Slashdot is powered by your submissions, so send in your scoop


Forgot your password?
User Journal

Journal Assmasher's Journal: Linux's biggest problem need no longer be a problem

I have to say that everytime I log into my Slackware or MDK 9.2 systems, I can't help but feel cleaner, more streamlined. Almost as if the sedentary developer's tire I've been growing for the past 5 years seems to disappear (somewhat.)

I've been working with Win32 and various *nix flavors since the early nineties (yeah, back in the Win32 'S' days, and early NT, lol) and I've always felt that way when using a *nix machine.

Cleaner, somehow healthier, as if I know that there is NO extraneous crap going on in the background, I can control everything if I wish and nothing gets installed that I don't wish to be installed.

I'm beating around the bush of the subject I really wish to diarize, but I just want to convey the actual 'feeling' of using Linux for me. Now, unlike many Slashdotters, I don't get hives when I run Win2K or XPPro; however, it just isn't the same. I feel like I'm always watching out for things when I get mail, or visit sites, or pulling things off the web to view offline. I just seem to always be looking over my shoulder and checking my system for spyware, trojans, making sure my AV software is up to date (even though it is set to 'automatic updating' I still feel I have to watch it.)

Basically this means I, later than many others, feel that Linux is TRULY ready to make the foray into desktop dominance. Server side? Not an issue, it is already there. The only reason M$ has ANY server installations is because companies haven't abandoned the M$ desktop yet. Believe me, there'll never be an IT department where the desktop machines run Linux and the IT servers run M$ OSes! LOL.

This brings me to the sticking point. What is it that keeps Linux from absolutely demolishing Windows XP (et al) right now?

Interfaces? No way. They're all pretty close to the same now.

Development tools? Not anymore. While we don't have an exact counterpart to VB (well, we sort of do), C/C++ development is trivial on Linux now.

Multimedia support? No way.

Stability? LOL! No way.

Security? While we're not as secure as we all like to pretend we are, we're still parsecs ahead of M$.

Ease of use? Not any more, with the exception of 'latest and greatest' hardware which is not a reflection on Linux but the marketplace. By this I mean that if we resolve the issues I do think we still have, this problem goes away.

Available quality software? Somewhat, many of the most important areas are covered by Linux, but there is still a gap between what kind of quality you can get on XP as opposed to Linux.

Now we get to the three things I believe are holding Linux back from replacing Microsoft's dominance of the desktop.

(3)Closed Source Stigma

Number 1, our dependency issues. I'm sure everyone has run into this one. "I can't have both programs installed and usable at the same time because their dependencies conflict." It is wonderful that there are a myriad of distributions, and it is wonderful that most of them include a default installation of development tools (unless you specify not to); however, we cannot in all seriousness consider our dependency problems to be a non-issue because users can just 'build the software' themselves. This causes havoc for numerous reasons. (A)People cannot simply rely on downloading and installing software (although this is often successful) through urpmi or other package management systems because there often need to be different packages for different distributions and also it is shockingly common to find to applications which cannot run at the same time because they require different versions of (for example) glibc. (B)Relying on our current model for dependency resolution (building it yourself) keeps commercial closed source products (such as games) from seeing linux as an attractive deployment environment. They sure as hell don't want to expose the source code for 5 million dollars (a normal dev cost for a modern game) worth of software to the world for free...

(2)Games, games, and games. Games are CRITICAL to wide acceptance of Linux. Most kids through their teenage years, that's all they care about (plus Instant Messaging.) Dependencies are a concern for games, but shouldn't be a serious one as most games can solve this through static linking; however, meeting driver requirements (such as simply having hardware acceleration of OpenGL) is still a potentially tricky issue for the vast majority of linux users. This is an issue mostly (imho) on the side of the video card companies (i.e. the install process for ATI's drivers to allow for DRI support) and somewhat on the kernel/driver side (as in 'perhaps there are some changes to facilitate making this more idiot proof for companies like ATI/nVidia.)

(3)The stigma automatically assigned to closed source products on Linux. Now Linux and OSS go hand in hand; however, the 'religious' nature of the many open source zealots actively discourages smaller companies, who don't know if they do or do not wish to support linux, from feeling a level of comfort and community acceptance. MAKE ANY SOFTWARE COMPANY WELCOME IN THE LINUX COMMUNITY, even (don't laugh [too hard]) Microsoft. We may wish for Utopia; however, many companies who don't have the deep pockets (nor political motivations) of IBM are not willing to give away software that they've paid their developers to produce. Don't shun them for this. Don't disparage them for this. THANK THEM for making software for us.

I really think that with the egregiously cheap costs of massive amounts of memory and HDD space available nowadays, static linking as much as possible would remove many of the problems associated with dependencies which plague our fine operating system.

Yes that makes upgrading more of a chore in the sense that each application would need to be upgraded/patched; however, we (now) tend to break more applications through upgrading an *.so than we fix.

Let's do what we can to make Linux distributions a level gaming platform. In other words, (not that I've got the cojones to do this) we should have our own 'DirectX' of sorts. While we have systems that handle parts of these, I would argue that these should be Linux ONLY tools for the same reason that DirectX is M$ ONLY. It is a key component in the war to bring games to an OS.

Lastly, let's cut back on the rhetoric about 'evil empires', 'Windoze people are stupid', 'M$ is evil', 'Open Source is the only way', and simply take the high road in all instances. Why? Because we KNOW Linux is better. It doesn't matter if some WinTroll comes and says "Linux is sh*t" and then we retort with an educated yet vitriolic assasination of everything and everyone windows because we've likely scared some people off about ourselves in doing so. Let's do some "turn the other cheek" reasoning. I've lost count of the times I've talked to another techie who was looking into deploying Linux and told me that they found the ngs and people disturbing and disparaging, even rude.

My $0.000002 for today

This discussion has been archived. No new comments can be posted.

Linux's biggest problem need no longer be a problem

Comments Filter:

"I will make no bargains with terrorist hardware." -- Peter da Silva