Microsoft flings the Windows Calculator source at GitHub Microsoft has slung the source code for the Windows Calculator onto GitHub under the MIT licence in the hope of building a community around it. Hopefully some kind-hearted dev will deal with the C and CE symbols not quite lining up with the backspace symbol on the keypad. It's the little things. The move, announced last night …
They've already announced plans to do exactly this with Edge, of course. Why reinvent the wheel by having their own proprietary rendering engine? If they use the open-source Chromium engine, they offload the development costs of the hardest part of browser design to someone else. By building their own UI around an existing Chromium base whose maintenance costs are zero, they can achieve the same thing for far less cost.
This would not have worked during the browser wars of the early 2000s. IE was proprietary and not very standards complaint by design... by conquering the browser market, they effectively replaced the official W3C standards with their own. What better way to conquer the web-design and server markets than to offer high-dollar web infrastructure that was designed to use the same closed standards as the browser everyone used?
Now MS knows they aren't going to achieve that kind of market domination in browsers again. They've struggled to surpass the market share of the discontinued, outdated legacy IE. Having their own rendering engine may help the web as a whole by forestalling a monoculture, but it doesn't do anything but cost MS money to try to write their own standards-complaint web engine that, if done correctly, will render everything exactly the same as a rendering engine they could be using for free.
That's exactly what MS plans to do, of course. They're adopting the Chromium web engine for Edge. What they're NOT doing, of course, is open-sourcing the Edge engine. They can offload web engine development to the open-source community without actually giving up any of their secrets.
The closest thing to this would be MS abandoning the NT architecture and adopting Linux as their kernel. It would require a compatibility layer to run existing Windows programs, and of course, there is one of those already. WINE is not at the level of compatibility that would allow a Linux-based Windows, but part of the reason for that is MS Itself and its refusal to fully document its APIs. It has to be more difficult to create a full x86 emulator for ARM than to create a Windows compatibility later for Linux, and they did accomplish the former.
I don't know what the long-term plan is, or even if they have one over at MS. What does seem evident is that they are not interested in continuing to develop Windows as it has been for so many years. Their head is in the clouds now, and no matter how many times they utter the inane "Windows is a cloud service now," it doesn't make it true.
I think most People would say no. Windows (H)8.x was an unmittigated flop. They already done did the 9x Series back in the 90s. So Windows X was the next logical step, and with the one hand they givith, and with the other they taketh. But, how many of the current installs currently live do you imagine were gifter upon their users, like a theif in the night?
Execpt the Theif generaly doesn't care to stick 'round much afterwards.
To avoid one of your biggest competitor to reach almost total control on the next standards?
Having Google controlling the browser is not better than having MS controlling it.
The Linux kernel is not the ultimate kernel as well, sorry - it has its drawbacks and outdated/bad designs too - plus that big limitation GPL is for many kinds of commercial development.
And Windows API are far better documented that they were years ago, but emulating another OS is far more difficult than you think, the API signature is not the biggest problem - how they are implemented, how they work and their interaction is the real issue. Low-level details could be critical - and many Linux low-end parts may work differently from their Windows counterparts.
Competition is GOOD - having only one kernel, one browser, etc. will just lead to worse products, not better ones. And we're seeing already the first symptoms of a far worse IT world, with too many just flocking to the same products - regardless of their quality and verstatility, often just side products of something else, with all the inevitable limitations.
A sizeable proportion of WINE incompatibility reasons are because WINE won't implement features correctly. Whilst WINE staging is a mismatch of compatibility fixes that work on only a limited selection of configurations (usually Linux with an Nvidia card), pure WINE is not immune from this.
It's clear that in a number of cases the WINE API has been implemented in a way that works 95% of the time, but isn't implemented in the same way as Windows as this would be tremendously difficult and/or break applications again that were working up until this point.
Also, to be fair, the Win16/32/64 API is rather baroque, and even some of the more clued up developers who understand a fair bit of Windows' architecture may be unaware of some of the nuances that affect compatibility. The documentation is there, but dependencies between APIs are always obvious.
Azure has been successful only because it was the only good path for MS developers and customers - as it supports the Windows ecosystem better than other clouds.
Look at IBM or Oracle - their clouds are not successful because they have no competitive advantage over AWS.
If MS renounces to its competitive advantage with Windows and its software - it will just become "another cloud" - and customers could easily switch to other providers, if they find the same software.
"Azure has been successful only because it was the only good path for MS developers and customers"
I don't know about "only". Judging by the extreme (and extremely insulting) hard sell that a couple of Microsoft reps gave to us about Azure last year, some of that success is likely to be from arm-twisting. Although that didn't work with us -- they managed to anger management so much they committed to never even considering the use of Azure in the future.
when I went here
https://github.com/Microsoft/calculator
I saw a screenshot of a 2D FLATTY McFLATFACE UI, and "windows 10 required".
I'm thoroughly crestfallen...
(and here I was thinking I could do something good with it)
Fortunately, there's still the 'gnome calculator' which looks 3D skeuomorphic on my Mate desktop
https://wiki.gnome.org/action/show/Apps/Calculator
I suppose THAT could be ported to windows, if it hasn't been already...
Hmmm, then again some things never change. looking at their GitHub info...
eg None at all.
Now that's not fair - their policy of "if it compiles release it to the public and see what breaks" is still a policy. Not a very good one but, like ISO 19000, it's OK to have a process that's defined as "we don't have a process" as long as you stick to it.
And they very much do.
The calculator in Windows has been flat for longer than it's had fake-3D buttons. It only got skeuomorphic in Windows 95 (it was flat before that), and went back to being naturally smooth with Windows 8.
By my maths that's nineteen years flat (and counting), and only seventeen curved.
Mines the one with a casio in the pocket >>>
"went back to being naturally smooth with Windows 8"
WHAT??? Where'd he go... damn, left already. I wanna throw things!
'naturally smooth'. you owe me a LUNCH. Mine just came up!
Seriously though you have identified *THE* problem I think. Prior to Win 3.x (and '9x) 2D FLATTY was the way things like the calculator were done. A 3D skeuomorphic UI is more intuitive, so it was done for '9x at least with the first major re-write [most likely].
And of course the Sinofsky+Larson-Greene "revolution" (2D FLATTY McFLATFACE) forced yet ANOTHER re-write to "that".
But 'calc' doesn't seem to get a whole lotta love, and so it's been relegated to what ended up on github, like some kind of token effort, by engineers who don't even like it nor use it.
V0.99 Used to be that software wasn't done until it had an email client.
V2.0 Now it it isn't done until it includes a dating app.
V3.0 Uploads your every keystroke to the data aggregation service of your choice to be packaged and sold off, then backed up by the 5 eyes and Russians and Chinese?
Which one comes first to Calc.exe?
V3.0 is already implemented.
No, I'm not joking, the calculator really does slurp.
Expect similar for every Windows 10 app.
Like with Notepad, the value and utility of Calc is that it's basic. Basic enough that I don't know why it needs a community around it. The only thing that's likely to do is to increase the number of features and complexity of the thing, which would ruin the main thing that I find useful about it.
Please, someone fix this blasted calculator. I know, this is not the right forum to request, so I'll explain what I find wrong with the calculator (besides it being way too huge). I'll look elsewhere for someone who can fix it.
There are times when you might want to spawn a second instance of the calculator, and sometimes you might want only a single instance, brought to the front of all other windows.
In Windows 10, each time you run the app (I have a calculator button on my keyboard), a new instance is launched, This is the default behaviour.
In Windows XP, each time I pressed the calculator button, it either launches the initial instance, or brought the existing instance to the front. This is the default behaviour.
In the work I do, I need always single-instance, as I'll be plugging in values from different programs. Pressing the calculator button (or a hotkey) could bring the calculator to the front, so I could enter values, before going on to the next program to add more values.
I would love to be able to configure the calculator for single-instance mode, with a button in the upper-right to create a second instance when required.
Suggestions:
(1) Use a third-party calculator program, or
(2) Use a physical separate calculator (you probably have a calculator in your mobile phone), or
(3) Pin Calculator to your taskbar instead of using the keyboard button, or
(3) Get a utility called "AutoHotkey", which can do scripted automation things in Windows. The simplest might be a script to run from, perhaps, your "calculator" key (AutoHotkey has key mapping skillz), with the function of finding Calculator running and pushing it to foreground, or launching it if not running.
I have a script managing the "Fitaly" on screen keyboard for me that has these features, mainly so that I can also put Fitaly ON the taskbar - by itself it either gets pushed over onto the main desktop, or the taskbar is drawn over Fitaly.
Here. Have fun with program lines that are for a computer I had 10 years ago, or debugging, or to do something else entirely that I no longer remember.
Note that this SCRIPT is made single-instance by an internal command (I think). IIRC running the script again kills the existing instance in this case. Times are set in milliseconds.
; AutoHotkey Version: 1.x
; Language: English
; Platform: Win9x/NT
; Author: Robert Carnegie
;
; Script Function:
; Set the Fitaly on-screen keyboard as top window.
;
#SingleInstance, Force
; MsgBox Version 13:40 11/11/2010
TimeOut:=1000
TimeCheck:=300
TimeAfterLaunch:=5000
Menu, TRAY, Icon, H:\Software\AutoHotkey\Scripts\Fit2a.ico
WinGet, WinTarget1, PID, ahk_class FitalyMainWindow5.02
; WinGet, WinTarget2, PID, Omnis
FocusID:=WinExist("A")
; WinSet, Top, ahk_pid %WinTarget2%
; MsgBox, About to perform WinSet, Top,, ahk_pid %WinTarget1% (Fitaly)
WinSet, Top,, ahk_pid %WinTarget1%
; Sleep, 500
; MsgBox, About to perform WinActivate, ahk_id %FocusID% (previous focus)
; WinActivate, ahk_id %FocusID%
; WinMove, ahk_pid %WinTarget1%, , 000, 000
; WinMove, ahk_pid %WinTarget1%, , 1445, 496
WinMove, ahk_pid %WinTarget1%, , 1445, 783
Loop
{
WinGet, WinTarget1, PID, ahk_class FitalyMainWindow5.02
If WinTarget1
{
IfGreater, A_TimeIdle, %TimeOut%
{
FocusID:=WinExist("A")
; WinSet, Top,, ahk_pid %WinTarget2%
; MsgBox, About to perform WinSet, Top,, ahk_pid %WinTarget1% (Fitaly)
WinSet, Top,, ahk_pid %WinTarget1%
; Sleep, 500
; MsgBox, About to perform WinActivate, ahk_id %FocusID% (previous focus)
; WinActivate, ahk_id %FocusID%
; WinMove, ahk_pid %WinTarget1%, , 000, 000
; WinMove, ahk_pid %WinTarget1%, , 813, 445 ; for 2013 touchscreen
; WinMove, ahk_pid %WinTarget1%, , 1445, 496 ; for Dell Optiplex All In One
; WinMove, ahk_pid %WinTarget1%, , 1445, 716 ; for Dell Optiplex All In One, bottom right
WinMove, ahk_pid %WinTarget1%, , 1445, 783 ; for Dell - Windows 10, bottom right
}
Sleep, %TimeCheck%
}
Else
{
; MsgBox, About to relaunch Fitaly
Run, C:\Program Files (x86)\Fitaly Tablet PC\Fitaly.exe
Sleep, %TimeAfterLaunch%
}
}
Color me cynical, but I fail to understand why anybody would care about this in the slightest. There are like a billion calculator programs out there, and half of them are already open source and cross platform. It's not like a calculator is complicated. I wrote one in VB3 when I was about 12.
It seems to me that this is the latest in a long line of empty gestures, where MS open sources something nobody asked for or cares about and makes a big song and dance about how they're soooooo into FOSS these days and soooooo not like the old, evil MS. Meanwhile you look at their calculator and it slurps your data and phones home, and they're yet to open source a single thing of any consequence (you might think that .net core is important, but we already had a more full-featured FOSS .net implementation before that - mono, making it redundant. Maybe you think visual studio code is important, but we already had a million full-featured IDEs. VS code running on Linux is merely an ad for windows - if you like this and you want all the features enabled, install Win10 and pay $whatever). MSSQL is arguably something that somebody might care about, but there were already better FOSS database engines out there.
I've been saying it for years, and I'll say it again: Let me know when they open source DirectX, then I'll give a crap. But that will never happen because 10 minutes after they do it'll be compiling on Linux and 10 minutes after that porting any game to Linux will be truly trivial.
(I'd *love* to be proven wrong about this, BTW, because it would mean many more games would get ported)
If MS want to prove how into open source they are, they'll need to actually open source something useful and important. Then maybe I'll believe them. But as far as I can see everything they've done so far is just spin. The sad thing about it is how many people are buying into it.
As a bonus, they've now offloaded the whole writing code thing to a team of volunteers.
Read this (linked to above) https://blogs.msdn.microsoft.com/oldnewthing/20040525-00/?p=39193 and think again.
Writing a calculator app isn't too difficult when someone has already done the back end work for you. It's a tad more tricky when they have not.
For DirectX under Linux there is already DXVK, and other alternatives. Few people have an interest in DirectX integrated into Unix, when there's already Vulkan and OpenGL.
There's a lot more than just DirectX involved in porting games to Linux. It also doesn't help that despite a lot of work by Valve and others, Linux Steam usage is still under 1% and going down relative to the increasing user population.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
