The Infamous Windows “Hello World” Program

http://www.charlespetzold.com/blog/2014/12/The-Infamous-Windows-Hello-World-Program.html

by johndcook • 12 years ago

134 89 12 years ago

And then MS came up with MFC which was slightly better. But only slightly.

Borland had a better solution if I remember, in their C++ Builder thing.

Another "adventure" I had in Win32, trying to make a dynamic dialog box (basically, selecting something in a list opened up a list of properties - like the Preferences dialog box in most browsers). It was some time before I figured out how to do it (there was a coordinate conversion needed)

bhouston• 12 years ago

Borland has OWL: http://en.wikipedia.org/wiki/Object_Windows_Library

Strangely it is still maintained by volunteers: http://sourceforge.net/projects/owlnext/

But it was always behind Microsoft's MFC in terms of features. And you had to drop out of it for advanced stuff and then you had to spend a bunch of time figuring out how it actually worked.

stormbrew4• 12 years ago

And you had to drop out of it for advanced stuff and then you had to spend a bunch of time figuring out how it actually worked.

This was really no less true of MFC, and imo the OWL code was a lot cleaner and easier to understand. In the end, though, they were both just wrappers around the windows api.

But the GP seems to be referring to VCL, which is what came with Delphi/C++Builder. One of the reasons OWL was usually behind MFC was that Borland basically abandoned it and their mainline compiler products for the Delphi line, and C++Builder was always a second-rate product behind Delphi as well.

userbinator3• 12 years ago

It was some time before I figured out how to do it (there was a coordinate conversion needed)

That sounds overly complex - all you need to do is to catch the listbox selection message, hide all the controls that shouldn't be shown with the current list selection, and show all the ones that should be. I don't see how a coordinate conversion would be needed for that.

raverbashing3• 12 years ago

I don't remember all the details, but it was more or less like this (with MFC)

I built the dialog using their tools, one for each "tab"

Then I attached that dialog on the space for it based on the selection, and the coordinate transformation (was a function in MFC)

Something like this http://stackoverflow.com/questions/2653130/how-to-get-size-a...

Hiding/showing might have been a good idea, but for more complex dialogs it wouldn't work

aurora72• 12 years ago

With the Win32 API, we saw some innovative use of C in action, such as the function pointers and the callback functions.

Jare• 12 years ago

I was using C function pointers routinely in 1991, way before I ever touched Windows. It was a straightforward thing to do after using them in assembly language throughout the 80s.

goalieca2• 12 years ago

What? Checkout the signature for the posix signal function. Notice the function pointer?

http://man7.org/linux/man-pages/man2/signal.2.html

"Signals have been around since the 1970s Bell Labs Unix and have been more recently specified in the POSIX standard." --wikipedia

Basically, they're a software interrupt handler. Not too different from assembly. I don't see why that would be innovative at all?

In fact, early C++ just compiled to C. You can write structs with function pointers and vtables all in C if you wanted. People have been doing that since the 80s.

gryph0n• 12 years ago

Forget the Posix/Win32 platforms, function pointers can be easily found in the Standard C library itself. I believe there are atleast 2 functions taking function pointers as parameters:

1.>'qsort' function takes a function pointer to decide how to compare your data in order to sort it, presumably using quick sort.

2.>'signal' function has a Posix origin, but it did manage to make it into the C standard. It takes a function pointer to decide which function to call on getting a particular signal.

fit2rule3• 12 years ago

For those of us watching from the Unix world while Microsoft fumbled its way towards world domination, it was an incredibly frustrating time to see the banality of their technology become 'the thing' in light of so many, many, forgotten and ignored mysteries. It was something like what it might be like to see a new religion form for your children while watching that of your parents die.

Thank Linus we still have an alternative to Microsoft, here and now! It is absolutely a huge gain for us that Unix was not eradicated by the feral, vermin new thing.

KC8ZKF• 12 years ago

Oh but there was VMS and Lisp and others that are gone, with all the ideas. All replaced by the Unix way. The Unix way is gone, too. Linus should be given some credit for that as well.

BTW, Gnu's "Hello" is 709k.

Someone1234• 12 years ago

Thank Linus we still have an alternative to Microsoft

Because BSD isn't a "thing?"

fit2rule3• 12 years ago

I doubt it would've been a thing if we didn't have Linux to make the idea popular to culture, but thats just my opinion.

Dewie• 12 years ago

So now we have Windows and Unix-likes... well at least it's not a monoculture.

ryandrake• 12 years ago

Brings back horrid memories of learning Windows programming back in the mid nineties. From the very first line of code, Microsoft had you doing it their (nonstandard) way, making you use WinMain() instead of main(). That style, too... Yuck! Even today, I can somewhat reliably identify folks who were raised on the Windows SDK by looking for lpszHungarianNotation and UNREADABLEALLCAPSTYPEDEFS.

Does anyone remember why the original samples allocated structures on the heap, did stuff with them, then immediately freed them, rather than just using them on the stack directly? Was that a thing? Were early Windows systems stack constrained or something?

MAGZine• 12 years ago

I used to have much disgust for hungarian notation. When I worked on my first big C++ codebase though, hungarian notation was standard, and I have to say, it made the code more readable. using 's' for static, 'm' for member, etc brought some extra clarity as to what was happening. It's not that we couldn't determine these thing for ourselves (VS has great code-crawling facilities), but its less time hovering, symbol searching, and jumping-to-declaration.

I don't use it out of choice, but it isn't so bad. Though, this was using a slightly HN slightly differently. complete variable types in the name i have different feelings for!

stormbrew4• 12 years ago

There's a limited degree to which it's useful in resolving scope and type ambiguities in C or C++, but excessive use could definitely render things much less readable. It doesn't help that some of those letters survived long past their shelf life (ie. 'lp' for long pointer, where in non-segmented architectures there is no far (aka long) and near pointer distinction).

These days in C++ I'd only ever consider using s or g, and never the type specifiers. I also largely consider signalling member variable vs. local variable to be pointless in all its forms (including _ suffix or prefix) now.

darklajid• 12 years ago

I think you're mixing two absolutely different types of Hungarian Notation.

No offense, but what you describe seems more or less like the reviled version (stuffing ~redundant~ information in the name, where you end up with strName and iCounter).

The more-or-less accepted version described in the GP is trying to fix a shortcoming of the language's type system. If you cannot tell your compiler that this is a (default example) pixel value and this is a color, because both are plain ints to it, you're trying to make accidents less likely by requiring the programmer to double-check the prefixes (which encode a subtype, a special type. Not int, but int-representing-color or int-as-a-bool-here).

reacweb4• 12 years ago

I have no strong opinion about hungarian notation, but I have noticed that hungarian notation haters were often english native speakers.

hnal943• 12 years ago

Joel tells a pretty compelling story about two types of Hungarian notation: http://www.joelonsoftware.com/articles/Wrong.html

amaks• 12 years ago

's' for static, 'm' for member is not a Hungarian notation. It's just an established prefix patterns. Some style guides use '_' as a suffix for class member variables, and g_ prefix for globals. It's just a consistency. Hungarian notation is a nightmarish replacement of the intellisense such as lpszFoo.

shmerl2• 12 years ago

Out of that I use m_ for members. But that's about it. It is very helpful when you read lengthy methods code. For actual names I use snake notation. I.e. something like

    int m_some_property;

Type in the name is definitely an overkill.

userbinator3• 12 years ago

I found it a lot easier to learn Win32 when I discovered that much of it was just MS' own styling and not actually necessary at all. A lot of the typedefs are artifacts from the Win16 days (e.g. you can just use a void* instead of LPVOID). This is a perfectly valid "Hello World" program, in a more traditional C style:

    #include <windows.h>
    #include <winuser.h>

    int main() {
        MessageBoxA(0, "Hello", "World", MB_OK);
    }

Depending on if you specify /subsystem:windows or not in the linker settings, you can even get a console window (and thus printf() to it, etc.) and use it for logging etc. alongside the GUI, something that isn't often mentioned.

Another thing that probably drives people away is the insistence on starting with a "whole window" app, one that uses CreateWindow/RegisterClass, handles WM_PAINT, etc. when for a lot of purposes, a "modal dialog-based" app using a template and DialogBoxParam would be sufficient, and writing one of those is certainly easier. I've made a lot of trivial GUIs of the "window with a set of buttons to do various things" type with Win32, and it definitely doesn't take all that much work. Nothing more than two functions, one main() and one wndproc(), switch on the message/control in the wndproc() depending on which button was clicked, and do something accordingly.

I think it's a bit of a shame that the majority of the documentation on it is both overly complex and nonstandard in style, because the true "essence" of Win32 is really a small and simple C API. It's easy to create really small programs with it, and it doesn't take much code to do that either. When I learned WinSock I wrote a simple netcat-like GUI terminal emulator and it was <10KB.

Were early Windows systems stack constrained or something?

Win16, certainly. All apps ran in a segmented address space and the stack was usually in the 4-8KB range.

eurekin• 12 years ago

That's very interesting.

Are there any tutorials/articles on how to do it?

vidarh• 12 years ago

My one development job involving Windows code was a cross-platform Windows / some-embedded-os-that-used-Win-32 / Linux app that ran an FTP server and SNMP server.

No GUI stuff, but all the differences between Win32 and Linux for the APIs I used were abstracted away with a few handfuls of typedef's and #define's.

ww520• 12 years ago

Stack was pretty small. Allocating structure on the heap instead of on the stack is pretty standard. We just need to understand the historic context before making judgement.

swalsh• 12 years ago

I first got into programming when I was around 13. I wanted to make games (little did I know how big an order that was!). So eventually I stumbled upon code that looked a lot like this, and I started hacking it out. I would tweak things until I understood why it was there, and then move on. I guess since I didn't know that something better existed it was the only world I knew. I had a lot of fun playing around. What a different world we live in now though. Stuff that would have took me weeks to figure out can be done in 5 minutes today.

AceJohnny2• 12 years ago

making you use WinMain() instead of main()

As I recall, there was a different entry point to distinguish DOS from Windows programs. A Windows program started in DOS would have a stub main that just printed out the "this is a Windows program" warning.

It's actually quite a reasonable segregation to do when you consider the different kinds of setup each type of program requires, not that there aren't other ways around it, of course.

stormbrew4• 12 years ago

That's a different thing, actually. There are two distinctions here:

- Win16/win32 programs in PE object files which could be: -- Console subsystem, where they'd have a stdin and stdout and all that, take a char* argv, their entry point was main(), etc. and if run directly would open a console window, and if run from a console would block the console with their own execution. -- Windows subsystem, where they would have the more elaborate entry point (WinMain), could get unicode argv, etc. If run directly they would not spawn a console, and if run from a command line would appear to return immediately. - DOS EXE files, which could obviously only be 'console' and only had one entry point.

PE executables (of either sort) had a stub DOS exe at their header that would print that message.

As another post pointed out, nothing was actually stopping you from using the gui windows apis from a console app, incidentally. It wasn't done very often though.

rickp• 12 years ago

Your stack, globals, and local heap were all allocated out of the same 64KB segment. Any locals in WinMain are effectively allocated for the entire lifetime of your app, and took up some of that limited space.

Normally, you'd save the stack space by putting the structure definition in a separate function, and you'd regain the space on return from the function. But for these Hello World apps, you're over-simplifying to death, trying to reduce the number of functions. I suspect the LocalAlloc here is vestigal, from an earlier hello app where the HelloInit function was inline.

atburrow• 12 years ago

Does anyone have a mirror? The link appears to be dead.

mgaut72• 12 years ago

I thought that was the joke....

csixty4• 12 years ago

https://archive.today/dmmJI

bluedino2• 12 years ago

I couldn't even imagine the breath of fresh air it must have been to have worked with Windows programming in the 80's and then discovered something like NeXTSTEP.

protomyth3• 12 years ago

It was pretty amazing although VB probably gave a lot of people the same reaction. NeXTSTEP and the Newton were probably the two biggest reasons I always hated Windows programming. Both had great interface ideas that Microsoft should have stolen.

I swear I muttered "OnOK" in my nightmares for years.

threeio• 12 years ago

I just miss the concept of Soup storage on the Newton ;)

dugmartin• 12 years ago

Coming from DOS programming where you had to draw your own "windows" using using extended ASCII characters and listen for interrupts to handle mouse clicks I thought Win32 API was awesome. I think NeXTSTEP would have blown my mind if I had chance to touch a NeXT box.

protomyth3• 12 years ago

You spend a goodly amount of time looking at the interface builder and trying to figure out why you did not have to add code to the button or subclass it. There is a definite "wow" moment.

pjbrunet• 12 years ago

Drawing your own Windows was great because every application had its own character and style. It was an art. Then Windows came and made everything boring. (I see Zuckerberg following Gates footsteps--let's make the interface boring.) My mind was blown by Amiga--the music, multitasking, deluxe paint, video editing, 3D animation, it was a machine for real professional artists that made Windows & Mac look like a Timex Sinclair.

lobrien• 12 years ago

Actually it wasn't as revelatory as you might think.

Windows programming was irrelevant in the 80s -- it wasn't until Windows 3 and 3.1 that it really caught on. In the DOS world there were lots of "character-mode windowing libraries" (I mean lots : I think we once reviewed 15 in a single issue of Computer Language) and they all generally had fairly straightforward programming models. People were much more focused on, for instance, memory management.

When NeXTSTEP came out, word traveled by the magazines and BBSes and Compuserve fora. You couldn't really communicate the NeXTSTEP programming model in pictures and so the only people who I think really got NeXTSTEP were people who'd already experienced Smalltalk! (Which, at the time, was at it's peak, with Smalltalk/V available on DOS machines and ParcPlace available on UNIX systems.)

rozzie2• 12 years ago

I've uploaded three versions of the missing source code from floppies, along with windows.h for each.

http://ozz.ie/h31100

chinpokomon• 12 years ago

Thank you, Ray. Any old compilers in your time machine as well? ;-) I had (incorrectly) presumed that this older K&R style of C predated Windows and was never ported from Unix.

stormbrew4• 12 years ago

Until a few years ago you could still find K&R declarations in the Ruby source code[1]. Wasn't until 1.9 that they switched things over to ANSI style, iirc.

So, believe it or not, you might still find it out there in the wild.

[1] eg: https://github.com/ruby/ruby/blob/v1_8_7_374/compar.c

rozzie2• 12 years ago

Of course. I've uploaded a UTILITIES folder that has all the build tools. In BIN, cc.exe is the compiler, but it forks the P*.exe's for each pass.

sneak• 12 years ago

Might it be worth having these historical artifacts mirrored in the appropriate places, like the IA or the Computer History Museum?

slipstream-• 12 years ago

Any chance of uploading 1:1 floppy images of those? Been looking for Windows Premiere Edition for a while, and now parts of its SDK (and a Win 1.0 beta's!) shows up, and from Ray Ozzie too!

CurtHagenlocher• 12 years ago

From what I remember, the equivalent code in something like Toolbox or Xt wasn't better or easier -- just different.

ultramancool• 12 years ago

Quite amazing to see how Qt compares to this these days. Can accomplish this in a few lines without any generated code or other nonsense.

wolfgke2• 12 years ago

Qt wraps both XLib and WinAPI. So the same argument can be brought for WinAPI, too.

pjmlp2• 12 years ago

Actually in the case of Xlib and Xt it was even worse.

Motif made it better, but still.

My exercise for fellow developers joking about the Windows API was to ask them how they would do similarly in Xlib/Xt/Motif.

turnip1979• 12 years ago

Brings back memories. I recall learning to write from memory the 40+ line win32 hello world from the Petzold bible. That's the way it was back then ... but you got super speed from it. Event loop was blazing fast. I discovered MFC a bit later and worked as a professional dev at an ISV that made use of it. I have fond memories of reading the v1 MFC manual - it was small enough at that time so that a person could understand the whole thing with relative ease.

Some interesting notes on what happened in a few years .. when I first saw .net Compact Framework, it blew my mind how easy it was to write .net code/WinForms stuff. Even though I was relatively poor, I bought an MSDN subscription on the spot and started to write mobile apps. This was early 2000s. If MSFT had paid proper attention to mobile devices, they could have rocked it.

I faced weird challenges. Companies would ask me for 20K to certify my free apps on their phones. It was disgusting. I sold a few apps on the hangango app store, and then got out of mobile dev (missed the whole iPhone revolution). Talk about bad timing.

leaveyou• 12 years ago

it brings back memories indeed.. me watching in disbelief how long, ugly and weird Microsoft C/C++ Win32 or even MFC programming for Windows was, compared to Borland Delphi or C++ Builder. I tried, I really tried to like it but in the end I gave up and I became a windows programmer much later with the arrival of .net win forms and c# (created by Delphi architect wikipedia.org/wiki/Anders_Hejlsberg - it's scary how scarce were the sane language/framework/IDE architects back then)

jmount• 12 years ago

Wow. Interpreting this under the "hello world is often a demonstration of a particular point" rubric ( http://www.win-vector.com/blog/2008/02/hello-world-an-instan... ) it looks like Petzoid was trying to demonstrate that a mere finite amount of code could launch a Window API client. Large was, of course, undesirable- but he was trying to see if the API costs were even bounded.

ChuckMcM2• 12 years ago

A story. As Petzold writes, the Windows version of "hello world" was amazingly baroque. It bothered me, a lot. I had come up in the four line hello.c world of UNIX, and later the make/test world of UNIX, and when I had left Sun the idea of writing code like the Hello World example was something I could not bring myself to do. I rebelled in a bunch of different ways, but the worst part was it started a period for me where I programmed a lot less, I was getting curmudgeonly because programming had become so screwed up.

The thing that changed for me was getting a chance to help a young man with his programming who had come up through Windows rather than something easier, and he was full of all the wonder and excitement I had had at his age for programming, and had no idea that he was running with a large handicap to his productivity. And partly because many of the "tools" you had to have like Visual Studio worked their butt off to make that gap small, from self writing code to documentation at your fingertips. I realized that if I didn't get the 'wonder' back of writing code I was going to become one of those guys who sits around grumbling about how in my day things were so much better and FORTRAN just worked dammit :-).

What I tried to learn from it was that systems with great generality suffered from usability efficiency. The reason UNIX hello.c was so simple is that UNIX programs, by and large, did simple things, algorithmic computation driven by interaction over a character based interface. One way to drive effecive use of a system was to either eliminate or otherwise hide the modalities that were "possible" but not probable in future execution of the program.

That has helped me see when I'm building something that is letting to much generality leak out to the surface.

jerf• 12 years ago

"What I tried to learn from it was that systems with great generality suffered from usability efficiency."

I'm entering "curmudgeon" age myself, at least for our industry, and what I'm getting curmudgeonly about are my fellow curmudgeons, and that's part of why. Oh, programming was so much better in the 8-bit era? So go back there then. It's still available to you if you want it! Oh, you don't want to take actions to match your words? Why not? Because, in a nutshell, no it wasn't easier. It was easier to do the things it could do, sure, but it can't do much. So stop telling me about how wonderful the 8-bit era was.

That's just one example. There's several of these recurring rants that come up here on HN periodically.

ChuckMcM2• 12 years ago

Well said, in particular this bit, "Oh, you don't want to take actions to match your words?"

There is a downside though when you get a curmudgeon with a fondness for the "old" days and takes action by re-implementing your build system to look like something you would read about in an IBM JCL handbook :-) I'm sure 8 character error codes were easy to remember but even better is a short sentence on what the error was.

weddpros• 12 years ago

It says "Service Unavailable"... Fun!

bkeroack• 12 years ago

I had that book (Programming Windows) as a preteen. This is actually Win16 API programming--the event loop gives it away. This was before Windows had preemptive multitasking (which came with Win95, IIRC). If you forgot to yield at the bottom of the loop you'd lock up the whole OS.

Fun times.

wolfgke2• 12 years ago

The same program for GNU/Linux/XLib is not much less complicated: http://www.paulgriffiths.net/program/c/srcs/helloxsrc.html

Under Windows it is easy to create a console application, which is similar to the usual C helloworld. But the application here creates a windows to show hello world - so it's fair to compare it with a similar program on GNU/Linux/Xlib.

EDIT: Yes, it is much easier to use something as Qt on GNU/Linux, but there are also abstractions around the WinAPI available, so using Qt on GNU/Linux for a helloworld example yields no fair comparison.

twic• 12 years ago

That's not fair. On Ubuntu it's part of systemd already.

dfox• 12 years ago

I would say that this X hello world is needlessly complicated by things that are not strictly needed like WM hints and font selection.

thorin• 12 years ago

My Master's thesis in electronic engineering was supposed to be based on a previous work written by a maths phd in c++ on XWindows. I'd only used C and Basic before and not done any GUI programming. I couldn't believe how much code was involved to set up a few simple windows. It really put me off and I pretty much gave up and started writing my own [image manipulation] code in C. I wish I'd had the sort of resources you can get for free nowadays available it might have been a fun project!

It was the mid 90s and I didn't have unix/linux at home. In those days you had to pay for a compiler on windows and I didn't know anyone running linux. Just a couple of years later it was so much easier!

breadbox• 12 years ago

Certainly there is advantage in having a simple hello-world program. The original purpose of hello-world, naturally, is just to have something to verify that your environment and tools are properly installed and configured before you go off and do anything serious.

But because Petzold used "his" hello-world program to introduce you to a bunch of critical, foundational concepts -- to show you everything you needed in order to get a functioning top-level application window open on the desktop -- I myself found it a great place to start. By forcing you to confront the message loop from square one, it really reinforced the idea that you aren't in Kansas anymore (and you can't do things like the Kansans do).

As Petzold hints in his article, the original version of the book (for Windows 2.x) didn't even start with that program, but worked its way up to hello-world through a sequence of 5-10 iterations that got you to that point over the course of an entire chapter. (Thus for example one iteration actually opened a top-level window but lacked a message loop, so the program exited and the window immediately disappeared after it was created.) It makes sense that he had to edit out such a lackadaisical introductory chapter as the subject of Windows programming got bigger and bigger, but I for one always liked that chapter.

forca• 12 years ago

I really miss the "old days" of programming. Things were simpler then. I fondly remember the early 80s programming on a Commodore 64, then moving to the early IBM machines and DOS.

Even though I favour a nix environment, there is more sanity in Windows programming circles methinks. I have been thinking about getting into some .NET development these days and moving away from doing nix stuff, altho I really love the BSDs.

moron4hire• 12 years ago

It's pulling teeth to get 3rd party library developers to make anything up to the standard of the BCL. A lot of times it feels like someone just copied a Java library into C# and did a bunch of global text replaces until the compiler stopped complaining. The BCL is very extensive, but if it's not already in the BCL, .NET is a pain.

billforsternz• 12 years ago

I remember back in the day being annoyed and frustrated that the two functions in the canonical Petzold "Hello World" were WinMain() and WndProc(). "Why not WinMain() and WinProc() ?" I used to silently scream. I must be getting more mature because I am not quite so pedantic about these sorts of things these days.

nevster• 12 years ago

As a unix/c coder in the early to mid 90's, the future seemed to be Windows and I dreaded having to learn that gobbledegook. Then Java came along and I never had to. Thank goodness.

goalieca2• 12 years ago

It's funny, but AbstractBaseFactoryLongName is less annoying than LPSTERVOIDFOO _PACSCAL nonsense.

megablast• 12 years ago

That version probably has as much relevance to real-life Windows programming as printf("hello world") has to real-life character-mode C programming.

A simple program has a lot of advantages, it shows you can actually compile and run a program. When you have never done so before, there are 100s of hurdles that regular developers would not even consider when starting out.

The Infamous Windows “Hello World” Program

Related Stories