Re: [Flightgear-users] pre-release 0.9.9 on Win32

[Eric Brasseur]

[EMAIL PROTECTED] wrote: Hello Kevin, I think that you spoke the truth but that the truth is not always what matters. I think we shouldn't care about who has "merits" and who is "responsible". The Open Source strategy is brilliant. To me FlightGear just has some problems of tactics. End users are a little dumb when it comes to compiling things, even with a neat procedure listed. Whatever precise and reliable the procedure, they simply don't understand the words. Even if they understand the words, they're afraid. They already have so many problems with their system... I once installed Cygwin and used it. It wasn't simple at all. It took me gigabytes of DSL bandwidth. The developers must adapt to the end users, not the opposite. It's quite possible that I'm not getting what you're saying. I read you as saying that for the most part, the FG-on-Windows user community isn't going to "get it" when it comes to making the kind of contributions to testing that developers need, and so the developers have to do whatever is necessary to make things easy for the Windows users to try the software out and report problems -- becasue if the developers don't do this, the Windows users just aren't going to provide that input/info. Do I have you correctly? Assuming that I do . . .first of all, it's important to understand that the project doesn't have a huge number of developers contributing; and of those that are, very very few of them (one or two?) are working on the Windows platform; and of those that do, they are limited in time available to contribute (like all of us). For example, I can't do anything to make it easier for Windows users because I don't own a Windows machine, I don't use Windows, and I don't have access to a Windows machine on which I could work on FG-related stuff. So what you're essentially doing is telling the very small number of Windows contributors -- people who probably got involved, and sustain their interest in the project, because of things they were excited about working on -- what they should be doing with their very limited time *instead* of working on what they find interesting and what keeps them enthusiastic about committing their limited time to the project. Can you see how that might rub some people the wrong way? And so developers tend to respond to stuff like this with the thing that first came to my mind when reading your post: "It sounds like this is important to you; so it seems to me that this would be a fabulous way for you to get started contributing to the project." I've seen this paraphrased earlier in this thread as "fix it yourself," which may make it seem like the poster is being blown-off; but I assure you that that's not the case. That is the path by which more or less *every* contributor to *every* open source project got started working on that project -- they saw something that they thought would be worth doing, and they started working on it. Jumping in and trying to make a contribution is well-respected and encouraged; telling other people to stop doing what they're doing, and do something else you find important but they may not, is bound to come across as rude. Particularly when one barely has any time to work on the project at all, one wants to work on things one has some enthusiasm for. FG has a fairly sizeable Windows user community. What it doesn't have is a sizeable Windows *developer* community. If it did, Windows-specific bugs would probably get fixed a lot faster, pre-release testing would be a lot easier for Windows users, etc. And that would be good, sure. But I myself am not going to be purchasing Windows XP and installing it on my machine, and learning how to develop on that platform. I want to work on the things that interest me. If I'm not doing that, working on FG stuff becomes a chore rather than fun. And believe me, I already have enough chores to do. In the end, the solution is for developers who care a lot about fixing a problem to fix the problem. That can happen either through the small number of existing Windows developers -- in which case it may take a while, because there's a small number of them and their time is tight like everyone else's. Or, it can happen because some Windows users who like FlightGear decide they care enough to do what other contributors all did -- roll up their sleeves and try to help get it fixed. So again: you're saying that it's important for the project to make it easier for Windows users to test the software? Making that happen sounds like a good way you could help out. -c Hello, Your assumptions on my assumptions are correct. But I'm not going to put my hand in the Windows stuff, for the reasons you mentioned above. The problem should be split in three nested parts: More persons need to test the binaries on different Windows platforms and hardware. v More persons need to become capable of compiling source code on Windows. v More persons need to become capable to work on the source code on Windows. To me the final suggestion you mention is a key to this. If installation of Windows pre-versions becomes easier, many will try them out. Some of them will get enthusiast enough to learn to compile, because they love to get an even more recent version. And some of these some will get their hands in the code. I'm a bit puzzled about the fuss around this because to me building such a pre-version is a matter of zipping the binaries directory, together with DLL files and a short README. That way you get sort of an ugly copy of FlightGear, in a few minutes time. Lots of Windows users are used to work that way because of the continual reinstallations of their system and the need to backup their software. An MS-DOS script can do the job to bundle such a ZIP file automatically. It can even send the zipped pre-version automatically on an FTP server. As time goes on and more people get involved I hope the install file of pre-versions will get cleaner and easier to install. One day it may become the standard way to install FlightGear. Easy to install pre-versions are necessary to attract more people to the Windows development but also simply because there are so many different Windows platforms. You will never get hold of enough Windows source compilers to test all possible configurations. You need to make binaries available. A good thing would be that Windows community becomes strong enough to offer technical choices in FlightGear to avoid compatibility problems in the Windows world. A second key is to become more offensive to request testers, whatever the kind of testers. I'm sure there exist persons who are capable to compile the source and who would love to contribute but they simply aren't aware of this opportunity. What about a banner on the FlightGear.org site? Eric

_______________________________________________
Flightgear-users mailing list
Flightgear-users@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-users
2f585eeea02e2c79d7b1d8c4963bae2d