If someone eats their dessert from a dirty plate, they won’t have a culture shock when they eat those desserts from a clean plate.

If you get used to skydiving onto poison ivy, you’ll likely won’t have a culture shock when you skydive onto soft ground.

I don’t recommend you eat from dirty plates or dirty plates for this purpose. I don’t recommend you skydive onto poison ivy or grow poison ivy for this purpose. It’s just as easy to skydive onto something else and it will be a better experience even if initially it will be a bit tough. Ditto for eating from a clean plate. If you aren’t accustomed to plates or eating it might be a little tough at first.

http://www.linuxtoday.com/news_story.php3?ltsn=2008-08-03-005-35-NW-MS-0000

http://www.linuxtoday.com/news_story.php3?ltsn=2008-08-01-028-35-NW-MS-RL-0004

>> On the server side, the purpose of FOSS on Windows is entirely so that sysadmins can set up a sensible system on a crappy OS that they were forced to use… and then, come cash crunch time, they can easily slide the OS out from underneath and just casually triple performance on the same hardware. I have done this trick and seen this trick quite a bit. It’s most satisfying.

Sure, if the WinFOSS exists I might find some use for it too if I had such a job.

On the other hand, if the WinFOSS doesn’t exist (or is very sucky), there is a greater chance I would not have to use the crappy OS but would be able to use Linux to access the LinFOSS. Sounds like a pretty simple decision for me, actually.

WinFOSS doesn’t cut it for me. Most of the time I just might prefer to use Monopolysoft tools. The thought I would be hurting my chances of being able to work on Linux because WinFOSS exists and is of OK quality makes me a bit queasy.

And why would I want good FOSS devs to waste their time?

>> And from a purely technical viewpoint: porting stuff cross-platform always results in a more robust application, as unexamined assumptions get examined. Porting from a Unix-like platform to a weird and crappy one like Win32 is an extreme case, but certainly counts, and typically results in better-architected apps with clearer separation between logic and platform/presentation.

Not true. Porting to a bad architecture interferes with good design practices.

What the developer does is find the design that allows each of these archs to be addressed with as little trouble as possible, but if one of these is broken anyway (not to mention shifty), why bother? Just go for the significant time savings and better overall code and design by focusing on the quality architectures. There are many architectures for which to practice that are better than Monopolyware and are less shifty if you are really after self-improvement and have the time.

Truly, you can’t be competitive with Monopolysoft if you play their games while they use their internal APIs. We might as well optimize for Linux so as to have a chance to kick hiney all around on the desktop. Users will see the final more optimized result on Linux and know it beats Monopolyware. But we have to focus on a solid Linux experience.

Don’t take Monopolysoft for granted. Let them make a fool of themselves, but always take them very seriously. One day all of those PHDs might decide to start working hard and playing less tennis (or count money or do whatever it is they now do). For that day, we will need all of the Linux focus and optimization possible. No time to waste on closed source Monopolyware.

If you have time to waste and don’t mind helping Monpolysoft hold their monopolies a little longer, well go ahead then. I don’t have that luxury or such a strong stomach.

>> So yeah, more free software on Windows! I’m a fan of the idea! I encourage it! Yay free software on Windows!

I think you know my feelings towards indulging Monopolysoft by/while sullying the work of those that work openly and license their code for the benefit of almost everyone [I say "almost everyone" because Bill Gates has enough money not to benefit from LinFOSS in any special way and actually seems irritated by the notion everyone else would have that access.]

>> You appear to be reacting with incredulity to my actual experience and assuming that if you assert loudly enough and question that I’ve actually used a modern Unix then reality will change for you. It won’t. You’re not responding to anything I actually wrote, you appear to be reacting to what you assume I wrote. You could try again, because I’m certainly not going to go through that tl;dr line by line.

Well, if you still want more clarity, just ask. I can try again.

There is no reason for us to fight. It’s Monopolysoft the one that doesn’t deserve 100 more opportunities to screw with us.

You appear to be reacting with incredulity to my actual experience and assuming that if you assert loudly enough and question that I’ve actually used a modern Unix then reality will change for you. It won’t. You’re not responding to anything I actually wrote, you appear to be reacting to what you assume I wrote. You could try again, because I’m certainly not going to go through that tl;dr line by line.

I very much disagree with you.

Most people that stick to Linux or to Windows have something that they really don’t feel they can get on the other platform. This is particularly important for Linux since it doesn’t come pre-installed and fewer locals are experienced with it (it’s the new kid on the block).

>> If someone’s living in Firefox on Windows, they’re going to have no culture shock at all going to Firefox on Linux on their Eee.

Most of these apps are not that difficult to pick up when you know the other. How else did Firefox get 20% of the market in the first place? The key is having something you can’t really get anywhere else. Firefox had/has that over IE. The more Linux has that is unique and valuable over Windows, the greater the odds that people will be willing to do what is necessary to use it. Eg, Microsoft now is messing with the boot loader. This will send a number of dual boot noobs back to Windows/Vista only.. But it would not if they “just had to” get access to that app.

What I find valuable is that 10 decent free FOSS apps only really available on Linux will create a strong desire to have Linux running somewhere. If these work fairly well (“good enough”) on Windows, why run Linux? I am speaking about the typical practical user that doesn’t care about freedom too much at this point and is just using the computer as a desktop. Also, $0 has value to those that like to be clear and legal. And everyone has a hacker inside that wants to create something different and share it. All of these Linux advantages disappear if there is a “good enough” port. ["good enough" depends on the user, but the more we work on wasteful Winports, the more users will fall into the set]

Honestly, I have not seen or heard about that “culture shock” you are talking about. Where you might see that is with sophisticated users that know really well some app that you can’t find on Linux. But these people would use Linux for good and special apps. They just might not dump Windows for a while (eg, for unique games).

Are you sure you have seen a modern Linux distro lately?

I think all FOSS app projects should create a distro just to feature that app: how to use it, tricks, tutorials, demos, etc, and keep the rest of the system as simple as possible.

>> On the server side, the purpose of FOSS on Windows is entirely so that sysadmins can set up a sensible system on a crappy OS that they were forced to use… and then, come cash crunch time, they can easily slide the OS out from underneath and just casually triple performance on the same hardware. I have done this trick and seen this trick quite a bit. It’s most satisfying.

You aren’t the typical Windows admin.

I don’t get it. You are allowed to switch servers but you can’t have Linux there to begin with and develop on it? Linux is free. It doesn’t take up much space and even comes in a LiveCD (a custom version can be set up to include all your fav tools). If there were challenges prior to the switcheroo (challenges that may have been used as an excuse for not having Linux) then the same issues will be there afterwards for the most part. A demo can be done on a LiveCD and spare PC.

Why would they want to use Apache in the first place? Well, free developed add-ons might be made for Apache. People want to tap into that. Diminish the motivation behind acquiring relatively greater time and effort savings, and why deal with Linux, a Windows admin will wonder? Again, you are an exception. Linux is still the new kid in town and has less developed tools in some areas.

A more likely scenario (not for you, but for others I think) is that knowing Apache runs on Windows will lead those that don’t want Linux (fewer things to have to learn) to use the Windows version.. especially if it is “good enough”. Without a “good enough” Windows version, Linux would actually be seen as a savior because of all the free goodies they could use and save time and effort.

On a practical level, Microsoft is missing the volume of decent free goodies users want. That is what Windows ports can give them.

Developers developers. Microsoft wants “freebie” coders coding for their platform. This takes time away from Linux and general app growth. It puts the project on the Microsoft controlled treadmill. It makes it easier for others to add Windows-only items (here Microsoft would spend more of their money).

If Windows is sooo “crappy”, I pity your company. You will likely go out of business soon. Your CEO should get fired for using such a crappy platform. Or maybe Windows is not that crappy as you pretended?

Fact is Windows is viewed as good enough in many situations by many. Linux needs motivational items in its corners. The last thing FOSS devs should be doing is wasting time on the treadmill while giving Windows value.

>> And from a purely technical viewpoint: porting stuff cross-platform always results in a more robust application, as unexamined assumptions get examined. Porting from a Unix-like platform to a weird and crappy one like Win32 is an extreme case, but certainly counts, and typically results in better-architected apps with clearer separation between logic and platform/presentation.

Linux ports to a ton of platforms already. Microsoft has full control over Windows. It’s a time sink running the treadmill.

Technically speaking, dealing with a changing and opaque platform that cannot be understood well is a loss.

When Windows becomes completely open source, this argument will cease to be.

Today, it is a technical disadvantage to fight the Windows quicksands.. the profiling you do today could change in an instant and you will likely never be told or be able to discover it.

>> So yeah, more free software on Windows! I’m a fan of the idea! I encourage it! Yay free software on Windows!

I feel quite the opposite. The more reasons people have to drive them to Linux the better. People adjust easily to similar interfaces, but will fight the obstacles only if there is something they think they will get that they basically can’t get elsewhere.

There is an equation pitting things like laziness against things like excitement. People require as much of the incentives as possible to increase the odds of overcoming the obstacles.

To Roy: Microsoft gets _zero_ influence on the coding of the httpd server. To say otherwise means you clearly don’t understand how Apache works and don’t know any of the actually committers. Besides, Apache httpd is licensed in such a way that Microsoft could extend it anytime they want without giving the ASF money. Beyond that, well, hey, “patches welcome.”

Anyone else want to try and spread some ill-informed conspiracy theory about this?

If someone’s living in Firefox on Windows, they’re going to have no culture shock at all going to Firefox on Linux on their Eee. Same for GIMP and OpenOffice. Users care about doing their stuff; if the free software is *clearly better* (e.g. Firefox), then handing them that Ubuntu CD when they’re sick of Windows’ flakiness means no disruption to their actual work and play.

On the server side, the purpose of FOSS on Windows is entirely so that sysadmins can set up a sensible system on a crappy OS that they were forced to use … and then, come cash crunch time, they can easily slide the OS out from underneath and just casually triple performance on the same hardware. I have done this trick and seen this trick quite a bit. It’s most satisfying.

And from a purely technical viewpoint: porting stuff cross-platform always results in a more robust application, as unexamined assumptions get examined. Porting from a Unix-like platform to a weird and crappy one like Win32 is an extreme case, but certainly counts, and typically results in better-architected apps with clearer separation between logic and platform/presentation.

So yeah, more free software on Windows! I’m a fan of the idea! I encourage it! Yay free software on Windows!