How Microsoft shattered Gnome's unity with Windows 95

There never will be a year when Linux conquers the desktop, because desktop computers are going to merge into tablet-style touch-driven devices and disappear. But desktop Linux was getting close, until Microsoft derailed it a few years back. The GNOME project’s recent release, GNOME 3.8, served to remind me of the significance …

[Richard Wharram]

To be fair to MS...

I think the Linux devs were only too happy to create a series of desktops. Without a chain of command led by a control-freak, the people who make Linux desktops will happily split off into groups with different visions. That's often a good thing as well as a bad one.

[Erwin Hofmann]

You are mistaken …

… Linux dev's (Kernel developers) are rarely concerned with a GUI (graphical user interface) and happily prefer the command line over any GUI … and, very important, Linux and GUI's are not the same. Linux-GUI's are only playgrounds, using the most widely used operating system in the world (Linux) to emulate and/or excel know GUI's like Windows, Android or Mac OSX. Oh, and another mistake (in the article) is the assumption Microsoft (Windows 95) invented the Taskbar. As I remember (being that old) Arthur 1.20 (later RISC OS), an Operating system from 1987, had that feature years earlier, called "the icon bar". Sure, Microsoft invented the "Start Button" ... wow … but that was merely an exercise to declutter the "the icon bar" (er, sorry, "the task bar") …

[Richard Wharram]

Re: You are mistaken …

I was using the term 'Linux devs' in a generic, non-anal sense obviously :)

[J.G.Harston]

Re: You are mistaken …

And the RISC OS desktop had a application launcher (Start Button-alike) if you chose to install something like Director. And what do you mean by "window switching". You do that by clicking on the window's title bar to bring it to the front, or just click in it's edit area to give it the input focus without the brain-dead MSWin lobotomy of bringing the whole window with it.

["Dead Eye"]

Re: You are mistaken …

Actually, RISCOS (the Acorn one, that is) had all of:

1) the "Icon Bar" with two types of icon that worked in from both ends

2) Window "Title Bars" with icons in them to expand and collapse the window, &c

3) A "Start" buttin, called the"Apps" icon (on the icon bar) that contained applications is the ROM

as a starting point but also applications from the boot disc that were added to this set.

This last bit is the one that differed a nit form Window because it wasn't at either end of the icon bar, and it didn't support a menu structure for the applications -- but it was there.

[DanDanDan]

Re: To be fair to MS...

This article is saying much the same as I said in the comments to an article back in February:

"Unfortunately, as well as eating itself, it also splits and forks itself. Web servers has always been a FOSS stronghold. Everywhere else that counts is full of crap competition: KDE vs Gnome - developer show-offs leading to no winners. LibreOffice vs OpenOffice - no winners due to diluted development.

Yes it's one of the best things that you can fork a project if it heads in the wrong direction, but it also dilutes development effort trying to do 10 things at once."

That comment received 6 thumbs up and 16 down. Diluting the developer effort is, according to this article, Microsoft's strategy. Funny that.

[Wayland Sothcott 1]

Re: To be fair to MS...

I don't see that Linux has to be number one. It's actually doing very well because of all the variety. However cloning stuff from Microsoft is simply following a company who makes good popular software. It's not innovating and producing something better. Honestly by now there should be one document format that can be worked on by one application which combines all the features of MS Office. Linux is not going to produce this first whilst it's Office is a copy of MS Office.

[1Rafayal]

Re: To be fair to MS...

Ohh, you are going to get a lot of downvotes for that one....

[JEDIDIAH]

You've got to be kidding.

Most of the basic UI elements from Win95 were not invented by Microsoft. They were cobbled together from a wide range of sources including CDE and other ancient window managers. A lot of the apparent missing prior art is MIA primarily because the author refused to really acknowledge the state of X before 1995.

The the first "clone" of Windows95 was nothing more than a theme for an existing window manager (fvwm).

A start menu is nothing more than an anchored app menu from any of the early window managers.

[TheOtherHobbes]

Re: To be fair to MS...

What's depressing is that there's been very little OS development since almost forever.

Apparently no one can imagine anything more original than a desktop with menus and icons over a pre-emptive multitasking system based on ideas from the 1960s.

And then you get Win 8, which takes half of that and breaks it.

Forking doesn't make something interesting if all you get are twenty variations on the same ideas.

What would a completely original take on interface design, networked file system design and data type sharing, Internet integration and distributed processing look like?

Nothing like Linux, OS X, or Windows, I'd bet.

[Richard Plinston]

Re: To be fair to MS...

> LibreOffice vs OpenOffice - no winners due to diluted development.

Actually not. Developments in either could be included in the other. They can be somewhat competitive in trying to outdo each other and that is a good thing, but in the end the code is available.

Same with desktop environments. It doesn't matter which one I use because applications will run in any of them (given I load the required libraries from the distro).

With, say, MacOS vs. Windows the choice must be made and stuck with. With KDE, Mint, LXDE, I can switch between them and not worry.

[Erwin Hofmann]

You are mistaken …

... how can someone give a "thumb down" for this post (except for the spelling mistake in: "known") ... seriously, what's wrong with it. There is noting emotional, non factual or irrational in it.

And for the endless pro and contra topic of variety, which is exactly the strength of Linux and its distributions (like it, or hate it) ... Linux makes a good operating system for your telly, the space station, mobile phones and pads, the Google search infrastructure as well as your desktop etc. etc. ... and if you don't like it, buy a Mac ... fair enough ...

[fishdog]

My head hurts ...

[Matt Bryant]

Re: fishdog

"My head hurts ..." I gave up trying to track all the forks in the Linux desktop years ago, so I can only sympathise with yourself and applaud Mr Poven's efforts! The only suggestion I can make is fuggedaboudit for today, it's actually sunny here in the UK, so I shall be scheduling a "team meeting" down the local watering hole instead.

[h3]

Windows 95

Windows 95 broke one of the most useful conventions. (Being able to type in a window without it being on top).

The RiscOS / X11 ways of doing it are different but still just as useful.

The Xmouse powertoy was never very good. (Funny thing is the reg hack for Windows 7 / 8 works really well.)

Don't think a task bar is that necessary. The middle mouse button is pretty much never used in Windows so using that to list the open programs would work fine.

If the start button never existed people would just hit the windows key (Or use ctrl-escape on an older keyboard).

[Philip Lewis]

Re: Windows 95

.. and introduced the abominable and terminally braindead concept of windows suddenly taking context and popping up at inopportune times. Keep typing and the next thing you know you have hit the "Y" key and something quite bad has happened.

[Ian 55]

Re: Windows 95

Is being able to type into 'beneath' windows that useful?

Here (MATE, but also with most other window managers, including MS's) it's easy to see what window will get my keystrokes - it's the one on top (and with a different coloured outline, the way I have it set up). It is trivial to give another window focus.

I'd hate to have start looking for where a mouse was before knowing what would happen when I started typing.

[Vic]

Re: Windows 95

> Is being able to type into 'beneath' windows that useful?

Yes. Massively so.

It allows you to control machines without having to dedicate much screen real estate to the control window (which, at that point, you're not that interested in, so long as you can get a command to it). That leaves your screen displaying other bits of the system that you *do* need to monitor closely.

> it's easy to see what window will get my keystrokes - it's the one on top

If that's the way you want to work, then that's just fine. Choice is good. Imposing decisions on other people is bad...

Vic.

[Bill the Sys Admin]

Re: I'm not sure Microsoft *has* won.

MATE is no where near stable enough to be appearing in RHEL. They still use a solid version of Gnome2 in RHEL. Why would you bother with MATE which is very new and unstable if you ask me. Certainly not suited to a enterprise environment currently.

[Random K]

Re: Bill the Sys Admin

Have an upvote on me. RHEL/CentOS with any GUI is bordering on silly. Having a new and largely untested desktop environment is just outright craziness. I would venture that MATE would stand a decent chance of fizzling out before Red Hat got through with the necessary evaluations. That said, in 3-5 years who knows. RHEL turns slowly but it does move.

[AJ MacLeod]

Re: Bill the Sys Admin

CentOS with a GUI is anything but silly - it provides a stable desktop setup designed with system-wide configuration in mind and is supported for a very decent length of time after release (which almost nothing else is.)

Funnily enough that's exactly what most businesses need and want, not stupid and almost entirely useless desktop effects, not the very latest version of this or that application. CentOS 6 is pretty close to ideal for this use case, not silly at all.

[h3]

Re: Bill the Sys Admin

So what if you are using any of the Applications that are GUI apps and really expensive.

Normally you can have them on Solaris / RHEL / Suse Enterprise.

(Seems like more and more of it prefers RHEL as well. Used to be mainly on real UNIX).

(i.e the people who the Nvidia Linux driver is actually written for : 3d / CAD / CAM / EDA)

Forget CentOS that should be no concern to Redhat.

Those type of apps (I know of EDA) are like £125000 a seat and they are run on fully paid for and supported RHEL. (And use the most expensive Nvidia cards). Wasting 50% of the 3d performance isn't going to sit well.

I would say that is pretty much all the people who pay for RHEL at a decent price who actually use it as a desktop.

They are going to have a new and largely untested environment gnome 3 and annoy massively the paid desktop users.

That combined with Oracle licensing being slightly more reasonable using Oracle's Linux variant.

So what are people running on RHEL headless ? (Not Oracle - DB2 might as well have it on AIX).

For decent support from Redhat you will need to be giving them lots of money.

No great advantage over CentOS unless there is something you are running on it that needs to be run on RHEL to not invalidate the support contract on the 3rd party app.

I think Redhat might have realised this a bit so we don't even have a RHEL7 beta yet.

[Michael Wojcik]

Re: Bill the Sys Admin

I would say that is pretty much all the people who pay for RHEL at a decent price who actually use it as a desktop.

And I would say you're wrong. We have plenty of customers who run RHEL headless, typically doing back-office processing, running migrated mainframe applications, etc. Hell, my teams run a number of headless RHEL VMs just for CI and testing.

And since all of the RHEL GUIs are X11 window managers, it's entirely possible to run RHEL headless and use X11 servers running on other machines for the UI. You can even use Cygwin-X or one of the other Windows X11 server implementations if you like.

[Yet Another Commentard]

Re: I'm not sure Microsoft *has* won.

I think the premise of the article was that we have reached "peak desktop", so a "victory in the desktop" would be hollow as it will be a declining market as we head towards slates and other non-desktop things. In that market, as things stand a form of Linux is very, very likely to "win".

You may disagree with that premise of course

[Wayland Sothcott 1]

Re: I'm not sure Microsoft *has* won.

Like most people I using various computing platforms. Certainly the smartphone is the most available since you don't need a desk or even a lap. However if you want to do some serious sit down work then you want a big desk and a big screen or screens, a proper keyboard and mouse or something even better and loads of computer power and storage and a decent printer. There will always be an advantage in having a better tool for the job than your competitors. If you can sit down to do the job then sit in front of the most powerful machine you can afford, not some dinky hand held.

[itzman]

Re: I'm not sure Microsoft *has* won.

Mate is stable enough here. Its a lot more stable than W95 ever was :-)

[Anonymous Coward]

Re: I'm not sure Microsoft *has* won.

> MATE is no where near stable enough to be appearing in RHEL.

Unstable? WTF?

I've been using Mint/MATE practically since it came out both at home and at work (as a systems programmer).

So far, I've had not a single problem.

What stability problems?

[eulampios]

Re: I'm not sure Microsoft *has* won.

Mate seems to be as stable as Gnome2x has ever been, if not more. Try it out for yourself.

[JEDIDIAH]

Re: I'm not sure Microsoft *has* won.

MATE not stable? You could have fooled me. Been using it since I upgraded all of my Ubuntu boxes to 12.04.

"Stability" is not something I would knock it for.

[TheVogon]

Re: I'm not sure Microsoft *has* won.

There is nothing much wrong with Windows 8 - it is rock solid and faster in key benchmarks like graphics and large file transfers than any other desktop OS.

The few things that some of the public are finding hard to cope with like no start button, and having to start in Metro mode are already fixed in V8.1 - public preview due out later this month!

[Dave 126]

Re: I'm not sure Microsoft *has* won.

>Windows 8 is a toy.

but some of the applications that run it are not.

[Anonymous Coward]

Re: I'm not sure Microsoft *has* won.

">Windows 8 is a toy.

>Windows 8 is a toy.

but some of the applications that run it are not.."

If by that you actually mean "but some of the applications that run on it are not" that equally well applies to Windows 7.

[RISC OS]

Re: I'm not sure Microsoft *has* won.

I use windows 8 for work, and there is nothing wrong with it.

[Trevor_Pott]

@RISC OS

Then you have an easy job. The kind robots will be doing soon.

[Trevor_Pott]

Re: I'm not sure Microsoft *has* won.

8.11 for Workgroups does not fix any bloody issues. A Start button that brings up the Start Screen? *bzzzzzzt* Wrong answer! Explorer (and so much else) still has Ribbon bars? *Bzzzzzzzzt* I could go on - at length - but 8.11 for Workgroups doesn't actually address any of the concerns that the general public raised. It was a shitty token gesture designed to seem like outreach without doing a goddamned thing to change the real issues.

8.11 for Workgroups is Microsoft's way of "doing something" that is in fact nothing so that they can get on their horse a month later and scream "but we did what you want!" They'll claim "persecution" and will start a P.R. war whereby they blame their opponents (Google, Amazon, etc) for "fighting dirty" by funding (or arranging airtime for) people who continue to highlight legitimate grievances with Windows 8, or the general "trustability" of Microsoft.

8.11 for Workgroups is a mirage. A handwave to befuddle the gullible and give them justification for a protracted campaign aimed at silencing dissent. Microsoft has thrown power users under a bus and done so on purpose. They've done it for the same reasons Apple has. It will come back to bit both of them in the ass in short order; on that day, I will give out free popcorn. Until then, well, Windows 7 doesn't end support until 2020 and Cinnamon works just fine for me...

[Matt Bryant]

Re: Trevor Pott Re: I'm not sure Microsoft *has* won.

".....Microsoft has thrown power users under a bus....." Yes, and the market for real "power users" is how big compared to the masses of average desktop lusers and home users....? I used to hear the same shrieking from self-proclaimed "gurus" with win95 and when Linux and the various commercial UNIX flavours got desktops - "REAL users do it on the command line!" Seriously, there was very little chance Redmond or the FOSS brigade could make one OS interface to please all users.

[Roland6]

Re: Trevor Pott I'm not sure Microsoft *has* won.

"and the market for real "power users" is how big compared to the masses of average desktop lusers and home users....?"

Whilst I know it has been trendy for many years now to focus on mass market share and/or total numbers of users, we shouldn't forget that companies such as MS still need to make money.

I suspect that whilst the OEM consumer market sales account for the biggest in number, I suspect that the enterprise market accounts for a significant proportion of revenue - remember enterprise licensing is an annual license and not a for life license. So if you want to run those old systems on XP and W2K VM's you still need to keep paying the annual fee...

The question is whether MS can make the right noises to pacify the enterprise market - remember they've got until 2020 (when Win7 extended support ends)...

[Anonymous Coward]

Re: Trevor Pott I'm not sure Microsoft *has* won.

> Yes, and the market for real "power users" is how big compared to the masses of average desktop lusers and home users....?

That "masses of average desktop users" is shrinking rapidly to swell the coffers of Apple and Samsung. Didn't you get the memo?

As a proportion of desktop users, the "power" user is becoming more and more significant.

Which is one of the reasons why a lot of people are wondering why Microsoft are in such a hurry to foist a touch screen mobile experience onto desktop users that decidedly don't want it.

[JEDIDIAH]

Re: Trevor Pott I'm not sure Microsoft *has* won.

> and the market for real "power users" is how big compared to the masses of average desktop lusers and home user

The IBM PC clone is turning back into what it originally was: a business machine. That may not mean "power users" but it does mean people that need to get stuff done. Given the number of PCs in corporations, that is no trivial number.

As the "toaster" users flee to tablets, the relative share of power users will INCREASE among PC users rather than go down.

[Jess]

the market for real "power users" is how big compared to the masses

If we eliminate those who can fulfill their needs with an iPad or similar, pretty significant, I'd say. (Especially since those have actually started to wise up to the fact, in their purchasing decisions.)

[Wayland Sothcott 1]

Re: I'm not sure Microsoft *has* won.

It does seem that Microsoft has thrown power users under the bus. I can't get on with Win8 and my stepfather gave his new laptop to my son and getting an Android machine instead. I still wonder if I just need more time to get good at Windows 8. You say they have done this deliberately but for what purpose?

When Labour appointed the wrong Milliband bro I saw parallels with Microsoft and Windows Vista and again with Win8. Some sort of deliberate concession to your competitors.

[craigj]

Re: I'm not sure Microsoft *has* won.

"8.11 for Workgroups"

ROFL - You are so clever. That was such a good joke, you felt it needed saying 4 times!

[Ottman001]

Re: I'm not sure Microsoft *has* won.

Can't help noticing that Richto, the formerly dominant pro-MS shill on these forums hasn't posted since TheVogon started posting in January. Both always use the "Eat This" icon and both have a similarly skewed pro-MS take on reality. Coincidence?

[Roland6]

Re: I'm not sure Microsoft *has* won.

"There is nothing much wrong with Windows 8"

Well having installed Classic Shell and ModernMix, plus done some additional configuration, I would tend to agree the Win8 desktop is very similar to Win7 albeit at times some weirdness does occur eg. I installed a Flip camera last night and Win7 didn't need a reboot whereas Win8 did...

[Richard Plinston]

Re: I'm not sure Microsoft *has* won.

> The few things that some of the public are finding hard to cope with like no start button, and having to start in Metro mode are already fixed in V8.1 - public preview due out later this month!

It isn't actually the 'start button' that people complained about the lack of, it was the _consequences_ of the start button, ie the start menu.

8.1 may have a button now, but it merely drops one into Metro, which is the cause of the problem. They want to avoid Metro and want the menu back.

MS on the other hand want to force Metro down users throats until they love it - or die!

[Shagbag]

Microsoft *has* NOT won. AT ALL.

The article is disappointing for its lack of balance. It's framed on the underlying assumption that MSFT's patents have some legal basis. This is fundamentally wrong. It is up to the patent holder to prove their patents, not for the accused to dissprove them.

The analysis should be one of "tell us which patents you allege have been infringed and then we can start talking about them".

It is an undisputable, historical fact that MSFT HAS NEVER

(a) publicly detailed the patents it alleges have been infringed; and

(b) never disclosed the terms of the so-called 'Patent Agreements' it has signed with Novel, et. al.

Defending a spurious claim involves costs - both in legal fees and management's required attention. It is an equally plausible argument that the reason by Novel, et. al. signed the agreements with MSFT is because the 'protection money' they pay to MSFT (if they do at all- we don't know as no-one has seen the agreements) is much less that the expensive US legal bill that would have to be carried through to settlement. With out knowing the content of these 'Patent Agreements' is CANNOT be said that they give support to MSFT's claims and it is a complicit media that does not point this out.

IT CAN ONLY BE DEDUCED that the reason MSFT has consistently failed to detail anything (other than the total number) about its patents is because even they do not believe they can be enforced - either because (a) they have no basis in law, or (b) they are trivial such that they can be side-stepped/worked around with some non-significant code adjustments.

The article fails to directly mention any of this. And for that reason it gets an....

... EPIC FAIL!

[Anonymous Coward]

Re: Microsoft *has* NOT won. AT ALL.

I agree with Mr. Shagbag.

Does the article have any basis for intimating that there is a cause/effect between GNOME's evolution and Microsoft's deals with some "Linux" companies? Or is it merely speculating on an apparent coincidence? This needs to be made clear.

[Liam Proven]

Re: Microsoft *has* NOT won. AT ALL.

Go on then. You offer a better explanation.

[a cynic writes...]

Re: Microsoft *has* NOT won. AT ALL.

It's framed on the underlying assumption that MSFT's patents have some legal basis.

No it's not. It's framed on the underlying assumption that veiled hints about litigation to come were enough for GUI projects who might be at risk to cover their arses.

MSFT didn't need to prove its case - in fact it didn't even need to go to court. It just needed people to believe it could and that litigation costs would be significant. There's not a lot of point fighting if the best you're going to manage is a pyrrhic victory.

[Liam Proven]

Re: Microsoft *has* NOT won. AT ALL.

@Shagbag: either you didn't read the whole article or you didn't understand it.

Its points, simplified enough for you to maybe have a chance of understanding them:

[1] It never came to court so we don't *know* the details. Therefore, we have to try to *deduce* them.

[2] So, what is in common between "the Linux desktop" (of the time) and Windows? I list a detailed selection of features.

[3] I then point out, with examples, that these are *all* derived from Win95. The existing similar concepts are too different.

[4] I then point out that after the legal threats, the KDE team caved & signed; the GNOME side didn't, then ripped & replaced their Win95-like features.

Sorry if this was too complex for you. Try Computer Active instead.

[fandom]

Re: Microsoft *has* NOT won. AT ALL.

"[1] It never came to court so we don't *know* the details"

Suing Tom Tom and Barnes & Noble doesn't count?

"[4] I then point out that after the legal threats, the KDE team caved & signed"

The KDE team? Are you sure?

"Sorry if this was too complex for you. Try Computer Active instead."

But you are good at self-deprecating humor, that's something, I guess

[InsaneGeek]

Re: Microsoft *has* NOT won. AT ALL.

As a unix user/admin for multiple decades you are completely ignorant of reality. Really look around at the patent fights that are going on, anybody who believes Linux, etc have no infringing patents is just plain stupid. Seriously just look at the things others are getting sued over and now tell me that there is nothing in Linux that infringes. From stupid bounce-back patents to online shopping carts to how music is sorted... heck just last week bunch of podcasters were sued for infringing patents.

The belief that MS has no case is simply void of any actual critical thinking and is more a delusional fantasy... seriously put the smallest amount of mental effort behind it and think for a second.

[David Hicks]

Re: Microsoft *has* NOT won. AT ALL.

Most of the patents are likely trivial and trivially worked around, which is why a list of them is never made available.

It's a big game of bluffing and legal fees.

[Shagbag]

Re: Microsoft *has* NOT won. AT ALL.

The WHOLE POINT is that because no-one (except MSFT) knows the details of the 235 (IIRC) patents any speculation about what may or may not infringe, and how easy or hard they may be fixed, is entirely that: SPECULATION.

To state in any way that MSFT's assertions have validity, has the same basis in fact as any statement that MSFT's assertions have no validity. Neither position can be proved or disproved until the substance of the claims are open to examination. Which they, currently, are not.

To write an opinion piece that gives undue prominence to one side of the argument is not right and that's why the article is a FAIL. EPIC for effect.

[Don Jefe]

Re: Microsoft *has* NOT won. AT ALL.

MS is not the only one who knows, that is not how business works. With secrets and IP a lot of times detailed information is kept strictly between counsel who assess the situation and advise the CEO and senior management about likely legal outcomes. The CEO will not know the details but will make a decision to settle or not based on the legal teams review. The confidentiality agreements are exceptionally detailed and restrictive so it is unlikely we will ever know the specifics behind what is a private affair.

This is not an MS thing or even an IT industry thing, not showing your hand to the general public is a strategy both sides participate in throughout the defense, pharma, petrochemical, and transportation sectors (and othets as well I would assume). There's really no reason to bash MS for not disclosing details. At the end of the day it is a private affair between two parties and is also common practice.

[Shagbag]

There's really no reason to bash MS for not disclosing details

WTF?

What planet are you on?

Are you a MSFT apologist?

Do they pay your salary?

Is your real name Florian Mueller?

MSFT made these allegations in the public domain. There is EVERY reason to call bullshit on them. The fact that they refuse to put any substance behind them (either providing any details, or initiating legal action) is a the clearest indication that the allegations are groundless.

MSFT's allegations that Linux infringes MSFT's patents is just another case of The Emperor's New Clothes.

[Don Jefe]

Re: There's really no reason to bash MS for not disclosing details

No. I'm a business professional who obviously knows significantly more than yourself about how IP really works. You are acting like a petulant teenager, railing against things you do not understand. You are citing a children's story in a discussion about complex property issues that you don't like. That's not much of a basis for anything, really.

[JEDIDIAH]

Re: There's really no reason to bash MS for not disclosing details

Ok Mr "I'm a business professional". Explain how defamation is not an actionable tort. Microsoft has made some serious accusations without basis in an attempt to damage the credibility and business of competitors. There's probably a specific Lanham Act cause of action in there.

Talking trash is a dangerous business in the real world where adults live.

Microsoft benefits from the fact that it is the industry bully and it's defaming a loose collection small entities none of which really have the clout or resources to mount a legal counteroffensive.

[Don Jefe]

Re: There's really no reason to bash MS for not disclosing details

Then they should act on it. This is the real world, where adults live and if you have been wronged there are avenues for redress. You're spewing probably and maybe and should. That's not the way it works Mr. Basement Legal Expert.

[JEDIDIAH]

Re: Microsoft *has* NOT won. AT ALL.

> anybody who believes Linux, etc have no infringing patents is just plain stupid.

No trivial device or software is free of problems with patent trolls.

You can get shaken down just for using your multi-function printer.

This is a much wider problem than the anti-Linux FUD you're trying to propagate there.

[John Smith 19]

What's with this "Apple invented" BS?

The PARC Alto "invented" most of this stuff. They recruited a bunch of HS kids as testers, some of whom went on to work for Apple (perhaps lured by the siren call of Jobby).

My jacket would be the one with a copy of Newman and Sproull in one pocket and "ALTO: A personal computer" in the other.

[wolfetone]

Re: What's with this "Apple invented" BS?

Sure didn't Steve Jobs tell Microsoft in the 1980's that both of them had been to Xerox and saw the Alto, when Microsoft were bitching about the use of Windows on the Mac GUI?

[Liam Proven]

Re: What's with this "Apple invented" BS?

You don't know your history.

I have specifically addressed this, with pictures for the hard of thinking.

Pre-Apple: no menu bar; no title bar widgets (often no title bars); no standard dialogs; no filer windows; no folder icons with icons for files inside them; no standard buttons (e.g. Yes/No, OK/Cancel); and enough more to fill another article.

All this stuff came in with the Lisa. Not the Mac, the Lisa.

You prove just two points: firstly, your ignorance of early GUI design and secondly, your bias against one company.

[Roland6]

Re: What's with this "Apple invented" BS? @Liam Proven

"You don't know your history."

Whilst I get the graphical parentage, I suspect we need to look at the CRT parentage where two schools of thought were at work: Firstly forms, which tended to put stuff across the top of the screen and had drop downs and secondly the keyboard/function key mapping, which tended to put the soft-functions at the bottom of the screen.

So whilst I would tend to agree about Win95 being the source in terms of GUI, in terms of layout design things do go back much further.

[John Lilburne]

Re: What's with this "Apple invented" BS?

You are forgetting the XEROX STAR which had menus and buttons and all the rest. Apple licensed a whole load of the star technology for the Lisa.

http://www-sul.stanford.edu/mac/primary/docs/starmac.html

But mostly the earlier stuff which you have shots off did things differently you manipulated the objects by dragging one over the other rather than by menu commands. But still the smalltalk environment had menus they were context sensitive popups rather than pull-downs. It just that knowing what application a window was part of, and the commands weren't that important. The whole point was that you dragged one thing to another thing and the system took care of it.

[Liam Proven]

Re: RISC OS and applications

Indeed; I have it running here and if you'd been paying attention you'd have noticed that I have written, well, pretty much all the Reg's RISC OS content for about the last 5Y now.

*However* this is not prior art:

[1] all the Apps icon does is open a folder;

[2] it's not a menu;

[3] it's not hierarchical;

[4] it's not extensible (AFAICR).

[J.G.Harston]

Re: RISC OS and applications

"[1] all the Apps icon does is open a folder; [2] it's not a menu;"

So? If you want a menu-ing Apps, use something like Directory

"[3] it's not hierarchical;"

Yes it is. As just another directory, it can contain more directories, just like any other directory, such as the directories that already are in it.

"[4] it's not extensible (AFAICR)."

Yes it is. *AddToApps <source name> <apps pathname>

[Ben Tasker]

There never will be a year when Linux conquers the desktop, because desktop computers are going to merge into tablet-style touch-driven devices and disappear.

Bollocks are they. They may not take as primary a role in most people's lives, but they are a long way off dissapearing, there are tasks that just can't be done efficiently on a touch-screen device.

Right rant over, I'll read the rest of the story now!

[Anonymous Coward]

The thought of coding on a touch driven device actually scares me.

[swampdog]

"The thought of coding on a touch driven device actually scares me."

Even simple file creation would be a nightmare. Ordinarily one would type..

$ touch balls

$ touch nads

..and so forth. Just what are people going to think if they see me feeling up a pair of animated testicles onscreen. Moreover, how would you tell the difference between the nads & balls. Wouldn't want to apply vim onto the wrong pair. Could be painful.

Hmm, now I'm thinking back to the days of hairy bison. I really do *not* want to be seeing what sex that bison was. Why has this popped into my head? Oh yes - I've got far too much work to do today so am solving the problem by doing none of it.

[Anonymous Coward]

« The thought of coding on a touch driven device actually scares me. »

It's not so bad actually, I've been doing that for a number of years.

Next time in the computer shop look for a section labelled "keyboards" and give one of them things a try. I tell you, they beat punched cards hands down.

Hope this helps.

[wolfetone]

What's the bloody point in doing that when a PC or a laptop can do that from the word go?

[Anonymous Coward]

Re really twattish touch driven device joke.

Aha. Aha. A ha ha ha. Aha ha ha ha ha. Ha. Aha.

Now kindly fuck off.

[Terry 6]

Actually, I used punch cards t code, as a kid. I kind of liked them.

[BobChip]

Desktops will never disappear

While I have no doubt that any number of small devices such as phones and tablets will be used to CONSUME internet content, they will not be used to produce it. Productive work will continue to be done on desktop workstations with large, non-touch monitors using a mouse or tablet.

Take CAD as an example. Would you rather fly in an aeroplane designed on a large screen where the designer can see the big picture and has precise control over what components go where, or one designed on a tablet with components shuffled by someone's fingers and thumbs?

Microsoft may have abandoned the workstation, but it remains the basic tool on which PRODUCTIVE WORK is done. Fortunately, there are alternatives.

[Number6]

Re: Desktops will never disappear

The danger is that if desktop machines become significantly less popular, the cost will increase and they'll become even less popular compared to tablets, and so on. We'll end up back in the bad old days where companies buy posh, expensive workstations for those who need them and the rest of us will end up with a tablet in a dock, possibly with a keyboard and mouse attached, and perhaps even a second monitor.

I much prefer a keyboard, mouse and monitor to a touch screen with finger marks all over it.

[Babbit55]

Re: Desktops will never disappear

Desktops will be around for a long time still, it is laptops that tablets will replace. The desktop will still be a main work device/ proper gaming device, with your "Lapblet being the portable consumption device.

[Gav]

Re: Desktops will never disappear

There's no point in looking at things as a either/or scenario. Tablets will not replace laptops or desktops. All that is going to happen is that touch screens will become an addition input device.

Sometimes they'll be on tablets and held horizontally, sometimes they will be on desktops and be used vertically. Users will continue to use whatever input device is most convenient for the task and circumstances. That includes keyboards and mice.

I've already caught myself using a "touch screen" where one does not in fact exist, (on a camera display) simply because it seemed to be the most natural thing to use at the time. In the near future it will be normal switch between them as suits.

One of Windows 8 failings is it has jumped to touch screen use before most people were used to the idea and, more damningly, at the expense of the other input devices. The "Start Screen", for instance, is not an efficient way to access things by either keyboard or mouse, but perfect for touch screen. So what Microsoft have failed at realising is that users wish to *add* to their input devices, not throw out the ones everyone's used for the last 40 years.

[JEDIDIAH]

Re: Desktops will never disappear

> The danger is that if desktop machines become significantly less popular, the cost will increase

Nonsense. In the old days the "off brand" machines were always cheaper. I bought my first non-PC in the 80s for $300. It took another 20 years for PCs to catch up to that kind of pricing. Meanwhile you have general purpose ARM devices being sold for $50. The world is full of microprocessors. Anyone can design and build their own variant of a PC and offer it up for the world.

A lot has changed since a really crappy 8088 clone would set you back $1000.

[Anonymous Coward]

Re: Desktops will never disappear

" Take CAD as an example. Would you rather fly in an aeroplane designed on a large screen where the designer can see the big picture and has precise control over what components go where, or one designed on a tablet with components shuffled by someone's fingers and thumbs? "

How large a tablet?

[Dave 126]

Re: Desktops will never disappear

>Take CAD as an example. Would you rather fly in an aeroplane designed on a large screen where the designer can see the big picture and has precise control over what components go where, or one designed on a tablet with components shuffled by someone's fingers and thumbs?

That 'precise control over what components go where' of which you speak is not achieved by fine cursor control, but by 'snaps' and 'relations' between 2D entities, mathematically defined, and by the use of 'mates' for parts. The idea is that you 'sketch' a form, and then add constraints until it is fully defined. In short, you shouldn't be relying on pixel-perfect cursor accuracy.

[John Sanders]

Re: Desktops will never disappear

"""In short, you shouldn't be relying on pixel-perfect cursor accuracy."""

You have a valid point, now try to do snaps on a tablet with your lousy fat fingers.

[JEDIDIAH]

Re: Desktops will never disappear

...it gets even better.

CAD applications tend to have a command line. They aren't just GUI driven. They just don't need a big display and a mouse for that "your grandma would be able to handle it" interface but they also need an actual keyboard for the really precise manipulations of the drawing.

It turns out that professional content creation is not a task for dummies or people afraid of their tools.

[Panicnow]

Apple didn't invent those

"features like menu bars or buttons on window title-bars at all – Apple invented most of that stuff with the Lisa and later Macintosh projects."

Er, Not Apple, Xerox maybe, but most of these innovations had parentage elsewhere.

(EEK am I the oldest still reading this stuff?)

[jubtastic1]

Re: Apple didn't invent those

Not Xerox, they didn't have menubars back then, and title bars were buttonless and clearly the inspiration for BeOS, I suppose it's possible that someone other than Apple created them but it seems unlikely that there's anything out there with those features that predates Xerox.

How about drag and drop, or being able to browse the filesystem? filetypes? updates to background / obsured windows? The thing about obvious ideas is that they commonly only become obvious the moment after someone has created them, there are billions of obvious ideas out there right now that represent untold fame and riches, realising them and making them work is hard, being labelled 'obvious' later is essentially the hallmark of fantastic ideas.

I guess I just find it disparaging when a whole load of original work is written off simply because it uses the windows and mouse concept from Xerox, the Apple developers didn't leave there with disks and code, just the basic idea, the concepts they added which have spread to all GUI's should be recognised just as much as the work of Xerox's developers in creating the basis of the modern GUI.

Back to the article, regarding NextStep , I recall a whole bunch of Docklets being ported over when OS X was first previewed, which were basically active notification icons, clocks, volume control, CPU usage, weather etc, most of these had menus attached to launch or change something, they fell out of favour so most of them are long gone but some degree of this remains in OS X with unread mail counts on the Mail.app icon, date on the Cal app etc.

[Liam Proven]

Re: Apple didn't invent those

You're wrong, according to all the research I've done.

I am of course entirely happy to recant, on production of cited, dated evidence showing these features in existing GUIs before the advent of the Apple Lisa in 1983.

It is one of the saddest aspects of the arguments over GUI development that so many people who weren't there and don't actually know pooh-pooh the massive amounts of original development that Apple did, turning Xerox' prototype Smalltalk-based GUI into a complete functional desktop layered on top of a hierarchical filesystem with discrete applications programs.

[Phoenix50]

So in summary..

...it's everyone elses' fault that Linux never conquered the Desktop.

And certainly not Linux itself.

[Anonymous Coward]

Re: So in summary..

Have to agree, this just read like a fanbouys tantrum. It has nothing to do with MS it was th fact that Linux was an nasty, horrible OS for end consumer user. It's greatly improved, but lets not kid ourselves, until recently it was a geeks and for the most hardend anti-ms users OS.

[wowfood]

Re: So in summary..

Agreed. I consider myself to be a fair bit of a geek, but it took me almost a decade to start using linux propperly. And even now I still have windows on my main PC for games (although that may change soon, fingers crossed)

Up until the last few years Linux was almost unusable for me. Even installing basic apps I needed to go trawling through the internet to find the specific command line needed to extract the package and installl it, and then I needed to find where it was installed to and use another command to create a shortcut to it.

Now it's so much easier. Although I've migrated over to MATE on Mint, I know not a huge leap from MS guis, but baby steps right?

Lets be honest here though, Microsoft didn't hold back Linux, at least not directly, linux held back linux along with devs. First it wasn't user friendly enough, now it's fragmented as heck. Originally nothing ran on Linux aside from special linux only apps, now more and more stuff is avaliable on the web, or on linux as well as on MS software.

I'd honestly love to see a few of these extra GUIs merged back together, if only to see a more concerted effort. I'd rather see 3 interfaces which are drastically different to one another, than 7 which are all very very similar.

Of course that kinda defeats the purpose of an open infrastructure I guess.

[keithpeter]

Re: So in summary..

"Up until the last few years Linux was almost unusable for me. Even installing basic apps I needed to go trawling through the internet to find the specific command line needed to extract the package and installl it, and then I needed to find where it was installed to and use another command to create a shortcut to it."

I haven't had to do anything like that since Ubuntu 5.10. Which distribution and which basic applications? I'm genuinely interested.

[Ben Tasker]

Re: So in summary..

@Keith

I can only assume that it was only in the last few years that he bothered to re-install whichever distro he originally installed!

[Chemist]

Re: So in summary..

"Even installing basic apps I needed to go trawling through the internet to find the specific command line needed to extract the package and installl it"

Well I've been using SUSE and then OpenSUSE since you had to pay for it and it's never needed any of this. Perhaps a little research ?

[sabroni]

Re: Perhaps a little research ?

yeah, maybe start by trawling through the internet to find the specific command line needed to extract the package and installl it?

[Chemist]

Re: Perhaps a little research ?

"yeah, maybe start by trawling through the internet to find the specific command line needed to extract the package and install it?"

NO !. I install all packages from a GUI and have for years ( certainly since SUSE 4.0)

By research I meant a little research would show plenty of Linux distros where all packages could be installed by GUI ( if that''s what you want ).

[Gav]

Re: Perhaps a little research ?

Thanks Chemist, for illustrating perfectly what has held Linux back.

Noob User: "I can't get this to work. It's too complicated and I can't find any proper documentation I can understand to help me."

Linux Guru: "You are doing it wrong and it's your own fault it doesn't work for you. It works fine for me and has done for simply ages."

Microsoft won easily because it understood that blaming the end-user is not a solution. If the end user doesn't understand the OS, then it is the OS's problem, not the user's. Make it easier and clearer. And for god sake, write proper end-user documentation.

Most of all, do expect end-users to magically become smarter because it suits you. Users are not ever going to change. That's the OS's job.

[Gav]

Re: Perhaps a little research ?

"do *not* expect end-users to magically become smarter because it suits you."

[Chemist]

Re: Perhaps a little research ?

I think it's perfectly reasonable to research an area, any area, before diving in. To just suggest that the command-line is needed to install a package on Linux without actually knowing that it isn't just shows a kind of arrogance.

[AJ MacLeod]

Re: Perhaps a little research ?

Have you ever actually tried to use any Microsoft documentation? I know I have and I've generally found it useless, requiring resort to the usual searching of mailling lists etc instead. Documentation for open source software on the other hand is admittedly of variable quality but perhaps somewhat ironically I've found the Gnome 2 System Admin Guide very good.

Microsoft provides practically zero support to "noob users" and I've seen hordes of them completely unable to make the latest version of Windows do what they want.

[M Gale]

Re: Perhaps a little research ?

Microsoft won easily because it understood that blaming the end-user is not a solution.

Windows 8.1.

[wowfood]

Re: Perhaps a little research ?

I honestly cannot remember which distro or which application it was. The application didn't come from a package manager, and the distro didn't come with any built in way to install the application.

I had to extract the application via the command line, manually find and install the dependencies and then insatll the application. It may have been an ass about face way of doing it, there may have been a package somewhere which had a very similar / the same appliation but this was a fair time ago, and the help on the linux forums was... well there wasn't any really.

I THINK the distro was Mandriva, and it was around 2000 I tried it if I remember rightly. It put me off Linux for a few years, but now I'm using Mint perfectly happily.

[keithpeter]

Re: Perhaps a little research ?

"I THINK the distro was Mandriva, and it was around 2000 I tried it if I remember rightly. It put me off Linux for a few years, but now I'm using Mint perfectly happily."

Excellent, Mint is a great choice, but most of the other main stream distributions have package managers and extensive repositories these days. 13 years is a long time in technology!

Thanks for coming back to us.

[Gav]

Re: Perhaps a little research ?

"Windows 8.1."

Which only goes to show just how badly MS have gone wrong with Windows 8 compared to previous versions.

(Yes, Vista was a mess too, but for different reasons.)

[Daniel B.]

Re: Perhaps a little research ? @Gav

Heh. It actually made me remember back when the god-awful Winmodems became common. I was saddled with one of those ugly things, so I go googling how to get the POS to work, and ended up reading stuff like:

"Oh, you gotta compile these drivers. If you don't know how to compile a Linux kernel, you shouldn't be doing Linux at all!" or "Closed source evil! We won't accept closed-source drivers even if winmodem manufacturers provide them!". Basically, the Linux community had a dark, newbie-hostile side to it and it showed in many places. Not to mention the pointless naming wars "EETS CALLED GNOO SLASH LEENUX!!!".

Though Microsoft is now losing the plot, given Windows 8. It seems they want to do a poor copy of Jobs' "blame the user" game, and they're losing it. They just don't realize it yet!

[eulampios]

@Gav

You're trying to make a very untrue claim.

If you go online, by just googling the error message or formulating (only roughly sometimes) what is not working or how you want it work for, you almost always find a solution, if someone has it already. Moreover, the fact that never had to sign in for a forum to ask questions but only profited from other people having done this job for me proves this point. Yes, there are differences in the *nix world, say with FreeBSD, you might be politely pointed at the pertinent chapter of the "FBSD Handbook". On Linux forums you get both: solution/recipe and a reference for (further) education. Along with the recipe if you'd like to receive a simple explanation of why it works, you can get it. Never does it get "dumbed down" for you.

Compare it with what you might get yourself with a Windows problem. First, it usually harder to troubleshoot it, like this enigmatic message: "Windows has encountered a system error f3-f100-0010 and will have to shut down" What the hell that hex number is? The devs or MS should know right? Since they are expert? What do you get? Nothing from both MS and OEM. They politely play football using you as their ball. Along with this you get the usual Windows panacea to reinstall the whole system. You might never get anyone to tell what the hell your problem was. Usually, it has very little to do IT knowledge: in 50% or more will tell you that it is virus (malware), so scan, delete some files, clean registry etc. If it doesn't help, get another AV scanner, and did you get told to reinstall Windows? It's like trying to understand the Kepler's Laws, their derivation from the Newtonian Mechanics by asking questions on astrological forums.

If you do get a qualified help (from MS or others) very rarely it will explain you what it is. It just takes you for a dumb person and goes to where to click, to press buttons and to select menus. While the Linux gurus would be happy to explain you their recipe, if you're interested.

[Cipher]

Re: @Gav

The reality is this (from my experience):

Official Microsoft answers to problems are either ignored or the user is read the MS script about malware, safe boot, driver updates and all the usual crap that has nothing to do with the problem. Private forums are your best bet, a well formed question with necessary info generally results in an answer. I work one of the larger sites, it is amazing what volunteers do to help the noobs...

Linux has its own problems, there are some good forums, but far too many sites where they expect you to be an expert first or belittle the user or ignore the problem. There is an air of elitism in many Linux forums, the high priesthood seems to wish to continue Vi/Emacs fanboi wars rather than help people. Worse yet, they often give recommendations for alternate solutions instead of addressing the problem. Telling someone they are using the wrong stove isn't an answer to why the stove they use isn't working...

[Philip Lewis]

Re: Support fora (ums?) was @Gav

"Worse yet, they often give recommendations for alternate solutions instead of addressing the problem. Telling someone they are using the wrong stove isn't an answer to why the stove they use isn't working..."

PRECISELY!!

The "forums" of the world are populated by people who by and large know nothing and sometimes have done something in their lives that remotely resembles your problem. They then offer one of the following scenarios.

a) Why do you want to that. I always do this (unrelated solution)

b) I did this once ... (unrelated story)

c) That is not "best practice" (cop out)

d) Answer a question in detail not related to the question asked (deflection)

e) ... and so on

I visit SQLServer forums (fora?) often, including those hosted by MS and the amount of totally useless crap there is astounding. I have seen threads 30-40 entries long where the OP continues to write "yes, well, that's interesting .. but I acually want to know the answer to the question". The OP gives up because the thread has long since devolved into something unrelated, uninteresting and completely (to the OP) useless.

Most people can't answer the question and even more can't understand it and even more of them do not bother to read it before they answer. Fact.

I don't know about Linux forums but I expect the same issue exists here.

As for Microsoft support? After getting threw the first 4 levels of idiots who cannot spell the product name (they are worse than illiterate) you may reach someone who just might be bothered to read your support request - or not. They will then tell you 10 things from their script which are not helpful and not related to the bug in question. Paid-for MS support is, in my experience, a complete fucking waste of time for any sophisticated and knowledgable professional. My experience with other large vendors is different. end-of

[JEDIDIAH]

Re: Perhaps a little research ?

> yeah, maybe start by trawling through the internet to find the specific command line needed to extract the package and installl it?

You mean like unzip or unarc?

I'd hate to see how this guy would react to the really old versions Windows.

Although it sounds like his brain would melt even if he encountered an oddball Windows archive format.

[itzman]

Re: So in summary..

Linux is not a window manager. Or a desktop.

What counts for 3rd party apps is a reliable X-window API at the least, and that tends to be common across all distros, and better a reliable graphics library, and that is getting to be GTK or one of a couple of others, all of which seem to be available on all distros.

Just because Apple and MS have decided to make 'look and feel' the defining criterion of brand differentiation, doesn't mean anyone else needs to, wants to,or has to.

Indeed for the power users - the generators of content, rather than the consumers, it is almost a matter of complete indifference as to how their application gets launched. What matters is the application, and its own menu system and controls.

[Anonymous Coward]

Re: So in summary..

+1 for writing X Window in the singular form!

[keithpeter]

Re: So in summary..

I suspect things will change over the next 10 years or so as we move to mobile devices and people get used to 'direct manipulation' interfaces. We already have Android based PCs on the way (no, android isn't GNU/Linux but it is not MS either so people get used to devices that are not MS on hardware that is not Apple).

@AC 07:48

I agree that the article was slightly odd but hardly a tantrum. I found the possibility that Gnome Shell and similar emerged as a patent dodge interesting. The claim has always been superior usability. Canonical do actually document some of their usability studies. All I can find from Gnome is some work by a summer intern.

The tramp: I have to use Windows at work. It works. They can't afford to upgrade the hardware though.

[Graham Dawson]

Re: So in summary..

Fanboy's tantrum? The author's article history makes it pretty clear where he stands on Linux, the GPL and open source in general and he is no fanboy. I really rather suspect you haven't read the article at all.

And how do you define "recently" anyway?

[Anonymous Coward]

Re: So in summary..

« ...it's everyone elses' fault that Linux never conquered the Desktop.

And certainly not Linux itself. »

I cannot speak for everyone else, but as a former contributor to the Linux environment (including in the KDE ecosystem), that was never one of my goals, nor of anyone else's I was developing with.

I always had the impression that the whole "conquering this / conquering that" came from computer magazine writers trying to evoke an emotional response rather than presenting useful information.

The whole Linux vs Windows, Apple vs Android, Boeing vs Airbus, This vs That thing sounds so incredibly childish to my ears. Can we please leave that kind of bollocks for drunken pub talk between the uneducated? Thank you.

[Anonymous Coward]

Re: So in summary..

I cannot speak for everyone else, but as a former contributor to the Linux environment (including in the KDE ecosystem), that was never one of my goals, nor of anyone else's I was developing with.

Agreed, personally I couldn't care less whether the mass market want it or use it. Works well for me, and I'll continue to use it and develop for it for as long as that remains true.

[McVirtual]

Fair play

Great article there, interesting read!.

Thanks

[Anonymous Coward]

Boooring, where is CDE?

First of all you hit a nerve with me by mentioning SunOS. Where you actually meant Solaris because that was the all-in-one operating enviroment, SunOS didn't include the X environment per definition. In fact; SunOS sort of got replaced by Solaris, up to version 4 it was actually based on BSD.

Another thing:

"Look at screenshots of long-dead early GUIs from the 1980s, such as SunOS or Smalltalk, and they don't even have features like menu bars or buttons on window title-bars at all"

Apart from the statement above Solaris *had* menu bars, if CDE was used. There were at least two graphical environments to chose from, by default OpenWindows, but as a counterpart we had CDE, which didn't provide one but several menu bars in the tray panel, one which inspired Xfce later on.

Don't cut Solaris short like this, please...

Needless to say but this undermines your statements. A statement which is flawed too:

"RISC OS already had its Icon Bar, but that doesn't have an app launcher or a way to switch windows."

And how exactly could Windows 95 switch Windows? Or are you now confusing windows with virtual desktops here, because Risc OS had plenty of options to use several windows side by side, in a way comparable to GEM.

Oh that's right; GEM... Even usable on CP/M and somewhat comparable to a mixture of Windows 3.1 and OpenWindows, which you claim wasn't prior to Windows 95.

Then how come that looking back at GEM reminds me heavily of a Windows hybrid?

I do agree with you on one thing though; it really seems as if the main desktop managers copied Windows. But there were plenty of environments which didn't, Xfce for example (before it became "cool" to use it, you know what I'm referring to).

[Anonymous Coward]

Re: Boooring, where is CDE?

SunOS had a GUI, and it had menu bars. It was called SunView, with a look and feel called OpenLook. When Sun replaced SunView with an X Window System based GUI, they created a library called XView that provided a way of porting existing applications and writing new ones that conformed to the OpenLook specs. It was also the first decent "open source" GUI toolkit, as Sun eventually released the source code and it was rapidly ported to other operating systems including Linux. I coded my first commercial GUI using it.

[Liam Proven]

Re: Boooring, where is CDE?

/Some/ OpenWindows apps had menus, not all. When present, they were implemented as a row of separate buttons, each of which opened a separate menu. This design was bolted on later - SunView lacked even this.

In fact, SunView originally did not even have things like maximise, minimise or close buttons. Example:

http://en.wikipedia.org/wiki/File:SunView-Desktop.png

[Liam Proven]

Re: Boooring, where is CDE?

No, I did not mean Solaris, because OpenWindows appeared in 1989 and Solaris in 1993.

CDE also did not appear until 1993 and contained elements licensed in from - guess who? Microsoft!

Fancy that, eh?

CDE itself was based in part on Windows 3 concepts, licensed from MS. I would welcome any traceable design lineage from CDE to Windows 95 /et seq./

Win95 task switching: jeez, did you actually understand *any* of the article? Switching windows is what the taskbar is *for*. That row of buttons between the start menu and the clock? That's the task switcher, that is. Those buttons with the names of windows on. Those.

Show me the equivalent in OpenWindows, or the Mac, or RISC OS, or GEM -- oh, wait, you can't, because *they don't have an equivalent.*

I am not talking about window managers or tiling windows. I am talking about a graphical mechanism for switching between multiple open windows, *even if they cannot be seen*, even if invisible, even if minimised.

You appear to have failed to understand the most basic points of the argument and you lack knowledge of the timescales involved.

[J.G.Harston]

Re: Boooring, where is CDE?

"That row of buttons between the start menu and the clock? That's the task switcher, that is."

No, that's the task *bar*, and the action of clicking on one of the items there is "bringing a task's window to the front and giving it the focus". Not "task switching".

"Show me the equivalent in OpenWindows, or the Mac, or RISC OS, or GEM -- oh, wait, you can't, because *they don't have an equivalent.*

"I am not talking about window managers or tiling windows. I am talking about a graphical mechanism for switching between multiple open windows, *even if they cannot be seen*, even if invisible, even if minimised."

That's not task switching, that's bringing a window to the front and giving it the input focus. Task switching is what the switcher does continuously in the background. I agree that if the RISC OS window you want to give focus to is hidden by everything else you can't directly give it the focus, but a) you don't have to bring it to the front to give it the focus, b) you would normally just push a few other windows to the back of the display stack until the one you are looking for is visible c) you could use something like WinTab to Tab between windows until the one you want appears.

[Schultz]

Short-term vesus long-term

I agree that fragmentation is bad in the short term. But it gives the users choice and can thereby ensure that the good (better call it 'successful') solutions survive and thrive. So in the long-term it should lead to benefits. It also keeps many users happy through the course of a Win-8 / Gnome-3 transition.

Let the best desktop survive.

[Anonymous Coward]

"Revolting"

"the users are revolting"

Yes they are.

[Anonymous Coward]

Re: "Revolting"

« "the users are revolting"

Yes they are. »

Good one! :)

[Steve Foster]

Re: "Revolting"

Drat, beaten to it (good job I checked, or I'd have looked like a right noob!). Have an upvote.

[John Deeb]

(round) OpenWindows

Hmm, but what about OpenWindows in 1990? Not sure what Windows 95 improved five years later fundamentally apart from using the latest graphic adapters better (which also caused it to be unstable to the point of being unusable for any business relying on PC's not crashing). Also Windows 95 put a somewhat fixed START button on the screen which many opted to hide again. Barely a standard desktop breakthrough.

OpenWindows workspace menu in 1990: better then taskbar?:

http://www.youtube.com/watch?v=haUnfHLw4sQ#t=301s

That said, apart from Microsoft being involved in the Unix desktop developments quite early, I think they did well to really find out what inexperienced people wanted to see on their desktop, how they would "feel at home" the quickest. It's not surprising therefore other desktop environment developers discovered the same round wheel. But should we really talk about licensing common sense designs again?

[Chairo]

WPS

Effectively the Win95 and WinNT4.0 desktops were very close to OS/2's WPS. As for myself, I found the WPS to be far more powerful and elegant than the Win95 desktop.

That might also be one reason why MS didn't start suing Linux vendors. IBM might have had some rights on the Win95 desktop scheme, as well. You could argue, that OS/2 was also done by Microsoft, but the WPS changed a lot for the 2.x versions of OS/2. And AFAIK MS already withdrew from OS/2 development at that time.

[MacroRodent]

Re: WPS

> Effectively the Win95 and WinNT4.0 desktops were very close to OS/2's WPS

Exactly. And OS/2 3 aka "Warp" certainly predated Windows 95 for years. I used it for a while at home (until giving up because some applications I wanted were not available for OS/,2 and did not run in its Windows emulation), and after that Windows 95 felt like a very poorly done copy.

So I think this puts a big hole into the theory proposed by the article.

[Liam Proven]

Re: WPS

I'm glad you liked the WPS. As an OS/2 user from 1.0 myself and an early-adopter of OS/2 2.0, I hated it. I thought it was an ugly, clunky mess.

However, it had little influence over Win95 -- if you feel differently, then please provided illustrated examples, with dates.

But yes, OS/2 and Windows share a lot of design elements, especially in window management, menu and dialog box design, because they were written by closely-cooperating teams with members from both IBM and Microsoft.

OS/2 inherited its window manager design from Windows 2. Windows 3 in turn took its Program and File Managers, its 3D title-bar widgets and other elements from OS/2 1.2 and 1.3.

OS/2 2.x was created after MS & IBM fell out and shows signs of IBM intentionally trying to do things differently, presumably for fear of litigation - e.g. the missing "Apply" and "Close" buttons in WPS dialogs. However, OS/2 contains lots of licensed Microsoft code, which is one of the reasons why it has never been open-sourced.

[MacroRodent]

Re: WPS

> However, it had little influence over Win95 -- if you feel differently, then please provided illustrated examples, with dates.

Oh, come on, I am not a professional computer technology historian. What really influenced what behind the scenes is beyond my knowledge. My memory is getting hazy on the details, but I definitely remember FEELING after installing Windows 95 that this is like a poorer version of OS/2. Its UI certainly felt more like OS/2 than like Windows 3.1. Maybe I should set up OS/2 Warp in some virtual machine to refresh memories...

[Anonymous Coward]

Thoughts

If MS could fragment so easily then maybe people wouldnt hate the windows 8 experience so much? Through all efforts to claim that linux is dead on the desktop it quite clearly isnt if there is a market for multiple desktop forks.

I run both windows and linux and believe in the right OS for the job. When windows 8 was released I read the reviews and bought a win7 laptop. I use mint with cinnamon because both have the interface I am most comfortable with. When a big entity say charge with their new and super idea I like that others can refuse and go their own way. Windows 8, gnome 3 and Ubuntu unity got slammed for their interfaces. And you can download some start menu addons for windows or you can completely swap out your desktop on linux. Having the option isnt a bad thing.

[Paul Crawford]

Really, it was MS' fault?

Having seen the attitude of most GUI developers, not just Linux but MS and others, where they jump from one way to another announcing that their way is the best, I have to wonder about the underlying idea behind this article:

1) Did MS' vauge patent threats actually matter or get taken seriously, or

2) Did the various Linux GUI designers simply spend too much time with their heads down their Y-fronts?

I liked Gnome 2 and it was such an obvious option for XP-escapees due to its similarities, and yet Gnome 3 was a re-write with the attitude of "users are dumb, lets make all our options dumber" (even if there were technical reasons for wanting to fix some G2 stuff) and the apparent desire to move things around for no obvious reason. See Linus' comments on Gnome 3 for further information.

AFIK Unirty was Ubuntu's attempt at a touch-friendly desktop for small devices with the modern p[iss-poor HD style screens, hence the side icons and default-to-full size operation.

So as afar as the fall of Gnome 2 and something sensible on the Linux desktop it is more a case of incompetence and managements problems of the Gnome teams that being forced off. The same irritating design decisions are also part of MS' TIFKAM cluster-fsck so it seems to be one of industry-wide phone/tablet fixation taking precedence over what a power-user's desktop should be doing.

[Joseph Lord]

Legally wrong

I don't think that laches applies to patents which you can enforce selectively and also wait until it is worthwhile. You may however act in ways that offer implied licenses.

Having said that Windows 95 patents will be expired by now or expiring very soon. They must have been filed before the technology was made public (certainly before the public beta was released) although they may have taken some time to be granted so the 20 years may start later than you would expect. Windows 3.1 and earlier related patents are probably all expired and anything added for Windows 95 will soon be gone.

The best thing about patents is that they do expire within a lifetime.

[Roland6]

Re: Legally wrong

I think we need to be circumspect about Microsoft's IPR and how it may or may not impinge on Linux distributions today and in the future. Whilst utility and design patents have a limited life, we shouldn't forget about copyright, which doesn't have such a limited life.

[Vic]

Re: Legally wrong

> we shouldn't forget about copyright, which doesn't have such a limited life.

No code was copied, so copyright doesn't come into it.

You can't copyright ideas, however often Oracle tries to...

Vic.

[Terry 6]

Usability

Any new OS requires a degree of learning.

I've moved from CP/M via most MS releases to Win 7 and use MINT on one of my boxes.

But the Linux "community" don't do any favours to ordinary users, or the Linux cause.

it's not just the matter of all the different versions each with it's own set of fans, but also the impentrability of it all for anyone who just want to use a computer that hasn't got Windows on it. - possibly because they just don't want or can't afford to fork out the cash to M'soft.

Trying to workout which version to choose is a job for a hobbyist ( or IT Pro), not an (even skilled ) ordinary user.

Then getting help to make the thing to work properly is another piece of hobby work: There are few if any simple guides. Usually the recommendation is to use the forums. Most users don't want to mess with forums at the best of times. (Those are also for hobbyists and Pros). And Linux forums are not the easiest to use. If you're lucky enough to get a single and consistant answer to a "How to" question it will almost always be in the form of jargon or technical instructions; whereas a Windows forum will tell you which buttons to press.

There's kind of religious attitude to Linux, with a priesthood, that has a heirarchy from Newbie to Guru ( even the word gives it away) and of course endless schisms and accusations of heresy (which we call "forks" but we know what we mean)..

[keithpeter]

Re: Usability

"Trying to workout which version to choose is a job for a hobbyist ( or IT Pro), not an (even skilled ) ordinary user.

Then getting help to make the thing to work properly is another piece of hobby work"

I'm not arguing with you, I'd just like to know what you think is wrong with the following advice for 'general' users...

"Download Ubuntu LTS, burn it to a DVD or copy it to a stick, and install it with default options. Then use Ubuntu Software Centre to add GIMP and the Ubuntu Restricted Extras package"

(One thing I would like sorted when we have all stopped arguing about the interface is sound. I'd like jack/pulse sorted out into a single sound system that just works. )

[Anonymous Coward]

Re: Usability

What is wrong with the following?

"Download Ubuntu LTS, burn it to a DVD or copy it to a stick, and install it with default options. Then use Ubuntu Software Centre to add GIMP and the Ubuntu Restricted Extras package"

You clearly haven't had the pleasure to support a totally computer-illiterate user (be it your aging mother or a technophobe). Either way, you would be doomed to fail as soon as you utter your first sentence, having omitted to define a number of concepts to which they are totally alien to: download, burn, copy, stick, install, options.

But you know what? these people still want to use a computer and Microsoft is not ignoring them, quite the contrary actually if you look at who Metro is targeted to.

[Anonymous Coward]

Re: Usability

"You clearly haven't had the pleasure to support a totally computer-illiterate user"

A user who would also be incapable if installing MS Windows.

[wayne 8]

Re: Usability

>A user who would also be incapable if installing MS Windows.

Who needs to install Windows? It comes on almost every computer sold. And funny thing the Windows computers are cheaper than the same hardware with no OS.

Linux, for the most part (my experience) requires the end user to interact with the computer at a level that a casual user does not want to act.

Like a high performance car that requires more skill, knowledge, and commitment than a commuter car. Many drivers just want to know how to open the door and start the car, get where they're going and shut it off and lock it up.

[Phill 3]

Re: Who needs to install Windows?

Every point you make seems to me to apply more to Windows than to GNU/Linux.

Who needs to install Windows?

I generally find that a new Windows PC will last about 18-24 months before needing a complete fresh re-install.

If all you do is remove the nasties (including bloatware & updaters) and update everything, then more like every 3-6 months. A fresh format/install/configure/activate/update can take 8 hours with a 2Mbps broadband for W7. This is how easy it is to own a well maintained Windows PC.

The issue of hardware pre-installed with Windows being cheaper should tell you something about monopolistic practices. Nothing to do with the pros & cons of either OS.

Have you seen a normal Windows user struggle with drivers, malware & the seemingly endless variety of updaters that come with 3rd party apps, constantly screaming for attention? The reason why average PCs are so slow & full of this stuff is precisely because the users don't want to deal with it. So they click randomly on anything, inevitable making the situation worse.

Hence I agree with your last paragraph. But honestly GNU/Linux is a breeze to live with compared with Windows.

[Anonymous Coward]

Re: Usability

> You clearly haven't had the pleasure to support a totally computer-illiterate user (be it your aging mother or a technophobe).

How do you know my "aging" [sic] mother is not a fucking computer scientist?

(requirements gathering hat off)

[JEDIDIAH]

Re: Usability

> How do you know my "aging" [sic] mother is not a fucking computer scientist?

I know a 90 year old retired accountant that doesn't even speak very good English that does well enough adapting to new tech. It's more about the personality than the skill set.

The whole GUI concept allows for exploration and discovery. You just have to be willing to use it. If you aren't willing to bother at all then no amount of shiny shiny is going to help.

[keithpeter]

Re: Usability

"You clearly haven't had the pleasure to support a totally computer-illiterate user (be it your aging mother or a technophobe)."

Yes I have. Such users do not install operating systems. They use what comes with the computer. My point a few pages up was that 'what comes with the computer' may well be Android/Sensible GNU/Linux/Windows/Mac OS in the future.

The post you are replying to was about the kind of people who put their head over the parapet and decide to try something a little different. Like the HND students of mine who use bootable USB sticks to run sagemaths and like the colleague who has an ailing Windows XP laptop and is trying Ubuntu.

My fault for deleting a paragraph in the post to which you are replying that made the segmented audience clear.

[JEDIDIAH]

Re: Usability

> You clearly haven't had the pleasure to support a totally computer-illiterate user

There should be a Godwin's Law variant for this as you are describing an impossible situation. Your "granny" isn't going to be any better equipped to deal with Windows. She will require an equal amount of hand holding and be just as confused by it as she would be by Linux.

You are Microsoft's unpaid support network as these people would not be able to cope without you around to pick up the slack.

[Vic]

Re: Usability

> You clearly haven't had the pleasure to support a totally computer-illiterate user

I have. Quite a number of them.

> Either way, you would be doomed to fail as soon as you utter your first sentence

...Which is why you wouldn't use that sort of approach with them. You'd give them a system and say "have a play - you're not going to break anything".

IME, those who are told something is going to be scary will find it so[1]. Those who are told they have the ability to do something will generally be able to.

Vic.

[1] I've taught a number of people to dive over the years; the exact same thing applies. If the instructor believes something is difficult, the students will find it so.

[Roland6]

Re: Usability @Keithpeter

"Download Ubuntu LTS, burn it to a DVD or copy it to a stick, and install it with default options. Then use Ubuntu Software Centre to add GIMP and the Ubuntu Restricted Extras package"

Re: What is wrong with the following advice for 'general' users...

All of it!

For the 'general' user, I would strongly advise purchasing a Linux magazine and running the included Live disk. If they are happy with which ever Ubuntu/Linux distribution they settle on, it is then worth taking matters further and considering doing a HDD install along with sourcing all the various necessary peripheral drivers.

Yes loading a Live disk does take time, but it is significantly less and less frustrating than starting down the install path and hitting problems that could of been discovered by running the Live disk...

[Jamie Jones]

@keithpeter Re: Usability

"(One thing I would like sorted when we have all stopped arguing about the interface is sound. I'd like jack/pulse sorted out into a single sound system that just works. )"

I'll get flamed for this, but why not try FreeBSD or pc-bsd (which is a more newbie/desktop orientated version) ?

FreeBSD has no problems with sound

http://www.pcbsd.org/

[FrankAlphaXII]

@Jamie Re: PC-BSD

PC-BSD's pretty good for a desktop OS, I'll hand you that. I mess around with it in a VM from time to time. I think its a cool project that is worth more attention than it gets. Its never once given me an issue in the VM, so I doubt running it on bare metal would be a problem either. Plus it uses ZFS without any kernel module bullshit. And other Window Managers work properly now since 9.0 came out, so you aren't virtually stuck with KDE.

Plus it has the entire FreeBSD ports collection available, as well as the PBI installers, there isn't a lack of software at all. Documentation's kind of sorely lacking but the community seems to be much friendlier than OpenBSD or FreeBSD's and happier to help answer questions, even relatively dumb ones with a better response than a dressed up "RTFM" response.

Its definitely different than Linux though, and some people may not like that.

[Jamie Jones]

@FrankAlphaXII: Re: @Jamie PC-BSD

Thanks for the info - I've never actually used PC-BSD, being a long time FreeBSD user.

"Its definitely different than Linux though, and some people may not like that."

That's why I like it :-)

[Terry 6]

Re: Usability

OK. You asked!

1.) "Ubuntu LTS" - LTS?

2.) "Install it with the default options". This ssumes user already knows how to install it. From Windows or not. Must they make their PC boot from a CD/stick: Can they?

3.)"Default options". Easy to do if you understand what it is or are prepared to take a leap of faith and go ahead and install it without understanding.

4.) "Ubuntu Software Centre". Where is that? How does the new user find it? Is it anywhere near the local shopping centre?

5.) "Gimp". I'd hope they know what that is, but they might not and is it really essential? And is it a bit too complex for many ordinary users.

6.) "Ubuntu Restricted Extras package". If you can't see why that might sound or even* be* scary, well...........

[keithpeter]

Re: Usability

@Terry 6

1.) "Ubuntu LTS" - LTS?

> Long Term Support current version. Linked prominently from Ubuntu.com home page

2.) "Install it with the default options". This ssumes user already knows how to install it. From Windows or not. Must they make their PC boot from a CD/stick: Can they?

> See earlier reply. I really regret cutting the paragraph from the original post about how most people just use what the device comes with, so that we are talking about a smaller minority that want to try something different for some reason. Having set up a Windows 7 laptop recently including playing 'hunt the drivers' I really think installing Ubuntu is easier!

3.)"Default options". Easy to do if you understand what it is or are prepared to take a leap of faith and go ahead and install it without understanding.

> Sensible defaults in most distributions I have tried/seen recently

4.) "Ubuntu Software Centre". Where is that? How does the new user find it? Is it anywhere near the local shopping centre?

> Its the button on the left near the bottom that says 'get new software' or something similar.

5.) "Gimp". I'd hope they know what that is, but they might not and is it really essential? And is it a bit too complex for many ordinary users.

> Photo editing. See note about 'ordinary users' above.

6.) "Ubuntu Restricted Extras package". If you can't see why that might sound or even* be* scary, well...........

> It is the US patent system that is broken - so distribution providers can't supply basic codecs with their standard distribution but have to make them available (nudge wink) in less obvious ways. Actually I forgot the Fluendo question in the installation process.

[Jamie Jones]

Re: Usability

"Then getting help to make the thing to work properly is another piece of hobby work: "

Why do people complain about their apparent difficulty in installing Linux - when comparing to MS windows - yet they always refer to a pre-installed windows.

Sure, it's easier to get a PC preinstalled with Windows, but if you want to do a fair comparison of Windows versus Linux, you need to compare either both systems as pre-installed, or both systems requiring installation

[John Lilburne]

Re: Usability

Why do you need to compare when for the majority of users MS comes pre-installed. When getting a new machine they don't say "Tell you what don't stick windows on it, I'll do it myself, naw scrub that I'll put linux on instead.". My colleague was saying the other day that Linux is a hobby it needs careful and continuous messing with, its just that you don't notice you are doing it.

[JEDIDIAH]

Re: Usability

> Why do you need to compare when for the majority of users MS comes pre-installed.

Sooner or later the bit rot will set in and they will need to reinstall.

The beauty of Unix (and this Linux) is that you can set it up and forget it about and leave it in a closet until you forget how you ever managed to install it to begin with.

It's Windows that needs constant maintenance just to remain safe to use.

[Vic]

Re: Usability

> My colleague was saying the other day that Linux is a hobby

My colleague was saying the other day that he's a really good driver.

He was talking bollocks.

Vic.

[Furbian]

A Windows 8 like Linux GUI...

... now that would interesting... though Ubuntu's Unity comes close to the 'flat earth, sorry, tile" (though scroll bars do appear eventually at the edges) design. If mimicry is the sincerest form of flattery, then Windows 95 did well, but Windows 8, seems unlikely there'd be a similar article about that years down the line.

[Tom 7]

To be more accurate

the big mistake was even considering following a leader who was only intent on stopping you using your computer properly.

OK maybe not maliciously but the idea that you can make a complicated things simple always was a nonstarter- to make it easy for some to use you have to make it hard to get to the bits that you deem complicated but others deem necessary.

In the end you end up with people arguing over whether the stirrups on the horse should be labelled using roman or comic sans while some of us are tinkering with matter transporters.

[JimmyPage]

A *single* example of how things work ....

I had my [virtual]* machine running Ubuntu 10.04 with GNOME. Upgraded to 12.04, and got Unity. Immediately discovered my "system monitor" taskbar applet stopped working. Why ? Because Unity panels apparently can't support animation.

End of line for system monitor then. As the discussion I briefly read surmised, it's simply not possible to code a replacement.

Just fundamentally change the framework for the desktop, and tell users to whistle ? Great strategy.

*Now you know why I did it in a virtual machine first.

[Anonymous Coward]

I switched to KDE back in 2008

And never, ever, looked back. The 4.0 series were a little rocky, but usable. And there is nothing that can match KDE in terms of customization. You can make it look and behave like almost anything you want. Oh, and Gnome apps work perfectly with KDE (please skip your rants about having to install Gnome libs on KDE to run these, unless you're looking for saving such small amount of disk space, in that case neither Gnome nor KDE are for you)

[Anonymous Coward]

Re: I switched to KDE back in 2008

KDE, the customizable Desktop Environment. Designed by and for anal retentives

[Vic]

Re: I switched to KDE back in 2008

> The 4.0 series were a little rocky

KDE 4.0 was *utte shite*. Really, it was. As a long-term KDE user, I could not believe how bad it was.

I stopped using KDE regularly at 4.0. I think it's time to give it another go...

Vic.

[Frank Haney]

Let's call the whole thing off

You say "fragmentation", I say "choice".

It doesn't scan as well as "tomato/tomato", it probably needs a bit of work, but your choice of words says a lot about your biases. "Fragmentation" has negative overtones. On the other hand, I like having a choice.

[Tim Parker]

"The problem is that GNOME and KDE and so on didn't. Their versions are precise replicas of Windows 95."

GNOME 2 was not a "precise replica" of Windows 95 and many of the changes were due to feedback from a lot of user interaction studies, paid for in part by Sun. One of the reasons I liked GNOME 2 was the very differences you seem un-aware of. However I think it's fair to say that KDE was, for the most part in earlier versions, a lot closer to Windows 95 in gross layout even though it had some very different interaction modes and features (particularly in later versions).

[Liam Proven]

Interesting. Do please elucide, with examples, ideally illustrated.

The differences seem minor to me - multiple panels, 2 by default, rather than one. (But readily merged, e.g. in Mint.) Multiple menus with apps, filesystem and system-admin separated out (as they were in early Windows 4 betas) - but merged back together in some distros.

GNOME 2 is more customisable, sure - but it did not do vertical taskbars as well as Windows, for instance, so it's not a complete win. It lagged badly in some areas. (E.g. try renaming or deleting a folder from the tree in the Nautilus sidebar. It's a magic, uneditable view, unlike in Windows Explorer.)

[Tim Parker]

"Interesting. Do please elucide, with examples, ideally illustrated."

Search for terms like usability study sun gnome and you should find a bunch of discussions. The collection of studies which I find easiest to dive into is at

https://live.gnome.org/UsabilityProject/UsabilityTests

In practical terms, the single most enlightening difference for me was the placement of the task menus and applets at the top of the screen by default. That may sound trivial (it is) and can be accomplished in various versions of Windows, but the important thing is not that you can get some of the gain by telling people to change the default layout but to do it yourself in the default layout. I spent a good deal of time after I started using it watching Windows users repeatedly dipping their heads down, chin on chest, looking for the start menu (KDE users too, or at least those that didn't just use the 'type to search' idiom which came a deal later)

"The differences seem minor to me"

They don't have to be large, that's probably another take-away from some of the studies - small differences in how things are presented can translate to big differences in user experience.

"GNOME 2 is more customisable, sure - but it did not do vertical taskbars as well as Windows, for instance, so it's not a complete win."

It's not about winning - I didn't even say one was better than the other, merely that it wasn't a "precise replica" of Windows 95, which is what the article claimed. Personally I preferred some of the changes that GNOME brought about eventually, most of the early 1.x series was clunky as hell, and I wouldn't have touched it with a barge-pole - YMMV.

"E.g. try renaming or deleting a folder from the tree in the Nautilus sidebar. It's a magic, uneditable view, unlike in Windows Explorer.)"

I don't have a problem with that, and curiously enough the "slow double click to rename" idiom I absolutely hated on Windows - especially as the default was to just select on single click of a folder, not enter. I was forever hitting escape to stop it trying to rename things when I get it event loop in a tis or I dared to double-click a fraction too slowly. I also know plenty of people you can't stand the way I have things set up as well. That's life. Whatever works for you.

As I said before, i'm not claiming it's better or worse than anything else - merely that it wasn't a carbon-copy of Windows 95.

[keithpeter]

Gnome Shell Usability

@Tim Parker

"Search for terms like usability study sun gnome and you should find a bunch of discussions. The collection of studies which I find easiest to dive into is at

https://live.gnome.org/UsabilityProject/UsabilityTests"

Thanks, but nothing there later than 2009 for Gnome Shell. Any references on decisions taken after that point (3.0)? Monitoring the devel mailing lists implies a picture of strong personalities simply doing stuff.

Aakanksha Gaur did a bit of work on later releases of Gnome as an intern, see

http://www.linux-magazine.com/Online/Blogs/Off-the-Beat-Bruce-Byfield-s-Blog/GNOME-Gets-Formal-Public-Usability-Testing

for summary. Her blog has evapourated but some of the material can be extracted from the way back machine.

Any other substantive references?

[Tim Parker]

@keithpeter Re: Gnome Shell Usability

"Thanks, but nothing there later than 2009"

Not sure I follow - this started off as a post about GNOME 2 not being a "precise replica" of Windows 95 and a mention of the usability studies used in the transition from 1.x to 2.x. Is there something you think that might be pertinent to that in posts from 2009 onwards ?

[keithpeter]

Re: @keithpeter Gnome Shell Usability

@Tim Parker at 1942: In my mind I was allowing the discussion to move forward to Gnome 3/Gnome Shell as that was the logic of the original article ('new' UIs developed to avoid patent issues). I have an interest in documenting the shifts in Gnome Shell. Ignore the question.

[PaulR79]

Microsoft won? That must be why we now have a Steam client for Linux

If Microsoft had won I very much doubt that a Steam client would have appeared quite as fast as it did in response to Windows 8. I'm pretty sure Valve would already have been working on it but as a lower priority until the mess that is W8 appeared. Even 8.1 is just a token gesture to try and look like they're listening (not a chance) since the returning 'Start' button just does what the Windows key does now - pulls up FAILTRO. It will still be the same confused mess that I, and others, hate.

Why do I call it confused? The best example I can think of is Windows update. I get informed of updates and click to view so it pulls me into FAILTRO but if I want to change what updates are installed it's back to the Desktop. It's like a tacit acknowledgement that even the makers are confused about which interface should be in control of certain tasks.

[1Rafayal]

RISC OS vs Windows 8 Desktop mode

As far as I can tell right now, there is hardly any difference between the RISC OS icon bar and the Windows 8 desktop mode. At least, up to and including RISC OS 4, dont know about 5 and six.

[JimC]

Have we forgotten IBM's CUA?

I remember that CUA documentation being a vital tool for a while. I still don't really get on with the damn silly ribbon stuff. 'Spose I'm going to have to stop using the old stuff that works, but I don't relish having to teach my 80ish mother a new interface.

[gmguy]

Re: Have we forgotten IBM's CUA?

Yes, it does seem that CUA has been forgotten/overlooked. CUA was amazing and ground breaking. It was a natural way of doing things and Microsoft followed it and benefited from it with thier ealier versions of Windows 3.x releases. The Workplace shell was the ultimate as far as ease of use and the new pardigm of contect menus. Windows before CUA was a confusing mess.

[Mage]

This is nonsense

There is actually minor difference between Win 3.x and Win9.x desktop.

The actual windows / applications and icons/menus on both are the same.

So what is the differences?

The desktop from Win9.x onwards is one giant global Program Manager group that can't be resized or minimised. You can create folder icons on win9.x desktop that are like Program manager groups so your main desktop isn't too cluttered.

The only real difference is the taskbar with start button. Which can easily be added as an application on a desktop Icon. Of course an Explorer window is completely different to File Manager, but that is really just a different File managing/Browsing tool integrated to the Desktop.

You can operate Win 3.1 desktop very like Win9.x except for Start Button/Task bar, is can easily be added. Nothing revolutionary or patentable.

Win 2.x Win 3.1, Gem, Llsa, Mac and others simply copied the Xerox project and each other.

MS or Apple don't deserve a single desktop patent, nor should anything like that be Patentable, only copyright protected if original.

Part of the fragmentary and Disparate nature of various UNIX/GNU/Linux desktops is due to underlying architecture of X-Windows being more flexible allowing any kind of Window Manager. The MS Windows API is quite different.

[Liam Proven]

Re: This is nonsense

As with many other posters and commentators on this issue, you appear not to understand the actual details that this debate is about.

If you consider the differences between Windows 3 and 95 to be minor, then you do not understand the sort of software design changes that are patentable and the elements of look and feel of which are made legal disputes. Seriously. The stuff that you casually dismiss as minor *is the basis of the whole discussion.*

So either go study up on the trivial differences or kindly leave the debate, because you do not understand.

It is also facile, glib and useless to idly comment that U, V X, Y and Z copied each other; the important part of of the discussion is to take them apart, look at the dates, and work out who took what from where and when they did it. Xerox did not borrow from the Mac; the Mac didn't exist yet when Xerox did its work. Lisa didn't borrow from GEM; it predated GEM. And so on.

[JEDIDIAH]

Re: This is nonsense

> If you consider the differences between Windows 3 and 95 to be minor, then you do not understand the sort of software design changes that are patentable

The fact that trivial nonsense can be patented does not alter the fact that it is trivial nonsense.

[Anonymous Coward]

Patent claims

Microsoft never revealed what the 235 patent claims were.

The reason for this was that all they wanted to do was instil fear, uncertainty and doubt in companies planing on deploying Linux. At the same time that you had this patent nonsense you had SCO (with some back-door funding from Microsoft) claiming to own all of Linux, sending out licensing demands and suing companies for using it.

If Microsoft had revealed what the 235 patents were then the developers would have worked around them (and/or had the patents revoked) and that particular piece of FUD would have been rendered useless.

[nuked]

Re: Patent claims

@Condiment.

See icon.

[bailey86]

Ubuntu should have kept their focus

I think Canonical should have stayed focused on the Desktop and Laptop market. Gnome 2 should have been kept and just made leaner/cleaner/better.

Unity *may* pay off - and colleagues seem to like it - but I've changed to Xubuntu and this has been great for me.

If Canonical had stuck to the desktop only idea and kept with Gnome 2 they could have left Android, iOS and possibly Chrome to fight it out on the phone, phablet, tablet space.

[Liam Proven]

Re: Ubuntu should have kept their focus

Another commenter who appears not to have read the article.

Just to help you out - the purpose of the article was to explain *why* Canonical and GNOME rewrote their desktops *instead* of comtinuing work on GNOME 2.

Perhaps you might write this on a sticky note, slap it on your monitor and then read the article again, all the way from beginning to end, while glancing regularly at the note to remind yourself of why you're doing so.

HTH. HAND.

[MACWINLINO]

SOFTWARE SOFTWARE SOFTWARE

An idea that almost seems alien to people!

It is not an OS that makes it viable to users. It is the software that runs on it.

If the Linux community would release more alternative software that was on par with its WIndows OS X counterparts more people would use it. This type of common sense no one seems to get.

Do you think so many people would use OS X if it didn't have an array or decent applications? Same can be said for Windows.

[AJ MacLeod]

Re: SOFTWARE SOFTWARE SOFTWARE

I think you'll find that that the "Linux community" (letting that one go for now) has produced a wealth of "alternative" software, much of which is so effective it's entirely mainstream and in use on all the main platforms around today.

The Linux desktop makes sense for a lot of people and many of us have been happily using it every day to get our jobs done for years - into the decades now even. If you really think that "we" have some kind of desperate need to entice Windows and OSX users to switch sides, you're entirely mistaken...

[MACWINLINO]

Re: SOFTWARE SOFTWARE SOFTWARE

Name one alternative package to Outlook (Evolution) doesn't count. How about a product that is polished and on par with iWorks/MS Office? Alternative doesn't necessarily mean better, and still to this day I haven't seen a proper "alternative" to Active Directory.

"If you really think that "we" have some kind of desperate need to entice Windows and OSX users to switch sides, you're entirely mistaken..."

Nope I don't think that at all, I was referring to the article published.

Personally I don't care what OS people use.

I use Linux, Mac OS X and Windows

[AJ MacLeod]

Re: SOFTWARE SOFTWARE SOFTWARE

Well, the thing is - lots of people (including me) find Outlook horrible to use due to its everything-and-the-kitchen-sink nature. If you must have a calendar function embedded in your mail client you can use Thunderbird + Lightning which works fine, though personally I'm more than happy with a web-based calendar (lots of open source options) and Sylpheed / Claws for email.

Active Directory... what are you using it for though? It's a bit of a sprawling mess in my experience and the documentation isn't good. There are open source directory services of various sorts around, but if you want an alternative sprawling monster to control everything from email address books to wallpaper settings you probably won't find one.

LibreOffice is just as polished as Microsoft Office (which again isn't exactly ease-of-use perfection IMO, particularly when you try to do more than the odd letter) even if it isn't a 100% feature complete clone. iWorks is in my opinion not nearly as good as LO. I personally use LO or LyX (which to my mind is vastly superior for many tasks) for document preparation and have never come across anything that they couldn't do. If you want DTP, Scribus is pretty good, though I don't often want to work that way and MS' own DTP effort is rather abandoned feeling anyway.

I think your "problem" is that you're really looking for a 100% clone of one or two particular applications - you're not likely to find that, and you don't really know that your workflow with said application is actually as efficient as it's possible to get.

To sum up; the "Linux Desktop" is every bit as usable and efficient as "the Windows Desktop" or "OSX Desktop" for most people, assuming someone equal levels of IT support; there are applications (open and closed source) available to do almost everything that the vast majority of people want to do. Doubtless there are corner cases where that might not be true and some people have invested too much time and effort into particular applications to consider changing to any other way of working but that's their own problem... it works for me, it has worked for me for a decade and a half and if it doesn't suit everybody I don't see that being a problem for me.

[JEDIDIAH]

Re: SOFTWARE SOFTWARE SOFTWARE

> Name one alternative

Why? You will just find some weak excuse to discount them all just as idiots like you have always done.

It doesn't matter if the license is open or not or what the platform.

You can't get over your brand fixation. Quality actually has nothing to do with it. Never has.

Anything that is not the herd choice will get shouted down. It's like you're a pod person.

That's one of the reasons I dumped Windows. I didn't need pod people screeching at me for daring to use an alternative to the herd selected brand.

[Jim 59]

Article reasonably well written but seems poorly researched, exhibiting almost troll-like qualities on occasion.

"desktop computers are going to merge into tablet-style touch-driven devices and disappear."

- No. Tablet / touch is a new market distinct from the the mature desktop market. Both will continue for the foreseeable future. Unless you want to use a tablet to write Reg articles ?

"[GNOME] takes the Linux and open-source community down a path of fragmentation they seem only too willing to venture down"

- Having more than one product to do the same job is known (outside the Windows world) as "choice". Some people like it.

"[Linux] infringing a worryingly precise 235 patents"

- of which Microsoft declined to name a single one. All American tech companies have hundreds of supposed patent infringements with all other tech companies, but never prosecute. This is a direct consequence of daft American patent laws in need of reform, discussed on every tech forum over the years. Get thee to Wikipedia.

Every reader will love this howler:

"Everything that unified GNOME 2, KDE, Xfce and pretty much every other desktop GUI in the world originally comes from Windows 95."

Er, no.

[Peter Simpson 1]

Er, no, indeed.

Motif window manager was my favorite. I think I even *purchased* a copy to run on an early Linux release.

I'm pretty sure that was pre-Win95. Before that I had run it on my Sun workstation.

I notice nobody has brought up the window managers or "root menu" on Unix systems. Linux didn't (necessarily) want to look like Win95, it wanted to look like Unix. At least, that's why I started using it.

[Anonymous Coward]

Interesting article, elegantly written. Staggeringly display of ignorance in the comments though - maybe Reg readers should stick to commenting on code- and science-related articles, as they certainly don't seem to have much understanding of UI history.

Personally I've never come close to grasping why Linux developers spend so much time and energy constantly recreating the Windows 95 interface over and over again. It's a terribly sad waste of resources.

[Phill 3]

235 patents & laches

If the claim of 235 infringing patents has now been rendered unenforceable under the legal principal of laches, then why has Foxconn, the biggest hardware manufacturer in the world, just signed a licensing agreement to pay Microsoft? Personally I was baffled as to why they would even before hearing of 'laches'.

[eulampios]

Re: 235 patents & laches

Foxconn, the biggest hardware manufacturer in the world, just signed a licensing agreement to pay Microsoft?

So, who is paying who and how much, what is covered? Until you successfully found all solutions of an equation you can't claim you solved it.

[JEDIDIAH]

Re: 235 patents & laches

> why has Foxconn, the biggest hardware manufacturer in the world, just signed a licensing agreement to pay Microsoft?

Appeasement usually seems like the easier, lest costly option.

[Anonymous Coward]

Re: Choices choices...

So then you get 10 developers working on 10 products used by 1 person each, instead of 10 developers working on 1 product used by 1000 people. Great.

[eulampios]

@dz-015 Re: Choices choices...

Or you have a dozen of developers and thousands of managers working on the NT kernel, while you find thousands of developers and a dozen of managers working on the Linux kernel. A small permutation indeed.

[Wardy01]

Success is a matter of opinion

Depending on who you talk to each person will say "this company / group is the best / winning".

It's all subjective and irrelevant.

Given that ...

Microsoft accept that they have their way of doing things and will do things their way no matter what, this results in progress being made and gives them this "on, off, on, off" habit in windows version like, dislike pattern out there.

Linux, like every other open source project out there is all about decision by committee and every stupid moron out there that has an opinion seems to get a say no matter what resulting in an endless loop of testing 1 fail after another.

It's about time open source projects started working a bit like companies and have some level of structured plan that leads down a defined path so instead of being stuck in the dark ages things can move on.

That's Linux's biggest problem.

Just build it and tell people to suck it up like Microsoft does!

Notice how the biggest complainers about "start buttons" or "tiles" are Linux lovers?

Yeh they can stay over there and mess with software that will always fail to be in the limelight.

As for "Linux is popular" (comment by someone else):

Android was built by Google not the Linux community, Google is a company, driven and with a precise path.

This is why GNOME will never be any good, it lacks direction from the leadership down.

Ubuntu however ... stands a good chance of really making it as a consumer OS IMO, it just needs a little cleaning up, well built with a direction and another plan based project, and why: canonical get it!

It's about time the Linux community started to grow up and stop thinking they own everything and have the right to demand crap backward changes every 5 minutes, when that happens we will seriously start to see Articles like this about Linux!

I feel for Linus Tolvolds having to put up with the crap he must get daily.

[Joe Gurman]

Linux conquering the desktop?

In what year and in what world? Ever?

Making such wildly unrealistic claims is yet another "feature" of the pathology of the Linux culture. That's culture as in Petri dish, as opposed to the culture of dedicated Linux devs.

[Anonymous Coward]

Yep, I've tried Linux approx. ever year since it was going to be the "new thing", and the GUI has always been lame, either a muddle or a poor knock-off of W95

[Protemo]

Why is the "laches" doctrine not applicable to submarine patents ?

[BadMoles]

Symantec did it before Micro$oft....

I was working for Symantec UK in the early nineties, we had the Norton Desktop for Windows running on Win 3.11 which gave you most of the nice UI features you later got in Win95.

[ecofeco]

Re: Symantec did it before Micro$oft....

Good call! I had forgotten all about that. But I don't think Norton was owned by Symantec at that time. (honestly not sure. maybe they had just been bought out.)

[Stevie]

Bah!

I will get excoriated for saying so, but I tried all these alternatives out at the time and I can tell you that although superficially similar in many ways to the windows 95 desktop, the other guys were missing the point by miles.

What was truly a game changer in Windows 95 was the context-sensitive right-click menus. They gave the pc-ophobe (me, in 1995) a way to educate myself in what I needed to do to get this blasted toy computer I'd been given out of the way so I could do my work (mainframe database administration) in a job I had had all of two hours.

That it took me 30 minutes to do so from "Oh christ, I've lost this job before I start" to "$$OPEN" - the start of my real work is a testament to the underlying design *whether or not the actual OS is/was crap from a comparative technology standpoint*.

That none of the alternative OS/desktops offered such richness was a real non-starter for me. I don't do operating systems for fun. They are there to allow me to do other stuff.

[ecofeco]

And they all stole from Xerox PARC

Did I also mention OS2?

Which came first? Warp or 95?

[Bob Camp]

This is a bad article. So Microsoft never sued, there's no evidence that the threat to sue actually changed the behavior of anybody, we don't even know what the patents were, yet somehow it's all Microsoft's fault? Maybe the tin foil hat is on a little too tight.

Let's face it, desktop Linux was/is not successful because you can't just build a good product and hope people will come. You have to go to every PC manufacturer, hardware developer, and software developer and SELL it. Microsoft did a great job at that.

Plus, even if the infringing features were ripped out later, Microsoft could have sued for damages caused by the previous versions. And according to the article, GNOME didn't radically change their GUI until 2011! Clearly the threat of a lawsuit way back in 2007 wasn't the main reason for the change.

[Major Variola]

TWM has a task bar

Tom's Window Manager for the XWS has an optional bar allowing you to select windows, kill tasks, etc.

That predates all this newfangled stuff

[Will Godfrey]

@Liam

Rule 1: If you want to be taken seriously as a writer never insult your readers.

Rule 2: See rule 1.

[Anonymous Coward]

open-source copied Windows 95?

"GNOME Classic mode brings to six the total number of GNOME desktops and takes the Linux and open-source community down a path of fragmentation they seem only too willing to venture down".

This is BS, most Windows 95 users would never have heard of Gnome or Linux, considering (dues restrictive and onerous Microsoft contracts with the OEMs) it's virtually impossible to buy it ready-installed on the high street.

[Anonymous Coward]

Fork Fork Fork

People "at home" don't want Linux - I deal with these people every day.

Businesses don't want to upset user bases and are 90% slow to update to even a new version of Windows.

Most will use Windows. I have worked in IT in multinationals.

Windows is the Magnolia of the OS World. You can preach revolution all you want, like an imam saying that the whole planet will become islamic - but it ain't happening.

To expect linux to take over desktops everywhere is like expecting people to stop using their cars and use unicycles instead......but what do I know? Has anyone tried asking non-techie computers users (apart from me)?

[Marshalltown]

Win95? - bah

I suspect the author may not be old enough to remember clearly. At the time Win95 came out, the most notable thing about the interface was the slavish way it copied OS/2 (3.0 IIRC). I actually had an argument with a Winfant that ended when I showed him my system and explained just how long it had been installed. Win95 was not particularly attractive. The capabilities of the interface were comparatively limited. It didn't have REXX. If it had a BSOD hick up, reinstallation might be your only choice, while with OS/2 you could step back through configuration changes and software installations until things worked. Files didn't remember what created them. The text editor didn't suck, and you could easily shift to another if you liked. Yeah, you can do that in Windows now, sort of, after a fashion. The thing is, it was default behavior in OS/2 in 1994! Unless you had a hardware failure, you never lost your system. Files remembered what created them. The text editor didn't suck, and you could easily change the default to another if you liked.

[eulampios]

are you serious?

but in 2007 there were only a few complete desktop environments – KDE and GNOME being the main contenders.

Let's see here LXDE 2006, XFCE (started in 1997, using gtk+ since 99),fluxbox was forked from blackbox in 2001, enlightenment started in 1997. So, what was the point of such misinformation?

So let's consider what they have in common: windows, obviously, with a menu bar inside each one.

So. let's consider windows or a window, did MS pioneer that? Who holds the patent and can subsequently sue MS?

Everything that unified GNOME 2, KDE, Xfce and pretty much every other desktop GUI in the world originally comes from Windows 95.

Are you suggesting that a Unix based kernel with a POSIX userland, multi-user paradigm, privileges, file permissions on fs, efficient multitasking etc came from Windows 95? This is a joke in response to yours ;-)

[AdamWill]

Reasoning a long way ahead of its evidence

This post seems to reason a long way in advance of its evidence.

It seems to attempt to tie Microsoft's 2007 patent sabre rattling to the desktop specifically, but there doesn't appear to be any evidence for that. The apparent quotation - "the Linux desktop including OpenOffice" - is not sourced, and so far as I can tell, neither the Reg article it links to nor the source *that* article links to actually contains that text. The source and my memory both indicate that Microsoft was talking about Linux or even F/OSS in general, with no specific focus on 'the desktop'.

Note that the GNOME "Big Board" effort and the independent 'Gimmie' project, which together basically kickstarted the discussion and prototyping work which ultimately lead to the development of Shell, both date to early 2007 (April or earlier): *before* Microsoft's sabre-rattling in May 2007. See http://blog.ometer.com/2007/04/03/gnome-online-desktop/ , which pretty much marks the start of the "Online Desktop" / "Big Board" efforts, and http://www.ubuntugeek.com/gimmie-a-new-panel-for-gnome-installation-in-ubuntu.html , an article about an early version of Gimmie which is dated January 2007.

I don't see any compelling evidence to suggest that the long process of discussion, design and prototyping which started in early 2007 and ultimately led to the development of GNOME Shell and GNOME 3.0 has anything to do with Microsoft's May 2007 patent sabre rattling.

[eldergeek]

These sorts of articles are annoying.

I've been a full-time Linux desktop user since 2006. I'm not an IT guy, didn't go to college (electronics trade school in fact but had no use for computers at the time) and didn't get online until early 1999, at which time I had a 256 MHz, P2 which consumed virtually all of my expendable income in purchasing maintenance, internet service, extra RAM, faster dial-up modems, spiffier sound cards, etc., ad nauseum. In 1990, my first (used) Windows machine (which never went online at all) cost me over 800 U.S. dollars and, even then, wasn't a particularly impressive machine.

At the moment, I'm typing in Firefox on an HP, Compaq 6710b laptop running Kubuntu 12.04. I bought it used for a little over 200 U.S. dollars and it's faster, with Kubuntu, than the much newer Windows Vista and 7 machines which I'm often called on to repair for my neighbors and friends. It's very annoying to read an article every few days which authoritatively informs me exactly how and why Linux is "never" going to make it on the desktop. I'm beginning to wonder if these sorts of articles haven't been written "to order", as it were, and paid for under-the-table by Microsoft as a component of their ongoing Fear, Uncertainty and Doubt (FUD) campaign against the Open Source world at large.

I realize that leading edge computing is in a state of rapid transition but anyone who believes that Linux isn't going to be "all up in the middle" of it is kidding themselves. I'll remind you that Android, which is but another form of Linux, virtually owns the smart phone market. Meanwhile, there are a lot of folks like me who aren't affluent enough to invest in leading edge computing systems while there are plenty of used P4 systems out there which are priced within virtually anyone's budget and will run Linux very nicely. In many large cities there are organizations which will provide underpriveleged families with a Linux system and pre-paid, high-speed internet service for free. I myself have given away about four dozen Linux systems (though folks had to buy their own internet service) since 2006. I've also spent less on computer hardware & software since 2006 than I spent per annum in the preceding decade and that includes the cost of the systems I gave away.

While you, and the apparent army of literate, like-thinkers continue to write authoritative articles telling me how Linux isn't ever going to make it on the desktop, every few days, I'm just going to keep on using, and very much enjoying, Linux.

Ask me how much I've spent on security (firewall, anti-virus & anti-malware) software since 2006. Zip, nada, nothing amigo (that's no dollars and no cents) and I've yet to have a virus or malware infect my computer system. It's the same amount I've spent on my desktop Operating System during the same time period and numerous folks who liked the idea of having their old computer become a boon to a student or a senior citizen have given me entire systems.

I think my $.02 is worth considerably more than your opinion. Nuff said.

[Wardy01]

Re: These sorts of articles are annoying.

My main use of my home computer is gaming and game development, an area that today I still feel linux isn't quite as strong as windows in (although admittedly is very close since steams push).

Whilst I agree with virtually every statement you make, I don't think linux is quite the "1 shoe fits all" type solution that windows has been all these years.

It's well known that Linux until android has basically been the domain of tech savvy user and for good reason, it's a different mind set and works in a tech savvy way.

An interesting question might be, how long before Linux gets a good clean standard for touch at a low level?

Considering Linux in general still has ongoing debate over desktops and desktop functionality I can see it being some time, this I believe to be Linux biggest failing.

Note:

When I refer to "Linux" in this comment I mean "Linux Distributions" in the general sense, because after all "Linux" is not an operating system, it is merely a kernel on which OS code sits.

[Oligova]

..what do you mean "quite old" ????

I remember GEM I remember devloping for win311 in the OS/2 win emulation as it waas more stable then win95 and I remember a lo of other stuff. So if I can remember all that how con you call me old old stuff doesn't have that much memory. pfff

:D

[sisk]

You say fragmentation, I say choice

I suppose the end result is the same either way. Desktop Linux remains the domain of those willing to take the time to learn how to use it and always will, so it will never have a chance to conquer the desktop. Everyone else remains at the mercy of Microsoft and Apple.

[RDW]

Microsoft faked the moon landings!

Did Gnome, KDE, Xfce and everyone else originally copy Windows 95? Yes, of course they did. This is blindingly obvious and hardly worth writing an article about. Do we need to invent a silly conspiracy theory to explain the current 'fragmentation' of the Linux desktop? Not so much. Like most conspiracy theories, this one falls apart on close (or even cursory) examination:

- What did Microsoft claim when they made their dubious statment about 235 patents? From a random CNN article:

http://money.cnn.com/magazines/fortune/fortune_archive/2007/05/28/100033867/

"But he does break down the total number allegedly violated - 235 - into categories. He says that the Linux kernel - the deepest layer of the free operating system, which interacts most directly with the computer hardware - violates 42 Microsoft patents. The Linux graphical user interfaces - essentially, the way design elements like menus and toolbars are set up - run afoul of another 65, he claims. The Open Office suite of programs, which is analogous to Microsoft Office, infringes 45 more. E-mail programs infringe 15, while other assorted FOSS programs allegedly transgress 68. "

So even if you throw away all Windows-style GUI elements completely, if MS is to be believed (hah!) that still leaves 170 patents they can supposedly sue your favourite Linux distribution over (and MS would presumably claim that even Unity infringes some of their 65 GUI patents, so you're probably still dealing with a couple of hundred). Redesigning the GUI doesn't make you substantially less 'liable' (if you actually buy into the MS FUD).

- If a Linux distribution switches to a new primary desktop, does that actually mean it's no longer 'infringing' the supposed GUI patents? Not really. 'Classic' modes that look suspiciously like Windows 95 are still available as lawyer bait, as are alternative 'traditional' desktops like Xfce in the distro's repository. The legal threat, if there is one, is no less than before. Incidentally, Redhat, which never signed a patent deal with MS, chose Gnome 2 as the default RHEL 6 desktop in 2010, 3 years after MS's posturing about the patents, and is still using it today, which ought to tell you something about exactly how seriously RH took the threats.

- Is there a more plausible explanation for the recent proliferation of DEs? Yes. Gnome developers got bored with the Win95 style GUI, thought they could do better, and were arrogant enough to ignore the wishes of a large proportion of their users by (partially) ditching the old interface. Canonical got upset by this, and re-purposed a netbook GUI as their primary desktop in a bid to differentiate Ubuntu from everyone else and make an interface that was equally irritating on a wide range of devices. All the other projects mentioned (MATE, Cinnamon, etc.) are simply attempts to restore sanity by returning to the Win95/Gnome 2 'metaphor' by one means or another. If Gnome had not deprecated the traditional GUI, nobody would have bothered forking or emulating Gnome 2. There's nothing about Unity and Gnome 3 that can't be explained by hubris, obsessing over tablets, and more or less misguided attempts to re-invent the wheel. MS itself has recently caught the same disease, but there seems little risk of anyone copying Windows 8. When someone (like Gnome!) pulls a stunt like this in the FLOSS world, the natural instinct is to fork or write something new, which is exactly what we've seen with MATE and the various attempts to tame Gnome 3 with alternate shells or addons. Projects then proliferate until natural selection kills the less viable alternatives off.

Software patents and threats of legal action have a lot to answer for, but fragmentation of Linux desktops? Yeah, maybe it was a 'controlled demolition', as the 911 conspiracy nuts say.

[RLWatkins]

CUA

I hate to bring this up,but they didn't descend from Win95, they descended from CUA, part of a standard which predated Win95 by a decade or so.

[sisk]

I'm quite certain that Linux (and every other OS on the planet for that matter) does indeed violate some of Microsoft's patents. I'm also quite certain that those patents wouldn't stand up in court. Just as an example, Microsoft holds a patent on deleting files. What judge isn't going to laugh that out of court if they tried to enforce it? I also strongly suspect that's the reason they never revealed which 235 patents were being violated.