The rise of OSX (remember, when it came along Apple had a single-digit slice of the computer market) meant that people eventually got used to the idea of a life with no desktop personalization. Nowadays most people don’t even change their wallpapers anymore.
In the old days of Windows 3.1, it was common to walk into an office and see each person’s desktop colors, fonts and wallpapers tuned to their personalities, just like their physical desk, with one’s family portrait or plants.
It’s a big loss. Android and Linux desktops still offer massive amounts of personalisation options – thank god – but the the other major platforms have all individuality stamped out of them. It’s boring.
I never bothered much with personalisation. What I would like is the ability to log on from anywhere in the world using any PC and get my desktop and apps and possibily even data (in decending order of being enabled). Really honestly I don’t have the mental bandwidth for meddling especially becaause, like games, anything created or learned at this level is instantly reset with the next version. One thing I do miss is proper human computer interface design and skeumorphic elements. Flat design doesn’t appeal to me at all.
One thing about the anologue world is things don’t change on you. They are more stable. More real. Computers can be too busy and arbitrary and change too much on the flip of a bit. They don’t always fit how the brain works nor are they always ergonomic.
The problem with changing appearance is that a lot of programs don’t respect altered color schemes. Keep the default scheme or the program will be unreadable. I had that problem with an employer 30 years ago. I wanted the program under development to use the user defined Windows colors; the employer preferred a color scheme specific to the app.
One video comes to my mind :
https://www.youtube.com/watch?v=5OGOwzGbnFw
Come on. You know Win 3.11 was the pinnacle, and it’s been down hill from there. 😉
Some of those are valid criticism, but it has a hard time maintaining momentum.
The mouse pointers are a push. The dino cursors weren’t that usable, and neither were any of the alt cursors outside of the jumbo or high contrast versions. I kind of prefer the updated circle pointers to the hourglass pointers. Especially the flat circle pointers.
Edge is better then IE. I used to get rid of all of the IE toolbars or use Phoenix to create a very similar interface.
The Windows command prompt has always been trash. There. is. no. excuse. for. the Windows Command prompt. There are some wonderful terminal emulators which they could have ripped off, but nope. MS ships this garbage.
I really hated the Windows98 startup screen. The new one is better.
To me Windows XP was the ‘pinnacle’ (or Windows 7 to a lesser extend) as the UI was absolutely clear of the intend, you knew what was clickable/selectable/editable, scroll bars were not hiding, shortcuts were obvious with the underlined characters, you could change the theme (with the right ux.dll patch) or even return to official “classic” theme.
Windows 10 is shit in every of these regards. Sure it improved in some fields (hope it did) but I cannot see a breakthrough improvement, a “revolution” over what was made before. It was an almost perfect desktop os, it fired itself in the foot which gave Linux some momentum in the desktop area. Sort of.
I miss those days when I was not cussing against that clusterfuck of a user interface that Windows 10 has became.
Kochise,
The first thing I did was to switch to classic on every new OS until microsoft deleted it. When I’m forced to use windows 10 today, I can say definitively that it has not grown on me. And not just because I’m some linux fanatic, but because I find it more difficult to see and use without the visual clues that operating system vendors including MS had worked so hard to standardize in the 90s and early 00s.
I know this is becoming ancient history at this point, but the willingness (and even forcefulness) with which microsoft would throw it’s own UI guidelines away was notoriously frustrating at the time and caused real usability impediments.
https://www.pcworld.com/article/2012202/how-to-shut-down-windows-8.html
They’ve since backtracked some of the worst design changes, which is good but I still feel they’ve devolved from their pinnacle of emphasizing clarity and usability to where they are at now. And while some people are fans of artistically flat featureless UIs, it’s kind of frustrating for those of us who want our computers to be clear & functional first rather than to make a fashion statement.
As someone who designs and programs user interfaces I completely understand why they went against customisation. Creating good user interfaces that are also at the same time customisable is incredibly difficult so I totally understand why people went away from that approach.
We have no specification judge whether their decision to allow customisation or not was a good ro bad idea. I can think of very few instances (close to or zero) where not hooking in with user customisation it was justified. Most apps aren’t actually that important even if the boss thinks they are.
Just using normal Win32 api, made your program’s colors customizable. This applied from Windows 3.1 to Windows 2000 at least, then to a certain degree all the way up to now. Not difficult at all.
With web applications, the only thing you have to do, is to give elements proper class and id names. Then people can use userstyles.
Neither of these are difficult.
Adding a switch for dark mode, isn’t difficult either. It’s just a separate theme.
Invincible Cow,
It depends a lot on the framework the developers are using. Windows components use windows styles by default. Take a visual basic application, everything gets built up on top of native components (yet these could be customized). A lot of applications that look out of place on windows are those that are created with non-native tools that render their own components from scratch (GTK/Java). However standard interfaces became uncool and non-standard components became a fad even for win32 applications. Think back to programs like winamp who’s interfaces are literal bitmaps and x-lite with odd shapes.
https://skins.webamp.org/
If I recall correctly, I had a gigabyte motherboard overclocking utility that was skinned and shaped as an animated combustion engine. Honestly I find such things annoying more than anything, but I also find all these damned LEDs inside the case useless too so what do I know? Haha.
If you build your GUI framework correctly you can easily switch between Win32 and custom interfaces including building an entire “lookalike” interface on an OpenGL or Direct3D surface if you want to produce a full featured interface for a game – just blast it to a texture. It’s a bit of work but not hard. Most GUI frameworks could do this but they’re too lazy.
You can easily do fully featured GUI apps in a game which work live either as a walk past item with your avatar or click to zoom in and use as part of gameplay. Hook in a scripting engine and you can do fully featured apps either native, custom, or in game. Examples to look at for inspiration are Borlands VCL object model and Autodesk Maya (previously) Ailas Wavefront Maya. VCL was a really nice piece of design. Also not many people may realise how much Maya was built around scripting. Apart from the GUI bit and core rendering engine pretty much everything in there was a script. So you could run something as an app or as part of a fullscreen app as an entity running within that app .You can either go down the VM route or go for compiled as well. Throw in script compiling and you can run an app on your Windows desktop as a standalone exe exactly as you would in a game. Again, it’s not hard. It’s just people are lazy.
The really ambitious might want to float this as a desktop on top of an OS or throw in an HTML parser…
HollyB,
Maybe, but it’s kind of a big ask for 3rd party developers to fully replicate the look & behavior of native controls. I’ve seen it tried. What usually happens is that developers copy a specific default OS theme at one point in time. But this theme can be wrong if the user upgrades or customizes their native controls. Programs done this way can have the wrong fronts, button styles, etc and may neglect to implement the full set of keyboard/interaction semantics that native controls have. But perhaps more problematic is that it stops working with accessibility tools like text to speech and context sensitive mouse acceleration profiles.
Consider that when win98 plus pack themed our desktops back in the day, most applications back then took on the new themes because they were using native controls. This would not work in applications that rendered their own widgets. It’s not the end of the world if they don’t look the same, but I just mention it because the end result can be very different that using native controls.
@Alfman
Relatively easily solved. Get all the dimensions, draw on a native surface, copy. But, yes without looking into things I can’t give specifics for OS changing implementations. Yes, accessibility is an issue and custom draw surfaces are a problem. I’m not religiously opposed to providing accessibility hooks if they are there. It’s been so long I can’t remember whether it could be done or not. The whole focus was to provide native gui and in-game gui with equal functionality so although it could be used on a framebuffer surface as a replacement that wasn’t the goal. (All this stuff could have been restricted in a configuration file so not a problem.)
The reason I built a gui framework with this kind of functionality was for a now long somewhat over-engineered lapsed game project so I had reasons to do it. The editor was to run on the same VM.
HollyB,
Why would someone want to go through the bother of implementing a native UI for the purpose of copying only to then render those pixels on a new surface? That doesn’t make sense to me. Either go for native UI, or you implement your own…this hybrid sounds like a lot more complexity and debugging.
Many controls don’t have static images to copy. You’ve got sliders that move, buttons that take focus and can change shape & apparent when clicked, color interpolated gradients, text boxes that have cursors and are editable in complex ways, drag and drop, possible mouseover effects, drop down menus and trees that expand. Windows also has standard dialog boxes that resize and navigate directories, etc. Hypothetically you could potentially copy all these things in real time to your own UI engine, but the question I can’t get past is “why?”
I’ve seen a lot of linux applications ported to windows that have awful file dialog boxes because they did not use native dialog boxes available on windows. Some toolkits have tried to clone these, but they always lack functionality such as not being able to navigate network shares. Sometimes the simplest solutions (use what’s already there) will produce the best results…
Of course, I know all about the dilemma for portable toolkits. Native controls are much less portable and toolkits generally aim to do things the exact same way on every platform, even if the result is inferior to native. There are always tradeoffs, but fortunately I think the greatest common divisor has improved over the years.
If you are focusing on a game specifically, well the usual route is just to rely on in-game menus/UIs/mechanics and not bother with native controls at all.. Games in particular can be a lot more creative this way. There’s not really any expectation from gamers for games to inherit desktop themes & customization. A lot of games are full screen anyways. I think most of us were thinking more about standard UI themes in productivity applications.
@Alfman
You have to bear in mind I’m writing from memory and it was 10-15 years ago.
It was a fully object orientated model. All the properties were exposed so the code to query stuff or do stuff was already there. The code to implement native, or surfaces which contained a GUI desktop like openGL or Direct 3D was fairly trivial.
Because creating a GUI object with the target set to native allowed doing a full application like a Maya style editor, for example, and with target set to a framebuffer surface allowed for a fully functional in-game GUI which could replicate an OS desktop in-game for artistic and gameplay reasons – in-game presentation of a desktop on a computer in a scene and fullscreen in game was possible.
The reason why is a design decision. The GUI object framework was part of a larger application framework which could either be exposed for a script or compiled script to create native applciations which ran on the VM, or provide a fully functional full screen composited desktop (before composited desktops existed) in-game or the same thing on a discrete surface in the game. The editor would be based on the native stuff. Gameplay and artistic content would use the in-game surfaces. An app could run native as a standalone or in-game. The HTML object could easily have handled text content like books. Then there is in-game email etc. Scaleability, and play while downloading and updating were design goals too. This last one was made a lot easier with Windows 200o and onwards as OS network and file handling changed.
Remember this project began before composited desktops and CLANG and LLVM and DevIL were a thing, and Java was big, and nobody had even thought of browsers with JIT engines. While the engine didn’t make a huge thing of shaders as they weren’t mature and the hardware wasn’t there shaders were only just beginning to take off Khronos didn’t exist let alone Vulkan. As for why? Lots of reasons which made sense at the time. It wasn’t hard work and the functionality was trivial to add as well as expanding and making realisable various game design goals such as interactivity, a richer world, real artistic content and lots of it. My favourite map design was an art gallery with voice lectures like you would experience walking around a real gallery. Just because you have action and adventure doesn’t mean everything else has to take a back seat. Anyway, it crashed and burned before it was anywhere near finished and Call of Duty took over the world. Why? Why not?
This is probably the last time I will mention it. I’ve got over it now. I just wanted to vent and share.
HollB,
Well, it sounds like you had a very specific project in mind that I’m not privy to. My points in this thread have to do with the pros and cons of native OS components versus custom components generally speaking for normal application usage. I have no way to know if my arguments are relevant to your game project. Without more information I’m not really in a position to critique it. I don’t understand what your goal or outcome was or even how it related to windows component theming, I’m kind of at a loss to be honest.
Ok, I do plenty of that myself.
I could be wrong, but the minimum customization was part of the value proposition of MacOS. Nothing could be changed, so everything was the same. It’s a very well done interface which gets out of the way, and I find very few reasons to gripe about it. Same thing with Gnome3, it’s a very well done interface which gets out of the way, and I find very few reasons to gripe about it.
The flip side was Windows which could be hacked up all day into something totally unrecognizable. Application developers took advantage of this, and they created some monstrosities. Windows media player, Realplayer, certain Winamp skins, and I want to say the original Trillian was really bad.
I took advantage of this and created some monstrosities with different theme packs and skinning. Part of it was for fun, but most of it was to work around problems with the Windows interface.
Forcing devs into a box and making them conform to human interface guidelines is a good thing. At the end of the day, we need to get work done. Computers are tools, and setting hard boundaries on interface design makes people’s lives easier (assuming the interface isn’t awful). It’s the concept of mise en place. Everything is where it’s supposed to be.
Something else to think about, is Windows has a brand and lifestyle to promote. I’m not sure what lifestyle that is, but MS needs Windows to promote Windows because people buy a Dell/HP/Lenovo/Acer and not a Microsoft, for the most part. The Windows brand needs to be in your face so we don’t forget we’re running Windows because at the end of the we don’t see an MS logo on our hardware.
Most people didn’t change their wallpaper or theme back in the day either. Most installs stayed bone stock. Or they installed the daily wallpaper malware thing that I don’t remember the name of, and I had to explain to them why it was getting removed.
@Flatland spider
I agree with your comments except the bit about logos. My life won’t suffer if I never buy or consume a product without an “in my face” logo ever again.
Who’s been watching French Cooking Academy on Youtube? Yes “mise en place” is a good concept. There are variations on this which I am sure everyone knows. Following on slightly speed comes with skill. I know Marco Pierre White has become a meme but oh my life does he know how to chop vegetables. He cuts garlic so fine it’s a paste and his onions aren’t far behind, and he’s like woosh done. Following on further, and it’s becoming orthogonal to the topic on the surface, there is the Zen concept of “learning by observation” and of course so called “mirror neurons”. you can learn a lot on a deep internal level and improve your own performance simply by observing a master of their craft and putting things into practice yourself.
I saw what you did there!
I prefer things without logos too. I still hold on to my youthful, punk anti-corporate ideals. XD
I was trying to find a rationalization for the Windows UI such a cluttered mess.
I’m not sure, but not me! I learned about mise en place in a cinema class, and that’s my reference for it.
I do like cooking though, and I’ve heard several chefs mention it in cooking classes and at bougie dinners. XD
I’ll have to check out the French Cooking Academy. Thanks!
This is true. There are little things we can pick up by watching those who are incredibly skilled. Somethings are hard to translate into writing, or they’re background information which people should already know. Such as, browning garlic. Lots of Italian pasta sauces start with browning garlic in olive oil, but what shade of brown? A deep tan, a nice gold, or hint of yellow good butter?
I remember the early 2000’s as an exciting time for Windows customization, creativity, and overall fun.
– Stardock’s WindowBlinds, CursorFX, IconPackager.
– Winamp skins.
– Custom mIRC themes.
– Alternative shells that were actively developed and maintained.
– Word Clip Art and Clippy. Rover the WindowsXP Search dog.
– BonziBuddy.
C++ Builder and Delphi were still a thing too before MS Office bloated like everest and Visual Studio took over and the C++ standards commitee went nuts and Windows needed a truck of hard discs to install. Yes the world was more fun then after the blout came more monopoly behaviour and more boat and walled gardens and assaults on consumer rights and recessions and pandemics and climate change and wannabe far right dictators.
For a sense of fun you don’t actually need or want to do something yourself but seeing what other people do and knowing you could too adds to a lighter mood. If I was schoolage again there is little or nothing about the current IT landscape that would make me want to sign up.