Microsoft's CoApp To Help OSS Development, Deployment

badpazzword writes "Microsoft employee Garrett Serack announces he has received the green light to work full time on CoApp, an .msi-based package management system aiming to bring a wholly native toolchain for OSS development and deployment. This will hopefully bring more open source software on Windows, which will bring OSS to more users, testers and developers. Serack is following the comments at Ars Technica, so he might also follow them here. The launchpad project is already up."

Comments

um...

[Anonymous Coward]

Serack is following the comments at Ars Technica, so he might also follow them here

Yes, I'm sure he's following all 0 of them...

I'll follow them here too. :D

[His name cannot be s]

Ask me about CoApp, I'll tell ya everything ya wanna know.

Garrett Serack
CoApp Project Owner

Re:I'll follow them here too. :D

[Meshach]

In the same vein feel free to ask me about Linux.

Linus Torvalds
Linux Kernel Founder

Re:I'll follow them here too. :D

[h4rr4r]

When will MS be pulling the rug out from under the community?

How much of a fight will we seen when someone tries to packup an app that competes with an MS product?

Re:I'll follow them here too. :D

[tomhudson]

Ask us about nuclear disarmament.
We'll tell you everything WE want you to know.

Barack Obama
Dmitry Medvedev

Re:I'll follow them here too. :D

[His name cannot be s]

Well, considering that I spent several months hacking thru red tape to get VP approval, and the enthusiasm that I've been getting, I'm pretty damn confident that we're clear sailing.

And given the first three targets that on my radar are PHP, Apache and Python (and the 40 or so shared library dependencies), and that's what I took to the VP, I'm fairly confident that's not going to be an issue.

And, on top of that, MS doesn't own the project, I do. "Shutting it down" is not an option for them.

Re:I'll follow them here too. :D

[DoofusOfDeath]

Ask me about CoApp, I'll tell ya everything ya wanna know.

How do I know that MS won't file a software patent related to this work?

Re:I'll follow them here too. :D

[vux984]

You do know that the work agreement that you signed during orientation stated that Microsoft owns any software that you produce on your own time, as long as Microsoft may compete against said software at some point in the future?

Perhaps that was some of the red tape that needed to be cut. Guess what, you can run things past management, and get legal to sign off on something that amounts to an agreement between the employee and employer that a given project belongs soley to the the employee. I don't know about microsoft specifically, but lots of companies are amenable to this sort of thing.

Sometimes there are legitimate concerns that have to be resolved... often its just a matter of jumping through the required hoops.

Re:I'll follow them here too. :D

[causality]

Ask me about CoApp, I'll tell ya everything ya wanna know.

Garrett Serack CoApp Project Owner

I'll bite. Given Microsoft's track record, particularly its embrace-and-extend tactics, its questionable business practices, its status as a convicted monopolist, its use of vendor lock-in, its related use of proprietary file formats, and the Halloween e-mails from top management clearly defining Open Source as an enemy, I have just one question: why should we trust them?

Most (nearly all) of the upper management people who arranged everything I just listed are still working at Microsoft.

"Fool me once, shame on you; fool me twice, shame on me."

Re:I'll follow them here too. :D

[theskipper]

But won't all this lead to competition for Sharepoint/IIS/Office for Windows users? Why would Microsoft want to sanction any alternatives to their proprietary offerings? In other words, what's the angle?

Re:I'll follow them here too. :D

[Red Flayer]

Hey Garrett, I'm cautiously optimistic.

But I really have to comment on one of the items from your announcement post, regarding the aims of the project:

Also be Windows admin friendly. Even if it's open source, you shouldn't have to be a developer to put Open Source applications on Windows.

Is this some kind of back-handed comment based on the general view at Microsoft about Open-source software, or the general view that MS would like to push out to userland? That people should use MS OSS because you need to be a developer to use it on other platforms?

Re:I'll follow them here too. :D

[Cryacin]

I'm Santa.

Ask me about anything to do with snow, the north pole, midgets, flying reindeer and flying midgets.

Re:I'll follow them here too. :D

[sortius_nod]

It all feels too much like a dirty beat up van in a shopping mall parking lot with "free candy" painted on the side...

Re:I'll follow them here too. :D

[Anonymous Coward]

Curiously, given microsoft's recent wrist slappings by the EU, fostering the development of "Competing" products could help microsoft, rather than harm it;

Take for instance, the recent "Browser choice" screen. If Microsoft had been fostering a package downloader at that point in time, then they would have not needed to do anything to comply with the EU. Their OS would already have IE by default, and "Offer" a nice little package handler for those "Other Browsers".

If the EU were to press, and try to stick MS with the stigma of not actually wanting any other browsers to run on their OS, by making users use a round-about way of getting their browsers of choice, MS could point the finger right back at how much capitol they invested into the alternative software ecosystem, and how they leveraged their power to help bring FOSS and the package manager to their OS.

In short, creating a package manager like this is a good way for MS to be more two-faced than ever.

Not that I am gonna complain; ALL corporations are two-faced, and a well supported package manager, and better acceptance of the win32 platform (Not just windows, there are attempts at FOSS Win32 platforms.) by the FOSS community is a good thing all around.

I just dont think MS is overly concerned that it will compete with their software ecosystem at this point, and is more convinced that government regulators are the bigger threat.

Re:I'll follow them here too. :D

[grcumb]

Ask me about CoApp, I'll tell ya everything ya wanna know.

Garrett Serack CoApp Project Owner

Okay, serious questions:

Assuming that you've looked at APT and similar packaging tools, and given that you're still convinced that there's a 'Windows Way' (your term) to handle deployment that differs from Linux best practices, how do you plan to address:

• Package Repositories - This is one of the main strengths of Debian and related distros. Do you think it's even possible to replicate this level of community control in Windows? I know you've mentioned decentralisation, but have you considered the implications of such an approach? What is the cost of failure to affect consistent, formalised management of package builds?
• Dependancy Management - This issue is largely done and dusted on Linux, but remains a dog's breakfast on Windows (albeit not as frustrating today as it was in the mid-90s). In the absence of centralised repositories and the Unix toolchain philosophy, how do you propose to cope better with dependancies?
• File locations - How do you propose to manage the proper placement of libraries etc. when the conventions concerning where to put such files are not nearly as well defined on Windows? I'm suggesting here that you need cultural leverage rather than technical answers. You need to change perceptions, not toolkits.
• Security - Do you think it's even possible to replicate one of the main strengths of Linux package repositories: the ability to curtail security risks such as malware and flawed code?
• Scripting Interfaces - Say what you like about make and other command-line utilities, but as a busy sysadmin, I consider GUI package management a waste of my valuable time. If I'm going to deploy regular security updates, for example, I want to know that I can script every aspect of the operation. Even the tab-completion features in aptitude make it many times more efficient than a point-and-click interface. What is the potential for scripted deployment/management of packages under your system? Why?

I guess it's clear by now that I'm suggesting that what Windows needs is not another new way to do things. Package management in Debian, for example, is vastly more advanced and sophisticated than anything on Windows, and yet you feel the need to do things the 'Windows Way'. Don't you think you'd be better off learning from others who have been dealing successfully with package management for over a decade now?

These are all serious questions and I expect to be challenged by your replies. I applaud your courage in taking on this huge task. I also think that you're going to need to learn a lot more humility than you've demonstrated so far if you want to achieve something better than a new brand of anarchy in packaging.

Re:I'll follow them here too. :D

[jtdennis]

I'm a Windows admin that's also comfortable in the Linux world, and work with someone that doesn't have Linux experience. I can tell you from experience that I am much more comfortable diving into config files trying to get something to work than he is.
Many people with primarily a Windows background expect config to be done via the UI and not via text files as many OSS projects are. I believe that this is what he was implying.

Re:I'll follow them here too. :D

[Culture20]

And, on top of that, MS doesn't own the project, I do. "Shutting it down" is not an option for them.

Microsoft CEO: "Buy him out, boys!"
His name cannot be s (16831): "Hey, what the hell's going on?"
"Oh, I didn't get rich by writing a lot of checks. Muhahaahahahaa!"

Re:I'll follow them here too. :D

[martin-boundary]

I'm Batman.

Uh. That's it.

Re:I'll follow them here too. :D

[h4rr4r]

At least you can diff a config file. Try that with a gui.

Text based config, with an option gui/wizard really should be the only way this sort of thing gets done.

Re:I'll follow them here too. :D

[Anonymous Coward]

What color is it?

Re:I'll follow them here too. :D

[His name cannot be s]

I use that all the time!

Re:I'll follow them here too. :D

[h4rr4r]

How are you handling dependencies?
Will this be the standard windows every app carts around all its own libs, wasted space and outdated/insecure funland?

Re:I'll follow them here too. :D

[His name cannot be s]

That is precisely the red tape that I had cut.

Microsoft has given me a signed contract that says that whatever I produce for the CoApp project isn't owned by them. They do get a license to everything I make (fair deal), but they don't own it in the end.

That, and I've also chosen the BSD license for it's do-what-the-f*-you-want spirit.

Re:I'll follow them here too. :D

[His name cannot be s]

No.

My intent is to completely do away with the practice of everybody shipping every damn shared library. It's one of the things that piss me off the most. I've got a very workable solution that uses WinSxS to cleanly handle this.

It is extremely important that there is a unified method for sharing libraries between apps.

Re:I'll follow them here too. :D

[Dragoniz3r]

Can it run Crysis?
Just kidding. Really, I'm curious how you're planning on exposing the dependency management to the developers. Would there be a GUI tool where you drag+n+drop dependencies to create a configure-script-ish data blob that the installer would use to figure out if you have everything you need? Would I be able to go ahead with compilation even if some dependencies aren't met, just with those features disabled? Imagine for instance that I'm building a media player and some codec libs aren't present. I'd want to be able to still build the player, just without the code to activate those codecs (let's assume for the sake of the example that I don't want to putz around making it check for codec DLLs at runtime and load them, for whatever reason).

Re:I'll follow them here too. :D

[h4rr4r]

If you only succeed in getting windows folks to learn this lesson you should be made a saint.

Re:I'll follow them here too. :D

[His name cannot be s]

Is this some kind of back-handed comment based on the general view at Microsoft about Open-source software, or the general view that MS would like to push out to userland? That people should use MS OSS because you need to be a developer to use it on other platforms?

No-no.. exactly the opposite

Have you tried to roll out some OSS apps on Windows?

On Linux it's two clicks, and BAM! Done.

On Windows, it's almost never that easy to setup OSS apps.

The problem I see is that it doesn't take a Developer on Linux to get Apache installed and configured. Why should it on Windows?

Re:I'll follow them here too. :D

[tpstigers]

So how come it's not being called CANE?

Re:I'll follow them here too. :D

[Anonymous Coward]

I'll tell ya everything ya wanna know.

This is nice and all for OSS projects, but when is Microsoft going to fix the Microsoft Download Center? Seems kinda odd that they would put effort into OSS dependency management when depreciated packages in the download center don't have links to the most updated version.

Re:I'll follow them here too. :D

[h4rr4r]

Because windows lacks proper package management?

Windows has some nice user stuff, but the package management situation is a total joke. Something like aptitude or apt-get or even yum would be huge.

Re:I'll follow them here too. :D

[His name cannot be s]

If it does, so be it.

I've spent the last couple of years at Microsoft working to make PHP better on Windows, and validating PHP apps including CMS systems like Drupal on Windows. Seems to me they want some competition.

Re:I'll follow them here too. :D

[__aasqbs9791]

I thought that was an attempt to get a stripper out of jail when I saw it.

Re:I'll follow them here too. :D

[VoltageX]

Please, please, please, please make it easy to roll a python app into an MSI.

Re:I'll follow them here too. :D

[Anonymous Coward]

I'm Santa.

There you are, you fat bastard. Where's My PONY?!?

Merry Christmas,
Virginia

Re:I'll follow them here too. :D

[His name cannot be s]

Assuming that you've looked at APT and similar packaging tools, and given that you're still convinced that there's a 'Windows Way' (your term) to handle deployment that differs from Linux best practices, how do you plan to address:

Yes, I've worked with APT and RPM for a very very long time now. The reason I'm convinced there is a 'Windows way' is because it's a different system that Linux; yes, I've learned a lot about PMS from Linux, and I know how to apply that knowledge to Windows.

Package Repositories - This is one of the main strengths of Debian and related distros. Do you think it's even possible to replicate this level of community control in Windows? I know you've mentioned decentralisation, but have you considered the implications of such an approach? What is the cost of failure to affect consistent, formalised management of package builds?

I have a plan for allowing any publisher to publish packages in the CoApp ecosystem, provided they meet two qualifications:
- They must be able to host their repository meta-data on an SSL protected connection.
- All packages must be digitally signed with a certificate that chains back to to a commonly-accepted CA.

Dependancy Management - This issue is largely done and dusted on Linux, but remains a dog's breakfast on Windows (albeit not as frustrating today as it was in the mid-90s). In the absence of centralised repositories and the Unix toolchain philosophy, how do you propose to cope better with dependancies?

I'm working with the developer of WiX to ensure that we can trivially build chained MSI packages that have the necessary smarts to properly manage this. Kind-of mixing in something like ldconfig with the Windows SxS library management.

File locations - How do you propose to manage the proper placement of libraries etc. when the conventions concerning where to put such files are not nearly as well defined on Windows? I'm suggesting here that you need cultural leverage rather than technical answers. You need to change perceptions, not toolkits.

Yes. The change starts with PHP, Apache, and Python, and the 40+ packages needed to build them (community members from each are already on board) Half of the project is setting some intelligent standards, and then bootstrapping the ecosystem with packages to enable other software to follow.

Security - Do you think it's even possible to replicate one of the main strengths of Linux package repositories: the ability to curtail security risks such as malware and flawed code?

Yes. By requiring code-signing (and I've got a plan for opening that up without cost for smaller projects) we can replicate the benefits of MD5 and PGP signatures found in the Linux world.

Scripting Interfaces - Say what you like about make and other command-line utilities, but as a busy sysadmin, I consider GUI package management a waste of my valuable time. If I'm going to deploy regular security updates, for example, I want to know that I can script every aspect of the operation. Even the tab-completion features in aptitude make it many times more efficient than a point-and-click interface. What is the potential for scripted deployment/management of packages under your system? Why?

I agree 100%. Scripting interfaces are an absolute requirement, and will likely come well before the GUI.

Think of it as a clean adaptation of the same concepts to the model that will be attractive to Windows developers.

I also think that you're going to need to learn a lot more humility than you've demonstrated so far if you want to achieve something better than a new brand of anarchy in packaging.

I apologize if I'm coming off arrogant. Frankly it's taken an extremely long time to convince the powers-that-be at Microsoft that Linux's package management is stellar compared to Windows. It's also not near as hard or large as it sounds, I'm walking on the shoulders of giants here, both in the Linux and Windows worlds.

Re:I'll follow them here too. :D

[tomhudson]

I'm Darl McBride, and I'm dyslexic.
So, Satan, I sold you my soul, and what the (*^*%^&^%$S did I get in return???
I'll see you in HELL!
You can have my soul when you pry it from my cold dead body!

Re:I'll follow them here too. :D

[stuckinphp]

Honestly, I'm just waiting for the MS fan boy kids they are cranking out of schools these days to start spilling details of this great new thing MS just invented called a package manager too all their friends.

Re:I'll follow them here too. :D

[His name cannot be s]

Dammit! I pooched some quote tags!

Sorry about that. Should have previewed. *sigh*

Re:I'll follow them here too. :D

[toastar]

At least you can diff a config file. Try that with a gui.

I do that that all the time, screenshot + OCR only takes me 30 secs to pop the data into excel.

office/windows is a pretty stable work platform, Linux I'm always spending 15 minutes getting a screenshot app working.

Re:I'll follow them here too. :D

[WrongSizeGlass]

I'm pretty damn confident that we're clear sailing.

I'm fairly confident that's not going to be an issue.

And, on top of that, MS doesn't own the project, I do. "Shutting it down" is not an option for them.

Um, then what are you doing wasting your time here on /.? Shouldn't you be locked in a caffeine fueled coding frenzy, programming until your fingers are bleeding? Open source software won't write itself, you know ;-)

His name cannot be s (16831)

Is that a hint? Does that mean it could be one of the other 25 letters? Or maybe one of the 20 remaining consonants?

Re:I'll follow them here too. :D

[Compholio]

Ask me about Grim Fandango.

Re:I'll follow them here too. :D

[CAIMLAS]

I just dont think MS is overly concerned that it will compete with their software ecosystem at this point, and is more convinced that government regulators are the bigger threat.

Yep, exactly. As it pertains to Sharepoint, the more something sucks, the more it seems to get broad adoption early on.

Quite smart of Microsoft to make Sharepoint a "killer app" at that - all-encompassing and utterly impossible to replace in whole.

Re:I'll follow them here too. :D

[martin-boundary]

Having read your blogpost, I can see what you're trying to do, but as a Linux/Unix developer, I have zero interest in running through Windows like hoops. *But* I do go to great lengths to follow POSIX standards, and make sure that my autoconf tarballs are clean, and I don't expect this to change any time soon (or even not so soon).

If your target audience is like me, then you're best off creating an automated conversion tool that can take a standard tarball and create an MSI package (or whatever) to your specifications with minimal human intervention. Ideally, this ought to extend seamlessly to the "make check" incantation, which is an important sanity check for cross platform development, since merely compiling the source successfully is no guarantee of correctness.

Note that doesn't mean that you have to accept *nixish directory names etc, it just means that when such a tool sees a standard tarball construct, it knows how to convert it to something sensible for the Windows platform.

As you pointed out yourself in the post, standard tarballs just work (mostly). You can gain a lot by reusing this property as a foundation for your project, rather than expecting people to adapt to your own design.

Re:I'll follow them here too. :D

[Lingerance]

I get the impression that's what this is doing.

Re:I'll follow them here too. :D

[His name cannot be s]

Um, then what are you doing wasting your time here on /.? Shouldn't you be locked in a caffeine fueled coding frenzy, programming until your fingers are bleeding? Open source software won't write itself, you know ;-)

I know!!!!

"His name cannot be s (16831)"

Is that a hint? Does that mean it could be one of the other 25 letters? Or maybe one of the 20 remaining consonants?

Well, ya see... with a five-digit slashdot-id I originally had "His name cannot be Spoken" as my name... then they did some database truncation about 12 or so years ago, and I lost some letters.

And ya can't change your name on Slashdot, and I didn't wanna give up my 5 digit ID. :D

Re:I'll follow them here too. :D

[h4rr4r]

Please tell me that is a joke.
How do you get a screenshot of all the submenus at once? Think IIS manager for an example. How does the OCR app know what is a titlebar and what is an option?

Re:I'll follow them here too. :D

[msclrhd]

What about libraries built with different compilers and compiler versions?

Problems include:
    1/ dealing with memory heaps from different C/C++ runtimes
    2/ changes in C++ standard library type representations
    3/ changes in exception handling logic; throwing an exception from a gcc C++ shared library to an msvc one
    4/ changes in behaviour and binary compatibility of debug/release versions, and components built with different flags/compiler settings (e.g. building without wchar_t and with wchar_t as a native type results in different ABIs when using std::wstring types (basic_string vs basic_string) for msvc compilers)
    5/ building with _DEBUG vs NDEBUG

These reasons and more are why binaries from different build settings, compilers and compiler versions are not mixed and one of the reasons why you wnd up with duplicates of the same shared library (especially on Windows).

Are you aware that:
    1/ Microsoft are stopping using WinSxS assemblies for managing the C/C++ runtimes as it is complex to manage and get right;
    2/ With XP, Microsoft were selling WinSxS as being able to deploy different versions of the binaries, but for Vista/Win7 they are now saying that WinSxS is for archival purposes (see the Engineering 7 blog)
    3/ It does not really work as intended in practice -- e.g. comctl32 version 6 is different in Vista/Win7 than in XP, yet the applications that reference the XP version use the Vista/7 version

Re:I'll follow them here too. :D

[His name cannot be s]

Please, please, please, please make it easy to roll a python app into an MSI.

One of the first people on board with the project is Trent Nelson; he's all about Python.

I think we're gonna cover that.

Re:I'll follow them here too. :D

[markdavis]

>That, and I've also chosen the BSD license for it's do-what-the-f*-you-want spirit. I think you had no choice to choose the BSD license instead of the GPL. Had you chosen GPL, it is likely the project would have been immediately rejected by Microsoft.

Re:I'll follow them here too. :D

[His name cannot be s]

As for the first five points, yes I'm aware of all of that, and I'm working to solve all of them. Some of them are not possible (mixing compilers has a lot of bad mojo) and some are solvable with some really good best practices.

1/ Microsoft are stopping using WinSxS assemblies for managing the C/C++ runtimes as it is complex to manage and get right;

Ah, Visual Studio is backing away from WinSxS. I read their justification. I didn't buy into it. I think it's a solvable issue.

2/ With XP, Microsoft were selling WinSxS as being able to deploy different versions of the binaries, but for Vista/Win7 they are now saying that WinSxS is for archival purposes (see the Engineering 7 blog)

Uh, what? I've been talking to the maintainer of the WinSxS system. He's fully supportive of my plans.

3/ It does not really work as intended in practice -- e.g. comctl32 version 6 is different in Vista/Win7 than in XP, yet the applications that reference the XP version use the Vista/7 version

It works just fine, as long as you use it correctly; if they didn't, it's not my fault. Some of the tools I'm building will make it easier not to screw up.

Re:I'll follow them here too. :D

[gzipped_tar]

It's 2010 and you are still doing *that*.

Re:I'll follow them here too. :D

[Cryacin]

FROM: Santa's Workshop helpdesk
TO: Virignia

We have finished the research on flying midgets. We are moving on to flying goats next, which should let us finally allow pigs to fly.

We thank you for your patience whilst we satisfy your mother's prerequisite requirements on when you can have a pony.

Re:I'll follow them here too. :D

[h4rr4r]

By killer you mean those that admin it wish to kill themselves right?

The thing even lacks the ability to use smtp auth.

Re:I'll follow them here too. :D

[Larryish]

Because Jesus made it that way.

Re:I'll follow them here too. :D

[Animaether]

My intent is to completely do away with the practice of everybody shipping every damn shared library.

If you only succeed in getting windows folks to learn this lesson you should be made a saint.

The major problem with this is that, as mentioned, Windows doesn't have a package manager, and Microsoft keeps telling developers that they cannot expect a user to have internet connectivity.

So when you compiled your application with Visual Studio 2008 SP1 with the ATL update installed - which means every user of your software will have to have the Visual C++ 2008 SP1 ATL runtime redistributable package installed as well, you're left with scant few options.

The most reasonable of which are:
A. If you're distributing something boxed, to include the redistributable package on the media (CD/DVD/USB stick/whatever).

B. If you're distributing something via downloads:
B.a. Include it because - again - you're not supposed to assume the user will have connectivity.
B.b. Don't include it, but detect whether the user has it installed and has internet access, and then offer to download it and install it (silently or otherwise).

Of course for option B.b., Microsoft further seems to suggest that you do not link to -their- download pages (after all, the URLs could change, etc.) but instead host the binaries yourself.

The only reason, thus, that Windows developers tend to include or download shared libraries at runtime, is simply because there -isn't- a package manager for Windows.

So don't blame the developers - blame the lack of a package manager. Which I fully welcomed the last time a topic hinting at a package manager popped up on /.
Unfortunately it seems like they would be two rather separate projects?
http://it.slashdot.org/story/10/03/24/189248/Microsoft-To-Distribute-Third-Party-Patches

Re:I'll follow them here too. :D

[TubeSteak]

I'm Alfred the Butler.
Would you like a cup of tea Sirs?

Re:I'll follow them here too. :D

[codepunk]

"On Windows, it's almost never that easy to setup OSS apps"

Probably because most of us do not even care if it runs on Windows or not.

Re:I'll follow them here too. :D

[h4rr4r]

While you are making miracles talk to somebody about getting us bash on windows. No powershell is not good enough. Bash and pipes, objects are no good either.

Re:I'll follow them here too. :D

[exomondo]

Because windows lacks proper package management?

Windows has some nice user stuff, but the package management situation is a total joke. Something like aptitude or apt-get or even yum would be huge.

RTFA? A package management system is exactly what he is working on.

Re:I'll follow them here too. :D

[Wumpus]

I'm pretty sure the titlebars are an option in some apps.

Re:I'll follow them here too. :D

[Anonymous Coward]

I'm Batman.

Ask me about young boys in green tights and masks and I will deny that I ever let them inside my "bad cave".

Re:I'll follow them here too. :D

[His name cannot be s]

think you had no choice to choose the BSD license instead of the GPL. Had you chosen GPL, it is likely the project would have been immediately rejected by Microsoft.

That's not true actually.

I didn't tell anyone what license I was going to use until a few days ago, by which time they'd already signed the agreement.

In addition to that; as a Microsoft employee for Microsoft, I've contributed code to GPL, LGPL, BSD, PHP and Apache licensed projects.

Re:I'll follow them here too. :D

[His name cannot be s]

It's 2010 and you are still doing *that*.

*sigh*

You know, that'd be funny if it was so damn sad. :)

Re:I'll follow them here too. :D

[Jason Earl]

I just want to make sure I understand what you are trying to do.

The aptitude and Windows equivalent of the Debian repository sounds like a no-brainer. Microsoft should have done this 15 years ago. There is nothing the slightest bit controversial there. Assuming you can get enough packages in your repository to make it worth anyone's time, this is sure to be a smash hit.

However, in order for your plan to work, you need a source of Free Software packages for Windows. Apparently you are planning on doing this Ubuntu-style (to the point where you plan on using bzr and launchpad.net). In other words you are going to have a shallow branch of every package that you wish to include in your repository, and each branch will contain the build information necessary to build the package and format it according to your specifications. In fact, it would appear that you are planning to build tools that take autotools-style build systems and turn them into something that Microsoft's tools can use.

That is a tall order. In fact, if your tools get even remotely close to being able to do that for the five (or so) variations of Windows that people find to be interesting, then I would not be surprised if the UNIX people don't borrow your tools to help them migrate away from Autotools. No one actually likes Autotools. And let's not even get started on what your system is supposed to do with projects that use alternate build systems like cmake, scons, or whatever else.

Of course, such a repository would be useful even if it only worked for a handful of applications, which you could probably handle in a mostly manual fashion. After all, most Free Software does build on Windows already if you poke it hard enough. It doesn't tend to build with Visual Studio, of course, so playing nicely there raises the bar considerably.

I do have one question. Why, exactly, do you think that this sort of approach is likely to be easier than doing what Apple did and simply exposing a Posix API that is actually useful?

Re:I'll follow them here too. :D

[h4rr4r]

RTFA? I see you 7 digit people still have not figured out that we slashdotters do not RTFA ever :)

Re:I'll follow them here too. :D

[CAIMLAS]

What existing package manager are you modeling your system after, if any?

Do you have a feature list which you will be implementing (or already have implemented)? For instance: pinning, multiple versions of the same package, multiple "satisfies" vectors, etc.

The lack of proper package management in Windows (for all applications, but particularly Microsoft software) has been a huge beef of mine for the better part of a decade. Glad to see Windows is finally getting it! It'll bring Windows' utility as a server OS up quite a bit (probably up to or above that of something like FreeBSD or gentoo).

Re:I'll follow them here too. :D

[toastar]

IIS?

ok I give, even windows people use apache.

Re:I'll follow them here too. :D

[Runaway1956]

So, what's going on? Has Microsoft decided to commercialize some of the open source concepts? Not a bad move, I guess. If you can convince people to only install software from safe repositories, infections worldwide will plummet. Of course, that won't stop drive-by infections, or "hackers" actively probing systems, but it would cut down on the number of people downloading trojans that claim to protect their computers from trojans.

I may break the rules here, and read TFA. No good slashdotter does that - but I've got plenty of good karma.

Re:I'll follow them here too. :D

[Runaway1956]

Let them do it. I have no problem with it.

Seriously, getting people to install software only from a safe repository would be the greatest security move that MS could make. There are so many security holes in Windows, but the vast majority of crapware is installed voluntarily by some lackwit who responds to a banner ad or something with lots of pretty images and empty promises.

I don't much care how Microsoft spins it, as long as they educate people, and motivate them to secure their machines.

A lot of fanboys have claimed that "Windows can be as secure as Linux". I don't entirely believe that, but we all know that Windows can be a lot more secure than it is!!

Re:I'll follow them here too. :D

[h4rr4r]

My guess is MS will never do that, it would lead to apps being far too portable.

Re:I'll follow them here too. :D

[forkazoo]

One of the first people on board with the project is Trent Nelson; he's all about Python.

I think we're gonna cover that.

I think I may be in your target audience. I'm a Linux guy. I am pretty comfortable developing and deploying in my home environment. I'm not an expert, but I can get by. If you can make it easy for somebody like me to make a .msi package that ordinary Windows users can easily install with a few clicks, then I would probably pay a lot more attention to that half of the world.

Re:I'll follow them here too. :D

[dudpixel]

but....but...this is slashdot...and you're saying Microsoft allows you to contribute code under the GPL?

that cant be true. it just cant be.

can it?

Re:I'll follow them here too. :D

[His name cannot be s]

That's the trouble with the six digit generation. Too Lazy.

Us 5-digit generation look down at you with disdain.

Now, get off my lawn!

Re:I'll follow them here too. :D

[His name cannot be s]

I do have one question. Why, exactly, do you think that this sort of approach is likely to be easier than doing what Apple did and simply exposing a Posix API that is actually useful?

Because, even if we could get a great POSIX experience on Windows, it leaves out Windows developers.

One of my goals is to get Windows developers in the OSS game.

On top of that, there is a hell of a lot of non-POSIX open source software on Windows that needs fixing too.

Look at it this way: Would you respect someone who told you the best way to get FireFox running on Linux was to use some sort of Windows emulation layer... Like WINE? no, because FireFox *can* compile for Linux. Same thing with nearly all Open Source I encounter. I want to get the OSS quality and experience on Windows to exceed commercial developers... it needs the most love.

Like I tell people:
Working as an open source software developer at Microsoft is like being a preacher in Vegas. I figure I'm in the single most important place in the universe that I can be.

Re:I'll follow them here too. :D

[dudpixel]

How do you go about handling different versions of a library?

Will we eventually see the day where Microsoft has a central location for shared libraries in Windows (writable only by "root") and also a decent package management system, you know, like apt/rpm?

This isn't a flame, just pointing out some things that would make Windows fantastic for me. I really really really love the directory structure and package management of linux, and the benefits that it brings. If Microsoft could bring some of that goodness to Windows, I may be tempted to switch...no really. Just think, it could reduce the "clutter" that inevitably builds up in a windows system over time (often requiring the 6-monthly reinstall), and updating your entire system would be possible from a single app. Sorry if this sounds like a troll - it really isn't intended to be.

Re:I'll follow them here too. :D

[Jason Earl]

Microsoft wants apps to be portable to Windows. If they could create a project that allowed people to migrate autotools built apps to a Windows build system that would be a huge win. Especially if there was no way to go back to autotools.

The danger, for Free Software anyhow, of this system is that currently most of the actual development of Free Software applications is done with Free Software tools, and those Free Software tools pretty much assume a Posix API. That is true even when most of the folks actually developing the tools are using Windows systems. Free Software developers working on Windows invariably end up with a whole pile of specialist tools like MingW and Cygwin that are basically designed around getting the GNU build toolchain to work on Windows.

If this project works, then the pressure for Windows to be Posix-like so that it can run the GNU build tools goes away. Instead you simply fire up this new Windows translator and kick out VS project files from your autotools files. Notice that this tool is very likely to be one-way only, autotools to VS studio projects. Unless it is perfect every time (and if it has to try and read autools files that is very doubtful) the Windows devs are very likely to simply use the tool once to generate VS project files and then modify the VS project files going forward. Microsoft is probably hoping that, at least in some cases, the folks maintaining the autotools part of the build system will get lazy and fall behind in the same way that many Free Software projects do now when it comes to building on Windows.

Personally, I doubt that Microsoft is going to get anywhere near that lucky. I also think that, to the extent it tries to build using Microsoft's compiler, it is unlikely to be very successful. If this tool created a straightforward way to simply building for Windows with gcc then lots of projects would be interested as it would free them up from building their own custom system for building on Windows. Currently there are nearly as many ways to build Windows binaries as there are Free Software projects that build Windows binaries. However, for many projects maintaining compatibility with Microsoft's compiler is a non-starter. They have something that works with Mingw (or cygwin, or djpp in some cases) and they are happy to stick with that.

Re:I'll follow them here too. :D

[dudpixel]

not with all menus expanded its not.

Re:I'll follow them here too. :D

[shutdown -p now]

In addition to that; as a Microsoft employee for Microsoft, I've contributed code to GPL, LGPL, BSD, PHP and Apache licensed projects.

Not that I doubt you, but it would be very nice to link to specific repository revisions for those changes. If anything, it would be handy for me to reference every time this topic comes up here.

Re:I'll follow them here too. :D

[starseeker]

First off, thank you for the work you're doing here - WiX and now CoApp look very very interesting in terms of reaching potential users of OSS apps on Windows more easily, which is definitely a worthy goal in my book. So thank you for working to help this happen!

I know for a lot of applications to be taken seriously as possibilities for Real Use in Real Environments (e.g. commercial) running on Windows is a check box that has to be checked. So since it has to be tackled by a lot of OSS devs, anything that makes that more straightforward is awesome.

My question may not actually be an "appropriate" question for the CoApp level of focus, but I'll try it out - you mention in your blog article the use of autoconf on Linux style systems. There has been some interesting progress on the CMake system towards generating cross-platform compile logic from a single "control file" - is there any work that could be done on either CMake, CoApp, or both to make a CMake defined build stand a good chance of cleanly fitting into the CoApp world, as well as the Linux autoconf world?

Re:I'll follow them here too. :D

[shutdown -p now]

Microsoft keeps telling developers that they cannot expect a user to have internet connectivity.

Well, that's certainly news to me.

If you ever tried creating a Setup project in VS2008, for a .NET application at least, it defaults to including only the web installer, which downloads the rest of the framework.

This doesn't add up. Can you provide a reference for "Microsoft keeps telling ..."?

So when you compiled your application with Visual Studio 2008 SP1 with the ATL update installed - which means every user of your software will have to have the Visual C++ 2008 SP1 ATL runtime redistributable package installed as well, you're left with scant few options.

You also have an option of statically linking with ATL (and MFC, and CRT).

Re:I'll follow them here too. :D

[anarche]

So read the previous posties?

Re:I'll follow them here too. :D

[shutdown -p now]

PowerShell will happily pipe text (or any random byte stream) if that's what your program wants for input and/or output, so I don't see a problem there.

Anyway, if you just want bash-win32, MinGW has that today.

Re:I'll follow them here too. :D

[Anonymous Coward]

Um... wouldn't that be a potential issue with virtually any software concept out there, invented by Microsoft or not, given the current trends exhibited by the patent office?

This is potentially in Microsoft's interest - more software available on Windows, and not uniquely available on Linux. This sounds like it is controlled by the individual - he appears to have done his homework. Let's give him the benefit of the doubt.

Re:I'll follow them here too. :D

[GaryOlson]

...look down at you with disdain.

So you push your bifocals down lower on your nose, tip your head down so we see the rapidly receding hairline from the shine on your sweating head, squint and frown at us because you can't focus more than 18" away and you still don't know who is berating you from your doorway so you call all of us "Sonny".

And I'll get off your lawn after you pay me for mowing it just like you asked. Did you forget where your drugs are again? Remember, the blue ones mean you will have a good time tonight.

Re:I'll follow them here too. :D

[k8to]

Why are you using MSI? Really.

It's a horror show of a backend, a crazy badly engineered database. Really, it's worse even than RPM.

You really should figure out what you would need, and design that, and then get the OS people to accept that you've replaced their mindless zombie subsystem with something non-awful.

Re:I'll follow them here too. :D

[GaryOlson]

I actually had bash working on Server 2003 once. I installed Unix for Windows, installed gcc in that horrid environment, configured and compiled bash. And it worked. Then I looked around and noticed I had but one shining clean tool in an utter wasteland devoid of resources. Then I had a beer or six and forgot the whole incident until just now.

Re:I'll follow them here too. :D

[Anonymous Coward]

I'm Alfred the Butler.
Would you like a cup of tea Sirs?

Wait! Wait!!
"Alfred?" If you're not Jeeves, then who then hell have I been asking questions to online?

Re:I'll follow them here too. :D

[masdog]

I've learned a lot about PMS from Linux

I learned a lot about PMS from my wife. Does Linux get PMS every 28 days, or just when you have to reboot?

/Yes...I know what you meant.

Re:I'll follow them here too. :D

[styrotech]

As an admin that maintains both Linux and Windows systems, this sounds really cool. Hopefully the guys writing the Tomcat AJP connectors for IIS will use it (that stuff can be a nightmare).

To me though the initial setup is never the main problem (except with AJP/IIS hehe), it's the ongoing maintenance and patching of 3rd party stuff that suffers the most on Windows.

Sure Windows Update / WSUS make all the MS stuff easy, but 3rd party Windows apps are a nightmare to keep up to date network wide. They all have their own separate update mechanisms that mostly require an admin being logged on to work.

I've love to see Windows Update and WSUS allow 3rd party repos (eg the equivalent of adding stuff to /etc/apt/sources.list) so that practically everything could be patched via Windows Update / WSUS without admin intervention on each machine.

I don't know if your work will end up tackling all that, or one day get incorporated by the existing patch mechanisms, but I can still dream :)

Best of luck anyway.

Re:I'll follow them here too. :D

[Anonymous Coward]

That, and I've also chosen the BSD license for it's do-what-the-f*-you-want spirit.

In the do-what-the-f*-you-want spirit, the BSD license is great. But doesn't hold a candle to http://en.wikipedia.org/wiki/WTFPL

Re:I'll follow them here too. :D

[Anonymous Coward]

Furthermore, From the conversations I've had with windows devs, they don't even know how their own software works.

Windows devs know they use SOAP, but don't know that its made of xml.

Linux devs know SOAP is made of xml, but are confused at how people use xml in the shower every day.

Re:I'll follow them here too. :D

[man_of_mr_e]

Well, I guess it's better than trusting someone who doesn't understand the meaning of the word "convicted". Or maybe doesn't understand the meaning of the phrase "civil suit".

In any case, "convicted" and "civil suit" are mutually exclusive.

Re:I'll follow them here too. :D

[Jason Earl]

Because, even if we could get a great POSIX experience on Windows, it leaves out Windows developers.

The problem with this, of course, is that no matter what you do Windows developers aren't going to be able to help the Free Software community because they have chosen to use tools that only work on one platform, Windows.

Heck, the reason that there is a Free Software community in the first place was because they built a set of tools that was specifically designed to be portable to a wide array of operating systems. If Microsoft did not go out of its way to be incompatible this wouldn't even be an issue. Free Software apps build fine on every other proprietary OS on the planet.

One of my goals is to get Windows developers in the OSS game.

It takes discipline to build cross-platform applications, and despite the difficulty of using autotools-style build systems on Windows it is still arguably the best way to do so. If Windows developers were interested in joining the OSS community then they would already be using OSS build systems.

This is not going to get Windows developers to join the greater OSS community, at best it is going to be a method for making it easier to use Free Software on a non-free system. At worst it might actually shift some projects from being autotools dependent (and thus easy to build on GNU/Linux) to being CoApp dependent (and thus impossible to build on GNU/Linux).

Of course, you are free to do what you want with your own time (and Microsoft's money) but don't try and dress it up like some sort of blessing to the folks in the OSS community.

On top of that, there is a hell of a lot of non-POSIX open source software on Windows that needs fixing too.

If your tool aims to make software written for Windows work on POSIX systems then forgive me for my cynicism above. If not, why should the greater OSS community care? If we were interested in Windows-only software do you think we would be bothering with autotools?

Your tool is likely to create a Windows-only sub-community of OSS developers that can borrow from the greater OSS community but that doesn't share with the Linux, BSD, and Mac OS X OSS developers that created the software in the first place. That probably sounds good to your employers at Microsoft, but it doesn't sound like existing OSS developers are likely to get excited about.

Look at it this way: Would you respect someone who told you the best way to get FireFox running on Linux was to use some sort of Windows emulation layer... Like WINE? no, because FireFox *can* compile for Linux. Same thing with nearly all Open Source I encounter. I want to get the OSS quality and experience on Windows to exceed commercial developers... it needs the most love.

People don't like Wine because they use it to run software that neither targets nor tests against Wine. I personally would be perfectly happy with Wine-based software if said software was a first class development target. If the developers ran build tests against winelib, for instance, and accepted bug reports based on wine-derived versions then I don't see the big deal. Heck, in may ways it wouldn't be any different than software that uses a JVM as a platform.

The problem with Wine is not that it is a bad platform, the problem is that it isn't a platform at all. Instead it is trying to mimic a platform. If Wine became the target instead of Windows (or even alongside Windows) then I am sure it could be made into a perfectly acceptable platform for developing GNU/Linux applications.

The problem, as I stated before, is that Windows developers aren't interested in creating cross-platform applications. Free Software developers have created the necessary tools (like Wine or Mono), but Windows developers don't seem to be interested in spending the little bit of extra time that it would take to use those to

Re:I'll follow them here too. :D

[weicco]

I have just one question: why should we trust them?

Answer is simple: maybe you shouldn't. But who is this we?

Re:I'll follow them here too. :D

[Anonymous Coward]

Why?
1) There are several perfectly good operating systems that already have excellent support for OSS. Why take an OS that actively tries to destroy open source and make it support OSS?
2) What's the point of a closed source operating system with open source software? One of the points of open source is so that there's no limit to what you can do except your own knowledge. Open source software on a close source OS just makes you slightly less limited then closed source software on a closed source OS. No matter what you hit the MS wall.

The real limitation of Windows, is Windows. When I've been forced to use Windows as my primary desktop I've tried to make it work as well as Linux. I can make Linux an extension of myself by modifying every bit of it to work exactly as I want to. In Windows I tried to set up a tiling window manager (something that has greatly increased my efficiency on Linux) and it worked a bit, but then broke because some window wouldn't respect the way the window manager (or whatever it had to be to work within Windows since you can't change out your window manager) worked. That's really just the most irritating thing that stuck out in my mind, but it just seems like no matter what I did things never quite worked all the way...

It got even worse when I tried to help manage a network of MS systems. Things work differently in win 2k then they do in XP, then they do in 7, then they do.... it's a disaster. With Linux you can just upgrade. Just throw the newest version out and get everyone going again. You could even use LTSP to set up thin clients on systems that wouldn't run the full OS. You can't do that on Windows without paying. You're stuck. Open source expands what you can do with Windows, but it doesn't fix the deep underlying problem: the OS itself.

CoApp will make Windows suck less then it does now, but it will still suck. So, why?

Oh yeah, one more question... how long you think until this gets modded flamebait?

Re:I'll follow them here too. :D

[IntlHarvester]

Not to break your bubble, but I don't see Microsoft ever really pushing a 'repository' distribution model, or mainstream Windows developers buying into the concept. This approach works for Linux (sorta) because the software is non-commercial, but wouldn't fly in the capitalist anarchy of the Windows ecosystem.

However, for system administrators and developers looking to install commodity software library XYZ, such a tool will be a godsend. I have been complaining for a long time that MS never really understood the 'lego block'-like approach to building up *nix systems, and the appeal that has to the hands-on users. I could even see commercial dev tool vendors jumping on board - there's a gazillion 'components' out there for MS developers, but installation and updating them all is a pain in the ass.

Glad to hear that someone is finally addressing this - it will make my life easier for sure.

Re:I'll follow them here too. :D

[Anonymous Coward]

In case a windows admin out there is wondering: diff means "display the changes between two text files in human-readable format." With such a simple tool you can get a quick answer to questions like "what settings are different between the test server and the production server?" or "what settings have been changed since the last backup?".

To take it one level further, with text-based config files you can use a version control system to keep track of changes to software configuration settings: who changed what and when. It is extremely useful when tracking down configuration problems. You can even do things like merge in selected settings from another server, leaving the rest of the settings intact.

Re:I'll follow them here too. :D

[msclrhd]

What I understand this meaning is that these changes will need to go in the upstream codebases to get any form of traction. As a result, if the Windows version does not have autotools support, or provide a posix compatibility layer because thou shalt do things the "Windows way", then you are going to have to complicate the upstream code.

APIs like Qt, and tools like CMake end up making this process easier, but that is a decision for the upstream codebase.

Alternatively, if the code is forked so that the Windows port is done in such a way that is incompatible with Linux and those repositories get sanctioned as the official repositories then they will stop working on Linux and Microsoft would have won. Remember when Balmer said that he wants all open source to be done on top of Windows. This helps his vision.

I also find it strange (or perhaps not, given some recent developments) that this project is being hosted on launchpad, and therefore falls under Ubuntu/Canonical's wing where their #1 bug is replacing Windows.

Re:I'll follow them here too. :D

[pinkushun]

I think what you're doing is great! It's a big task though, fixing what your predecessors tried to destroy. Good Luck =)

Re:I'll follow them here too. :D

[lbschenkel]

Ask me about Loom.
-- Pirate Cobb

Re:I'll follow them here too. :D

[aBaldrich]

The point they're trying to make is: if it was GPL, Microsoft wouldn't be able to get a package managing software developed for free. Please understand our point of view. Whenever I see "Microsoft" and "Open source" in the same sentence, I'll always remember Halloween Document 2. I am worried, because MS is embracing OSS.

Re:I'll follow them here too. :D

[SanityInAnarchy]

- They must be able to host their repository meta-data on an SSL protected connection.
- All packages must be digitally signed with a certificate that chains back to to a commonly-accepted CA.

Why do you need both? Why not just sign the repository meta-data the way you're signing the packages?

Among other things, this would mean it's easier to accelerate with proxies, cheaper, and less resource-intensive. About the only downside I can think of is the possibility of a MITM presenting an older version of the repository (which would still be signed), but they could just as easily block the SSL connection, so I'm not sure what that buys you.

Also (and I should probably ask this elsewhere), there are multiple open source package managers for Windows, though I don't know how good any of them are at this point. Why start from scratch -- what's the problem with any of these? (Cygwin is obviously not a great fit, but the rest?)

Re:I'll follow them here too. :D

[Sun]

I have a plan for allowing any publisher to publish packages in the CoApp ecosystem, provided they meet two qualifications:
- They must be able to host their repository meta-data on an SSL protected connection.
- All packages must be digitally signed with a certificate that chains back to to a commonly-accepted CA.

Doesn't seem like a very good solution. The whole point of APT is its ability for ANYONE to open repositories, including digitally signed repositories.

If, for whatever reason, you don't like PGP, that's fine. Go with X509. Just don't force a SPECIFIC root CA - allow the package user to choose which is his CA of choice (one or more). This way, for example, a company can set up a local repository to push to its own employees.

Same goes with where you host this. Your answer did not make it clear whether any server can be configured, or just MS's servers.

Shachar

Re:I'll follow them here too. :D

[timmarhy]

you don't need to because you can see what's selected in a gui, rather then wading through a screen full of text?

Re:I'll follow them here too. :D

[Errol backfiring]

I agree 100%. Scripting interfaces are an absolute requirement, and will likely come well before the GUI.

What scripting language is there to be used? I am still waiting for "DOS" to be anywhere usable without installing all unix command-line tools. Any alternative to DOS is too short-lived or too unsafe to use (really, the fact that web scripting languages have gotten file access is still the worst bug in Windows. Especially security-wise).

Re:I'll follow them here too. :D

[ppanon]

Well, ya see... with a five-digit slashdot-id I originally had "His name cannot be Spoken" as my name... then they did some database truncation about 12 or so years ago, and I lost some letters.

Hmm, if that were the case shouldn't your currrent name be "His name cannot be S"?

And ya can't change your name on Slashdot, and I didn't wanna give up my 5 digit ID. :D

I can understand that. If nothing else, it's a freebee achievement point. Pull this OSS Windows package manager off and maybe Cmdr Taco should personally go into the database and fix your account name to whatever you want.

Re:I'll follow them here too. :D

[ppanon]

Some of us 5-digiters still have all our hair, it's just starting to go gray is all.

Re:I'll follow them here too. :D

[GigaplexNZ]

Sure they could. It would just be dual licensed under GPL and whatever MS wanted. The only limitation is that any 3rd party contributors would have to agree to having their code be distributable under all the licenses to be accepted upstream. Qt is an example of this sort of thing.

Re:I'll follow them here too. :D

[Richard_at_work]

Will we eventually see the day where Microsoft has a central location for shared libraries in Windows (writable only by "root") and also a decent package management system, you know, like apt/rpm?

Like the Global Assembly Cache? Central location (C:\Windows\assembly\) and only writable by the admin. It also handles versioning of assemblies as well.

Re:I'll follow them here too. :D

[Richard_at_work]

How do you know the same isn't true of any other software you use?

Re:I'll follow them here too. :D

[Anonymous Coward]

How to write a question that isn't going to be answered, in 1 easy step.

Re:I'll follow them here too. :D

[Richard_at_work]

IIS is a bad example since all of the GUI configuration tools actually run off of XML configuration files (web.config, metabase, machine.config et al).

Re:I'll follow them here too. :D

[Anonymous Coward]

Does your project relate to WiX in any integrated way? Which btw was MS first OSS project.

How integrated is CoApp to VS 2010 and msbuild? For example, compared to Azure platform, to which a deploy from VS is embarrassing easy.

Does apps distributed with CoApp require expensive authenticode certs?

Re:I'll follow them here too. :D

[ThatsNotPudding]

What is Microsoft paying for human souls these days?

Re:I'll follow them here too. :D

[Anonymous Coward]

I'd like to see an answer to this.

As Admiral Ackbar would say: It's a trap!

Re:I'll follow them here too. :D

[Anonymous Coward]

He said "They do get a license to everything I make (fair deal)" ... so why does it matter whether it's GPL or BSD or what? MS would still get it developed for free and be able to use it. If he's the author, he could release it under the GPL if he wanted and give it to MS under a different license. Not that it matters since it'll be BSD anyway... just sayin...

Re:I'll follow them here too. :D

[Anonymous Coward]

You tell him that, as if the Free Software apologists didn't do the same with Linux-only software.

I don't want to be locked into a kernel that includes binary blobs and NDA-based code and its nonstandard extensions any more than I want to be locked into Windows.

You name autotools, they are one of the worst offenders.

Re:I'll follow them here too. :D

[Kazymyr]

Have you tried to roll out some OSS apps on Windows?

You know, just yesterday I was looking at what was needed to compile OpenJDK on Windows.

I got a massive headache about 1h into it. :)

Re:I'll follow them here too. :D

[Anonymous Coward]

From what I gather his package manager will run via IIS/ASP.Net which you will use to install Apache & PHP!

Re:I'll follow them here too. :D

[TheRaven64]

Seems to me they want some competition.

No they don't. They make most of their money from two things: Windows and Office. Everything else exists to make these two products more competitive. The only reason that Microsoft develops things like Visual Studio, for example, is to encourage third parties to add value to their platform.

IIS does not make money for Microsoft. If Apache is a better web server than IIS on Windows, then that's fine, as long as you're still paying for the Windows license. Especially if you're using the ISAPI module so you can't take your code elsewhere.

Same with PHP. They don't care if you're using .NET or PHP to write your web apps, as long as you're serving them from a Windows system.

While I wouldn't describe this move as evil, it's disingenuous to describe it as anything other than self serving. Microsoft got its current position in the OS market by fostering third-party development. Aside from some shady business deals with OEMs (and a few former competitors), they have always been very active in encouraging people to develop for Windows. Microsoft, I'm sure, would love it if you didn't run anything other than Microsoft software on your computer, but they are rational enough to understand that this is not possible for most people and so they put a lot of effort into making sure that the non-MS products that you want have dependencies on some MS products (usually Windows, sometimes Office, and sometimes on something else that depends on Windows).

Re:I'll follow them here too. :D

[TheRaven64]

Well, ya see... with a five-digit slashdot-id I originally had "His name cannot be Spoken" as my name... then they did some database truncation about 12 or so years ago, and I lost some letters.

MySQL: Because your data probably wasn't that important anyway.

Re:I'll follow them here too. :D

[Anonymous Coward]

did you ever try the print screen button on your keyboard?

Re:I'll follow them here too. :D

[Anonymous Coward]

Err, Powershell? Windows scripting host?

Re:I'll follow them here too. :D

[LinuxIsGarbage]

Look at it this way: Would you respect someone who told you the best way to get FireFox running on Linux was to use some sort of Windows emulation layer... Like WINE? no, because FireFox *can* compile for Linux.

Except of course the best way to get Firefox running on Linux is to use WINE http://tech.slashdot.org/article.pl?sid=09/02/13/0058251

Though kludges like CygWin has brought the dependency hell headache over to Windows where multiple cygwin applications try to load different versions of cygwin.dll. Anyways, best of luck. Even though you're going to run into a lot of "It's a trap!"

Re:I'll follow them here too. :D

[DAldredge]

Be quiet newbie.

Re:I'll follow them here too. :D

[RazorSharp]

Preachers are in high demand in Vegas for shotgun weddings. So what's your analogy supposed to mean, that you're being exploited and disgracing the cloth you're cut from?

Perhaps you should hold your tongue when you come up with similes you think are clever. Really, I don't doubt that your intentions are good, but so were Howard W. Campbell Jr.'s. (http://en.wikipedia.org/wiki/Mother_Night)

Re:I'll follow them here too. :D

[Nesman64]

I'm Spartacus.
Don't ask.

Re:I'll follow them here too. :D

[Anonymous Coward]

The GUI config values have to be stored somewhere, either in a file or the registry, so you can compare them either way.

Re:I'll follow them here too. :D

[Jason Earl]

I agree that this project is potentially very dangerous to GNU/Linux as a platform. However, the only thing that Canonical has to do with this is that the hacker in question is using launchpad.net to host his code (or lack of it actually, so far we have a license and a stub readme file).

I like bzr and launchpad, so I can see why he made his choice.

I do not think that Canonical should get involved. Launchpad would be less useful if it tried to regulate what sort of Free Software could be developed on launchpad. If CoApp is to be stopped it should be stopped by not being adopted by upstream projects until such time as it provides some benefit for cross-platform compatibility.

For example, most Free Software projects that create Windows binaries have a build system that they use for Windows and an autotools one (or perhaps some other Posix tool) that they use for everything else. The Windows build system is usually a brain-dead cousin to the autotools build system primarily due to the fact that allows the project to target gcc as the compiler even on Windows. So while the project does have to have Windows compatibility baked in, it does not have to be compiled with Microsoft's compiler.

Of course, what Windows devs invariably want is VS project files and compatibility with Microsoft's compiler, which adds a great deal of complexity. If, however, CoApp was a tool that allowed VS to play nicely with gcc then it would be a different kettle of fish. Windows devs could have their precious project files while still targeting gcc. Sure, it would still be two build systems, but that is what we have now anyhow. The difference is that it would be two standard build systems that every project could use instead of each creating their own custom build system for Windows.

This does not appear to be what CoApp is trying to do, and probably for the reasons that you point out. Microsoft is not interested in making it easier to make cross-platform applications, it is only interested in borrowing code from cross-platform applications and turning the code into Windows-only code.

Re:I'll follow them here too. :D

[h4rr4r]

That is an interesting button you are wearing.

Re:I'll follow them here too. :D

[Deosyne]

Well done. Blatantly misinterpreting an obvious analogy and then rolling right into a Nazi parallel. You might just have a future with FOXNews.

Re:I'll follow them here too. :D

[TemporalBeing]

Look at it this way: Would you respect someone who told you the best way to get FireFox running on Linux was to use some sort of Windows emulation layer... Like WINE? no, because FireFox *can* compile for Linux.

Except of course the best way to get Firefox running on Linux is to use WINE http://tech.slashdot.org/article.pl?sid=09/02/13/0058251

Firefox typically works better under Linux than it does Windows WITHOUT any WINE. But then, may be your retrying to imply that if it's drunk it'll work better yet?

Re:I'll follow them here too. :D

[Anonymous Coward]

Linux I'm always spending 15 minutes getting a screenshot app working.

In the default Ubuntu installation, you can press print screen and it'll ask if you want to save it in a file or copy to clipboard. Really can't be easier.

Re:I'll follow them here too. :D

[Animaether]

alright... rephrase.. there are articles on Microsoft.com that point out that when you are building your installer, the default (with the latest set of Visual * products) will download prerequisites over the internet, and that this could possibly maybe not likely but hey here you go... cause the installation to fail.

If you try to install the program on a computer that doesn't have Internet access, the installation could fail.
To prevent this, you can package prerequisites such as the .NET Framework redistributable files along with the program.

Subtle - but there it is. Now guess what.. unless you -do- expect every single one of your clients to have an internet connection even though you're selling a boxed product (so you have no real reason to believe that they do)... *plonk* ...the installer just sits there and does nothing.
It gets better when you don't build in a method for your installer to include the prerequisites from a drive so that - if the user managed to get the prerequisites from an alternate location to their machine - the installer will still just sit there and do nothing for it insists that it must get the prerequisites off the internet.
One can always choose to not care and tell people to get internet, I suppose.

You also have an option of statically linking with ATL (and MFC, and CRT).

Which flies in the face of one of the things that this guy wants to address.. everybody including the libs.. and that does include statically linking. Worse yet - if Microsoft patches something, your app will need an update and that update would need to be sent to your clients in order to fix things.

It's an option - just not a very good one.

Re:I'll follow them here too. :D

[causality]

I have just one question: why should we trust them?

Answer is simple: maybe you shouldn't. But who is this we?

Initially, I didn't consider this an important point. It seemed like mere semantics to me, like I could have put that several different ways without taking away from the meaning of my post.

You've made me reconsider that, however. It may have been a poor choice of words. When I said "why should we trust them" I meant it in the sense of "why should anyone trust them".

I believe that's reasonable, really. If I don't know anything about a company, I can give them benefit of doubt as a default position. When I observe that the company frequently says one thing while doing another, I now have a reason not to trust them. Now my default position is to consider everything they say suspect without some hard proof that they are telling the truth.

Companies that want to be regarded as trustworthy will be careful not to deliberately engage in business practices that tarnish their reputation (honest mistakes are understandable, if corrected). In my opinion, Microsoft is not such a company. So I personally do not trust them.

Re:I'll follow them here too. :D

[causality]

Well, I guess it's better than trusting someone who doesn't understand the meaning of the word "convicted". Or maybe doesn't understand the meaning of the phrase "civil suit".

In any case, "convicted" and "civil suit" are mutually exclusive.

Actually I wasn't asking you to trust me. I don't want you to take my statements on faith. I would prefer that you maintain a healthy skepticism and evaluate for yourself whether my reasoning makes sense. Thus, if you disagree with me, it would be not because you trust me or don't trust me, but because the reasons I gave would not affect your perception of a corporation. If they wouldn't, I'd like to know why. A company's well-established negative track record certainly does affect how I perceive them; if it has no such effect on you, I'd be interested in hearing about this.

If a company knowingly does something that causes harm to people, the government taking them to court generally only results in a fine. An individual who does the same thing might face criminal charges.

In light of this, I believe "convicted" is an acceptable term. You may disagree, but you now know my reason for selecting this word.

Really this is semantics. We can argue about whether "convicted" is too strong of a word but it doesn't change the fact that various courts around the world have found Microsoft to be engaging in illegal business practices. Those practices were found to be abusive. That was one of my reasons for distrusting them. Whatever you like to call the process by which abusive practices were determined, I don't trust companies that engage in them. So you see, in a discussion about reputation and trustworthiness, that really is just semantics.

Re:I'll follow them here too. :D

[Anonymous Coward]

"convicted" is about as acceptable when used in a civil suit as "thief" is when talking about copyright infringement.

I don't know where you are on that side of the coin, but I don't think using loaded terms that are not actually true adds anything to your argument, and does in fact detract from it since it shows clearly and without doubt where your bias lies.

Re:I'll follow them here too. :D

[man_of_mr_e]

By the way, the term "convicted" isn't even an acceptable term using your logic, because civil lawsuits require an great deal less burden of proof than criminal proceedings. This is why montetary fines are all that can occur.

So basically you're saying that same thing as saying "OJ is a convicted murderer", which he's not. He was found civilly liable in the deaths, and he probably is quite guilty, but he's not a "convicted murderer". If he were, he'd be in prison.

I despise when people use such emotionally laden terminology, and especially despise it when it's not, in fact, accurate.

Re:I'll follow them here too. :D

[weicco]

Maybe things are different where you live but I've worked with Windows, Visual Studio, Sql Server, almost anything that comes out of MS for 10 years now (plus some 5 years as a hobby). I've never felt that MS has screwed me or the companies I've worked for.

I've done some Open Source stuff also. Not any biggies, I've written some applications and released them under BSD and lately MIT licenses. All of them are written for Windows with C, C++, C#/.NET. Some of them don't run on Mono I think but probably most do. I've never felt that MS has screwed me here either. They've even provided free tools for me (Visual Studio Express, Sql Server Express).

So personally I trust them on this one and on many other cases. I don't see how they could pull the plug on this or why they'd do it when they clearly has some synergies here. Your experiences seem to have differed and its totally up to you make your own mind, of course. And here's the reason I took on this subject: I don't like when people speak for me, no matter are they wrong or right. That's why I questioned who's this "we" you are talking about :)

How Long Before ...

[WrongSizeGlass]

... MS pulls the plug on this and leaves OSS developers hanging high and dry? Or worse, pulls some slight of hand with licensing, copyrights or patents and forces OSS dev's to stop in their tracks waiting for MS's next move?

Re:How Long Before ...

[ChrisMounce]

Maybe they recognize that there's a ton of open source software that people really want to use, and that easy installs of OSS on Windows adds value to Windows.

Like how they contributed some Linux stuff a while back to make it easier to run Linux in a VM... with Windows as the host machine (I'm not clear on the details, so I'm probably getting the terminology wrong).

Re:How Long Before ...

[Anonymous Coward]

How can the? Its an OpenSource project to use readily available to all tools. They may pull their guy off of it but they cannot shut it down because they don't own anything in it. . . Its generic MSI projects with FOSS software in it.

Re:How Long Before ...

[DigiShaman]

Not only that, but they have two giant companies to compete with. Namely Apple and Google. This industry is cut-throat, and the loss of momentum is extremely dangerous.

Embracing OSS while at the same time keeping control over the direction of your platform is the name of the game.

Re:How Long Before ...

[aBaldrich]

Do you know what was the first thing I thought when reading the article and the project's launchpad? Halloween Document II. Why does Microsoft need to support Free Software? I mean, they claim to have everything they need, their new shiny should 7 have it all.
This article's summary should be something along the lines of: Microsoft embraces OSS. How long before they extend their dirty tactics to OS? I don't trust them.
The jewels of OSS were built because we, developers, needed them. We needed an OS and made Gnu, and Linux. We needed a web server and made Apache. We needed a GUI and made GNOME, KDE and Compiz. We needed a web browser and made Firefox.
Now Microsoft needs package managing software, and they really believe we are going to build it for them. Laughable.

Re:How Long Before ...

[street struttin']

How long before Microsoft uses this to deploy MS Office and it's support software?

Re:How Long Before ...

[shutdown -p now]

... MS pulls the plug on this and leaves OSS developers hanging high and dry? Or worse, pulls some slight of hand with licensing, copyrights or patents and forces OSS dev's to stop in their tracks waiting for MS's next move?

It's in Microsoft's best interest to make sure that as many OSS packages as possible run well on Windows. Keep in mind that it's revenue from Windows licenses that brings in a huge chunk of cash, and if, say, someone doing PHP development can be convinced to host Apache on Windows Server rather than Linux, and maybe even MSSQL rather than MySQL, it's a huge win already. Pressing people into going for the entire bundle - i.e. either IIS+ASP.NET+MSSQL - or nothing at all, hurts the bottom line (unless you're Apple).

In fact, Microsoft specifically works on supporting certain popular FOSS software on Windows - for example, here is the project dedicated to supporting PHP.

Re:How Long Before ...

[pinkushun]

Microsoft could be trying to take control of the Windows FOSS culture, which they ironically tried to screw over and kill in the beginning.

Re:How Long Before ...

[LordLimecat]

How long before they extend their dirty tactics to OS?

Yes, and then they will simply EXTINGUISH open source! And then happiness, and love!

How exactly can one "extinguish" a philosophy or idea, pray tell?

Re:How Long Before ...

[tepples]

How exactly can one "extinguish" a philosophy or idea, pray tell?

It would start with patent lawsuits against the organizations that make the philosophy or idea practical.

Microsoft has been surprising me lately

[calibre-not-output]

They seem like less of an evil empire now that they're doing some good stuff once in a while and Google is being more blatantly monopolistic.

Re:Microsoft has been surprising me lately

[h4rr4r]

Do you get paid to spout this?
This is a method to get OSS apps on windows, to make sure you don't dare use another OS.
Interoperability would be making their apps run on an OSS OS.

Re:Microsoft has been surprising me lately

[Beelzebud]

Do you?

Re:Microsoft has been surprising me lately

[Anonymous Coward]

I got REALLY angry at Microsoft what I found out about the requirement for signed kernel mode drivers.

(occasionally, but not often, I need to write a kernel mode driver to get my job done)

Then, I got Win 7...It works well, not that incompatible,pretty too, much of the stoopid stuff from previous releases is reduced...

Hey, maybe Microsoft isn't all bad.

Now Apple...closed iPhone, closed iPad...maybe THEY are the new EVIL EMPIRE?

Re:Microsoft has been surprising me lately

[nurb432]

If you look close enough what they are doing is quietly pushing people towards windows only platforms. So really, they aren't acting radically different, just being a bit more subversive about it.

Re:Microsoft has been surprising me lately

[DAldredge]

I like Windows so please harass me too!

Re:Microsoft has been surprising me lately

[Anonymous Coward]

I like Windows so please harass me too!

You like WHAT?!
  Don't you know that every time you boot into Windows, an angel gets set on fire? Do you like doing that, you malevolent bastard?

  Stop your evil ways, or when you die you'll go to GNU/Hell! Where your PC speaker beeps all day long, every day! Where all HTML must start with an unclosed blink tag! Where EDLIN is the good editor you'll get if you kiss the sysadmin's ass enough! Where all your printer drivers require you to sign away a testicle! Where you must work 19 hours a day bugfixing Visual Basic code written by retarded monkeys, only to have them make printouts of your corrections and defecate on them at the subsequent five hour meetings!
  Repent! Repent!

Re:Microsoft has been surprising me lately

[thoughtsatthemoment]

What's your reaction if MS port visual studio to Linux?

Re:Microsoft has been surprising me lately

[h4rr4r]

Awesome, more software is great for the platform. I won't use it, only because the tools I already like and use are fine for me. I have friends that would love to have Visual Studio on Linux as it would remove their last need for windows.

Re:Microsoft has been surprising me lately

[__aasqbs9791]

I like Windows so please harass me too!

I'm pretty sure you get all the suffering you need already.

Re:Microsoft has been surprising me lately

[phantomfive]

I can't help but think you are incredibly naive. Have you been paying attention to how Microsoft is suing third party controller makers? Or how they are subverting their own standards? Or if you are in Europe, how they are heavily lobbying your representatives to hurt open standards? If you live in the US, don't worry, they are working to influence your representatives too. Not to mention they stand firmly committed to helping out a truly evil empire (yeah, saying 'evil' is a bit much but a government that censors political speech and has secret trials for people they don't like isn't exactly nice).

I mean, this is just in 2010. You shouldn't have particular love for any company, but claiming that Microsoft has changed can only be done by ignoring the facts and reality. Don't do that.

Re:Microsoft has been surprising me lately

[Anonymous Coward]

I see what you did there.

Re:Microsoft has been surprising me lately

[RMS Eats Toejam]

Yes, for-profit companies want people to use their competitors products. It's a great way to stay in business. Because it needs to be said and you need to hear it, "Go fuck yourself."

Re:Microsoft has been surprising me lately

[ClosedSource]

I assume your friends would use Visual Studio exclusively to create Mono apps. Otherwise, I don't see what use they'd have for Visual Studio without Windows - unless they don't test their software.

Re:Microsoft has been surprising me lately

[h4rr4r]

The software I am speaking of actually runs on linux, but their company standardized on VS for development and the build process requires VS.

Re:Microsoft has been surprising me lately

[shutdown -p now]

I assume your friends would use Visual Studio exclusively to create Mono apps. Otherwise, I don't see what use they'd have for Visual Studio without Windows

Visual Studio has some good C++ development tools - the debugger is very nice, with visualizers for STL containers etc.

Also, in VS2010, the code completion engine is driven by a full-fledged C++ parser, which can handle any convoluted mess you can throw at it, and still produce you correct symbolic information and completion list - I've tried it on Boost.Lambda, and it handled that without a hitch. The only IDE that can match that on any platform is KDevelop 4, but the release date for that is still a month off (VS2010 will be out on Monday).

Re:Microsoft has been surprising me lately

[ClosedSource]

I like visual studio. I guess I don't understand how you avoid creating Windows-specific binaries.

Re:Microsoft has been surprising me lately

[Anonymous Coward]

As the author has said, the VPs signed off on his project meaning that he owns all of it and MS just gets a license to use it. He's also released it under the BSD license. He works for MS and it's for MS, but it's not owned by them and so they cannot stop him (except maybe if he's doing it during work hours and they don't want him to). I'm trying to not be naive here, but it seems like it's fairly free from the company's control, and coming from a Linux user, it would be nice to have a unified package system even if it's just for OSS.

Re:Microsoft has been surprising me lately

[shutdown -p now]

I assumed that the hypothetical scenario talked about here is if everything that's shipped a part of Visual Studio (which includes the C++ compiler and libraries) would have been ported to Linux.

Re:Microsoft has been surprising me lately

[Anonymous Coward]

Don't forget about them taking over certain standards bodies.

Re:Microsoft has been surprising me lately

[Anonymous Coward]

Presumably they build the Linux binaries separately.

Re:Microsoft has been surprising me lately

[calibre-not-output]

Do you get paid to spout this?

Of course I do! It's the only explanation there could be for me disagreeing with you, right?

Re:Microsoft has been surprising me lately

[calibre-not-output]

As far as I'm concerned everything should be cross-platform. So my reaction would probably be positive.

Re:Microsoft has been surprising me lately

[calibre-not-output]

I didn't claim that they changed. I said that they seem less like an evil empire every day. This is a comparative statement, and largely due to Google's recent privacy issues and (though I didn't mention it in the parent post) Apple's absurdly restrictive mobile platforms. It also happens to be the case that Redmond's PR office seems to be doing a better job at promoting specific ventures lately.

Disclaimer: All my computers run exclusively on Linux (Gentoo, Linux Mint and Crunchbang), my favorite search engine is Google and I own an iPod.

Re:Microsoft has been surprising me lately

[calibre-not-output]

I stated an honest opinion and got modded down as a troll. I call shenanigans.

Re:Microsoft has been surprising me lately

[phantomfive]

Ah, I see, so what you are saying is since other groups have become more evil, it's become more of an evil trifecta, or an evil realm of competing feudal lords? I suppose that is reasonable.

Re:Microsoft has been surprising me lately

[calibre-not-output]

You could put it that way, I suppose.

Re:Microsoft has been surprising me lately

[Grishnakh]

Their company requires them to use VS to build software that runs on Linux?

Wow, that's pretty stupid.

Re:Microsoft has been surprising me lately

[h4rr4r]

It is indeed very stupid. All of the developers either run linux in a vm or linux on the desktop and windows in a vm. Since having access to both is pretty critical.

nothing MS does is good for anyone but Microsoft

[kawabago]

This is no different.

Will it run on Linux?

[thoughtsatthemoment]

Do I have a legitimate reason to ask this question or not?

Does it come with configure?

[Anonymous Coward]

I am not interested otherwise.

Finally, some sort of package management!

[Anonymous Coward]

The #1 thing I wish windows had would be some kind of package management like apt-get: a place where I can go to update everything at once (of course, being able to install from it is a natural progression)

Maybe you could spin it to management as an "App Store" competitor?

Anyway, this sounds like a great idea. Looking forward to how it turns out!

wholly native toolchain

[nurb432]

Native to what OS? Let me guess, windows.

Re:wholly native toolchain

[Dragoniz3r]

Is there something wrong with that?

From TFA:

• Place binaries, libraries and header files in a logical and consistent location
• Facilitate sharing of components and allow multiple projects to easily both participate and consume them
• Allow for upgrades and patching of both libraries and applications
• Be Windows developer friendly. No forcing of building using ‘make’, but rather taking advantage of the nifty IDEs we already have

Clearly these are horrible goals and this should by no means be done, simply because he is planning to do it in Windows.

Re:wholly native toolchain

[h4rr4r]

All but the last one are fine. I have some windows boxes I have to deal with and I sure as hell do not want to be stuck using some GUI IDE just to build the latest $foobar.

Re:wholly native toolchain

[His name cannot be s]

All but the last one are fine. I have some windows boxes I have to deal with and I sure as hell do not want to be stuck using some GUI IDE just to build the latest $foobar.

Use of the GUI ain't mandatory... it's just that in order to get Windows devs on board, it'll have to have one.

The core bits will all be able to be command-line driven.

Re:wholly native toolchain

[h4rr4r]

This is the sort of thing that makes the FSM smile upon you and shall earn you an afterlife spent near the beer volcano and stripper factory.

Re:wholly native toolchain

[codepunk]

That is some funny shit!, even though I could give two shits about anything that runs on windows.

Re:wholly native toolchain

[CAIMLAS]

You seem to be speaking as if this project is well beyond the officially-sanctioned-with-Microsoft stages.

"Unofficially", how mature are your efforts to making sane package management in Windows a reality? 6 months to public release? A year? Two years?

Re:wholly native toolchain

[weicco]

I sure as hell do not want to be stuck using some GUI IDE just to build the latest $foobar

You are absolutely right. But luckily there are also command line options. For instance I write every piece of code with Visual Studio but for the actual builds I use automated command line scripts. And in fact, since Visual Studio Express (which I use at home) doesn't support publishing ASP.NET apps from GUI I wrote a one-liner to do it from command line.

This project sounds something like what I planned couple of years ago but didn't have time or motivation to start it up. I'm just glad that someone has :)

msi sucks!

[Anonymous Coward]

MSI installers suck. Why would we want that kind of crap coming with FOSS?
You can't run two msi install processes at the same time. So why would we even want that on Linux.
We already have rpm, deb, slack packages, nix and zeroinstall. WTF would we want another install method to juggle and one paid for using dirty Microsoft money. FAIL.

Re:msi sucks!

[owlstead]

Indeed, FAIL to RTFA, mods, AC's post should be modded into oblivion.

Re:msi sucks!

[LingNoi]

You can't run two msi install processes at the same time. So why would we even want that on Linux.

Eh, since when have you been able to run two deb install processes at the same time? Oh that's right, never.

Re:msi sucks!

[Anonymous Coward]

Why would you WANT to install two things at once on any platform? Do you enjoy bizarre race conditions and impossible to reproduce brokenness?

"bring OSS more users"

[nurb432]

And more *windows* users, more windows license, more vendor lockin, and fewer alternative OS's.. Ya, real nice of them to 'help' us out. No thanks.

Re:"bring OSS more users"

[Anonymous Coward]

Open source applications can be ported to other OSes. Once people are using entirely an entirely OS solution, they can be brought to an OS OS much more easily. Sure it'll bring more licenses to Microsoft for a while, but eventually this will help out open source.

Re:"bring OSS more users"

[Anonymous Coward]

And entice as many F/OSS developers as possible into dedicating their limited time to the MS ecology, rather than to competitors. I think Garrett has noble intentions, but his superiors have other motivations.

Re:"bring OSS more users"

[im_thatoneguy]

And more *windows* users, more windows license, more vendor lockin, and fewer alternative OS's...

Until Windows users realize that all their favorite apps run great on Linux as well as Windows.

If chrome runs on Windows and Linux and you just use Chrome most of the day then it becomes trivial to switch over to Linux since your app will look relatively familiar. The largest obstacle to Linux adoption besides its contempt for its users is the lack of applications people are familiar with. If someone got used to pidgen then they would be less likely to revolt when they tried using Linux.

Re:"bring OSS more users"

[shutdown -p now]

Until Windows users realize that all their favorite apps run great on Linux as well as Windows.

There are way too many "favorite apps" that will never run great on Linux, the most noticeable category being games.

Of course, this project won't really change this, either for better or for worse.

Re:"bring OSS more users"

[GF678]

Until Windows users realize that all their favorite apps run great on Linux as well as Windows.

A lot of the apps I use are open source and run on Linux. The reason I continue to use Windows (Windows 7 specifically) is because I prefer Windows to Linux as an operating system.

Windows runs cooler on this laptop because it has better chipset drivers I suppose. It also runs cooler because the proprietary ATI drivers for the graphics chip have proper power management, whereas the open source drivers in Linux do not (unless you use the proprietary drivers in which case you lose the benefits of an open source system). I also have better battery life due to the above two aspects. PLUS I have the benefit of using any commercial software which does not have a Linux equivalent/version and don't have to wrangle with Wine or virtualization.

Believe it or not, some people do prefer the Windows way of doing things than Linux.

Re:"bring OSS more users"

[LordLimecat]

So youre saying that "improving Windows is bad because it makes windows better"? Or that somehow Windows having a large market share will mean less competition? Have you been under a rock, do you not know about linux?

Re:"bring OSS more users"

[Anonymous Coward]

Even more fun is getting monthly updates for your your favorite OSS application on Linux, while the Windows builds are updated one every two years.

Remember that cute little phrase the DOJ found?

[Anonymous Coward]

*cough cough*

http://en.wikipedia.org/wiki/Embrace,_extend_and_extinguish

*cough cough*

History is not on your side. I hope, for all of us, that your intentions are noble. If they are, I hope those who back you and/or succeed you hold to that ideal. Thankfully, even if they're not these programs will live on in their desired format on other software platforms.

Best of luck.

Why only open source?

[iYk6]

This sounds like a package management system for Windows, along the same veins of dpkg/apt and rpm/yum. Windows has been lagging in this area for years, and one of the reasons that it is so insecure and slow is because every program either runs in the background at startup, or doesn't get updated on a regular schedule. That wasn't my question, just how I view the situation.

Why limit this to open source? It would be great if the users could update every program easily and painlessly, at least the ones that use this new system.

I am assuming that this system will allow easy and painless upgrading like on most Linux systems. Is that true? Will it have automatic dependency handling and command line installation?

Re:Why only open source?

[acsinc]

I second the question about limiting to open source. A good package management system that can could make using SxS painless would be awesome in an enterprise environment.

Since this is open source and .msi based I assume you will be leveraging WiX somehow?

I hope this isn't going to be a big collection merge modules with duplicated component guids...

Re:Why only open source?

[His name cannot be s]

Why limit this to open source? It would be great if the users could update every program easily and painlessly, at least the ones that use this new system.

I'm Busted. It isn't really restricted to Open Source... but that's my mission. Commercial apps will be able to play just fine in this ecosystem.

I am assuming that this system will allow easy and painless upgrading like on most Linux systems. Is that true? Will it have automatic dependency handling and command line installation?

Yes. Painless and automatic dependency handling, and yes command line tools. You are singing the chorus to my theme song!

Re:Why only open source?

[His name cannot be s]

I second the question about limiting to open source. A good package management system that can could make using SxS painless would be awesome in an enterprise environment.

I agree. it ain't really limited to Open Source

Since this is open source and .msi based I assume you will be leveraging WiX somehow?

Yes indeed. The author of WiX is on the mailing list, and a personal friend. He's very excited about all this too.

I hope this isn't going to be a big collection merge modules with duplicated component guids..

Nope. I don't believe in merge modules. I believe in a system that works.

Re:Why only open source?

[Jenming]

Linux does not have painless upgrading. It has painless updating to whatever has been put in apt-get/whatever. This is often (always?) not the most recent version of the software. I guess its something but on Windows if you let the programs update themselves whenever you use them they are more up to date, though it is a pain in the ass.

Re:Why only open source?

[Anpheus]

Powershell support with proper verb-noun command structure and piping support?

I'm disappointed that I even have to add the last bit, but Powershell stuff from Microsoft differs greatly on supporting the implicit standards. There are some modules and snapins that are really good, there are some that are... not so much.

Re:Why only open source?

[Animaether]

Will the underlying mechanisms only be available to Windows Installer-based delivery systems, and not to alternative installation systems such as InnoSetup, NSIS, and the remainder of the usual suspects?

Re:Why only open source?

[h4rr4r]

And you get to have 1000 updaters all running on startup, each dragging along who knows what shared libs that instead of being properly shared are whatever version the app maintainer used.

Some distros do package up the latest and greatest. Normally though they use the latest update to the version of the app they shipped with, which makes sense from a support point of view.

Re:Why only open source?

[Lingerance]

> It has painless updating to whatever has been put in apt-get/whatever.
Depends on the distro. Most mainstream distros have horribly out-of-date software (by choice). There are distros that do keep near-bleeding edge software in their repos, Arch Linux is one such distro, I've seen packages appear for new releases within a few days, at worst I've seen a package only a few revisions behind.

Re:Why only open source?

[h4rr4r]

I have been meaning to try Arch, do you have any input on quality? What audio system most folks use, etc?

Re:Why only open source?

[Lingerance]

I doubt you'll see this, but, I use Arch on all of the systems that support it (7 now) and the only problem I've found is you get broken dependancies easily, like chromium uses libpng, it gets upgraded but libpng doesn't and chromium won't start because it can't find the newer version of libpng. Other than that it's pretty stable, and if you constantly upgrade to the latest stuff you shouldn't havethose problems. Personally I use ALSA.

Re:Why only open source?

[Anonymous Coward]

It mostly "just works", just follow the very complete tutorials.

It is quite fast, and the only drawback I've found is that from time to time updates can break some config files (or rather, don't modify them to make them work with new versions of the software) because of its rolling release, few mods to upstream packages approach to package management, but it's not nearly as bad as it sounds.

I once read somewhere that Arch is for Slackware users missing a package manager and for impatient Gentoo users.

http://wiki.archlinux.org/index.php/Arch_Compared_to_Other_Distributions

Arch's wiki is incredible, only surpassed by Gentoo's.

Re:Why only open source?

[h4rr4r]

Thanks, I guess I will try it in virtualbox first. Debian has me spoiled on lack of dependency hell.

Re:Why only open source?

[WeatherGod]

Why limit this to open source? It would be great if the users could update every program easily and painlessly, at least the ones that use this new system.

I'm Busted. It isn't really restricted to Open Source... but that's my mission. Commercial apps will be able to play just fine in this ecosystem.

Heh, good luck with convincing other Windows developers that commerical apps can be delivered by a package manager. I have lost count the number of times that FUD about apt and yum has surfaced here on slashdot. If you can get this to work with third-parties, this might make apt and yum more appealing to commerical vendors in the Linux world.

As a Linux zealot who has to administer a few Windows machines, you have my full blessings! Good Luck!

Mozilla Firefox

[Culture20]

If Mozilla can't figure out how to make an msi on their own, maybe they could get some help.

Re:Just like the other vendors

[Anonymous Coward]

Cygwin!? NOOOO!!

Cygwin BAAAD!

MinGW all the way.

I dont fucking CARE how browken FORK is! I like my systems to NOT run compatibility wrappers just to display text to the CLI, DAMNIT!

Oh, you want to use "$*random *nix based library" to do what you want to do in your program? Here's an idea-- create a native toolchain, with native dependencies. What? That would require you to make binary compatibility a REAL issue on linux? )(sarcasm)Oh no-- not that, anything but that. (/sarcasm)

Don't take this the wrong way-- But how would you feel if windows was the OS de-jour of the FOSS community, and all the foss developers insisted that you lug around WINE in order to use their free software? In fact, you know what-- I seem to recall hearing this exact complaint about Linux-Native games. Linux users want linux native, that doesn't run inside a compatibility shell, like WINE, while the game developers say that would increase development time without financial benefit. (Sound familiar?)

Is it so hard to understand that the feeling about native software could be, you know-- mutual? If you are going to develop on win32, and do so seriously, and not as an afterthought, then perhaps you should, you know-- use native libraries?

Just a thought.

So, again:
MinGW, NOT fucking Cygwin!

PLEASE!

The question we all really want to ask

[Anonymous Coward]

Does this service run on dead babies, or are they still alive and crying as you shovel them into the gaping maw of endless darkness that powers it?

Re:The question we all really want to ask

[shutdown -p now]

It's neither, but you really don't want to know how kittens are utilized in the process.

When will Miguel be porting it to Mono?

[Anonymous Coward]

... because such actions are perfectly inline with his every other "contribution" to Linux and Open Source.

Re:When will Miguel be porting it to Mono?

[Jason Earl]

The tools are apparently going to be written in C#. So I suppose that much of the work is already done.

Re:Just like the other vendors

[h4rr4r]

Cygwin at least gives you a usable CLI environment on windows. Who installs this type of server software and does not install Cygwin?
A server without proper gnutools is painful to administer.

Re:nothing MS does is good for anyone but Microsof

[arndawg]

But that doesn't mean they have to HARM someone else. Some good can come out of it as well. Every company is looking after its own interests so I don't see your point here.

Re:nothing MS does is good for anyone but Microsof

[teknopurge]

yet on the other hand, everything google does is good for everyone.


spare me.

It won't....

[Darkness404]

which will conversely bring OSS more users, testers and developers

Not really. most people who test/develop OSS software already do it, or will do when they have free time. As for users, there are about 4 types of users for any Windows program.

A) The person who uses whatever something that is forced on them. Such people will blindly use IE, Firefox, Opera, whatever as long as a boss says they must use it or it comes pre-installed.

B) The person who thinks that they get what they pay for. These are the weird people who we see -buying- boxed software, thinking that for some odd reason if they spend $30 on an obscure paint program they will get a better experience than The GIMP (note that a lot of these people wouldn't, say, buy Microsoft Office and Adobe Photoshop, but rather buy things like Lotus Notes and Correl Paint)

C) The specialist. Generally these are people with high skills who -need- a certain program and know it. These people may have tried OSS alternatives and found them lacking or need obscure programs that OSS doesn't offer.

and finally, very, very, very few people fall into the last category which is people who use the "best" programs and are average users.

This is not going to convert the other 3 types of users which are the majority. Until Dell, Gateway, HP, etc. all start making OSS be default, people in group A aren't going to use OSS. Person B isn't going to think the program is any good if they don't spend money which defeats the purpose of OSS. And people in group C aren't going to use OSS because there is some things that are so obscure that no OSS developer would develop or use.

Re:It won't....

[shutdown -p now]

Actually, this will be mighty handy for developers trying to use OSS libraries on Windows. Right now, it's a mess, if you've got more than a few, and they have mutual dependencies - you get all kinds of wonderful problems with precompiled binaries, such as having them compiled with different compilers (MinGW vs VC++), or with different compiler switches that break ABI compatibility, etc.

And compiling from the source is fun because you have to deal with all the trivial things such as include & library paths by yourself. Oh, and don't forget that a lot of OSS stuff has makefiles generated by autoconf, and many autoconf scripts just freak out on MinGW right away.

Re:Just like the other vendors

[Gadget_Guy]

Gee everyone else figured out a long time ago: give away the compiler.

How is this relevant to this discussion? You are (at least) 8 years too late to be pushing this line - Microsoft has been giving away compilers for a while now.

Maybe this will be a boost for gcc when everyone can see first hand how bad the Microsoft C++ compiler is.

And how bad is the Microsoft C++ compiler? Do you have any specific claims?

BEWARE !!

[Anonymous Coward]

This is the pusherman !! Run like you have never run before !! RUN !! RUN !! RUN !! Run Far, Far away from this pusherman !!

Re:Just like the other vendors

[shutdown -p now]

Gee everyone else figured out a long time ago: give away the compiler.

Microsoft Visual C++ compiler has been free since 2003. It comes as a part of Platform SDK these days.

Of course, this is just a command-line compiler. You can also get Visual C++ Express for free if you want the IDE. This doesn't have MFC & ATL, but you can combine it with PSDK to get a full-featured native development environment; and, of course, you can use it with any third-party framework, such as Qt or wxWidgets.

Maybe this will be a boost for gcc when everyone can see first hand how bad the Microsoft C++ compiler is.

What exactly is bad about VC++ compiler? Can you be more specific? Are you unhappy about it not supporting C++ exception specifications (which no-one uses anyway)? Do you have a problem with optimization quality (in my experience, VC++ inlines things better and deeper than g++)?

Re:Just like the other vendors

[jonwil]

Microsoft has given away its C++ compiler for years now. They started giving it away as part of the Windows SDK and as part of the Visual C++ Command Line Toolkit. These days they give it away as part of the Express Edition of Visual C++. The compiler in that version is identical to the version included in the paid version of Visual C++, as are the C and C++ run-times. You dont get all the fancy stuff like MFC in the free version though.

The info on this specifically says its DIFFERENT to Cygwin (which is a translation layer to allow Unix-esque apps to be compiled on Windows), Mingw (which is a way to use GCC to build Windows apps) and Microsoft Systems For Unix (which is a posix compatible subsystem for Windows that allows specially written apps to run)

As some have said, the biggest obstacle these guys are going to face is the body of OSS software that wont compile in anything other than GCC (FFMPEG is one fairly big example)

Conversely

[Culture20]

This will hopefully bring more open source software on Windows, which will conversely bring OSS more users, testers and developers.

No.
hopefully cause less X, which will conversely bring more Y.
hopefully bring more X, which will conversely cause less Y.
hopefully bring more X, which will similarly bring more X
Yes, similarly.

</nazi style=grammar>

Re:Conversely

[His name cannot be s]

I'm busted.

I dunno man, I wrote that blog post a really long time ago, and then got stuck in red-tape. It's possible I never even proof-read it.

*sigh*

Re:Conversely

[julesh]

</nazi style=grammar>

<nazi style="xml">
You can't include attributes in a close tag. Also, you need quotes around the value of the attribute for it to be confirming XML.
</nazi>

&winkingsmiley;

Re:Conversely

[Anonymous Coward]

"This will hopefully bring more open source software on Windows (voice in your head: boo!), which will conversely bring OSS more users, testers and developers (voice in your head: yay!)."

Gag, cough, bullsheet

[Anonymous Coward]

You know the drill.

Re:nothing MS does is good for anyone but Microsof

[Anonymous Coward]

"spare me"

ROFL

Spare ye not! Let em roast! :O

The largest issue

[codepunk]

One of the largest problems to be faced with this endevor is that I a open source developer could really care less
if my software even runs or compiles on a Windows machine.

Re:The largest issue

[Frenchman113]

One of the largest problems to be faced with this endevor is that I a open source developer could really care less
if my software even runs or compiles on a Windows machine.

It's great to see how enthusiastic you are for this! Not like all those guys who couldn't care less!

Re:The largest issue

[krelian]

But for some reason Linux users really like to complain how some software won't run or have a linux version.

Re:The largest issue

[shutdown -p now]

One of the largest problems to be faced with this endevor is that I a open source developer could really care less if my software even runs or compiles on a Windows machine.

Well, we could perhaps start with finding out how many FOSS developers want their software to compile & run on Windows, so as to see how large of a problem it really is. So, let's see:

- Firefox
- GCC
- Perl
- Python
- Ruby
- PHP
- Gtk+
- Qt
- KDE
- Apache
- MySQL
- PostgreSQL
- SQLite
- ...

... right. Definitely a very large issue.

Re:The largest issue

[styrotech]

And then there are smaller projects with less resources that are just combinations of the above list who can find packaging for Windows a real pain eg:

The Python Qt bindings
The Perl Qt bindings
The Ruby Qt bindings
The Python gtk bindings
The Perl gtk bindings
The Ruby gtk bindings
The Python MySQL module
The Perl MySQL module
The Ruby MySQL module
The Python Postgres module
The Perl Postgres module
The Ruby Postgres module
etc etc etc

I reckon the smaller open source projects are the ones that will potentially benefit the most from this if it works out. And it can only be a good thing for everyone if the devs can spend less time packaging and more time working on the actual code.

ANother ploy just like .nyet

[WindBourne]

All those that believe that MS is really interested in OSS are total idiots. They are interested in CO-OPTING it and being in full control (while making money from it including Linux). This is simply another part of their plan.

Re:ANother ploy just like .nyet

[msclrhd]

You mean CoApp-ting ;)

Let me right the first line of code for you.

[codepunk]

I am feeling generous this evening and decided I would donate the first line of code to this
fine project. I relinquish all copyrights on the following line of code, feel free to do with
it as you wish.

#include "ie6.h"

Re:Let me right the first line of code for you.

[Anonymous Coward]

"Let me right"?

Get back to kindergarten where you belong.

True OSS tools for both push and pull

[Pav]

If you're looking for a truly OSS tool for full desktop management (but push instead of pull) check out OPSI - it even deploys the entire Windows OS (via PXE), and afterwards allows management of apps. If you're looking for a simple graphical apt/rpm for windows both Cygwin and KDE have had their own versions working for years.

As an aside, OPSI can be used as part of a stack to replace an entire (Microsoft?) corporate software stack. Check out GOsa - this is the software that the City of Munich is basing much of its Linuxification effort around, and it is also used in other cities and organisations around the world - check out the GOsa website for a list. GOsa manages clients/servers via LDAP and RPC, and OPSI is just one of the stack of software it can manage via its web-GUI. The others include Samba + PDC (achieved using the GOsa goPDC scripts), groupware (choice of Kolab, phpGroupware etc..., or a 3rd party LDAP-aware groupware eg. SOGo), DNS, DHCP, Nagios, OPSI+FAI (for client system management), and a lot of other software I can't even remember. GOsa + supporting software is HARD to set up (especially due to out of date and missing docs), but I'm one of a team of 3rd party guys trying to document it better... check out the docs/scripts in the GOsa contrib section and visit us on #gosa on freenode. (Most of the guys are in Europe so keep this in mind when picking a time to visit the IRC channel).

Re:Just like the other vendors

[shutdown -p now]

Visual C++ has had correct - i.e., standards compliant - scope for variables declared in a for-loop since VC++2003 by default (you can still have the old behavior explicitly enabled by compiler switch).

Before that, you could control it with a switch, though default was non-standards compliant, and MFC headers wouldn't compile if you turned it on - which shouldn't really concern you if you're compiling portable code, right?

In practice, this means that there was only one release of VC++ which was non-compliant by default - VC++2002. The one before it, VC6, was released in 1998, before the final ISO C++ specification came out, so it's kinda silly to hold it against it. If you recall the original story, the "wrong" behavior was actually part of the draft spec at some point - they've been going back and forth on it.

Also, it is really a minor problem by itself, since you can trivially work around it by doing:

#define for \
  if (false); else for

or compiling with the equivalent -D compiler flag. This will ensure correct scoping, and will not affect anything otherwise (the compiler will, of course, optimize away the always-false branch).

In contrast, g++ 2.95 (which was the stable version of g++ until mid-2001 - assuming you consider g++ 3.0 stable) didn't even have proper namespace support - it did parse namespace { ... } and using declarations correctly, but pretty much just ignored them, and just dumped all identifiers into global namespace. That is something that is not anywhere near as easy to work around.

There's at least *1* fatal flaw with this concept

[Johnny Loves Linux]

The reason why most Linux distros have some kind of package management is because there's so much software with inter-related dependencies that it's a pain in the buttox to try to manage all those dependencies by hand.

One key point about the Linux package managers is that they are needed to manage *all* that open source software. Why open source software? Because no (or rather very few) proprietary companies provide proprietary software for Linux. So if you want some kind of functionality under Linux your best option is to write an open source version because there is very little proprietary software for Linux.

Why then is package management bad for Windows, if it's so useful for Linux? Well, because package management is effective if you have can pull source code, compile, and determine dependencies. That means an open source license for the source code. That means that Microsoft is telling it's huge "ecosphere" of proprietary software vendors to "Piss off you sod!", because *their* software doesn't fit the package management model. That leaves 2 possibilities in the Microsoft world:

1. The package manager is successful and really takes off. Which means most software is open sourced and provided packaged by Microsoft and the proprietary vendors are so screwed out of existence, as they are no longer needed if you can just do an "apt-get install" on Windows without the vendors.
2. The independent vendors realize the implication of this package manager and what it means to their bottom lines and either threaten to leave the Microsoft world for Apple OS X or Microsoft drops the package manager idea.

Re:nothing MS does is good for anyone but Microsof

[im_thatoneguy]

Nothing MS does is good for anyone but Microsoft this is no different.

Yeah like when they added desktop search so that I could find files easier. That did nobody any good except for Microsoft... somehow. Or when they sold me Halo. I really wish I could have benefited from that purchase somehow but sadly only Microsoft had any fun.

MS's new ways of competing is brought to you by

[ClosedSource]

the US Department of Justice with the generous support of Sun Microsystems, Oracle, IBM, Netscape, and Novell.

This seems like an opportunity

[digitaltraveller]

To build a whole new package management system that works across Linux and Windows.

Any chance you could build something that uses a git / msysgit / jgit backend, to allow for rewind-fastforward versioning of apps?

Re:Just like the other vendors

[Anonymous Coward]

GNU tools don't actually NEED cygwin though, as they can be compiled with MinGW instead, and then be fully native.

EG, "LS, BASH, DD, and pals" can all be compiled as native executables that run in the win32 subsystem, like they are supposed to, and not need to hook an outside DLL to do their work.

That is the way it SHOULD be. You really should only have a Cygwin dependency if you ABSOLUTELY need something that MinGW simply cannot provide (Like GNU style FORK). Otherwise it is not the correct way to go about it.

Now with more end quote tags

[Anonymous Coward]

Assuming that you've looked at APT and similar packaging tools, and given that you're still convinced that there's a 'Windows Way' (your term) to handle deployment that differs from Linux best practices, how do you plan to address:

Yes, I've worked with APT and RPM for a very very long time now. The reason I'm convinced there is a 'Windows way' is because it's a different system that Linux; yes, I've learned a lot about PMS from Linux, and I know how to apply that knowledge to Windows.

Package Repositories - This is one of the main strengths of Debian and related distros. Do you think it's even possible to replicate this level of community control in Windows? I know you've mentioned decentralisation, but have you considered the implications of such an approach? What is the cost of failure to affect consistent, formalised management of package builds?

I have a plan for allowing any publisher to publish packages in the CoApp ecosystem, provided they meet two qualifications:
- They must be able to host their repository meta-data on an SSL protected connection.
- All packages must be digitally signed with a certificate that chains back to to a commonly-accepted CA.

Dependancy Management - This issue is largely done and dusted on Linux, but remains a dog's breakfast on Windows (albeit not as frustrating today as it was in the mid-90s). In the absence of centralised repositories and the Unix toolchain philosophy, how do you propose to cope better with dependancies?

I'm working with the developer of WiX to ensure that we can trivially build chained MSI packages that have the necessary smarts to properly manage this. Kind-of mixing in something like ldconfig with the Windows SxS library management.

File locations - How do you propose to manage the proper placement of libraries etc. when the conventions concerning where to put such files are not nearly as well defined on Windows? I'm suggesting here that you need cultural leverage rather than technical answers. You need to change perceptions, not toolkits.

Yes. The change starts with PHP, Apache, and Python, and the 40+ packages needed to build them (community members from each are already on board) Half of the project is setting some intelligent standards, and then bootstrapping the ecosystem with packages to enable other software to follow.

Security - Do you think it's even possible to replicate one of the main strengths of Linux package repositories: the ability to curtail security risks such as malware and flawed code?

Yes. By requiring code-signing (and I've got a plan for opening that up without cost for smaller projects) we can replicate the benefits of MD5 and PGP signatures found in the Linux world.

Scripting Interfaces - Say what you like about make and other command-line utilities, but as a busy sysadmin, I consider GUI package management a waste of my valuable time. If I'm going to deploy regular security updates, for example, I want to know that I can script every aspect of the operation. Even the tab-completion features in aptitude make it many times more efficient than a point-and-click interface. What is the potential for scripted deployment/management of packages under your system? Why?

I agree 100%. Scripting interfaces are an absolute requirement, and will likely come well before the GUI.

Think of it as a clean adaptation of the same concepts to the model that will be attractive to Windows developers.

I also think that you're going to need to learn a lot more humility than you've demonstrated so far if you want to achieve something better than a new brand of anarchy in packaging.

I apologize if I'm coming off arrogant. Frankly it's taken an extremely long time to convince the powers-that-be at Microsoft that Linux's package management is stellar compared to Windows. It's also not near as hard or large as it sounds, I'm walking on the shoulders of giants here, both in the Linux and Windows worlds.

Works fine for me...

[Anonymous Coward]

You know it's funny, I've installed several OSS apps on Windows and it's never really as hard as this article makes it out to be. I've installed Pidgin, Xchat, SMplayer, Handbrake, Virtualbox, and more. In fact it was just a matter of double clicking an exe file and they install just like anything else. Wouldn't the problem be more of the laziness or disinterest of developers for not bothering to create a Windows installer? I hardly think they would bother having their code signed for inclusion in your repository if they've never been assed to port it in the past.

Re:Works fine for me...

[spongman]

have you ever tried to build any of those apps on windows?

good luck with that.

ACK! BSD?

[Anonymous Coward]

You must not have any RED blood in you, Amerikanski !! Me neither, only Vodka !!

Re:Just like the other vendors

[k8to]

It's not that the compiler needs to be free (it is), it should be part of the standard OS install. Perhaps an optional component, but a single click during install at most.

Re:There's at least *1* fatal flaw with this conce

[shutdown -p now]

Well, because package management is effective if you have can pull source code, compile, and determine dependencies.

I don't see how this follows. Most Linux distros do package management on binary level, not source level, and, in fact, can happily package closed-source software.

Why does it matter if application A and library B are open source or not, if there is a known dependency from A to B?

Visual C++ redistributable libraries license?

[Anonymous Coward]

I assume you are going to compile things with Visual studio, meaning all C++ programs will have to link against the Visual C++ redistributable libraries. Since these are not actual system libraries, you'll have trouble linking and distributing GPL programs against them. Since it seems you actually talk MS management, could you get try to sort out this issue? Otherwise, I fear you will not be able to distribute GPL'ed programs.

Re:Just like the other vendors

[spitzak]

Before that, you could control it with a switch, though default was non-standards compliant, and MFC headers wouldn't compile if you turned it on - which shouldn't really concern you if you're compiling portable code, right?

I have to strongly object to this phony argument. Too bad because most of your posts are quite good!

You seem to think "portable" means "it does not contain any calls that don't work on both platforms". This is an artificial description purposely designed so that "portable" is equivalent to "useless", which does sound like typical Microsoft word twisting.

"portable" code looks like this:

      portable-code;
  #ifdef _WIN32
      xxxxx
  #else
      yyyyy
  #endif
      more-portable-code;...

Note that "xxxxx" will certainly require the windows.h header file to be included!

This is no joke, for many years I had to keep fixing for loops because they would not compile on Windows. In most cases this was by far the most likely reason the code would fail to compile. So I think your attempt to defend Microsoft's old behavior is rather dubious.

Why most free software is already on windows

[SpaghettiPattern]

Every time I am forced to do something half-smart* on windows in invariably install Cygwin. It sort of covers all the crucial gaps. As soon as viable I take anything I need away from that god forsaken platform, process it and send the results back.

* half-smart: E.g. diff 2 files, edit 1 file, strip off \r, add \r, analyse XML, beautify XML, search files, run fortune

Sssshhhhh!

[poptones]

Don't speak that in public! The PITA toolkit install was one of the prime motivating factors for me personally to migrate from the hell that is windows a decade ago now. I'm proud of the contributions I made to avisynth, but the problems with brittleness and cleaning up from infections and having to spend two freaking days trying to rebuild a development system after a reload was just too much to endure.

Not to mention the problems with dependancies. It seemed like every six months MS would release some new jscript revision that broke previous jscript features, meaning anything I tried developing using those tools was so tied to one specific configuration it became a support nightmare.

This project seems to me like it's a decade late. Good luck with all this, but I can assure this is one former Windows OSS contributor who will never go back to the dark side.

Re:Just like the other vendors

[Anonymous Coward]

(Posting anonymously because I somehow managed to exceed 50-posts-per-day limit for my karma.)

Correct me if I'm wrong, but why would "windows.h" have any problems with for-loop scoping either way? IIRC, the problem was strictly with MFC and ATL, for both of which there are much better options available, in any case.

I may be wrong here - this problem is a really old one, and I only vaguely recall last time I hit it. If I am (or if you really need MFC/ATL), more general solution in this case - since you're already using the preprocessor - is to use that "#define for ..." trick for your code, but skip it for "windows.h". VC++ has a non-standard way of doing that in form of push_macro and pop_macro pragmas, so you can do something like this:

#ifdef _WIN32
    #pragma push_macro("for")
    #undef for
    #include <windows.h>
    #pragma pop_macro("for")
  #endif
portable-code;
  #ifdef _WIN32
      xxxxx
  #else
      yyyyy
  #endif
      more-portable-code;...

Not very nice, but it's still not that much boilerplate, and it scales well. You can do even better by making your own header which just does #include "windows.h", wrapping it in push_macro/pop_macro, and then using that everywhere.

I don't dispute that it isn't problem, anyway. I do recall it being rather annoying back in the day, but then doing C++ back then was generally annoying, because standard compliance was lacking all over the place. I recall being similarly frustrated by e.g. lack of std::vector::at() in g++ standard library - God knows why it wasn't there. Or, getting back to VC6, advanced template magic such as partial template specialization was very much hit-or-miss. Oh, and no RVO, which really is a perf killer. And so on. I dare say that, against this background, the for-scope issue is really just a minor part of the overall bleak picture, with a relatively trivial workaround.

Re:Just like the other vendors

[mandolin]

What exactly is bad about VC++ compiler? Can you be more specific? Are you unhappy about it not supporting C++ exception specifications (which no-one uses anyway)? Do you have a problem with optimization quality (in my experience, VC++ inlines things better and deeper than g++)?

VC++ doesn't support variable-length arrays. IOW it pukes on

void foobar(int i) { int myarray[i]; }

Variable-length arrays are part of the C99 standard. That's 10 years old at this point and it's a pattern I employ often. I am unsure of the status of this in VC++ 2010 (coming Real Soon) but last I heard, it still wasn't in.

Re:Just like the other vendors

[Gadget_Guy]

It's not that the compiler needs to be free (it is), it should be part of the standard OS install.

There are two compilers that ship with Windows: C# and Visual Basic. I learnt C# mainly because it was installed on every Windows computer, which is handy if you are working on someone else's system and need to knock up a quick program.

If you need C, but can't handle downloading and installing it then you should probably consider choosing another language as the installation will be the least of your problems!

Re:Just like the other vendors

[shutdown -p now]

VC++ doesn't support variable-length arrays ... Variable-length arrays are part of the C99 standard

Yes, VC++ does not support C99 - it sticks to C90 with a few common extensions (such as // comments). Heck, it even won't let you declare variables anywhere except the beginning of blocks when compiling .c files.

However, C99 is not a subset of C++. The priority for VC++ is to support C++ - which is why VC++2010 adds a bunch of C++0x features (auto, decltype, lambdas, static_assert, rvalue references). It does have a few C99 features, to the extent C++ TR1 requires them - most notably, stdint.h.

That's 10 years old at this point and it's a pattern I employ often.

The standard is 10 years old, yes, but it is relatively obscure. Even GCC doesn't claim full compliance yet, and that's saying something. Speaking of VLAs specifically, they have been broken in GCC until very recently - 4.4 still listed them as "broken", and only 4.5 declares full support - and the latter was released on March 31, 9 days ago.

I haven't actually seen any code using VLAs in the wild, likely precisely because of that - it's simply non-portable in practice. So far it seems that any practical use of C99, by and large, is restricted to stdint.h (immensely useful), "restrict" (can be very handy for particular applications), and free-form comments and variable declarations.

By the way, the topic of C99 compiler support in general, and VLA support in particular, is, in fact, a favorite one for flamewars on comp.lang.c.moderated.

screenshot to groklaw please

[Anonymous Coward]

A screenshot of the signed agreement is needed for legal purposes, in case you are fired by a nutcase Microsoft manager.

His name cannot be s (16831)

[kubitus]

Garrett Serack is pretty good with words. So are many other M$ officials, execs.

Lets see what deeds and results come out of his declared good intentions

forgive me if I am sceptic, the road to hell is plastered with good intentions

Code Review

[Anonymous Coward]

#include <ie6.h>

There, it's fixed. Whoohoo, we're already on revision 2!

Name Fail

[BlindRobin]

They could have been more ummmm.. direct and named it CoOpt. No ?

Dammit!

[badpazzword]

Just when I thought I had managed in the impossible feat to make a /. submission where nobody complains about the summary... :(

Re:Dammit!

[badpazzword]

Welcome to Slashdot. You must be new here.

3 E's

[Slash.Poop]

Embrace
Extend
Enhance

Dear Santa...

[tepples]

I'm Santa.

Ask me about anything to do with...

I've read rumors that your toy delivery operation would run more efficiently if you ran it hub-based like FedEx. Give a sleigh and a set of reindeer to each of your shopping mall representatives and have them deliver the presents locally. I suspect you already do this; is that true?

My second question: Where does your organization get the money to pay for the raw materials and the elves' wages? I don't see how your theme park outside Evansville pays for everything.

Re:Dear Santa...

[RockDoctor]

My second question: Where does your organization get the money to pay for the raw materials and the elves' wages? I don't see how your theme park outside Evansville pays for everything.

Four words : "tooth", "fairy" , "overpriced" and "dentistry"

This is odd...

[joedoc]

...there's a lot more ice here in Hell than I would have imagined.

Authenticode for individuals?

[tepples]

All packages must be digitally signed with a certificate that chains back to to a commonly-accepted CA.

So how does an individual developer of free software get such "a certificate that chains back to to a commonly-accepted CA"? The Authenticode CAs that I checked tend not to issue certs to individuals. Must every developer of a Windows application form a partnership or LLC? And must every developer pay upwards of $160 per calendar year (source: Comodo) for the privilege of releasing packages or updates in that year? That's even more than Apple charges for access to the app store on the iPhone.

Re:It is obviously a trap!

[miknix]

The way I see it, this might possibly happen:

1) Some OSS developers will, by curiosity or any other reason except need, try targeting windows using this package management and deployment system.

2) Lots of bugs and/or feedback start to appear in their tiny OSS project, coming from people running Windows.

3) Developers are blinded by 2). They sudden realize that there is a *huge* new ecosystem that they can support.

4) The primary target of Development changes to Windows. Developers abuse of hackery to properly run their software on windows. Heck! Some even buy/install windows just to test their software on it.

5) After a while, some software that used to be cross-platform is now full of dirty-tricks to run on windows. This obviously damages the stability on other platforms.

And why do I think this might happen? Because we, OSS developers, despite enjoying developing it; we enjoy more when others use our software. Worse, what happens if this OSS software management and deployment system for windows introduces some kind of Apple-store framework on it?

Cygwin

[Anonymous Coward]

What's wrong with Cygwin?

Re:Just like the other vendors

[spitzak]

We were not using MFC or anything, and I agree I can't see how anything in windows.h would be a problem (a for loop would have to be a macro and we could then avoid calling that macro most likely).

We certainly stayed with this behavior until long after they had added the switch to the compiler, some of that may have been intertia and an unwillingness to modify the rather complex and error-prone makefiles. I do remember it being the primary problem when we tried to port software from Linux to Windows and that in a few cases the hurried editing to fix it would break, mostly by hiding a parameter with a local variable with the same name.

Re:Just like the other vendors

[tepples]

You are (at least) 8 years too late to be pushing this line - Microsoft has been giving away compilers for a while now.

The "Windows Mobile SDK" for Windows Mobile 5 and 6 explicitly didn't work on the Express version of the compiler.

Re:Just like the other vendors

[TemporalBeing]

Gee everyone else figured out a long time ago: give away the compiler.

How is this relevant to this discussion? You are (at least) 8 years too late to be pushing this line - Microsoft has been giving away compilers for a while now.

True, but they limit what you can do, namely how well you can integrate new things into VS.

Maybe this will be a boost for gcc when everyone can see first hand how bad the Microsoft C++ compiler is.

And how bad is the Microsoft C++ compiler? Do you have any specific claims?

Quite bad when you start comparing it to standards. Ever wonder why MSVC doesn't support int32_t, etc? Until they fix basic things like that, it will continue to be a broken compiler.

Re:It is obviously a trap!

[tepples]

Worse, what happens if this OSS software management and deployment system for windows introduces some kind of Apple-store framework on it?

Do you mean like Software Center in Ubuntu 9.10? If not, what "Apple-store" aspects are you talking about?

Re:Just like the other vendors

[mandolin]

Thanks for the thoughtful and comprehensive reply.

I realize and understand that C99 is not a C++ standard and that VC++'s priority is C++ support. Of course, my priorities are different. I shouldn't claim that VC++ is bad for everybody, but it is bad for me.

Speaking of VLAs specifically, they have been broken in GCC until very recently

It is true VLAs are/were 'broken' in GCC in the sense that they are not fully C99 conformant. That has been documented in the info pages for a long time. However, speaking practically, for the simple declarations I have used, I have never had a problem. Perhaps I ought to double-check that :P

it's simply non-portable in practice

I agree that non-portability across compilers would be a problem, except that of course GCC itself is so incredibly portable, and is also the preferred compiler on many non-Windows platforms.

FWIW, I am not a C99 stickler -- I only care about the parts of it that I use, and I specifically don't worry about free-form variable declarations because there is an easy fix -- declare your variables only after a block begins.

VLAs have no "easy fix" -- the closest alternative is alloca(), which my manpage claims is "buggy" on many unspecified systems, and of course, it's not really portable or standard either. Using malloc() is slower, uglier (since if you're doing it right you'll want to check the return value), and has the potential to fragment your memory.

I apologize if these points have already been debated to death in the flamewars -- I don't read that group. I also expect you are already aware of these points and were simply trying to make yours, in which case, this was a superfluous post. Oh well :P

Re:Just like the other vendors

[shutdown -p now]

VLAs have no "easy fix" -- the closest alternative is alloca(), which my manpage claims is "buggy" on many unspecified systems, and of course, it's not really portable or standard either. Using malloc() is slower, uglier (since if you're doing it right you'll want to check the return value), and has the potential to fragment your memory.

alloca() is actually decent option on all mainstream architectures - the problems appear only when you delve into really exotic stuff. It can be easily broken when implemented as a true function (playing tricks with the stack pointer), but that same gcc actually implements it as an intrinsic (i.e. the output just directly adjusts the stack pointer as needed) pretty much everywhere. That is rather foolproof. If you go by the basic rule of thumb of only ever using it in a variable initialization (which is consistent with limitations of VLAs), it can't really go wrong.

I imagine that we will get something comparable (but standardized) in VC++ when it gets into C++ - e.g. the std::dynarray proposal. They'll have to wrap up C++11 first, though...

Anyway, IMO, main benefit of VLAs isn't even in dynamic stack allocation - it's rather in painless handling of dynamically sized arrays, especially multidimensional ones, when passing them as arguments, e.g.:

void foo(int w, int h, float a[w][h]) {
  a[3][4] = 1;
}

In C90, since you can only get a pointer through, you have to do all addressing arithmetic by yourself:

void foo(int w, int h, float* a) {
  a[3 * h + 4] = 1;
}

The verbosity increases exponentially as more dimensions are added, while VLA solution takes care of all that automagically.

Yeah, I don't really see a fix for that short of supporting them on the level of type system, one way or another. I'm not aware of any proposals to have something similar in C++, either (though there, of course, you'd likely just do a heap allocation, and wrap the arithmetic in a class, like boost::multi_array).

Did Serack have to duck any chairs?

[jonperez]

Did Serack have to see Ballmer wreck a few chairs first before he got the green light?