This is for all the deep geeks out there. Hey look more original content!
Keep in mind this is certainly biased and purely my own opinion, but I thought I would share my beliefs on where Linux is and is not fitting into the business world. Either way ya’ll can chime in or tell me where to go!
Let’s start out with some of the benefits of Linux:
- Typically it’s a pretty rock solid platform that relies on proven technology and engineering know how.
- Definitely up and coming. A lot of younger people and seasoned professionals have a strong interest in Linux which will only help to bring it into the main stream.
- From a programming prospective it is extremely flexible in that someone that knows what they are doing can rip the kernel apart and make it do exactly what it needs to do.
- Availability of robust, free applications has made it an ideal platform for application development.
Draw backs to the platform.
- The prize of what will eventually become the “standard linux platform†is huge and right now there are simply too many flavors of it floating around. Before Linux hits the big time there needs to be some big time consolidation of resources and know how.
- Lack of documentation and version control. There is not a single Linux flavor that even comes close to the level of documentation and version control that the big boys have (i.e. Microsoft and Apple).
- Product maturity. Yes, I know, Linux born from Unix is old, in fact a lot older than any Microsoft Platform, but it simply has not gone through the paces of an OS that a few billion people have tried to break.
- Lack of professionally trained support and engineering folks. I’m not saying that they are not out there but their numbers are not nearly as large as those who support a main stream OS.
- The reliance on browser based front ends is a mistake. An application that functions within an existing GUI shell performs so much better than some of the goofy and awkward web based apps out there. It was a good idea in 1999, but I think those days have passed.
Where Linux is going.
- A truly dominant player will emerge in the next few years. Maybe Red Hat, IBM, Sun, or Novell. Who knows, might be some kid messing around in a basement right now that will come out with the winner. Either way unless this infighting comes to a stop and a clear leader emerges the OS is going to continue to trail the pack.
- Companies will look to license Linux kernels on which they can build a robust, controlled platform for in house developed applications. This is already starting to happen and where the big money for Linux is.
- Backend systems. More and more I think you will see applications relying on a Linux backend that has a user friendly Windows based interface. Again there are examples of this in the market place but this is the ticket to the big show.
Where Linux is not going:
- Linux is not going to conquer the desktop world anytime soon or within the next 5-10 years. The reason why has nothing to do with how good the operating system is or how bad the competition is. This is a matter of training, application availability and the number of capable support folks out there. It’s just not going to happen. The business world has invested heavily in Microsoft based systems and despite the rhetoric are quite pleased with how those systems perform. This is all about economies of scale. For the foreseeable future a large Linux based desktop system does not come anywhere close to being as cost effective or well managed as a Microsoft based environment.
- Linux has largely grown up in an open source community and while this is not going to change any time soon. Sooner than later people will realize the value of their own code when others starting getting rich off it. This is totally going to change the dynamics of the platform. While this has brought much attention and interest, ultimately people will want to start getting paid. I’m not entirely sure where this will lead but something is going to change whether good or bad.
: End Rant (hopefully a good discussion to follow)
I will comment. I promise, and as I think I am one of the three resident linux users, and probably the one who has spent the most time infront of linux recently, I have definate oppinions about your comments. However, I am going to let them “stew” so that I don’t bring forth the instant criticisms of your statements. Tomorrow or this weekend I will give an informed response.
I am a relatively new user to Linux and would appreciate your comments. Unfortunately, right or wrong a lot of people in my circles share these opinions so if you can change my viewpoint who knows!
So my analytical side began thinking about each point listed and began checking off the ones I agreed with and the ones I disagreed with. Then answering the benefits/drawbacks etc. topics listed. It was the beginning of an epic, which many people on/off line debate til they are blue in the face. Finally a response that represents how I think was born.
At the end of the day its about solutions. Money solutions, time solutions, tech admin solutions, client usage solutions. I believe there is no all encompassing OS solution for everyone and every situation. For some Linux is affordable and easy, however a MCSE, Novell admin, or a Apple admin may not agree. Further more the opposite may be true for others. A couple examples of products with different types of users and strengths come to mind.
OpenOffice or Microsoft Office
Tivo or MythTV
Many tools with many solutions, choose your path.
First off, I must preface this with the fact that I am a Linux user. I had my first foray into Linux back in the early 90′s, and I have picked it up and put it down many times over the last decade. So, I know full well where its been, and I can see where its going. That said, let me say that I think some of the statements that you make in your discussion are right on point. However, I think that some are irrelevant, and some are just wrong. I think that your synopsis of the benefits of Linux is pretty good, however, you are missing the fact that Linux is not what people want to give you, it is what you want it to be. And, I think that is the basis that will show how a latter statement is wrong. Well, not wrong, but misguided. When you talk about there being too many linux distributions and this being a bad thing, I think that is an uninformed statement. You also talk about how one clear leader will come out. I think that has already happened, although there are 3 clear leaders. Redhat, Novell (Suse), and Debian are definately the three largest distributions. And they are that way for a reason, and anybody new to Linux should try one of those first. When I started in Linux, I tried Slackware, and I wouldn’t suggest that on anyone who hasn’t had over 3 years experience with Linux. I understand how the wealth of distributions (>300+) can be a problem for the people who are just considering the switch, but those people should be more drawn to the names they know. And, I for one would champion anyone thinking about switching to try out Ubuntu as it is by far one of the easiest distrobutions to get into, and its live CD is a great way to get to know the system with no irrevocable changes. However, the other distrobutions get into very tightly defined functions, such as EpiOS, which is a whole Linux variation derived around the Via Epia Motherboards. Like I said, not for everyone, but for someone who knows enough to specifically buy an Epia Motherboard, it could be a great choice. Also, there are many distributions that are targeted specifically at the embeded market, as chances are very good that you have at least one, if not several linux applications running an embeded system in your house, and don’t even know it. When was the last time you checked the system software in your TiVo or in your CableModem, Cell Phone, or your Router? Chances are very good that something is running Linux, and you don’t even know it.
Moving from that argument to the argument about Linux becoming Closed Source, or Proprietary software, it will never happen. Very simply, the GPL forbids it. And, if it were to do it, it would inherit the same problems that every other Closed Source Operating System has. With Linux, the community can come together to diagnose and fix problems in the code. Take for instance the quick response that was seen when Coverity released their findings of over 900 flaws in 32 open source projects, and they were all fixed within two weeks. This level of response has never been seen by any closed source system. And the reason is that the resources aren’t there. With open source, the code is out there so that anybody can check it, verify its integrity, and repair it. You know that Linux will never “Phone home” without telling you because if it were to do that, somebody would find it in the code and point it out. That option is not available in a Closed Source system. Now, don’t take my comments the wrong way and assume that I don’t think developers should be paid for their work. Rather the opposite, I think developers should be compensated, and should be compensated well. And by producing varifiably good code, they will be. However, I think the business model of Linux is better. The business model used by Redhat is a great one in my oppinion. They have a product that they sell, at a minimal price, but the real bulk of their income comes from supporting that product. And, their symbiotic relationship with the Fedora product is interesting, because they have offloaded their R&D costs to the community, who gets minimal support for their efforts, but is backed up by the large corporation. By going with Fedora, you arguably get the bleeding edge products, in a potentially unstable system, but there is no cost involved. If you want support and proven stability at the expense of the bleeding edge features, then the appropriately priced Redhat System is for you. The biggest detriment to Open Source Software is the way that the US Patent Office hands out patents for software, and the way that it stiffles innovation. By leveraging the patent office, big business can keep the smaller open source developers from actually coming out with new code, which is one of the reasons that Corporations like IBM, Redhat, and Novell are essential to the success of Linux. It is their power in the world of patents and protecting Open Source software that will accellerate it beyond what is available through other providers.
I will conceed that the situation of documentation is a place that severly needs help in the Linux world. However, I will draw version control away from that argument. I feel that the version control in the Linux/Open Source world is far superior to anything from Microsoft/Apple, and mainly because applications can usually be upgraded from version to version, with very little effort. In fact, on my Ubuntu system, I meerly have to do a single terminal command (accessible through the GUI if you want) to upgrade every piece of software on my system. “sudo apt-get upgrade”. And, that is benefitted by the fact that the Linux world has so much software that its very difficult to find a task for which a piece of software doesn’t exist. The big three distributions have repositories of software, usually precompiled for the common systems (x86), ready to be installed over the net for free with just a few mouse clicks. That is far superior to the idea of going to a store and spending tons of money to get software. I am not against paying for software, but I would love to see some sort of repository based internet install come around for Microsoft/Apple. With the permeation of Broadband, I think that would really help any system.
I will just disregard the comment about “reliance on browser based front ends” because I don’t think that is in any way a Linux issue. Maybe clarification would help, but I think that is a problem / benefit of any system. I am considering GMail as a viable example of the way it can be good. But, I also see Windows Explorer as a bad implementation of the same thing but tethered to the local machine. And, for what its worth, I don’t think that any system has ever come close to matching the excelent column view found in OS X. When that gets added into Gnome or KDE, then there will be a real winner there. Which I should comment on. In the multititude of Distributions, there are also a variety of Window Managers, with the two prevelant ones being Gnome and KDE. There are also more lightweight ones which are useful for older machines. I do think your idea of merging is a good idea here. I would like to see both Gnome and KDE fork, with their respective offshoots merging into one Window Manager that would be able to present itself to the new user. And, as the WM is the first, and more than likely only thing that a new user will really see, this helps to make any of the Linux Distrobutions differences more transparent after the switch.
In the next 5-10 years, I think that Linux will make significant growth into the desktop market, more and more as people like my mother begin to switch to Linux. I think that Microsoft’s bloating OS’s and unrealistic minimum system requirements are going to help drive this to some degree. And as it moves into the home sector, people will become more willing to accept it at work. I think that it would be a stretch to say that the majority of the market share will move, but I think that advances in Open Document formats is helping, and the continued expansion of government utilization of Open Source Software alternatives will help to expand the market. I also think that in the next few years, we will see another killer virus (not that I will have anything to do with it, nor do I advocate it at all), that will help to put Microsoft onto shakey ground. I think in this position, they will make significant advances in security, exceeding the changes in Vista, but I think that it will shake consumer trust enough to give other systems a look.
I have tried to answer most if not all of your comments, but, I am sure something probably slipped through. I hope I haven’t come accross as a “Linux-Fanboy”, but if I did, oh well. And, if I didn’t mention it, while I used to think that the abolishment of Microsoft was a good thing, I don’t think it is. Without the competition, I don’t think any system will become better. I do however wish that some of Microsoft’s practices would change, but I think the same thing for practices of Apple and Linux as well.
Where is Zogel’s input on this???
Thanks to 2487 for a thoughtful critique of Linux and to Mr Tex for you well thought out response. I agree with Mr O that it comes down to workable solutions. Obviously the solutions are many for the admin and growing for the desktop user. But for linux to make headway on the desktop there are a few basic needs.
1) A GUI comparable or perceived to be comparable to Windows or OS X.
2) A pre-configured PC sold at Circuit City or Best Buy or elsewhere
3) Marketing of pre-configured PCs
Without these OS Linux will stay in the server room.
Mergers are not the answer. The answer comes in standards implemented by many companies. As Mr O says “… there is no all encompassing OS solution for everyone and every situation.” As an example, the Microsoft monopoly has homogenized and hurt the tech industry, shutting down new technologies with their threats or with their buying of new and competitive technologies. We need an open industry, which Linux companies brings to the table, where a company competes on the merits of their products, not the force of their monopoly or threats by their lawyers.
[Quote]We need an open industry, which Linux companies brings to the table, where a company competes on the merits of their products, not the force of their monopoly or threats by their lawyers.[/Quote]
I think a sideproduct of that is the support for alternative architectures as well. I have been considering a Carputer as a free-time project for a while, and I have no doubt that I will run Linux on it, but the decision is really x86 or PPC? And unfortunately it seems more and more that the x86 architecture has stiffled support for other architectures. It was widely heralded as a good thing when Google released the Beta of Google Earth v4 for Linux. But, its a shell script which runs an installer, and is all packaged up in a binary format. They don’t say that you have to be running x86, but when you are running something else, you begin to understand to look for that i386 early on. Its unfortunate, because there is a lot of hardware that isn’t supported, but that little detail right there makes it easier for Windows Converts. But, any company which supports multiple platforms will definitely be more highly regarded. Even if they choose to use a crippled licensing version but to allow the source that can be built on platforms they don’t have access too, (Alpha, etc.)
As far as the GUI goes, I don’t see why that is an issue, There are two main DEs, which will work well for most anyone. This is my oppinion, but if you are more familiar with the Mac style of DE, then you will prefer Gnome, and if you are a Microsoft User, then you will more than likely prefer KDE. The benefit to the DE scheme that Linux has in place is that you can have any DE you want installed, and from the login screen (thats where users like my mom will get into trouble) you can choose whichever one you want, and the other ones won’t be affected by this choice.
I agree that a major retail chain needs to stock Linux based computers, but I just don’t think that is going to happen. I think that Macs have a hard enough time in that environment.
As a sidenote, I was looking at the top of the line Dell XPS Starfighter or whatever it is, and noticed a little link, definately not calling attention to itself on the bottom of the page, to “Build Your Own” or something like that. Basically you pay the same amount of money, and they just ship you the parts, and you assemble it yourself. Well, its the same amount, but there are pieces missing, which isn’t that cool, but what is cool is the only OS offered in that configuration is FreeDOS. A $0 OS that “will not utilize the hardware in the computer”. Seems like though they know people are going to pirate their MS OSs on it, they are making small steps towards allowing people like me not to have to buy Software I don’t want.