Author Archives: Bavo Luysterborg

The Windows 10 Feedback Hub Generator

I had some free time on my hands, so I built this: Feedbackhub.link.

The Windows 10 insider program is great, the ability to give feedback is awesome, but the Feedback Hub is… less than awesome.

There are numerous issues with the hub, this simple website solves one of those problems. It will generate an HTTP link from those unwieldy feedback-hub:? and insiderhub:// links, the generated links work like any web link so they’re easy to share on social media. When you click one of the generated links you’ll be automatically redirected to the Feedback Hub app.

Feedbackhub.link currently supports three types of links: Feedback links, Announcement link and Quest links. Some examples:

Go ahead and share some feedback!

UPDATE: Windows-Feedback:? links for feedback items are now supported t0o.

The Windows 10 Icons Database Version 2.0

A month before the first anniversary of The Windows 10 Icons Database, I’m proud to announce that today a brand new, shiny version 2.0 of The Windows 10 Icons Database has launched, check it out at Icons.Bav0.com.

The new homepage

What happened to V1?

Yeah, I guess I have some explaining to do here…

No post-RTM icon changes were added to the database even though there were a lot of new icons in TH2, so why weren’t they added? Well, most of these icons were in files other than system32.dll and imageres.dll, and neither the original back-end, nor the front end were designed to handle icons in a lot of different files, so adding them required a complete rewrite (or a quick and dirty patchjob, I guess) of the site. Spare time to maintain projects like these was scarce back then, so no more updates were added for the remainder of 2015 and the first months of 2016. In February I finally found some time to (every now an then) start coding the new version and 2 months later it’s finally ready.

What’s new in V2?

Let’s start with the most important thing, the database is brought completely up to date with all 33 Insider builds, the coverage was also expanded from only system32.dll and imageres.dll to over 60 dll and exe files.Improved icon pages

Adding all these new icons and files called for a better way of organizing and loading icons, whereas the old site would load all icons on page load (making pages very slow), this new version will load a list of files and only load the icons inside once you expand a file section.

The improved filtering at the top of a builds page allows you to show all icons or only those updated in that specific build (as before), but it also allows you to compare any 2 builds, useful for comparing e.g. the latest Redstone build against the November Update.

IconsDBV2Pillsn

Build pages now feature some more info, like the full buildstring, (fast ring) release date etc.:IconsDBV2BuildInfo

The last big change is in the download system, upon clicking one of the download buttons you’ll be presented with a bunch of options:

You’ll find the same improved filtering criteria, the ability to download icons as png files (in addition to the original ico files) and the option to only include specific files in the download.IconsDBV2DownloadOptions

There are some other minor additions like a full changelog, an FAQ page and countless other tweaks and bugfixes.

Some stats

Version 1 of the database ended up hosting 558 different icon variations in 10 builds, version 2.0 launches with 2624 variations in 33 builds.

In the past year The Windows 10 Icons Database served over 12.000 downloads.

These were the 5 most popular icons:

Click for all versions

This PC

 

Click for all versions

(Empty) Recycle bin

 

Click for all versions

Disk drive

Click for all versions

(Filled) Recycle bin

Click for all versions

Downloads folder

What’s next?

I promise that going forward I’ll do a better job of keeping the database up to date, I’m sure Redstone will bring us some interesting new icons… ;)

I also have plans for more minor improvements as well as major new features, but I’m going to hold off on the details for now (just in case it takes me another year to complete those features).

 

Converting any Chrome extension to Edge (in theory)

We already established Edge extensions and Chrome extensions are very similar, so, with Edge extensions now available in build 14291, how exactly would one convert an existing Chrome extension to run in Edge? It’s hard to give a complete guide without official documentation (We’ll have to wait for Build to get that), but from digging through the demo extension provided by Microsoft I’m fairly confident the guide below is a pretty good start.

Porting extensions with a minimal amount of work is definitely possible, as demonstrated by F.B. Purity and the wonderful Tweeten. I’ve tried to port a couple of popular extensions myself, with mixed results, most likely because not all the planned extension APIs are included in this release.

I suggest you give the steps below a try yourself and leave a comment on your experience.

Step 1: Grabbing the Chrome extensions source

This one’s pretty easy and well documented all over the web. Every Chrome extension is identified by a unique ID, you can find the id of any installed extension in the Extensions window, or as part of the URL on the Chrome Web Store:

https://chrome.google.com/webstore/detail/adblock/gighmmpiobklfepjocnamgkkbiglidom

To download the .crx file that contains the extension code you simply replace ID_HERE in the following url:

https://clients2.google.com/service/update2/crx?response=redirect&prodversion=38.0&x=id%3DID_HERE%26installsource%3Dondemand%26uc

Step 2: Extracting the crx file

.crx files are actually just zip files (well, almost…), extract it using your favourite zip tool (rename it to .zip if necessary).

Step 3: Modifying manifest.json

By now you should have the source files of the extension you’re trying to convert, let’s start the actual modifications. The manifest file only needs one additional key:

"minimum_edge_version": "33.14281.1000.0"

If your app uses localization it looks like you’ll need to add one other key, “locale“. Chrome uses a “default_locale” key, but going by the example extensions provided by Microsoft Edge uses this additional key. So if there’s a “default_locale” key present in manifest.json you’d better add a “locale” key with the same value.

Step 4: Modifying the javascript

This is probably the most important step, Chrome’s APIs are accessed through the chrome.* object, judging by the extensions provided by Microsoft, Edge uses either msBrowser.* or browser.*, you could go and replace all occurences of chrome with msBrowser, but the easy (and more complete) way to handle this is to insert this piece of code (pulled directly from Microsoft’s code) in any file that uses Chrome’s APIs:

if (typeof msBrowser !== 'undefined') {
 chrome = msBrowser;
}
else if (typeof browser != 'undefined')
{
 chrome = browser;
}

This should make Edge’s APIs accessible through the chrome object.

Step 5: Perform some permission magic

The extension should run in Edge now, but when you actually try and load the extension you’ll receive this super helpful error:

Edge-Error

This one took me quite a while to figure out actually, the Learn More link suggests it’s a permissions problem, and suggests running the install script again. After some digging through said installer I found that among the code to extract the extension to the downloads folder, there’s a single line that modifies the Acces Control List for the newly created directory: Edge-permissions

To get your extension running in Edge you’ll need to set the same permissions for the folder you’re using, simply download or create a .cmd file with this line and run it from inside the directory you want to set permissions on:

icacls %CD% /grant "*S-1-15-2-3624051433-2125758914-1423191267-1740899205-1073925389-3782572162-737981194":"(OI)(CI)(WDAC,WO,GE)"

Loading the extension in Edge should work now.

Be sure to share your experiences by leaving a comment below!

Running Microsoft Edge extensions in Chrome

Earlier today a landing page for Microsoft Edge’s Extensions went live ahead of the official launch, it includes three downloadable extensions: Microsoft Translator, Reddit Enhancement Suite, and Mouse Gestures. For now the extensions are simple .exe files that extract the necessary files for sideloading into Edge to your Downloads folder. It’s currently impossible to get these extensions working in Edge (in Insider builds that is), but since the upcoming extension system in Edge is rumoured to be compatible to Google Chrome’s system I decided to try to load them in Chrome to find out just how compatible they really are. Turns out they’re nearly 100% compatible.

edgeExtensions1Microsoft Translator

First up: Translator. After running the exe file and looking at the extracted resources it pretty much matches what you’d expect from a Chrome extension, there’s a manifest.json, an options page, a background page, a _locales folder.

I loaded the folder in Chrome (using the “Load unpacked extension” button on the extensions page) and got this error:

edgeExtensions2

Easy enough to fix, just add  “manifest_version”: 2, to the manifest. Retry.

edgeExtensions3

Welp, that didn’t go as expected. Looks like localization is broken, or is it? Some localization files in the extension (\_locales\XX\messages.json) are empty, those obviously aren’t valid JSON and can’t be parsed by Chrome. Let’s delete all folders with an empty messages.json file in them and try again (or simply delete all folders except _locales\en):

edgeExtensions4

Success! The Edge extension loads without any further problems and looks like any regular Chrome extension. (Notice the two Edge-specific keys in the manifest though.) The Translate button shows up in the toolbar and translates the page when clicked.

This slideshow requires JavaScript.

Mouse Gestures

Next up: Mouse Gestures, this looks like a really interesting one. After adding a manifest_version and removing empty locales this extension too loads without any further issues.

edgeExtensions6

The first run experience works as expected:

The settings page works:

edgeExtensions8

Sadly i couldn’t get any of the gestures to work, maybe it’s an Edge thing that doesn’t work in Chrome, maybe I’m just not bright enough to figure out how it works… There’s one feature I got working, it’s only mentioned in the changelog, drag to open a link in a new tab:

Reddit Enhancement Suite

I applied the same changes as above to try and make this one work, but I ran into a rather unexpected issue:

edgeExtensions10

A folder (vendor) was missing from the downloaded Edge Extension, super weird. Luckily RES is open source, so I downloaded the missing folder from the Chrome extension on RES’s GitHub page and placed it in the folder for the Edge extension. I also had to fix the manifest file, it listed vendor/snuownd.js  instead of vendor/snudown.js. Then I had to copy over these files from GitHub to the modules folder: (seriously, who packaged this!?)

  • usernameHider.js
  • userTagger.js
  • voteEnhancer.js
  • And the entire \modules\hosts folder.

It finally worked.edgeExtensions11

What this means for Chrome -> Edge porting

While this Edge->Chrome compatibility doesn’t guarantee complete compatibility in the other direction, it sure looks like that’s what Microsoft is aiming for. The source code for these Edge extension pretty much looks like the original Chrome code with some additional code to replace “chrome” with “msBrowser” and 2 minor additions to manifest.json.

No, the Pi isn’t a full Windows PC, but yes, it will have a GUI

IoT-rp2-banner

Earlier today, Microsoft announced that it will be releasing a free version of Windows 10 for the new Raspberry Pi 2. This is definitely an exciting opportunity for makers building all sort of cool stuff, it’s an interesting move by Microsoft, it’s a lot of things, but it certainly isn’t a cheap PC.

“It’s not clear exactly what version of Windows 10 will be available,

(…)

With the pricing of the Raspberry Pi 2 and Microsoft’s free copy of Windows 10, you could have a full PC for just $35 later this year.”

Tom Warren – The Verge

No.

It’s obvious that the Pi will run the Internet of Things (IoT) version of Windows 10, right now that version doesn’t have a GUI. The version for the Raspberry Pi will be different from the current version targeted at Intel’s Galileo boards, but that doesn’t necessarily mean it will have a full Windows Shell, no desktop, no start menu!

“But it will run Universal Apps!” That’s right, it will run universal apps and it will even support a graphical interface according to Raspberry Pi founder and CEO Eben Upton:

Cross-device compatibility will be there thanks to Microsoft’s “Universal App” strategy, based on the Windows Runtime platform (once known as Metro). Although the desktop might not be available, the Windows build will support visual applications.

“So it’ll run full Windows 10, no?” Not really, it won’t run Windows in the traditional sense, there will be no familiar Windows UI, no desktop, no start menu (or start screen for that matter). Again according to Upton, the Pi won’t be a full Windows PC:

…they [Microsoft] have the mindset of Raspberry Pi as a device that you deploy to, rather than Raspberry Pi as a standalone computer.

You will build a universal app in Visual Studio and then deploy it onto your Pi, When you power on your Pi it will essentially boot straight into your app. Perhaps it’s best compared to Windows 8.1’s Assigned Access, which allows you to lock down a user account to a single app, only instead of a locked down user account you get an entirely ‘locked down’ PC.

The Pi 2 with Windows 10 won’t be the $35 PC some are claiming/hoping it would be, but it will definitely have many other exciting applications. All those ATMs still running XP could be replaced with a Pi running an ATM app, for example.

We’ll probably get all the details we need at WinHEC

How to enable the Windows 10 Notification Center

First of all: I’m merely re-posting this trick, all credit goes to Adrian (@adeyblue)! Most of the info below is simply taken from his readme.

Here’s what the notification center looks like in build 9841:

How do you enable it?

  1. Download the hack: 9841NotificationCenter.zip
  2. Copy the correct notificationact.dll (x86/x64) to C:\notificationcenter\
  3. Run notifyactids.reg and notifyclassids.reg to add them to the registry. (It’s possible you have to take ownership of HKLM\SOFTWARE\Microsoft\WindowsRuntime\ActivatableClassId and HKLM\SOFTWARE\Microsoft\WindowsRuntime\CLSID, use something like SetACL or RegOwnershipEx)
  4. Now simply run notification.exe and the notification center will pop-up.
  5. You can also run notification.exe from the command line and append a number to select a corner:

notification 0 – top left
notification 1 – bottom left
notification 2 – top right
notification 3 – bottom right (default)

Known bugs

If you double click on a notification or long single click, it’ll crash. This doesn’t have anything to do with the hack, the bug is in Windows.UI.Shell.dll. It tries to convert one interface to another, the conversion fails but the code doesn’t check for this and assumes it will succeed. When it tries to use the result, it goes boom.

Stuff it could in theory do, but doesn’t do

The notifications popup can be shown on any monitor, though here it defaults to the one where the console window opens.

Also, I believe the code for the notification icon tray is present though it requires somebody more versed in WinRT to figure out how to connect it up to explorer, if indeed it can be. If you have a dissassembler and the Windows.UI.Shell.dll symbols, you want to be looking at ??0OverflowIcons@NotificationCenter@@Q$AAA@XZ which is where the object is created and the pointers to all its vtables assigned. The class ids of interest are NotificationCenter.OverflowIcons & NotificationCenter.TrayItemView.

 Why doesn’t the notification center work out of the box?

The activatable class ids of the center in Windows.UI.Shell.dll aren’t registered in the parts of the registry mentioned above (hence the scripts). Also, the dll lacks a DllGetActivationFactory function which is needed to actually create those interfaces once they’re registered (hence the dll).

To solve this, the registry scripts point Windows to notificationact.dll which acts as a surrogate. Its DllGetActivationFactory uses a nasty hack to call the correct interface creation functions in Windows.UI.Shell.dll.

Explorer doesn’t contain any code to use Windows.UI.Shell.dll directly or indirectly. Neither does it contain any code to start the Notification controller server, though bizarrely it has the call to shut it down.

Building from source

The download includes the source code fro this hack (thanks again, Adrian), here are his notes on building/modifying the code yourself:

Anything above VC 6 should be able to build the code since I’ve inlined and GetProcAddress-ed the newer stuff that’s required. Uncomment the VERBOSE define in both files to print status messages, the build command lines are at the top of each file.

If you need any help, or you have any other cool tricks, let me know in the comments below or on Twitter (@Bav0)

“Windows 10 Partner Guidance” reveals some more details on Windows 10

Yesterday Microsoft talked about Windows 10 (Threshold) for the first time (click here to rewatch the event if you missed it). Thanks to a confidential partner document posted on MDL Forums we now have a some more details regarding the preview and Windows 10 in general.

First off, the document clearly states: “Microsoft also shared that a Technical Preview will be available on October 1 around 9 a.m. for pre-release testing, evaluation and feedback”. There was some confusion about when exactly the Preview would become available, now you know for sure.

The document then goes on listing the same details and new features already explained at yesterday’s event, but end with a rather interesting Q&A section. Below is a selection of answers that provide new info, confirm existing rumors or are just interesting in general, be sure to check out the full document for more answers.

Q: When will the next version of Windows be made generally available?

We look forward to the next version of Windows being generally available over the summer of 2015.

Q: Will Windows 10 be available for free to existing customers?

Microsoft has not shared any information about pricing at this time.

Q: If I buy a Windows device this holiday season, will it be upgradable?

It’s our intention that the vast majority of PCs and tablets bought this holiday will work great on Windows 10.

Q: Should I download the Tech Preview?

Tech Preview is designed for PC experts who are comfortable downloading unfished software, preferably on a secondary PC. This is not a developer preview designed for app compatibility testing. While we do expect many developers to download the Preview, we fully expect them to run into compatibility bugs at this early stage.

Q: What are the system requirements for the Windows 10 Technical Preview?

If you want to test the Windows 10 Technical Preview on your PC, here’s what it takes:

  • Processor: 1 gigahertz (GHz) or faster
  • RAM: 1 gigabyte (GB)
  • Free hard disk space: 16 GB
  • Graphics card: Microsoft DirectX 9 graphics device with WDDM driver
  • A Microsoft account and Internet access

The Windows 10 Technical Preview can be installed on PCs meeting system requirements running Windows 7 or newer

Q: Can I install the Windows 10 Tech Preview on a small form factor Windows tablet or a Windows Phone?

The Technical Preview is designed specifically for x86 based systems and is optimized for non-touch, mouse and keyboard driven systems. Windows on ARM-based processors continues to be an important element of our strategy for phones and tablets, and we’ll have more to share on the Windows experience for mobile devices at a later date.

Q: Do you think the next version of Windows will have a better adoption rate than Windows 8?

We’re pleased with the adoption of Windows 8 and Windows 8.1. According to Microsoft, Windows 8 has surpassed 200 million licenses sold, and there are more than 6,600 Windows 8/8.1 certified products available to customers.

Why I like BlackBerry

There, I said it, I like BlackBerry! My smartphone of choice is a BlackBerry Z10 and it has been so for over 1,5 year, not an iPhone, not Android, not even a Windows Phone!

Why did I decide to buy a Z10, you ask? To be honest it wasn’t really a conscious choice, I tumbled into the BlackBerry 10 ecosystem after BlackBerry sent me a free Dev Alpha device to build and test apps for the upcoming BlackBerry 10 OS, then I traded it for a Limited Editon Z10 which I still use today. But that’s not important, it once was just another freebie, but now I don’t know if I could ever live with another smartphone OS.

BlackBerry 10 has nearly all the features (actual features, not gimmicks) competing OSes offer and it’s only getting better with regular updates, but most importantly it has several important features others don’t have that I could never live without.

The App gap & Android runtime

The big thing differentiating different devices, for most people at least, is their app ecosystem, iPhone and Android users have all the apps they could ever want, Windows Phone has, well, fewer apps. Blackberry 10 has even fewer… unless you count the millions of  Android apps ever made!

While BlackBerry managed to get some big apps in their BlackBerry World store (EverNote, SoundHound, Angry Birds, Real Racing, Asphalt…) and some pretty good 3rd party alternatives to Twitter, Snapchat, Instagram etc. that doesn’t come close to what Windows Phone has, let alone Android or IOS. But BlackBerry has one very clever trick up its sleeve, the BB10 Android Runtime!

The Android runtime started as an easy way for Android developers to bring their apps to BlackBerry, they just needed to repackage their APK into a BAR (BlackBerry’s app container) and submit it to the store, the device emulates a full Android install to run these repackaged apps. Soon end-users started to repackage APKs for their own use and sideloaded them to their device (yes, it was really that easy). With a little bit of work BlackBerry 10 users now had access to a big part of the Android ecosystem too!

The 10.2 OS update included a bunch of improvements to the runtime so nearly every Android app ever could be run on a Blackberry 10 device (albeit with some quirks here and there). But there was more, gone were the days of repackaging APKs, end-users could now install APKs directly from their device! One could even install the Amazon app store APK and download apps through the Amazon store and while it isn’t possible to run Google Play directly, many alternative, BB10 native apps exist, like Snap, that allow users to browse and install apps from Google play.

Having access to the 2 major Android stores right from you device was obviously a major step in the right direction, and with the upcoming 10.3 OS release blackberry is taking that experience to the next level by including the Amazon store out of the box. There really is no app gap when it comes to BlackBerry 10!

The signature BlackBerry keyboard

When people think BlackBerry, they think keyboards, for many years the outstanding physical keyboards on BlackBerry devices made them the device of choice for business users. BlackBerry still sells devices with a physical keyboard – their Q-series – but they also managed to offer that same great typing experience on their full-touch devices.

The iconic frets between the horizontal rows of keys do work, even on a touchscreen, that little bit of extra space between keys really helps diminish typos.

The autocomplete and next word suggestions are some of the best I’ve seen, they are far more accurate than the stock Android keyboard or Windows Phone keyboard. Predictions are shown in-letter, you simply swipe up to complete the word you’re typing, far more efficient than a suggestions bar above the keyboard if you ask me. It’s a simple concept, instead of dividing your attention between the keys and the suggestions, you just focus on the next letter you’re going to type, if the word you need is suggested, instead of pressing the key you simply swipe up, a fluid typing experience. On top of that, because suggestions are spread over the keyboard, you get a lot more suggestions at once instead of the limited number in the suggestions bar.

keyboardglanceThen there’s the way BlackBerry 10’s keyboard learns how you type, unlike stock Android you don’t have to manually save every new word, it automatically learns new words and the context you use them in as you type and will suggest them later on. It learns  your personal writing style and your specific sentences. If there’s a special sentence you use a lot, BlackBerry 10 will know when you’re about to type that sentence and after you type in the first 2 or so words you just go *swipe* *swipe* *swipe*, and there’s your sentence, no actual typing involved!

Some other nice features include voice dictation built right into the keyboard and the ability to swipe left anywhere on the keyboard to delete the last word you typed.

The BlackBerry 10 keyboard is clearly better than the IOS7 or stock Android one. There might be a better custom Android keyboard out there somewhere and IOS8 includes a new keyboard with autocomplete suggestion (seriously, every other OS has had this for years now, c’mon Apple) and support for custom keyboards, but for now I’m very satisfied with my BlackBerry keyboard.

Oh and Swype keyboards? I never mastered those :P

Gesture interface

Next up is the user interface, when you take a quick look at a Blackberry 10 phone you’ll notice it has very few buttons (unless you’re using a Qwerty device ;) ). There’s no home or back button on any of the new phones and no on-screen keys either, most of the time you navigate the Blackberry 10 OS using gestures. This is a pretty steep learning curve, yes, but once you get the hang of it it is very intuitive, fast, fluid and productive, you’ll be swiping your way through the day. A gesture based interface allows for a slew of new features that just aren’t possible when you’re restricted to 3, 2 or even 1 button(s). A great example are the different peek gestures that allow you to take a quick peek inside the BlackBerry hub, but more about that later.

Productivity

BlackBerry’s devices have always been productivity masterpieces and BlackBerry 10 is no exception! At the heart of this productivity is the BlackBerry Hub, a central hub with all your emails, texts, IMs, tweets etc. It’s nothing like other platforms where you get a notification saying “You have X new emails” which then takes you to a dedicated email app, in stead the Hub will aggregate every single item in one place so you don’t have to divide your attention between different apps. The Hub itself isn’t even an app, it’s an integral part of BlackBerry 10, it’s best described as an extra panel to the left of your homescreen.

Hub-Location

If you want to unclutter things you can always filter based on different types of notifications, like emails, tweets, …

hamburger

You’ll notice there’s a special category called “Priority Hub“, this is an automatically aggregated list of notifications that are important to you, this can be emails you’ve flagged, but also replies to conversations you started, notifications from “priority contacts” or messages containing specific keywords.

The beauty of the Hub lies in the fact that no matter what you’re doing on your phone it’s only one swipe away.

Hub-Gesture

You don’t need to open the Hub to find out what notifications out have, you can simply swipe up and hold at any time to see the notification charms or swipe up, to the left and hold to take a quick peek inside the Hub.

Peek

There are tons of clever tricks that help you get work done faster using the Hub, I won’t list all of them here, but the pinch filter is a great example. Notifications don’t automatically disappear when they’re read, but you can pinch to filter on unread notifications.

pinch

The future of the BlackBerry 10 OS

Blackberry 10 already is a great smartphone OS and the future looks bright, with OS10.3 bringing a lot of improvements and new features.

The biggest feature is a personal assistant, IOS has Siri, Android has Google Now and Windows Phone has Cortana and now BlackBerry 10 will get “BlackBerry Assistant” (catchy name, I know). Judging by the Inside BlackBerry blogpost, BlackBerry assistant will have much of the same features as the other assistants, let’s hope it performs just as well too…

assistant-main-use

The other big change is a complete UI redesign. BlackBerry 10.3 will have a flatter look, much like Microsoft’s Metro Modern UI, Google’s “Material Design” and Apple’s IOS 7 redesign. The update includes flatter icons for the built-in apps and a new, more minimalist design for the app bar with a “signature action”.

Other changes include bundling the Amazon appstore by default, an updated camera UI, an extra row of apps on the homescreen and a new gesture to reach the quick settings panel from anywhere in the OS. For a full list of changes in 10.3, head over to the Inside BlackBerry blog.

quick-settings-103

My Windows Threshold wishlist

Windows “Threshold”, the next big version of Windows, is coming, some even say a preview is due later this year and that the official Threshold announcement could happen as soon as WPC (July 13-17). I figured this would be a good time to list some of the features I’d like to see in Threshold…

What we already know

Windows 8.1 Update is a first step towards Threshold, with the right-click menu’s on the start screen, title bars for Modern UI apps and the ability to pin and show Modern apps on the taskbar. One strangely absent feature is the ability to have desktop shortcuts to Modern apps.

Microsoft already “demoed” some of the most important features in Threshold at BUILD 2014. We know Threshold will include a start menu and Windowed Modern apps and while that’s definitely a good start, I really think Microsoft can do more in Threshold.

vlcsnap-2014-07-05-17h24m01s66

UI overhaul (yes, another one)

“Metro” was a radical design change for Windows, but the desktop UI wasn’t updated at all. As I’m a big fan of the Metro design principles, I’d like to see a complete desktop UI overhaul in Threshold. (Some recent posts by Neowin and WinBeta actually confirm that current builds have a redesigned desktop)

Below are some mock-ups by Jay Machalani from his “Fixing Windows 8“-post. Although Machalani’s post is mostly about fixing the ModernUI-desktop schizophrenia, it has some very good designs for the future of the Windows desktop, quoting Machalani:

I did some simple obvious modifications to our beloved environment. The Desktop will now use the color from your personalization settings. In the future why not sync the color between your PC, tablet, phone and Xbox; that would be awesome! The time, date and icons got a little bigger and makes it easier for the users with a super small/dense screen like me with my Surface Pro 2 or even my Dell XPS 12.

thresholdconcept5

thresholdconcept4

thresholdconcept1

thresholdconcept2

You can see updated icons, a new flat look for window borders, title bars and the minimize/maximize/close buttons. Look at how Machalani integrated the charms into the windows for the metro apps, genius!

The thing I like most about his mock-ups is the taskbar, the padding around it is subtle yet beautiful and the way running apps are indicated is clean and simple. It could be further improved with notification indicators for shortcuts to Modern apps.

Notification center

Notification centers have been the subject of a lot of buzz and rumours lately, on IOS, Windows Phone etc. so why not on Windows? Modern UI apps already have an advanced, robust notification API, why not display a list of these notifications in a single place?

Notification center concept by Jay Machalani

Notification center concept by Jay Machalani

Cortana

This one is obvious, Windows Phone 8.1 shipped with Microsoft’s new voice assistant Cortana, I’m sure it would work great on Windows. I’d like to see some proper, OS-wide Cortana integration in Windows. It could be perfectly integrated with the existing speech recognition that’s already available since Windows Vista.

speechrecognition

Tabs, tabs everywhere

We have had tabbed browsing in every major browser for years now and it works great, so let’s add some of that productivity in other places like the file explorer. There are already 3rd party solutions, like Clover, that will add tabs to the Windows File Explorer, but it would be nice to see a native implementation (and a proper UI for the tabs).

Clover running on Windiws 8.1 (Image courtesy of Softpedia.com)

Clover running on Windows 8.1 (Image courtesy of Softpedia.com)

Another place where tabs *might* work is the desktop, so you can organize your desktop into different sections like “Games”, “Productivity” etc. This might turn out to be a bad idea, though…

Bring back desktop gadgets (sort of)

We all know the desktop gadgets from Windows Vista and 7 and we all cried when they were discontinued with Windows 8, right? Okay, maybe not, the start screen basically replaced the gadgets’ functionality with the introduction of live tiles, but with the advent of a ‘mini’ start menu in Threshold where are we going to keep all these live tiles? I think you should be able to put live tile gadgets on your desktop.

“Continuity”

The last feature I would like to see in Windows is what Apple likes to call “Continuity”, that’s right, Apple! Cherish this moment, Apple fanboys, because I rarely ever admit OSX has cool features Windows should ‘borrow’.

The upcoming iOS 8 and OS X “Yosemite” will work together like never before, you can answer phone calls and respond to text messages using your iPad or Mac, all without ever touching your iPhone.

calls_hero

This would work great in a Windows – Windows Phone combination, Microsoft could even create an Android app to pair your Android phone with your Windows PC. (An IOS app would be difficult because you can’t achieve the same, deep level of integration you can on Android, too bad)

Another cool feature is using the seamless content hand-off between IOS and OSX, you can start browsing on your iPhone, pick up your iPad and continue right where you left off, or you can start composing an email on your phone and finish it on your Mac. You can even start working on a document on one device and quickly switch to another device using iCloud.

handoff_gallery_icloud

Windows and Windows Phone are already getting closer and closer to each other using the cloud power of OneDrive (IE tabs, for example, sync across devices) This functionality would look great in Office 2015, the Office “Gemini” apps (Modern UI apps), and the iPad and Android versions of Office 365.

This is all I could think of right now, let me know, on twitter or in the comment section below, what other features you would like to see in one of the upcoming versions of Windows.

The truth behind Microsoft’s start menu ‘demo’

Lately there has been some discussion on MDL forums about whether or not the start screen and windowed apps Microsoft demoed at BUILD 2014 were a real build or just mock-ups. Let’s settle this once and for all!

There is something important everybody seamed to forget in this debate. That screenshot that’s all around the web is actually taken from the BUILD keynote livestream, go ahead and re-watch the keynote: http://channel9.msdn.com/Events/Build/2014/KEY01 (the start menu demos start around 2h08m)

It’s not a screenshot, it’s actually a video, so most likely taken from a real build! But there’s more, take a look at some sreencaps from the video below and pay close attention to the date and time in the bottom right corner:

vlcsnap-2014-07-05-17h23m16s131

vlcsnap-2014-07-05-17h23m46s176

vlcsnap-2014-07-05-17h24m01s66

vlcsnap-2014-07-05-17h24m26s63

You’ll notice the date is the 2nd of April ’14, the exact day of the BUILD keynote and the time goes from 10.11AM to 10.12AM in the video! So it’s a live demo? No! The keynote started at 8.30 AM, the demos are shown at 2h08m into the keynote, so around 10.38AM, the time is off by almost half an hour!

Then there are the differences between the first two screencaps and the last two: different battery levels, different running apps, the first 2 desktops don’t have a watermark…

The windowed apps look photoshopped, they’re both active, the mouse cursor in the Bing search box doesn’t blink and there is no interaction in the windows. In the last 2 screencaps IE and Mail are not even listed on the taskbar.

Conclusion: The start menu is probably taken from a working build, but the 2 windows are probably pasted in in post-production. The clocks are faked to try and make it look like a live demo, but the keynote’s schedule didn’t line up with the demos as expected. The only question is why would Microsoft go through all this effort to make us believe it’s a working build?

Thanks to all MDL users, especially sM4llziE for helping me analyze the demo.