface.

Generally, what I find saddening, is that these idea threads always seem to get a lot of a very emotional posts. This thread being the most constructive of the lot I have read, with more then couple posters giving proper arguments on the problems as well as on the most fitting approaches. The forum should have the thanks button for them because otherwise the negative posts can get too much value in the readers minds. I would think that overall, the experience of the seasoned users would be better used on guiding the new people/future developers on how to best complement the system with new and friendlier tools. Regardless of the interface style. I'm looking at adduser() as an example. It is a dumbed down and good interface to add users, yet there is the pw() as a full control interface.
 
I do admit that when FreeBSD's popularity does rise it is going to be very hard for the developers (and community) to decide on which features to include from all the many, many requests and patches.
This may sound draconian but where I believe the line should be drawn (and likely will be with FreeBSD) is directly behind requests from Administrators and Developers. My reasoning for this is because the requirements of casual users change too much with the drunken sway of social computing trends of the current day (Just look at how even the mighty consistent Windows UI has been crippled with this recent tablet fad).

naali said:
I'm looking at adduser() as an example. It is a dumbed down and good interface to add users, yet there is the pw() as a full control interface.
Click to expand...

This is a good example. I use adduser() because it is indeed simpler. However, if this traditionally UNIX util was dramatically changed or got replaced entirely... I would swiftly switch to using (probably scripting around) pw() directly (and accept the steeper learning curve proving that I am not averse to change) rather than play catch up with whatever trendy replacement it was given.

To clarify, I am really not afraid of change. Otherwise I would never have migrated away from Windows all those years ago. What I do not appreciate though is this recent trend of "trying things out because we think experimental == modern " which is plaguing the open-source world recently.
I am happy to miss out on cool features if it avoids any breakages in existing software, documentation or workflow.
 
CreativeGPX said:
To make the OS more widely appealing requires more volunteer-hours than the community has. To get more volunteer-hours for the community requires the OS being more widely appealing. It's a circular problem which is broken when some new developer comes along and starts a usability team. ;)
Click to expand...


So how does someone who doesn't program, but understands programming very well start a usability team? Because we need one.
 
FreeDomBSD said:
CreativeGPX said:
To make the OS more widely appealing requires more volunteer-hours than the community has. To get more volunteer-hours for the community requires the OS being more widely appealing. It's a circular problem which is broken when some new developer comes along and starts a usability team. ;)
Click to expand...


So how does someone who doesn't program, but understands programming very well start a usability team? Because we need one.
Click to expand...

You just start it. Emphasis on you. Don't fall into a trap of thinking that development of FreeBSD is centralized and you have to ask the powers that be if you can do something. Just do it and make your work available to others.

As for the work you would be doing, I would say that it would be mostly documenting what (and possibly why) is not working very well or is hard to use in the existing base FreeBSD and the add-on desktop environments and to write the evaluation documents based on the gathered information. You'd have to be decent technical writer and know what is a good bug report.
 
Every day, I conjecture to myself, more would discover and use FreeBSD if someone cobbled together an installation and use flowchart, asterisked to enterprise uses in the margins maybe, similar to those quickstart 8-pane foldouts one obtains when purchasing printers, routers, etc. This from receiving similar materials within shareware packets back before the days of common Windows usage... as well as wishing for the same in wide usage, say one for Zfs, so one would not have to skim through forum (not this one, in this context) posts asking questions one might have already have answered from a PDF printout... If one has read this far, one might surmise that one for Wireless networking would be welcomed by many, more directly relevant to the title of this thread, if not its purpose.
 
Going back to the original idea, I would agree that having some 'tool' to make setting up network connections would be a nice addition, even to the 'base'.
It wouldn't need to be a big GUI tool, but would need all necessary options to be asked for & then set up for the less experienced user. I think a curses based script may be adequate, as it wouldnt be used often by most people, & those that did need it regularly would come to know it intimatedly. As has been mentioned before, it would be just as easy to run under a GUI via an xterm.
 
jb_fvwm2 said:
Every day, I conjecture to myself, more would discover and use FreeBSD if someone cobbled together an installation and use flowchart.
Click to expand...

You might well be onto a good idea here. After all, unlike Linux, FreeBSD does have stable workflows and unless Launchd, Wayland and other massive breakages start to slip through, this kind of documentation would remain relevent, up-to-date and worthwhile.

Now we just need to find someone with a passion for this kind of documentation ;)
 
Wi-Fi utilities including wired networks shouldn't take much resources since similar applications are already working in mobile devices such as iPhone and all sorts of droids or tablets. The base code should be accepting command language as pure bsd for those do not need a GUI. Nevertheless the base app should be easily adapted to work under a GUI for further inclusion into desktops.

Any design or architecture on Wi-Fi utilities including wired networks should start with assuring Mobility while wired network is using only the discovery part to end into a Gate. Such starting point is to invoke a session based on which Mobility Connection is even achieved while moving from one point to another. Since the “Mobility Connection Session” is based on DHCP the wired network will very easy take advantage of such same session used in mobility called Discovery (DHCP and/or DNS implementation). The Property of such connection as we know can be further edited by user.

Reading: https://books.google.com/books?id=y...maintaining a connection while moving&f=false


Some thoughts on coding (not necessary complete but good enough to influence):

Discovery Function (M,W) to achieve best network discovery accepting user input of either mobile(M) or wired(W); Returns a list/array (every item in the list contains Property as another array of data pertinent to connecting including if required a prompt for user/pass or if it is a session based connection which may include credentials) – The M-Discovery may be static or “ongoing”. The later meaning is looking for next connection. But the next connection may be in a direction predicted by GPS or relay detection part of a network… and so on as much our minds can cover. – Of course this function only creates temporary items that are deleted if not selected. – Discovery also means what works for your adaptor among the findings. “The adaptor compatibility issues” but that shouldn’t stop us to at least see the case listed in gray and saved in a temp database for further process – this phrase calls for an intermediary function that will further filter results based on internal adaptor(s) and so the discovery part is dissected into a network analyzer of available networks that further are filtered if compatible with the adaptor(s) .

Connector Function (Process Selected Item from Discovery Function) returns a configuration e.g. adapter settings available for user editing. This function should have a section to auto-manage temporary adapter settings stored in a default directory. “Creating and saving a network connection is not necessary forever” – may allow a history of 30 days depth and last three if any are not deleted.

The GUI part (create new, edit/delete existing) would use the (a) Discovery Function to show a list of available or potential connection and should user select any listed items will then; (b) process in background the Connector Function then to: (c) screening a particular folder (where “Connector Function” stores its adapter configurations); make active, and create a list based on which user can further edit Properties of the stored Items.

REM: old ports may not apply all standards of discovery have been implemented recently. Imperative that is, to rewrite the whole code and only get inspired by what was done before in old ports.

Key Notes: Network Discovery, Network Analyzer, Network Adaptors

Hope it will help someone. -- AHR 20160327
 
kpedersen said:
You might well be onto a good idea here. After all, unlike Linux, FreeBSD does have stable workflows and unless Launchd, Wayland and other massive breakages start to slip through, this kind of documentation would remain relevent, up-to-date and worthwhile.

Now we just need to find someone with a passion for this kind of documentation ;)
Click to expand...

Be careful what you wish for. People tried to do the same thing for Linux and although well intentioned, it didn't go the way they thought it would.
 
Cthulhux: "Be careful what you wish for"
https://forums.freebsd.org/threads/44211/page-4

Plain and simple -- I am only trying to solve problems as I am running into -- nothing more.

Apparently I could see ramifications based on my writing but please believe me I had no idea what Berkeley Packet Filter is doing before the writing.

... and the beginning is posted up here:
https://forums.freebsd.org/threads/55651/

... and you can replicate as much as one can wish, while me I am just becoming familiar with FreeBSD forum.
... Yeah I am getting too much help …
 
Well, if I scratch my mind I could say two big words: INVENTORY and RETOOLING! – There is nothing wrong in serving desktops while the black screen is available.

And, lets not forget what free….(BSD) means: FREEDOM!
 
You must log in or register to reply here.
Back
Top