Linux, from a Windows users

[Skeith]

I use a mix of Windows and Linux. I dual boot Gentoo and Windows on my main machine for games, and I use Kubuntu otherwise for computers I don't spend a lot of time on. I appreciate the value of Linux and open source in general, but I'm not going to shred my XP CD's anytime soon. I am not a very advanced user, but I find open source software very interesting and read up on many projects.

My biggest problem is there seems to be so much effort working towards the same goal, but on different projects. Separate projects that would benefit from collaboration most often don't and instead form a competition with one another. I love competition, but when you see a project that has an awesome Feature A, while a similar project boasts a great Feature B you start to wonder. I like the idea of larger projects like Telepathy. Everyone works together on the backend features and people can write GUI's for them using qt/gtk, but everybody works together and benefits. The free software community has enough coders and testers to create products that put any closed source solution to shame. But any number of problems fragments it into sometimes half a dozen projects with almost identical goals.

I'd also to say Linux really shocked me when I found the terminal to be the best way to get things done. Master a few commands and its smooth sailing. Gentoo gets an unfair rap for being difficult, but its main difference is emerge replaces apt-get.

[Teg_Navanis]

You're making a good point, one that has also been made the main theme in Shuttleworth's keynote speech at the Linux Foundation Collaboration Summit today:

http://www.linux-watch.com/news/NS8470376604.html

[zeeR]

That's an excellent point, something that if changed, would make a huge difference on the quality of open source applications. Why don't coders of Banshee, Exaile, Listen, Quod Libet, XMMS, BMP (there are so many players...) etc., join forces and make a great, powerful media player? It's good to have freedom of choice, sure...but you have so many choice, that sometimes you need several applications because each has its good things and bad things, and damn, it's open source, why does nobody tries to join these apps into a big one? Five media players can be good, two main media players with features from all the others can be GREAT! (I'm just giving an example, of course )

[arvevans]

In the early days of UNIX & POSIX, thinking was focused on making small self-contained applications that ran very well, and very fast and did small tasks. By using the "terminal" or shell scripting one could link these simple applications together to form larger program entities for doing effective work with limited computing resources (computers were small and slow in that era). Some remnants of that mindset is still present in today's Linux systems, as you can see when using the command line interface.

Now that processing platforms have become faster and more reliable it is possible to make larger applications that include various user interfaces (GUI's video, audio, etc.). The folk at Redmond chose to utilize this increased capability by trying to make large subsets if common-use processes that are tied together with wrapper packages to form user applications. This worked to some extent, but eventually the inclusion of large sets of helper subroutines to do small tasks made the software large, unwieldy, hard to maintain, and susceptible to computer virus problems.

While there are Linux applications that follow the Redmond method, many other applications are developed pretty much as stand-alone efforts. This has several advantages (1) the applications are easier to port to many operating systems...including the Redmond one, (2) debugging and maintenance of code is easier because the developer owns all the code used in the application, and (3) there is a larger sense of accomplishment and pride from writing one's own application from A to Z without relying on or being confined by the work of others.

Is Linux the best way, probably not. Is it the only way, definitely not. Is it effective, definitely yes. Each code development method has it's own pros and cons, and it's own group of avid supporters. With Linux you have the advantage of using any of several applications that may accomplish much the same task, but you are not forced to use any particular one of those applications. You are free to choose. That seems to be the crux of the Linux advantage...you can choose. You can even choose to not use Linux, or to use it only part of the time.
_._

[Tux Aubrey]

arvevans@earthlink.net has a good point. Diversity can look messy and many of us do like the neatness that comes from standardisation and lack of choice.

On the other hand, I actually prefer this diversity (and redundancy) and believe it is better for even slight variations to be tested in the user market-place and undergo rigorous testing rather than have a board of "experts" decide what is "good" before it sees the light of day. There is plenty of evidence that the "board of experts" approach is not necessarily (or ever) dedicated to the user's needs. The GNU/Linux way is much closer to an organic evolutionary model that the Redmond/Apple way. Variations and mutations get a thorough work-out and the fittest survive - rather than have users adapt to an environment determined by corporate bureaucrats. Messy?, yes; Difficult to dive into? - Absolutely! but more likely to result in a ongoing evolution of a superior, robust and genuinely better system. (IMHO)

But of course, this model depends on having lots of users with a great diversity of needs willing to live in such a world when there is a more comfortable alternative. Mark S and the OP are obviously keen to see a little more standardisation in pursuit of a bigger user-base and I guess we do need to balance the boat a little in pursuit of that goal.

[blah blah blah]

My biggest problem is there seems to be so much effort working towards the same goal, but on different projects. Separate projects that would benefit from collaboration most often don't and instead form a competition with one another. I love competition, but when you see a project that has an awesome Feature A, while a similar project boasts a great Feature B you start to wonder. I like the idea of larger projects like Telepathy. Everyone works together on the backend features and people can write GUI's for them using qt/gtk, but everybody works together and benefits. The free software community has enough coders and testers to create products that put any closed source solution to shame. But any number of problems fragments it into sometimes half a dozen projects with almost identical goals.

well duh

[Tundro Walker]

OMG! You use Gentoo!? Like no way!? Get out of our forums! You use a completely different OS!

(someone whispers in my ear) ...Linux...really...? (whisper, whisper)

Oh, well then in that case, you can stay.

(Hehe, you know why Linux isn't more popular? Because it's so sub-divided amongst different distro's. That's both a boon and a bane, because it allows greater evolutionary paths, but fragments the user base so they clash with each other sometimes and might not support one another. It's trippy.)