Overall what makes linux so good compared to M.S. and what are the bad sides of having linux as a O.S.? Added that I have been running Red Hat linux and Windows ME on different computers. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Ignore me if I sound Dumb!

Linux pro points:

Excellent operating system
YOUR operating system, not MS.
Open source
Free
Choice of tools
Quality
etc.
etc.



Linux neg points:

Unavailability of commercial games



The only thing that is preventing me from exclusively running linux, is the fact that commercial games are not available on the whole for linux. Linux can do everything I need it to do, with the exception of playing commercial games (with the exception of WINE, but I mean it cannot play them without wine).

I personally use Gentoo, which is not only a fantastic distribution, if you don''t mind compiling from source and configuring the system yourself, but it also has a brilliant community along with it.

Stu

On the plus side for Linux, Samba provides a cheap (see also: free) domain controller (Ironically enough, for Windows machines).

Driver support in linux irritates me, as well as many other people I am sure, but that is also due to the lack of commercial support for linux. Drivers tend to have to wait for the opensource community to develop and test them.

Also, if you are trying to make money, not to say it is not possible, but linux development at present is not the most financially sound option for commercial software development.

Gamedev''s AI Auto-Reply bot.

The Good:

The Bad:
Kernel Panics
Rebuilding from source
Limited games
No unified vision
"Just good enough"
Zealotry
Hypocrisy

Linux has its good and bad Points for an O.S.
Overall linux has problems and solutions compared to Microsofts problems and solutions.

But is linux good for Programming games?
or is it best to have both O.S. for options?

I would love to hear more opinions on which is best!

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ignore me if I sound Dumb!

Good:

Open Source: You can modify the source code if you need something you can''t get otherwise.

Free.

All the power of Unix philosophy without commercial requirements (see free).

Bad:

Open Source: Often missing the final polished touches that make a professional software package. (Read: not all, put your flamethrowers away).

Free. Driver support from hardware vendors is lacking due to any $ motivation for many companies.

All the power of Unix without consistent feel.

Linux is a great movement. I love the idealism of it, but I don''t know if it can withstand the long test of time. The industry is moving towards useability more than features now, most people only use their computers for a few basic functions and want to spend less time fussing with features and more time getting things done. Linux is horrible in this regard. People do amazing work building software, but often have to stop sort of doing the really hard part of polishing it. Plus, it often relies on work of others to build full functionality, and a consistency in usage falls behind getting it to work. What''s the old joke about a platypus being built by committee?

Most of the stuff people rave about Linux for can be found with more consistency in other vendors. Stability is just a Nix thing (BSD, Solaris, Linux). Flexibility and Power come from the Unix Philosophy of software design.

Still, I have to say again, I love Linux''s idealism. I hope it does work. I use it all the time. GPL isn''t my favorite thing though. I think the GPL was formed under the pressures of intellectual property it''s trying to fight, and so suffers from it a bit as well.

As for Nix vs Microsoft....

Nix is much easier to tweak as an admin, and so easier to secure. MS is improving, but it''s an uphill battle on their design and useability.

Microsoft is great for running user based networks. The tools are much simpler to use, easier to integrate (so long as you use MS software *Grin*) and also have a lot of support from different companies and MS.

Nix would really benefit from a strong Policy based system that ADS now employs, particularly if a vendor like RedHat could build a management system that could provide the strong measure of admin control, system monitoring, and security LDAP + WMI and User/Group Policies provide.

All the same functionality is available on Nix, but if someone were to come to you and say, "I want all this in 3 weeks" you will probably have to go with MS.

Still, with the way the market around the world is backlashing towards MS, expect some smart programmers to start building better tools that make managing Unix systems easier. (Read: Admin doesn''t have to create a set of custom tools for every environment).

I''m still waiting to see what the next tech wave will be. I think there is a huge push towards useability. Look at Longhorn, MS next release for XP. They''re pushing features that are mostly found in Apple OS X. I think you''ll see more integration between software packages and hardware devices. (Better camera, video, TV input etc) and more focus on users not having to understand how it works, just doing what they want.

Linux doesn''t go in that direction. Though I think Linux has a huge role to play in the growth of the embedded market. It''s open, customizeable nature makes it a perfect fit. As a result, you''ll see more software on Linux that works and feels like "it just works they way your parents want it to." At least I hope so.

Interim

Interim, I''ve found your analyses so far to be excellent. I do feel it necessary to point out some "clauses" here and there, though.

quote: Original post by Interim
Free. Driver support from hardware vendors is lacking due to any $ motivation for many companies.

Additional problems include the driver implementation model, which until recently required drivers to exist within the kernel source directory (thank God we have loadable modules or we''d be completely screwed). Another restriction is that a basic driver category interface must exist for a device driver to be usable with the kernel or the kernel must be rebuilt.

Forunately, no more. The 2.6 kernel driver core model replaces virtually all of this with a much more modern approach. All devices fall into classes, and all basic classes already have their kernel hooks. (If new classes are created then the kernel will need to be recompiled, but let''s take this one step at a time ) There''s also a new filesystem, /sys that hierarchically represents all of the system features - /sys/driver for all drivers, /sys/device/class for all device classes and so forth. There''s also a project in the works to allow driver components reside anywhere in the user filesystem (so if you''re the only user on your system who requires a Logitech Wingman driver, or something more appropriately esoteric, it can reside in your home directory). These developments in tandem could help to alleviate the driver problem.

Of course, the fiscal motivation problem for device manufacturers and vendors still exist, but as more people use Linux and pressue vendors, the motivation will arise. Plus, the Linux crowd are happy to do the work for you, so long as you give them technical specifications.

quote: All the power of Unix without consistent feel.

Here I absolutely agree, except that I disagree that it''s an unqualified con. Unix is a command-line based environment. X Windows is really a separate system grafted on top of it, and the various window manager environments are more separate systems built on top of that. It''s a recipe for diversity, which some may interpret as anarchy. Personally, I want a system where the developer does not determine my widget set or the appearance of the applications, but merely the layout and functionality. No system in exist actually does that (since Windows and Mac do not present the opportunity of multiple WMs, the issue does not arise). The first individual or group to develop an accessible and sustainable system with this property will have my undying devotion.

quote: The industry is moving towards useability more than features now, most people only use their computers for a few basic functions and want to spend less time fussing with features and more time getting things done.

Again, I agree. Again, however, I need to point to the organic rather than "organized" nature of Linux development. Essentially, somebody needs to be bothered enough by this to either do the work himself or get others to do it. Until then, developers will simply continue to scratch their own itches.

I think the open and customizable nature of Linux, even compared to other Unices, is what will cause it to thrive in the long run. Somebody somewhere is bound to some day grow tired of the Windows-like paradigm of current Linux distributions and embark on constructing something different (okay, I''ll admit that I hope to be that somebody). The possibility of doing that doesn''t exist with anything other than Linux or one of the free BSDs, and the probability of it being done with Linux first is high because the real driving force of Linux is not technical but social. BSD development is organized and efficient, with seemingly little rancor or redundancy. Linux development is chaotic and anarchic, with huge personality-based spats (remember Reiser railing about reiserfs being sidelined in favor of ext3 because of its author?) and giant flame wars, extreme redundancy and massive overlap. But through a process of Darwinian evolution (peer review) and the power of numbers ("given enough eyes, all bugs are shallow") a pretty good system has emerged and will continue to emerge. It is for this reason above all that I adopt and recommend Linux.

Linux is not for everyone. It most certainly is not ready for mass market desktop use, and probably won''t be for a while (in fact, I predict that Linux will never beat Windows at its game and will only gain desktop presence when it offers something completely different from Windows).

The Good:

Customizeability. (You can use one of several window managers, and even with these, many have great theme support. That''s not even mentioning the customizeability of other applications or the many different flavors of Linux and UNIX out there).
Free. (Both as in speech and as in beer. Actually, I don''t really care so much that it''s "free as in speech", but it''s nice to be able to look at the source if you want.)
Stability. (Even compared to Windows XP Pro, my linux systems have always been more stable.)
Speed. (For me, linux boots and reboots faster than Windows. Many programs also run faster, but under WinXP, there is much less disparity between the OSes in this sense).
I could go on, but these are the big points.

The Bad:

Driver support. (Granted, I don''t have too many problems, but there are a few devices which I don''t have great times with. Generally, though, the more common the hardware, the better the drivers)
X. (I just plain don''t like it. It''s slow. However, starting with the 4.0 release, they have imporved it greatly.)
I could also add "not newbie friendly", but there are many easy-to-use distros out there like SuSE and Mandrake. But they are still generally harder to use than Windows is, just because of the learning curve if nothing else.

Yeah, I can''t disagree with you disagreeing with me. =) I sort of wrote that with a little play on words that don''t fit the arguments in full.

I think the driver issue isn''t as bad as most Linux opponents point out. It''s really just manufacturers doing numbers, "I got 98% on Win32, 2% on Linux. We got 2 months to market, just do the Win32 drivers." I know this happens with my company with Linux projects we push for. It comes down to time and what is the best way to hit as many customers as once. Linux just doesn''t have the market numbers yet to drive the business cases.

Still, we love Linux for that, right? You chaos example is perfect. I know I love it, particularly since I consistently learn more about computers with Unix/Linux than I do with Win32 platforms.

Problem with the organic vs unorganized is the lack of application interfaces. Linux/Unix is great on the CLI in this regard. It had a simple philosophy of specific tools chaining together. I don''t know if it''s the complexity of the GUI, but it seems to fall apart there. Mac is great in this regard, Windows is ok. Mac incorporates things like Spell check into just about every application since it''s available as a "freebie" in their application framework objects.

MS has Longhorn coming where they focus on this sort of concept as well, better organization of information and useability. I think Linux''s nature fights this since it doesn''t have central direction. Again, nothing wrong with this, but people are trying to push into the desktop market with Linux and missing this growing trend.

I do see Linux being a place in this trend, but not as a desktop, but as the OS on your PDA, your phone, or as the OS running your toaster and microwave. It''s organic, compact nature is excellent.

I think you''re right on about a new paradigm. I think people are starting to realize I don''t need a laptop that can do it all since I spend most of my day reading email, using IM, and writing word docs. In the past, I had to have a top of the line computer for games and email. Now, many are more likely to just use their same old computer for email and just get an XBox or a PlayStation2. Many of used to have a Phone and a PDA. Now we just get a Phone that has a built in PDA.

I think the idea of the perfect desktop might be nearing an end (nearing as in making technology more modular instead of central with a computer). Once you pull games off the PC, Email (now on your PDA), IM (also on your phone), etc, some people may just need something simple, or even just use something like ASP services.

I don''t know if this will hold, but it seems logical as a next step. Linux is great in this regard, being so modular and organic as you noted. Problem is, it built by committee, which is good, you get a solid design. I think I used the idea of a the joke "the platypus was built by commitee". It''s a very functional animal, but man, it sure is one wierd looking thing =)

But one place where Linux can take over regardless is in the back-closet of IT departments where this doesn''t matter to most users. Each Linux install does a task or two, does it well. But it binds with the rest of the technology with protocols and the internet for the most part.

If you want some weird service to interface with another, odds are someone in the Linux community has provided a unique solution, if not, do it yourself.

But on the desktop, you don''t need organic or dynamic, you need consistent and seamless. I think you hit it on the head when you said "The first individual or group to develop an accessible and sustainable system with this property will have my undying devotion." But, they have to find a system to merge all this dynamic work into an organized, useable paradigm.

PS Thanks for the 2.6 info. To be honest, I haven''t been watching upcoming Kernel changes like I did in the past and this is good news to me. =) Also good to have a non-flame war discussion of Linux pros and cons. Often hard to talk about it''s weakness and strengths without ruffling feathers, and it helps to point out the weaknesses to fix them.

Interim