'Windows 9' LEAK: Microsoft's playing catchup with Linux New videos of a "Windows 9" variant have emerged, and to this hack's eyes they look to have brought Windows up to speed with tricks that various desktop flavours of Unix have had for a decade or more. The feature in question is being described as “multiple desktops” and looks an awful lot like the “workspaces” that have been …
Before CDE HP-UX had them in VUE, and I'm sure they'd been around before then. Now HP's VUE is where MS got 3d effect windows from, HP & MS had agreed on making various things common in appearances and many of the keyboard shortcuts. These then became available in Windows 3. So its taken them a long time to implement the rest... like 25 or more years.
That was already done: a suit against RedHat about Virtual Desktops was dismissed when the 'prior art' was shown in court, running in an early 1980's Commodore Amiga!
But you may want to read about Xerox Rooms for Windows, (c) 1989-1992!
http://toastytech.com/guis/xrmw.html
This one is running under Windows 3.1, but I remember that it could run unde Windows/386 2.01, also.
"But you may want to read about Xerox Rooms for Windows, (c) 1989-1992!"
Very interesting! Thank you for posting!
I never heard of Xerox Rooms for Windows before. (Nor of its companion, Xerox Rooms for X Windows (sic).
http://toastytech.com/guis/xrms.html
I clicked on the first screen shot and verified, it really does say "Rooms for X WindowS" — twice! — instead of "Rooms for X Window", as it should.)
On balance I prefer the implementation of virtual desktops in PC Tools for Windows, from Central Point Software (may it rest in peace — gone, but not forgotten!).
http://toastytech.com/guis/cpdesk.html
Judging from the descriptions and the screen shots of each, and my memory of using the "Central Point Desktop" in PC Tools for Windows ("PCTW"), the latter was a more polished and fully developed implementation of virtual desktops than that in Xerox Rooms. In particular:
* Each individual desktop in PCTW was fully, easily and awesomely customizable, using menus right there in the desktop (and using right-click menus as well, if memory serves). (According to the link Anonymous Coward provided above, Xerox Rooms required use of the Windows Control Panel to change things like desktop backgrounds. Its own built-in controls were limited.)
* And PCTW provided — in my opinion — a better mini-view of all open desktops, and more and better (on balance) ways to navigate and switch among them. (You can switch desktops either by selecting a different desktop from a pull-down menu in the top right corner of the screen, or by clicking on the appropriate image in the mini-display of open desktops. I forget what that display is called, but it is fully customizable, resizable, reshapeable (e.g., in the case of 6 desktops: 1x6 horizontal, 2x3, 3x2, or 6x1 vertical), and repositionable.)
(Click on the links, read the descriptions, view the screen shots and judge for yourself, if you are interested. The fact that PCTW provided a full-blown replacement for Program Manager — whereas Rooms for Windows did not replace Program Manager at all — and the integration of PCTW's multiple desktop capability with all the other features and aspects of its replacement shell, give PCTW a huge advantage.)
But the two implementations (Xerox's and Central Point's) appear to be quite similar, and more alike than different. More proof that virtual desktops were perfectly possible under Windows 3.1 or earlier!
So why has Microsoft waited until now to offer them itself?
>>"Does it have some fanzy transitions ? Probably in 10 more years ;-)"
Ugggh. I hope not. I remember all those wibbly-wobbly windows in KDE and rotatable cube desktops. It was fun to see for about five seconds and then you turned it off.
This is a little late for me as these days I meet my needs with multi-monitor set-ups and I have 24" monitors, too. But it used to be the biggest thing I missed when going from using my *NIX box to someone's XP/Vista machine with a single 15" monitor. That felt really constraining. Nowadays it's much less of an issue. But still nice to see.
I could do without the Start Menu back. I wish MS would for once in their existence have the guts to stick with their vision despite angry internet commentators. XB1 - entirely digital with discs only as a distribution medium, share games across the entire country without ever meeting. "Nooo - we want to exchange grubby and breakable plastic and connecting once a day for five seconds even over a tethered phone is too much for us!". Simple swipe down in Metro to close an app. "Nooo! We're confused without a little minimize icon. Give it back to us! (even though it's meaningless in an environment where you switch between apps rather than a windowed environment you have to close a program out of the way)". Hit windows key and type the first few letters of the program you want / move the mouse a minimal distance to select the much larger target of grouped icons, about forty to a screen? "Noooo - we want to navigate up and down a small hierarchical menu for the dozen or so programs we commonly use. It's always been that way and should be that way forever".
MS - great ideas, backbone of a jellyfish.
We get it, that's what you like. But trust me, a lot don't.
My daughter had a new laptop last week, she tried to get used to it, she even liked some of it. In the end I installed an add on that gives her the traditional menu and/or the Metro interface.
And that is what most people do want. The choice.
Actually that is one of the problems. When MSFT say "this is the new Windows and it works in a slightly different, but much faster, way" all the WiFanBois scream "where's my start menu button?".
Henry Ford once said that when he first demonstrated the Model T a lot of people would ask "where do I put the hay in?"
Mm: "...what most people do want. The choice."
No. Impossible. You'd have to invent a check-box technology, and something like a Registry or similar data structure to store the user's preference. This is all completely impossible. Computers OSes must contain hard coded decisions, and it is impossible to offer any user options whatsoever.
Right?
</Sarcasm>
small hierarchical menu for the dozen or so programs we commonly use
So let's slap a full-screen window in their faces to they can find them better?
I don't think either solution is all that great. I think Linux has it right with the categorised menu*, rather than each company (and therefore peice of software) having their own sub-menu.
In windows, I almost always have to search in the start menu. In Linux, it's easy to click on it, when I'm in "mouse mode". I wish Windows could copy that, but I guess it's too late.
*I know you can make it happen in Windows, but you need to manually organise it. I gave up on that since Win95.
"MS - great ideas, backbone of a jellyfish."
No. Appalling ideas, arrogant until it hurts them. The customer is always right[*] - they're the ones paying for and using the damn thing. They will change whatever it is they have to in order to make more money, regardless.
Perhaps if you didn't blindly cheer Microsoft's crappy ideas so then you wouldn't need to U-turn when they do.
[*] Please don't waste my time with a Ford mis-quote.
>>"Perhaps if you didn't blindly cheer Microsoft's crappy ideas so then you wouldn't need to U-turn when they do."
I haven't U-turned in the slightest. I wrote about how I liked Start Screen when it was the way things were done, I'm writing about how I like it now when MS appear to be retreating on it. I have always been consistent except during the developer preview before I'd gotten used to it.
I am always consistent. It is MS that have changed directions which is why I now criticize them whereas before I was saying how good it was. This doesn't match up with your insults about "blindly cheerleading" at all.
And no, the customer is not "always right". Anyone who has ever worked in programming for a week knows that.
> And no, the customer is not "always right". Anyone who has ever worked in programming for a week knows that.
If you presented a prototype/beta to a customer, and they said "No, it's a piece of fucking shit, I want a bell on it" - would you insist it's the best thing ever, and force them to use and pay for it?
Would you say to your client "You're just holding it wrong, dumbass! You just don't get it, do you? I've spent months working on this and you will use it because I know better"?
When they are paying for it, you advise them, then it's up to them to take your advice. You then just give them what they want.
Anyone who's been a developer / software engineer for at least a year knows that.
The customer is NOT always right. Especially not in groups. We only need glance at the bland, boring and inane releases of the focus group driven car designs during the turn of the millennium.
At the end of the day, the final decision needs to be made by one person who is brave enough to stick to a vision. It's what saved Apple from their floundering inwardly collapsing business.
I see by your attitudes you are clearly mired in the 90s, and still hold resentment against 'Microshat' for what you deem to be his evil deeds that held down the "clearly superior" operating system, which would have OBVIOUSLY been the dominant OS of choice in business...
Except, that isn't the case, is it? Apple had Jobs (on and off, and it really shows his influence was what made the company a success), Microsoft had Gates, who, admittedly, did let the sheer scale of the monster he'd created get the better of him for a while, but in 2002, really started to turn it around.
Linux has what? Linus? Not really. While I may have respected him some time back, he's merely a self centred bully, and really doesn't have the vision or the power to pull the meandering behemoth in the right direction.
"Linux can now support 1024 CPUs!"
"Great. Will it finally work with the wifi card in my laptop?"
It's truly ironic how this article has picked up on a somewhat redundant and gimmicky feature and said "Hey, we had this for ages!" Personally, I look at the Linux GUIs as "nearly there", and "not quite". And, it's not just me. Corporations aren't stupid. There's a reason they pay gigantic licence fees for Microsoft products.
Having multiple desktops is cute, but Microsoft's Clipbook algorithms have been so far in advance of everybody else', we don't even think of it as a feature anymore. The Linux community should be collectively hanging their heads in shame. It's over a decade since I had to admit that it was the best at copy and paste out of all the OSs, and they've remained ahead of the game since.
"What do you mean, I can't right-click an image in a webpage, copy it, then flick to a remote desktop, and paste it directly into a random third party application running on a machine the other side of the world?"
Linux has a great, and brilliantly designed core, and given its royalty-free, which has allowed it to survive almost exclusively as the core that runs the Internet Of Things but guess what? Microsoft are catching up, FAST!
Linux might always have the Free thing, but as we are now seeing, the vulnerabilities in IOT devices are starting to become a real problem, and when it comes to security, Microsoft have been leading the world for quite some time...
"Linux has a great, and brilliantly designed core, and given its royalty-free, which has allowed it to survive almost exclusively as the core that runs the Internet Of Things but guess what? Microsoft are catching up, FAST!"
The freedom aspect is primarily what has made Linux successful here - the 'royalty-free' aspect is simply one of many consequences of that. It's the fact that there are no restrictions on how you can use it, and that you can modify it in any way you like. If Microsoft are going to go anywhere of consequence in the IOT space they need to radically overhaul their licensing models and really let things go. They have taken a few steps in that direction over the years but I don't think they have the cojones to take it as far as it needs to be.
"Linux might always have the Free thing, but as we are now seeing, the vulnerabilities in IOT devices are starting to become a real problem, and when it comes to security, Microsoft have been leading the world for quite some time..."
I would argue that rushed implementations and marketing trumping engineering would be the bigger factors here. There are excellent security features in both Linux and Windows but it seems that most of these problems are due to these either not being implemented properly or at all in favour of development time.
>>"The freedom aspect is primarily what has made Linux successful here"
Well no, the free (as in Beer) aspect is an enabler of success. In that if it cost a lot of money that would obviously have held it back. But it's not the reason GNU/Linux has been a great success. It's because it's very capable of doing the job. Evidence: it is free as both a server platform for websites and as a Desktop OS for end users. But it is massively more successful as the former. Ergo, whilst its low cost helps allow it to be taken up, it's the tremendous suitability that has really established it. Also, anecdotally, I work with clients who would be perfectly willing to use Windows Server and pay the costs if that's what the admins told them was needed, but the admins have said "we're putting in some new CentOS servers" and the higher-ups have simply nodded and signed off pretty much the same way they would have if the admins had said "Windows Server, sign here". I genuinely think that the cost (free as in beer) aspect of GNU/Linux is a secondary factor for its success to its actual capability and reliability. And this is coming from someone who likes Windows Server 2012 quite a lot, btw!
>>"It's the fact that there are no restrictions on how you can use it, and that you can modify it in any way you like."
Well, the GPL2 does have restrictions and they're every bit as enforceable as any other copyright-based law. Also, there are practical restrictions on what you can do. The vast majority of Open Source code works in a friendly and compatible way without forking, especially something as fundamental as the Linux kernel or the GNU tools.
"If Microsoft are going to go anywhere of consequence in the IOT space they need to radically overhaul their licensing models and really let things go."
I actually don't think that's necessary. People are willing to pay an extra for things if they like it. Technically OSX is free, but in practice you pay for it as part of buying a Mac. Microsoft suffered badly as a consequence of the Race to the Bottom that most OEMs pulled between 2000 to 2010. It took a while to come home but when it did, Macs savaged Windows laptops badly and Google found a way to monetize people, meaning they could attack the bottom end as well. MS and its OEMs only good way of fighting back against this is not to cut costs, but to raise quality. They're not going to win on price because they can't and because people are actually willing to pay for quality (laptops are being bought and kept for much longer now, as the pace of technology power is hitting Good Enough, and so seen as more of an investment worth paying for).
So changes to licencing the mobile area can help encourage OEMs, but overall, MS need to focus on high quality. Anything else is a strategic mistake, imo.
MS products frequently fail to work with modern hardware, and MS's worthless and expensive "certification" process doesn't make it any more likely that a third-party driver will actually work. MS' marvellous clipbook algorithms are copied wholesale from Gnome circa 2003....
MS are (roughly) five years behind Apple and more than ten years behind Linux and BSD. They will continue to play "catch up" for the rest of their (short) existence. Large corporations, governments and other institutions are now asking why they are blindly paying exhorbitant licence fees to MS for products that NEVER work properly - ever more of them are migrating away from Windows.....
Just remember: Microsoft have NEVER released ANY product that works properly.
Just remember: Microsoft have [sic] NEVER released ANY product that works properly."
DOS 6.0 and especially 6.2 both worked quite well. (Scandisk was rock-solid.) So did Access 2.0 (live views, and the incorporation of the Rushmore indexing technology acquired with the Foxpro purchase) and some early versions of Excel (pivot tables, etc.). And for that matter, even Multiplan was actually pretty good in its day. Microsoft tended to do good work when it faced serious competition and was not already the market leader.
There also was a lot that was good in early versions of Windows NT (4.0 and earlier — especially 3.5), and in Microsoft's contributions to OS/2. (HPFS for sure. The serialized message queue, not so much.)
But in the past 20 years . . .
H4rm0ny, I really can't tell whether that third paragraph is a joke or not, because nobody with half a brain could possibly hold that view.
Swipe down - will that scroll? Will it close and lose my work? Minimise and keep my work? How do I tell what will happen before it happens?
An icon that I can click on or poke has a thingy called a tooltip. Once I've read that tooltip, I know what it will do. (Or rather, I should).
Gestures cannot be labelled, and as they depend on context they are completely undiscoverable.
Watch this video. Labels matter.
The "half dozen" programs I commonly use pop up in the first level of the Windows 7 Start menu.
The hundred or so programs that I don't use very often but still need to have show in a hierarchical menu structure that lets me put "like with like".
I can already hit "windows" then type to search. Guess what - it simply doesn't work. It's a fundamentally bad concept because it does not match with how people think.
For a concrete example - I use IBM ClearCase. The application for it used to be called "Remote Client", and it's now called "ClearTeam Explorer". If I search for "IBM", "team" or "clearcase" it's not found.
With the old name, searching for "remote client" or "explorer" it's found, along with half a dozen other programs with almost identical names - with the old name I had several where the only difference was the icon.
How do you search for something when you do not know what it's called?
On Windows 7 I can follow the menu Programs > IBM Rational ClearCase > and bingo!
At home I can search for a fork - kitchen > cutlery drawer > bingo!
>>"H4rm0ny, I really can't tell whether that third paragraph is a joke or not, because nobody with half a brain could possibly hold that view."
I make a joke here about one day a year. Today is not that day.
>>"Swipe down - will that scroll? Will it close and lose my work? Minimise and keep my work? How do I tell what will happen before it happens?"
You can tell because it does the same thing as last time, and every time. Might as well ask how you would know what the little flat line is supposed to do if you've never clicked on it before. Swipe down close, swipe from left to switch between apps. Or just Alt+Tab if using a keyboard. Very easy, no need to clutter screen with window bars or icons - more screen real estate.
>>"Gestures cannot be labelled, and as they depend on context they are completely undiscoverable."
There are four directions you can swipe. That's pretty easy and once learned, you know. There are a dozen things just as unknown to you about your current OS but which you assume are obvious because you've grown used to them. Where is the tooltip on your double-click?
>>"The "half dozen" programs I commonly use pop up in the first level of the Windows 7 Start menu."
Start Screen holds five or more times that without having to resort to nesting or scrolling. Thus is better for anyone who uses forty or less programs regularly.
>>"The hundred or so programs that I don't use very often but still need to have show in a hierarchical menu structure that lets me put "like with like"."
They are still easily findable with the Start Screen either by scrolling down to the full list or simply typing. It is far better to optimize for the 90% of the time than the 10% of the time, so all those "hundred" programs (seriously) aren't cluttering up your normal usage. I'm a power user and I use about twenty programs routinely, and that's significantly more than most people. So why make people hunt for them in a menu with a small target area?
>>"I can already hit "windows" then type to search. Guess what - it simply doesn't work"
It does. I do it all day long. Win key, "ex" and return, I'm in Excel. Four key strokes quickly entered in less time than it would take to reach the mouse.
>>"How do you search for something when you do not know what it's called?"
In that minority case, you scroll down to the full list of programs and read.
>>"How did you "discover" these gestures? By clicking your mouse at random until it did something? Did you have to google it? RTFM?"
How did you discover double-clicking did something? How did you discover holding down the mouse button on a window bar and dragging moved the window around?
Honestly, you know you can swipe / drag from the side, I do, everybody here does. But I forget - the attitude on El Reg that we're all special people with great technological gifts. Perish the thought that Ordinary People could learn to do this quickly and easily.
How did you discover double-clicking did something? How did you discover holding down the mouse button on a window bar and dragging moved the window around?
Sigh, strawman...
In program manager, I clicked it once - it didn't work. Clicked twice... tada! Dragging the title bar for some reason was obvious to me. I had never seen or used a GUI before, but I read that windows can be moved about and sized - figuring out how to do it wasn't difficult.
Now, mouse/finger gestures just aren't obvious, unless you randomly drag/stroke until something happens? They might be cool for extra functionality, or when there's already a well established method to perform the same task.
Perhaps I'm just thicker than the average Windows grunt but even though, until recently, I've been developing GUI apps since 1992/3 (for Win3.0) I couldn't figure out how to get rid of metro apps (or shutdown) without slamming every key until something happened. Yep, I've been out-smarted by Windows.
If my software required some special knowledge for basic operations, it would get uninstalled and they'd move on to competing products.
Making swiping/swirling gestures was tried in web browsers in the early naughties I believe. It failed because it's non-intuitive. Why ANYONE would think it;'s a good way to do basic tasks in an OS, I have no idea.
And please don't assume anyone's using a touchscreen, they failed for desktop computing in the 90s because of 'gorilla arm', which is why they'll fail now
Double-clicking and dragging windows around with the mouse have been pretty universal GUI interfaces since the 1980's when mouses came into general use. I am pretty sure they were documented in the printed manual that came with the Windows floppy disks that you bought at egghead back then as well. Sort of like knowing how how to dial a telephone number. What, what, phones don't have dials anymore? How do I call someone?
By contract, these "Gestures" are not universal, not obvious, nor is there a nice printed manual that came with Windows to describe them.
>>"Double-clicking and dragging windows around with the mouse have been pretty universal GUI interfaces since the 1980's when mouses came into general use"
Not what I wrote. My point was that double-clicking is not inherently discoverable. Most mouse use isn't. My point is that they're known through familiarity not their obviousness and therefore supposed lack of discoverability of the active sides in Windows 8 can't be dismissed for a reason that is common to both.
Also, it's "mices". ;)
I don't know where you learned about user interface design, but they taught you some bad things.
Microsoft has pretty obviously mislaid any good interface designers they had, and you can tell that by the way the UI changes with every release. Hell, the UIs aren't even consistent among Office products (Visio is a perfect example). Yes, I realize Visio was purchased in 2000 from Visio, Inc., but you'd think 14 years would be long enough to get its UI into conformance with Microsoft's nominal "standards" (which they seem to revise way too frequently).
UIs should be simple, obvious and unchanging. You can always add optional features, of course, but the basic UI shouldn't change. And when adding features, DO NOT enable them by default. Because many of your users will be confused. Those of us who use computers in our daily work, don't like to have to waste time learning a new UI with every OS release. We're trying to use a tool, not enhance our Windows experience. We most certainly don't like to play hide-and-seek with gestures, until we find the one that does what we want.
OK. Rant over.
"Swipe down close, swipe from left to switch between apps. Or just Alt+Tab if using a keyboard. Very easy, no need to clutter screen with window bars or icons - more screen real estate."
Invisible UI is very obviously NOT GOOD.
What's the thought process here--"hey, there isn't any indication of what's going on with the system or what you can or should do with the software or how to do it, but we've reclaimed 2% of the screen real estate!" Has there ever been a case in software design priorities being so completely backwards?!
I have Windows 8 on my HTPC. I mostly just use it to run a web browser (Chrome) in the Desktop mode. I try to use my mouse as I normally would. I would say about 2-3 times per day, my mouse movements get misinterpreted as swipes and cause things to happen that are completely foreign and undesirable to me. I get UI bars inexplicably appearing from all different directions and I have to figure out how to make them go away before I can continue doing what I wanted. I don't see how anyone could justify this UI design as good.
It's pretty easy to turn off many of the mouse/swipe movements in the UI if you dig a little. Windows 9 is said to be more aware of the computer's capabilities as well. A computer with no touch and only a mouse should be recognized and behave differently than with those features.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
