Microsoft's Must-Have List for Windows 9 | Hacker News Reader

Microsoft's Must-Have List for Windows 9

pcmag.com

23 pointsmichaelwww71 comments

The "miserable counter-productive "ribbon" interface on Microsoft Word" was the result of a lot of actual user testing. I could understand people disliking it just because it was new, but it's been years now... How long does it take people to realize a better interface is more productive?

Edit: oh I just noticed it's Dvorak. Never mind.

This is pretty much my exact reaction to the article. That was the exact line that gave me pause, then I looked up at the byline and saw Dvorak's name.

I do find his pining for the failed WinFS particularly odd, though. It's something that was pretty much intended to back up the kind of anti-desktop model he's complaining about.

I feel so much better now, I did the same thing. It's such a weird list, too. Doesn't he know about putting more than one app on the screen in Windows 8.1?

The new fs is all about putting your data first. The new UI is all about putting apps first.

CAN WE GET RID OF ALL CAPS IN MENUS?

All caps is yet another poor choice. Microsoft seems to have a knack for deteriorating as many things as it improves in each new version of Office or Windows.

I've been using the ribbon UI in Office frequently for years, and still find its usability inferior to the old menu system. Aside from its poor usability, I dislike the visual clutter of having all the functions visible all the time. Personally, I'm less productive thanks to the ribbon UI.

Well, what do I know... I'm just one data point... probably an outlier.

You know you can disable the ribbon so it's not shown unless you click the "drop" it like you would a menu, right?

The thing I hate about the ribbon is that it's not linear - you can't scan it like a drop-down menu. It's a lot harder to find something if you don't already know where it is.

IME there is a strong divide in opinions on the Ribbon interface. Almost all "everyday users" I know came down in favour after getting used to it. Most of the "power users" found it frustrating.

Given that the former group is much, much larger in practice, it's hard to blame Microsoft for making the switch, even if it does invite strong and vocal criticism from some people with legitimate gripes. I suppose this is the inevitable burden of designing some the most widely used business software on the planet.

Honestly, I think MS should've learned from what they did on the desktop: just give me a big searchable menu and the power to pin things onto the bar. Everybody's workflow is different.

And while the groupings of various formatting and editing tools is sensible, their constant mucking with the "File" menu is inscrutable and frustrating.

I used to always make a custom toolbar with the items I frequently used. Then later, when I had to, I would go into the menu to find the other items I needed. It was nice having familiar icons for my most-used functions and then having text-based labels neatly organized in the menu for the other functions.

The way the ribbon had it all (or lots of items I didn't regularly use) always visible irritated me and made it hard to quickly find the things I wanted without wading through the things I didn't. I've gotten used to it somewhat and I no longer keep Office 2003 installed, but I still end up searching for ways to do things now and again.

You can still add a bunch of functions to the "quick access toolbar" which is still open at the top of the screen even if you minimize the ribbon.

I know, I am a dinosaur that forgot to die but... you'll take Office 2007 from my cold dead hands. My wife, however, left it because it doesn't open newer Word formats; she uses OpenOffice.

I really still didn't get the "better" on the Ribbon interface.

Wait, how do you go from "result of a lot of actual user testing" (PM who had the new ribbon designed, sold internally and implemented hosts and evaluates "user testing") to treating it as a truth?

My impression is that by the time companies are large enough to have "actual user testing" going on, they have lost the ability to make something truly nice and coherent, the naive "ribbon" being a prime example. I know it's been years, but I haven't gotten my gunslinger muscle memory from the old menu-days back.

The menu system was hardly "coherent". At least functions are nicely sorted into meaningful groups now.

The new Ribbon is basically a menu that is built out of extendable XML parts that can grow and shrink on context.

I don't see why its bad at all frankly. It is most definitely a win from a user-experience perspective. Even more so... its a win from a developer perspective.

A single XML file can serve as a consistent menu across old-style C++ COM programs or even be used in the new .NET / C# programs. You no longer have to worry about resizing icons to the user's screen, Ribbons automatically stretch or shrink as necessary.

Just from an API perspective, they are far more powerful and consistent way to communicate between the user and the program.

--------------

Menus of old can be well designed, but they won't be consistently designed across all programs.

That said, the older menu system of MS Word was well designed. The real issue was that they totally reorganized everything in a feeble attempt to get people to start using the more advanced style options of MS Word.

"Properly" using MS Word involves using it more like LaTeX or HTML: you use semantic markup, and then later style your document. (or use a style that is consistent with the rest of your office).

But people would rather just hit the bold button or italics button. They don't want automatically generated table of contents or figure charts... they want to manually type that stuff in... like a typewritter from the 80s or something.

I don't know, I can agree that the new Ribbon interface makes it harder to do certain things in Word, but that was unfortunately the purpose. People need to learn the semantic-markup style of Word and start using better methodologies to organize their Word Documents. But no one seems to put in the time or effort to learn how to use the programs they use every day.

Anyway, I've put in the effort to learn the "new way" Microsoft wants you to use MS Word. And I personally agree with Microsoft, documents are much more consistent, and much easier to update styles consistently across the whole document when you focus on styles first and foremost. Hiding the Bold and Italics buttons (and several other popular... but "wrong" features) was a wakeup call indeed.

The problem is that most users don't like being told that "they're using the program wrong". Part of it is truth (really, the new way is better). Part of it is business based though (Using a document editor as a fancy typewritter means you're ignoring a lot of useful features... and are probably better off using the much simpler Google Docs, or OSS like Abiword)

I stopped using MS Office entirely with the introduction of the ribbon. Granted, that's about when OpenOffice.org was about on par. I now use LibreOffice which I find, in many ways, superior to Microsoft Office.

And don't even get me started on the complete idiocy of ALL CAPS MENUS. It's demonstrably proven that all caps has a big negative impact on readability. To include ALL CAPS as part of the UI is just asinine.

Is it? Shortcuts are not discoverable anymore, most actios are hidden within tiny icons (or dialogs full of tabs, oppened from actions hidden within tiny icons), often there is no logical grouping of the actions (ok, maybe the old menu wasn't better), ribbons appear and disappear out of nowhere, and there are hiden ribons that take a Google search to get.

I quite like the "ribbon" idea, but the Office's implementation? I see no reason to think that it's more productive, and I certainly do not feel more productive on it even after years of training.

By the way, MS also had lots of actual user testing supporting their decision to take the start menu away and maximizing every window.

Am I the only one who hates the old Start Menu? I find it tedious and useless. I'm all for re-adding a start menu, but it should be one that doesn't suck, like the one in Kubuntu.

Also absent from this list is an updated console. Windows 8 still uses the old DOS shell. You can't even resize the window past 80 characters. It's embarrassing.

>You can't even resize the window past 80 characters.

Or, you can change your defaults...

Right click the top bar > Defaults Window Size > Width > 'What ever you want'

Oh, and use powershell if you want more.

Windows shells and the cmd.exe are a crap when you compare against anything that you have in GNU/Linux or *BSD.

bash, zsh and fish are far better and more intuitive that Powershell. Plus cmd.exe is a joke if you compare against Konsole, Yakuake, eterm, etc....

I think it's terrible as well. I have 2 21" monitors and I need to look for my apps in the bottom left corner of one monitor using tiny icons. It could be improved greatly, and still be in Desktop mode. Yes, the terminal is terrible as well. And the new Powershell can't be easily resized either.

Microsoft is trying to push Powershell as a replacement for venerable old cmd, but it's not really a good one - Powershell is an attempt to bolt OOP onto the shell language and interop with the .NET framework.

The problem is that it's horribly confusing to work with - the syntax is bizarrely inconsistent as you switch back and forth between "commandlets" and .NET methods, and the part of the power of plain text is that it's self-documenting and forces simplicity of API. Powershell allows full OOP, which doesn't map well to the need for things to be simple when doing command-line scripts. Also, the Powershell interface lacks the various tools you'd expect from an OOP IDE that let you learn about the objects/methods/whatever.

I do like the absense of the start menu. To me, the big failing of Win8 wasn't replacing the start menu with a fullscreen pseudo-desktop, it was the schizophrenic relationship between the fullscreen pseudo-desktop and the "classic" apps, as well as the bizarre "charms" things and the lack of a conventional taskbar when using fullscreen apps.

Desktops and larger notebooks have enough real-estate on the screen for an always-visible task-switcher, and the "Charms" thing is too touch-oriented.

It's a bit off-topic, but you can actually make the window larger than 80 characters if you want. You cannot expand it beyond the maximum amount of characters horizontally but you can increase that to anything you want; it's not limited to 80, really. :)

More on-topic: I don't quite hate the Windows 7 start menu, but it could be made better. Unfortunately, Microsoft chose not to improve on a good concept and, in my opinion, ruin things with their new interace. It is one of the main reasons I don't want to switch to Windows 8. I've been a Linux user at home for a long time now, though, it's thankfully only my work laptop that I need to have Windows on. I have a game computer that has Windows XP and will probably be upgraded to Windows 8 but I couldn't really care less about the interface there: I just want to start games on that machine, nothing else.

You can't even resize the window past 80

You can if you know those good ol' mode-commands. But yeah, no drag'n'resize like on all other platforms unless you install third party software like MinTTY and ConEmu.

You not need to use "mode" command to do it. If you search in the options of cmd.exe, there is a way to set the width in characters, but cmd.exe can't remember it for the next time that you open it (so is useless by laziness). ConEmu is far better that cmd.exe, but I really wish having Konsole or Yakuake running in Windows.

What did you miss in ConEmu, interesting?

The old start menu may not be perfect but the new start screen is definitely a step backwards. Over the life of a Windows installation, many, many programs tend to be installed, and almost every single one of them dumps a bunch of icons onto the start screen, which quickly kills its utility for visually identifying an application and launching it. After a while there is so much crap that the only way I ultimately launch anything is by typing to search. At least the old start menu had the concept of hierarchy, which meant I could identify a single start folder for an application, rather than having to wade through the vomit of 3-6 secondary icons that every installer also feels the need to dump on the start screen.

8.1 doesn't push new apps to the screen by default. They appear on the "all apps" screen, where they are indeed grouped (by the old start menu folder presumably, e.g. "Microsoft Office").

Also you can remove apps you don't want from the start screen. It's not supposed to include everything on your PC. It's supposed to be the things you use frequently. They only did the auto-pinning in 8 to increase discoverability, but it added more clutter than value.

Am I the only one who hates the old Start Menu?

I didn't hate it, but in Windows 7 I found a better alternative that almost made it redundant.

I have found myself pinning the 20-or-so programs I most often use to the new-style Taskbar instead. I also use Jump Lists and various related tricks all the time. I frequently run many applications at once, I open lots of command prompts and Explorer windows, and with 4 megapixels of real estate on my primary monitor alone, I want these functions one click away a lot more than I need a few extra pixels across the bottom of my screen.

Given that easier-to-use alternative, about the only things that still make me open the Start menu on Windows 7 are searching for programs I don't use regularly and accessing control panels.

My biggest pet peeves in the Windows 7 UI are also dominated by how quickly I can get things loaded so I can start doing something useful:

1. You can't pin starting directories for command prompts the same way you can for Explorer windows.

2. Opening a command prompt from an Explorer window isn't on the context menu by default.

3. You can't pin a Control Panel icon that pops up a menu for accessing the various panels.

4. Running things as Administrator is still clunky, particularly command prompts.

5. I often want to start some combination of applications I use frequently, position their windows in the same way, and then give a specific program the focus ready to start work. This ought to be a trivial use of any operating system's scripting facilities, but it is difficult-to-impossible to set this up for access with a single click from the Taskbar without writing real code using system APIs to do it.

If Microsoft fixed these frustrations and otherwise left Windows 7 exactly as it was, I would buy Windows 9 at full price for every PC I use in a heartbeat. There are several other things they could do that would make me pay much, much more: grown-up command line tools, a robust and standardised framework for installing/updating/uninstalling applications, a more transparent and robust model for controlling access to system resources -- basically addressing the major weaknesses that Windows has relative to Linux or OS X today for power users.

In contrast, I can't imagine any development of the direction they've been going with Windows 8 that would motivate me to run it, and we actively seek vendors who will provide Windows 7 instead for our new computers at present.

I didn't read the by-line initially, got about two paragraphs in and thought "Jesus, is this by Dvorak?" And it is. And I smiled.

I remember howling with laughter at his anti-mac articles more than a decade ago, and this is funny for the same reason. Dvorak has always been utterly incapable of thinking from any perspective other than his own. For example, his hatred of the full screen mode for applications is almost certainly born from being a power user with a multi-monitor setup or a large hi-res display (that's how I see things too, but I also understand that the UI needs to work on things like a Surface or an ultrabook). Oh well, it was fun reading him again anyway.

As many other people have pointed out here, the main thing MS needs to fix is the pricing point and the ridiculous number of options. It's an OS, stop with the Ultimate Edition nonsense and just sell one version of it at the lowest price point that works.

They did. There are two versions of (client-side) Windows 8: Home and Pro.

Almost everyone wants the cheaper Home edition. Businesses who want Group Policy and Domain Joining will spend the $50 extra for Pro.

Allow me to take this opportunity to address something related to a minor point from the article:

>To me an app is a small bit of code for an iPhone to do something cute. Are we going to call the massive SAP suite an "app?"

Lately I've been hearing many people complain about "serious" desktop applications being called "apps". Do you think this is a valid complaint? (Myself, I'm not sure; I won't disclose my reasoning right now so as not to skew the conversation from the start.)

Interestingly this is something I was thinking about the other day. I was trying to explain to my parents that they wanted to download and run a program on their Windows company - I called it an "app". They were confused as that is the term for things that you install on a smart phone.

Like yourself, I'm not sure. It does seem a bit of a disservice to call something as complex as AutoCAD or Photoshop etc by the same term as a stupid fart app. But then, it is an accurate term - they are both applications, so I tend to use it as such.

In my mind, app is simply short for application. I think it has good connotations that come along with it as a result of mobile apps, such as being a single contained package that doesn't interfere with other apps, but we've seen that on OS X for years anyways, so that's hardly something new to desktop computing. There are complex, CPU intensive apps that are developed for tablets to do things like 3D modeling, are those any less of an app than a desktop version of the same thing? Does anyone have a problem with those being called "apps" vs "desktop applications"?

There was a time that I would have agreed that an "app" is, in essence, a minimal application designed to perform a simple, narrowly circumscribed task. The distinction was natural because this was about the level of sophistication of most software that the average user would encounter on a mobile device, all of which were sold through "app stores." (And app stores sold only apps, thus defined.)

Since then, however, I have found that the definition has more to do with the purchase/installation process than the functionality of the product itself. The emerging definition that I perceive is that an "app" is something purchased and installed through an "app store." App stores are environments that greatly reduce the normal friction involved with purchasing and installing an application such that this can all be accomplished without leaving the app store interface and without the user's being required to enter any configuration information. (This step, if it is necessary, is typically reserved for when the app is first run.) Uninstallation of an app is similarly streamlined (or, at least, a user expects it to be...). Apps are also not typically conceived of as running in windows. They are presented full-screen with no OS-chrome. This is an obvious hold-over from the demands of the mobile-device environment.

An obvious counterpoint is Steam, which would seem to meet all of these criteria. This anomaly is perhaps explainable on the basis that it pre-dates the modern app store and, thus, users' expectations of it are not framed by the mobile environment as they are in other ecosystems. The fact that it typically is used only for PC games helps with this differentiation. It is worth pointing out, though, that Steam does sell non-game software, though to my knowledge it is not terribly successful in doing so. This lack of success might be attributable to its failure, in this capacity, to comply with the symbology of the app/application distinction that users have come to internalize.

No, I think that complaint is ridiculous. They were called apps long before the iPhone came around.

Depends on who's talking.

Amongst developers, using the word 'app' to describe any kind of software seems like the navy referring to everything as a 'boat'. Using more specific words improves the conversation and our thought processes.

But when talking with end users, it seems to make sense to classify their system into 'Apps' and 'files'. The average user still gets confused by the difference between Windows, Word and a Word Document. If saying "Apps are things you find in an App Store" helps end users be more effective with their computers, then it makes sense.

At least in Mac OS, they have always been called Apps, and with Mac OS X (2001) they even adopted the ".app" extension (Mail.app, Safari.app.) Steve Jobs used to call them apps in public all they way back to the 80s, even during the NeXT days when they only focused on 'serious' enterprise software.

Windows users call them programs and I guess after the 'there's an app for that' campaign, everyone associates apps with smartphones but in reality the term is almost as old as the PC.

The .app extension has been around since the first preview release of NeXTSTEP (1988) [1].

[1] http://www.flickr.com/photos/15578792@N02/2461295650/in/phot...

Rarely would I agree with Dvorak, but his conclusions here aren't too far wrong.

If you live with a Chrome OS device and Windows machines rapidly the Windows machines get relegated to "legacy apps only" status (this coming from someone that really doesn't like the overall Google direction). MS need to face the fact they've lost the client war, and focus on their true killer apps of Excel and Exchange. Exchange should be the core of their business, not Windows.

They probably look at Blackberry as a lesson of what happens when a company focuses on Enterprise and not Consumer.

Consumers come to work and bug their IT departments to use what they are familiar with, forcing the pure enterprise players into obsolescence.

That's not quite right, because as long as Exchange plays well with the devices people are using who cares what it's running on? The money is made selling the server anyway.

iOS adoption in big companies has been driven, to a large part, by the fact integration with MS servers wasn't so hard. MS have been playing Windows platform lockdown for too long and, with the exception of the Exchange group, forgotten the embrace and extend idea, which currently finds itself at home more in Mountain View.

A nice rant from John C. Dvorak.

First though surely MS have to work out what a Windows PC is for in 2014 and beyond. (I accept their true response would be something like "make money on every PC sold like we always did").

Chromebooks have demonstrated a minimalist PC for our age. Now what value can Windows deliver on top of that?

I think that is difficult to answer from a consumer's point of view. There might be a business angle but (for me) it is still not clear.

Ms Office, really. Take that away and businesses will move to other OSes. Everybody is tired of the constant poor quality of Microsoft products.

I've made this argument before, but I've always found Microsoft's recent obsession with killing the desktop as hugely harmful to their own business.

The fact is, Microsoft has seen this future coming for some time. They got themselves into antitrust trouble for going after Netscape and Java so aggressively in the 90s because they were afraid of OSes becoming commodity software.

So, how have they reacted to a world where operating systems are increasingly commodity software? By throwing their competitive advantages in their bread and butter market -- the enterprise -- out the window (no pun intended).

Applications have to be rewritten to run on Metro (and thus the cheap RT devices they so desperately want to compete with the iPad), central management of RT devices is not any real improvement over other devices, and whereas in the past they made the mistake of trying to shoehorn a desktop OS onto a touch device (e.g., XP Tablet Edition), now they've gone the complete opposite direction and tried to make a touch-centric interface designed for small screens in a world where 24-27" displays aren't uncommon.

While Microsoft needed to do something to stay relevant, they've so badly mismanaged the transition that I think the end result is going to be an acceleration in line of business applications moving to web applications that run anywhere. Again, that's something that's already happening, but the entire W8/RT thing has done Microsoft absolutely no favors.

It's funny that a company can be so lost while there are so many technology folks and enthusiast that are clearly laying out the solutions for them. It's going to be very interesting to see what happens to this company over the next 10 years.

Except that those "solutions" might not necessarily work either. I had even read on HN people arguing that the solution would be MS to fork android (something that we had seen already) or just stay still and don't do anything (again we had seen that before too)

The way I see it, MS started to get lost once they "accomplished" their mission "computer on every desk". At that point they were just like "Now.. what?".

>"Chromebooks have demonstrated a minimalist PC for our age. Now what value can Windows deliver on top of that?"

It depends on your definition for "value" here. For instance, for me "value" would be to create content in a comfortable way even if I am offline. So, since I can't edit a video, develop an app or edit a photo being offline during my commute to work, it is very unlikely that I will get one, and hence they have very limited value "to me".

6. When you hit the windows button, and then start typing the program you want to launch, DON'T FUCKING SEARCH THE FUCKING INTERNET AND SLOW THE WHOLE PROCESS DOWN. Jesus.

I just double checked and I don't have any issue with this in 8.1.

Microsoft's Must-Have List for Windows 9:

1. Customers

He missed a big one: price. It needs to come down. Windows is now much too large a percentage of the PC price.

I haven't read a John C. Dvorak column for ages! We're talking over 10+ years ago - I'd no idea he was still writing. It'd be hilarious to have Dvorak running GNU/Linux at the end of 2014!!!

I recently started reading him again after a 10 year break and he is crankier than ever and off-base much of the time, but I thought he was right on with this article. It would be really fun if he switched to Linux desktop and started writing about it. He'd be doing a great public service!

jeez., basically this list is all about reverting to the old stuff without really suggesting anything new both UI-wise/UX-wise or the fundamental aspect of what to expect of an OS in 2014/2015....

https://www.readability.com/articles/6w09uyph

1. Fix the file system. The company promised it would develop a database-style file system that would allow for easy and super-fast searches for keywords within the structure of the file. This means it should quickly retrieve every .doc file containing the words "Android OS." The biggest problem with such a file system would be the indexing costs. It is likely that a gigabyte of storage fully indexed as a real database would take 2GB or more in overhead. So what? Terabytes are cheaper than ever. You can always make this sort of file system an option if necessary for the cheapskate who thinks paying more than $300 for a PC is spending too much money.

2. Fix the Networking. How did the Microsoft networking subsystems deteriorate with Windows 7 and above? I have a legacy NAS from Cisco that no OS since Vista can access on my network. There are a few supposed patches that kind of work. Why do I need to patch anything? What's the reason these legacy devices cannot be used anymore? They don't work because nobody wants to write the code to make sure they still work.

3. Bring back the distraction games. What happened to some of the classic games that came with the OS? I'm talking specifically about Spider Solitaire. Where is it? Whatever happened to Pinball, for that matter?

4. Scale everything. This full-screen idiocy must go. I have complained bitterly about the stupidity of full-screen programs from their early pre-Windows 8 conceptualization. Full-screen programs harken back to DOS. Having a bunch of these running and switching from one to the other is nothing more than DOS task switching from 1985. Check the calendar, people. Why isn't everyone in the tech community squawking about this incessantly? Besides praising this lunacy, now some users are giddy about the new "half full screen" mode. It makes you wonder where everyone with a brain went.

5. And finally, put back the real Start menu. Microsoft can simply go buy the Classic Shell code and use that if it wants a good implementation. The reason for the Start menu was for convenience, nothing more. Taking it away was like that "Ribbon" interface—it's a way to impede an efficient use of the product. Why do this? Why encumber the user? Everything just takes longer than it should.

Windows 9 will be the end of the road for Microsoft if it keeps digging the same hole to nowhere. Everything the company does now seems to be part of some unspoken weird or secret theory stemming from smartphone UIs. It is all so inappropriate for the desktop computer. Microsoft is simply ruining the industry with this tactic, as we see with PC sales. My suggestions above, if implemented properly, would stem the decline.

>I have a legacy NAS from Cisco that no OS since Vista can access on my network.

My guess is that is a security/encryption setting issue with your NAS. Windows won't send a login request to a server that uses weak standards unless you change registry settings. This is done for your protection.

It's like complaining that you can't get leaded gasoline anymore. Complaints like that make this dude come off like a short-sighted, self-centered idiot.

That specific example might be, but Microsoft seem to have been in denial about some adverse side effects of the networking changes in Windows 7 since about five minutes after the launch.

I've seen an office with a hybrid network where numerous machines running XP, OS X and Linux all talked to each other quite happily, at the expected network speeds over wireless and wired connections, all working as you'd expect. However, most of the the Windows 7 machines (both desktops and laptops, from a variety of vendors) transferred data at a fraction of the speed the network could support for no apparent reason.

At some point, you have to accept that the obvious common factor is your new operating system and you've broken something, but MS never have.

Being that I run mixed Linux/Win7 networks with SMB filesharing and don't have speed transfer problems, it seems to be a sysadmin problem.

Update your NIC drivers and make sure your processor can handle encrypted sessions at full speed. If you want to downgrade to NTLM you are free to edit the registry keys to do so. The rest of us don't want our passwords and traffic easily crackable.

Thanks for the advice, but the company in question did a lot of work in the networking field, so it is highly unlikely that the poor performance was caused by either a lack of competence on the part of their sysadmins or simple oversights like not updating drivers. Various third party software running on the Windows 7 machines seemed to achieve the expected throughput just fine too, so the finger was definitely pointing to something in Windows being the culprit.

(Really obvious example: An rsync-based tool used to copy files to a central server prior to backing up would do so at roughly the expected speed from all of the Win7 PCs, but dragging and dropping the same files between the same machines in Windows Explorer would take an order of magnitude longer. Just to be clear, I'm talking about testing with new files where rsync would also have to transfer them in their entirety here, not a speed up because of the rsync algorithm itself.)

Pinball at least was some ridiculous hand-crafted assembly written for fun by a dev I'm sure I remember reading that on here or Reddit.

I can't find that source on my horribly censored network I'm afraid.

http://blogs.msdn.com/b/oldnewthing/archive/2012/12/18/10378...

1. Yeah WinFS would be nice but it seems to be out of reach still from a development point of view. Perhaps if it were a totally new file system and not having to maintain some kind of backwards compatibility with NTFS? Who knows I am not a file system developer but I do know Microsoft put some serious development work into WinFS back in the pre-Vista days and it was just too difficult.

2. God yes! Why the hell did they get rid of the connection center they had in Vista and 7 and turn it into a god damn side bar thing!?! Networking is painful on 8.

3. I miss the games too. It is bullshit that Microsoft made them ad-based Modern apps.

4. It is depressing to see how far things went backwards with these shitty fullscreen apps in Windows 8. They are nice on a tablet yes but on a desktop or laptop?! Horrible.

5. My biggest issue with the start screen is that they removed features that were present in the start menu. If the start screen were a full replacement to the start menu it wouldn't be such of an issue but how it is at the moment isn't nice for anything other than just launching an application. I don't agree with the ribbon comments. I find the ribbon to be fantastic and always have. I love Office with the ribbon and hate having to use LibreOffice with all the options lost in menus and sub-menus. The ribbon makes so much more sense and is far more user friendly and is one of the best UI features to come from Microsoft in years.

About #1, it just takes something like Beagle (http://en.wikipedia.org/wiki/Beagle_%28software%29) integrated with the file explorer, MS does not even need to touch the filesystem. But I never tought it was that usefull, and I can't get all the hype around the concept.

About #3, yep, XP's pinball is just great, I'd expect the number of good games to go up in newer releases, not down.

I think they could avoid the Start Menu. What do you want from the start menu? The ability to launch and search for programs. Launch is on the Metro screen now - that's covered. A full-screen start menu like the Metro desktop is fine - the start-menu kept growing absurdly anyways.

I think Metro could be made to work if you made sure it was operable with a mouse - that is, no auto-hiding, no gestures. Give me a single, consistent always-visible taskbar instead of schizophrenia and that "charms" crap, and on-screen always-visible buttons to launch an app-search-bar or to hit up the

I can see the logic of moving away from the old days of letting windows overlap (including the start-menu) - this served no purpose. Why would I want to partially obscure another window? But having more than one window visible at once isn't optional. Give me the means to slice up the desktop like the old Photoshop cut tool.

The Metro UI could be adapted to play nice with the desktop, but they have to admit that it's a desktop and desktops have different needs from touchschreens. Unifying this mess isn't out-of-reach if they are willing to compromise on unifying all the details.

>I think Metro could be made to work if you made sure it was operable with a mouse - that is, no auto-hiding, no gestures. Give me a single, consistent always-visible taskbar instead of schizophrenia and that "charms" crap, and on-screen always-visible buttons to launch an app-search-bar or to hit up the

I can live with this on the desktop, but what's particularly insane is they went with this same interface for Server 2012.

Fortunately, I got out of dealing with Windows Server before then, but I hear no end of complaining from friends that still do about how infuriating trying to invoke corner-based gestures through an RDC window is.

I don't consider corner-based gestures to satisfy my requirement of "operable with a mouse".

That's my point. Move away from gestures and auto-hiding crap and metro would be workable for the desktop. It's perfectly reasonable to auto-hide stuff on a 7" screen and have it always-visible on a 23" screen, or at least give it a button instead of a gesture.

Give me that kind of thing and Metro is workable instead of "omg revert this crap".

I just generally despise gesture based interaction, but Metro makes some sense on Windows Server. Strip away the gestures and leave just the tiles could be one way to go on the server. Normally you just login to do a quick modification and having the server tools right on the tile desktop makes it pretty easy to find.

Reply on news.ycombinator.com