GUI philosophies

[_The Doctor__]

What are the different schools of thought reguarding graphical user interfaces on computers of today and yesterdays systems?

[Nebulon]

Do you mean formally, or can this be a casual conversation?

 

Personally, all I think an operating system needs to do is cover the basics in an efficient manner (copy, paste, move, delete, rename, manage memory, and provide useful information about files and privileges).

 

As for interfaces in general, I know that there are people that think touch interfaces are THE future. I'm not so sure about that. I think touch interfaces are more of a component of the future since the mouse and tablet still has a place in the future of computing. Consider also that touch has been done in the past (along with light pens). Then consider the ergonomics of it all (dead-arm from raising your hand to the screen all day).

 

I can see touch interfaces being great for the types of things you typically see them used for on phones and tablets today (the web, certain types of games, videos, etc...). I don't see them being as good as a mouse or tablet for high-precision work like 3D modeling/animation and CAD applications.

Edited October 27, 2014 by Nebulon

[_The Doctor__]

Both formal and casual so we get a complete picture

 

I agree that the touch interface is being overly implemented and I can give and example of it at the local 'convenience store/mart'. I used to get my stuff and the clerk would be ringing me up as I approached the register and I would have the cash ready... gone in a flash! Now the touch key scanner combination has been installed and there is no keyboard at all... the clerks have to wait for me to get the the counter and scan everything followed by hunt and peck sessions on the screen. When I get a total and it is time to cash out... another touch and peck session! Each one of the tools can make things faster... we just need to have all of the tools and use the right tool for the right job... lets say the touch screen goes down... no keyboard and that counter location is down... or the scanner is not working... either the Point of sale is down or the hunt and peck heirarchy of touch screens makes it hard to do... all the options should be there so if any one thing fails there is more than one choice to get the job done... I like all of the devices.. keyboard mouse, pen, touchpads, tablets and touch screens. But still to this day it appears most POS systems are so stripped down and rely so heavily on touch scan that they hurt productivity.

 

A grocery store has the same problem.... I used to go to a grocery called Aldies ( a german company ) ... It was the best and again the clerk knew the prices and ran that keyboard looking down the conveyor belt and had a total ready to go when I got to the register.... same thing happened... they installed scanners and such... now I wait 10 minutes longer as each item is scanned or the clerk transitions to selec the multiple function when I have more than one of the same item..... there is no choice because of the system insists on it.... If I only had one item it would not matter done in a flash... but it is a grocery store... almost every one has 25 50 75 items.... don't forget the dreaded it won't scan no price comes up scenario where the manager and IT department need a 20 minute pow wow on how to enter it and get you on your way!

 

The GUI on the screen can help, a picture can be worth it's weight in gold, a familiar icon to launch something by touch on approach, as you settle in to use that keyboard and mouse or sketch on the pad. I prefer a GUI to be heavy with keyboard shortcuts, yet sprinkled with touch and mouse where it makes sense. I use the tools to help myself or someone unfamiliar with a system to find their way along as they learn the shortcuts. Cutting, pasting, selecting files I prefer a mouse for. I even like guesture based items to switch to another task rather than mousing for it or a function key... I use thumb gesture on the laptop pad to do that. I all makes sense and is efficient. I abhor using just a touch or just a mouse. Keying of some kind without covering up the screen with my finger/hand or smearing things up is a must!

[_The Doctor__]

Please don't bring thermal receipts back please! but they did.... these things dissappear or turn black after just a few months...

I remember that sentiment/ feeling as the touch screens came back into fashion after having failed the first time.. and now it's everywhere.

 

I prefer a GUI system with a watchdog timer to keep things going should something fail but not a strict time slice driven system. I want what I am doing to take precedent and give up the excess I am not using to be sent to prioritized tasks and pre-empting me only to handle something critical and telling my why. Currently you have to dig deep to find why a windows GUI pauses stutters and chokes. It should say I am indexing drives and give me a choice to delay it.. or it should say critical i/o is taking place so I don't worry what is going on. Then launching resource meter or sysinternals tools to find out what's going on to deal with it.

[UNIXcoffee928]

The book to read, before anything else is:

 

 

The Art of Human-Computer Interface Design by Brenda Laurel

 

 

 

[atari8warez]

mouse and tablet still has a place in the future of computing.

 

I don't see them being as good as a mouse or tablet for high-precision work like 3D modeling/animation and CAD applications.

 

Even the rodent is evolving and beginning to like being "touched"

 

 

[ataridano]

Here are my thought, in random order:

 

 

 

Having the GUI merged with the OS is fairly logical, yet I feel an argument can also be made for having the GUI as a separate library or program, especially in a single-tasking system. An example being DOS programs loading their own GUI at run time.

 

 

 

I agree with others here that touch is great and has it's place, but can't really replace other forms of input in the long term.

 

 

 

It's nice to have a GUI for basic file management, but I think a command line has to be optional as well.

 

 

 

One thing that Windows has been good about (one of my few compliments) is having a way to access most menu options using the keyboard. A GUI is still useful without a mouse or trackball.

 

 

 

This is an Atari forum, so I assume that a GUI on the 8-bit specifically is relevant too The problem with something like that on a machine with limited resources has always been the overhead. I'm not saying it can't work, but just that there will always be that temptation for the coder to get rid of the GUI so he has more resources for his code. On the plus side, having a GUI available can make GUI-based programs much easier to write because many of the I/O functions are already available.

 

 

 

Another A-8 consideration is the limited screen size and the tradeoff of features vs. available space.

 

 

 

Last comment: One thing I've noticed in recent years is the move to flashier neater looking desktops. While it may be cool, often I find it slows things down or clutters things up. I'm still trying to decide what my ideal desktop looks like

Edited October 28, 2014 by ataridano

[JD6502]

A couple of decades ago, as part of some science fiction ideas, I envisioned a touch display replacing the mouse. The concept was to add a small touch screen near or perhaps on the keyboard. Imagine a slightly over size iPhone. Stuff like pop-up and drop down menus, icon bars, task bars - all,or most, of the controls would be on this touch screen, the main display would mostly just show what you were working on. I can't say that I thought it through very far, but the modern touch interface is pretty close to what I had in mind, except that data entry would use voice or keyboard. I didn't think of tablets - Who would want to plaster their grubby fingerprints all over the screen? How would you keep it clean?

[+Stephen]

I'll tell you one thing - touch screens are utter SHIT for gaming. There's also many things I can do with a single line on the command prompt that would take hundreds of clicks. Why is there no way in Windows Explorer to do a "rename *.nfo *.txt"? This isn't advanced stuff.

[_The Doctor__]

well if you drop to the command prompt an beg pretty plz you get close

[phoenixdownita]

I'll tell you one thing - touch screens are utter SHIT for gaming. There's also many things I can do with a single line on the command prompt that would take hundreds of clicks. Why is there no way in Windows Explorer to do a "rename *.nfo *.txt"? This isn't advanced stuff.

Ahem

http://www.howtogeek.com/111859/how-to-batch-rename-files-in-windows-4-ways-to-rename-multiple-files/

[flashjazzcat]

Ahem

http://www.howtogeek.com/111859/how-to-batch-rename-files-in-windows-4-ways-to-rename-multiple-files/

Where does it explain how to use Windows Explorer to rename *.NFO to *.TXT?

[phoenixdownita]

Where does it explain how to use Windows Explorer to rename *.NFO to *.TXT?

I was thinking ... the part where it tells you to open a command prompt and type "ren *.NFO *.TXT"

 

Was I out of place?

Edited October 30, 2014 by phoenixdownita

[flashjazzcat]

I was thinking ... the part where it tells you to open a command prompt and type "ren *.NFO *.TXT"

 

Was I out of place?

No - but clearly Stephen already knows you need to use the CP for that and was simply pointing out why it's impossible in Explorer. So I didn't quite understand the point you were making. Sorry.

[snicklin]

I believe that the concept of touch will die out relatively soon, maybe in a decade or so. In its place will be pointing lasers at screens or the mind reading techniques, which will probably be a bit more refined, although still not perfect at that stage.

 

Compared to yesterdays systems, I don't think that we've actually progressed very far. The mouse has been around since the 70s, which I guess was the decade of the joystick also. Touch was around in the 80s but never took off due to its price and lack of accuracy (but did exist) and so I don't think things have moved that far.

 

As for reading the brain, they need to be trained and although I've never used such a system, my bet is that if you have a sneaky sexual thought in the middle of what you're doing, it'll mess everything up that you're trying to do.

 

Maybe we'll need to design some brain filter systems, possibly more for some people than others.

[flashjazzcat]

Regarding touch: despite the fact I have reasonably slim, guitarist's fingers, I bloody hate the interface. I have a 7" tablet I never use and a 4.8" Android Smartphone whose predictive, auto-correcting keyboard struggles to make sense of the nonsense I generate using the on-screen buttons. Meanwhile, I'm typing this at the PC at about forty words a minute. Of course the tablet and phone are woefully cheap and slow, so perhaps some premium hardware would result in a better user experience. In any case, when I use touch, I do it because can't reach a real keyboard.

 

I agree the desktop metaphor hasn't really changed much at all since the mid-Eighties. I think on the desktop PC in particular, no-one really knows where to go next with it and how to make big changes. But the thirty-year old desktop metaphor works just fine for now, in the absence of anything better. Since MS now tacitly concedes that trying to enforce touch on the desktop PC was an unmitigated disaster, I think we can all agree that's not gonna be the successor to the mouse and keyboard any time soon. When I can "think" complex code passages into an editor, then maybe (although - as snicklin says - results like "LDA #1 / STA BOOBS" are a distinct possibility). I don't want to speak them into a mic. At the moment, there is no faster and more accurate way to enter text than a keyboard. Sloppy, mangled forum posts across the Internet are at least partly caused by the proliferation of on-screen keyboards.

 

This thread is starting to sound like it belongs in "Computers, Science, and Technology", although I'm well aware of its original purpose.

Edited October 30, 2014 by flashjazzcat

[barnsey123]

Let's talk UI (will come onto GUI later). Years ago, when DOS 6 PC's with green screens were all the rage, my boss and one of his trusted advisers came up with the brilliant idea of speeding up data input by equipping the PC's with a barcode reader (a laser or IR pen hooked into the PC). EVERYTHING that users ever had to type were printed onto barcodes that were stored on laminated sheets of paper and held in ring-bound folders...can you see where this was going?

 

In those days, my opinion on such things was quite well respected - people used to listen to me (bizarre, I know) - so I was asked about this brave new barcode-centric world and I was expected to give it Barnseys Green Light. They were shocked when I told them that IMO, it was a complete turkey and I outlined all the flaws in the plan. It didn't go down well and in a heated debate I was put under immense political pressure to give the plan my support and get the users onboard. I could not agree but the plan went ahead regardless.

 

It failed...and it failed big time...there were, in fact, real tears of frustration from the users (they went so far as to disable the keyboard in the software to force users to "get used to it"). Management could not ignore the failings and so v2 was released. V2 was a masterstroke. The most commonly used barcodes (e.g. "08", "04", "AUDIT" etc) were removed from the folders and sellotaped to peoples desks. The trouble was, the list of "commonly used" codes differed from desk to desk dependent on function and rather than maintain separate lists (which of course would be inefficient) we had a "master list" of EVERYBODIES common codes all covered in sellotape and taking up lots of desk space. Of course, this slowed people down still further as they first had to scan the common list before leafing through the folders. Nightmare.

 

The project was eventually dropped and besides a few red faces, irreparable damage to reputations and careers, me becoming a pariah in the eyes of management and thousands of pounds in lost productivity and nervous breakdowns in some staff, no real harm was done.

 

So, what's this got to do with GUI's? Well, the important thing is not the "G" in GUI, it's the "UI". The example above describes how the inappropriate use of new technology can destroy the UI and make people less productive. Touch screens are a case in point, yes touch screens are very useful (zooming in on maps on an ipad is very intuitive) but touch screen keyboards are mostly horrible and slow you down. GUI's are a great tool for the compartmentalization and vizualization of applications but they are not an end in themselves. IMO, it's the combination of all these technologies that is important where the whole is greater than the sum of it's parts. GUI, Keyboard, Mouse, Touchscreen, Touchpad, Command Line, Scripting Language and of course the applications. Combine these parts well and give users control choices and things become flexible, fast and user friendly. Even barcode readers have a place as long as they are used properly...

 

Back to the Atari GUI, I'd like to see ARREX like wot we had on the Amiga...@FJC I want a set of proposals on my desk first thing Monday. Alright?

[atari8warez]

 

Edited October 30, 2014 by atari8warez

[The Usotsuki]

I still see ProDOS-16, GS/OS and pre-8 MacOS as the pinnacle of GUI design.

[phoenixdownita]

I'd go one level deeper.

Before GUI and even UI there's UX, in the sense of User eXperience.

Until we understand what is it that we want the User to eXperience the rest (UI/GUI or presentation layer in general) is kind of secondary.

 

I believe we managed to get by for decades focusing on UI (and GUI) thanks to advances in technology but made relatively little investment in UX per se.

 

UX itself does not live in a void and is bound to the scope of whatever the UX is supposed to be for.

 

The bar code example cited above is a case in point in which they focused badly on a non scope, eliminate typing, rather than on the real scope they were after, which apparently was to speed up data input .... once you focus on the real deal I believe that a UX in which you need to sift through lists of bar codes etc.... would obviously be against the scope.

 

Once the UX is defined, then UI and GUI can be brought in.

 

Historically we went the other way, like the renaming file example to change extension. Why would an extension even be needed? It seems like a way to classify content by "possible reader", which is a somewhat usable proxy for actually describing the content type. So a .pdf file which ties to a reader (or set thereof) really is used to convey both the fact that is some form of Text + Images and at that in a particular file format.

The fact that the file format itself is smack in the face of the user is somewhat subpar for good UX ... when I drive my car I don't need to know much of any details about it with the exception of the specific fuel it needs (at least for the time being) and rules/law of the road .... not much about the technology of the car itself is smack in my face when I steer that wheel etc... the UX is pretty streamlined and functional.

 

Granted the UX with a car has been made so it is pretty simple and hides many of the details that in quantity probably pale in comparisons with a computer inherent complexity ... or do they?

Maybe the real issue came from allowing everyone to do everything at a low level (files, memory, pixels and a heap of other details) with the result that there's no cohesive UX for a computer in the general sense, just a bunch of cobbled together partially functional UIs.

 

Am I making any sense?

[576XE]

As I can understand GUI is just a way to communicate comp with some man.

 

If so, It MUST NOT carry some another distracting meaning.

Never.

 

Just remember...

There was OS/2 with it's very sophysticated object oriented WPS (WorkPlace Shell)

It's technically perfect but... Where is it now?

 

I Remember only one GENERAL WPS modification .

 

It was German program which did all that do modern "Image Maps" in HTML.

If I want to go to Cinema I go alone the drawn street to some building determined as "Cinema", but not to the "Saloon".

 

And that's all. I even can't find this prog in Internet!

 

There are nothing above beautifying of meanings.

 

And Win8 withit's touchscreen is NOTHING more that LCARS of 60-th...

Edited October 30, 2014 by 130XE

[flashjazzcat]

Back to the Atari GUI, I'd like to see ARREX like wot we had on the Amiga...@FJC I want a set of proposals on my desk first thing Monday. Alright?

Heh. You might just get SymBASIC one day, if the stars are in alignment.

 

I still see ProDOS-16, GS/OS and pre-8 MacOS as the pinnacle of GUI design.

Yeah - as far as making best use of a mono display, System 1-7 is hard to top.

[kogden]

Early MacOS got a lot of things right from a UI standpoint. GEM wasn't terrible but I think MacOS actually did a bit more, especially after things moved into the System 6.0.8 era w/ MultiFinder, etc.

 

The sad thing is that all these things Apple got right people consider "too hard and confusing" now. People want Angry Birds and Facebook appliances, not computers.

 

And in the past people used to think of the Mac as a "toy" because it was an easy to use machine for stupid people and thus not a "real" computer. I don't get it.

Edited October 30, 2014 by kogden

[+wood_jl]

This thread is starting to sound like it belongs in "Computers, Science, and Technology", although I'm well aware of its original purpose.

 

Indeed, and your original A8 GUI thread shall now "rock on," as a result of recent circumstances. I am delighted, and I'm quite sure other share the sentiment!

[Kyle22]

Yes, hopefully we can keep the "on topic" discussion in its proper thread, and let the evil troll(s) post here instead of bothering and harassing others by PM. The baddie(s) will be dealt with and won't threaten any of us in the future

 

We all know who they (he) are (is).

 

Anyway, I absolutely love the concept of a GUI on Atari8. Written in 100% Assembler, that is awesome. If only M$ had written Win in Assembly...

 

I propose a project (that would take a team of programmers YEARS to complete) to re-write Windows XP into pure x86 assembly. It would be absolute paradise for the gamers