Now that Mac OS’s Leopard is out, I figured I’d make a long overdue post about the new operating system in my life.
No, it’s not a Mac (as if the title didn’t give that away already).
A few weeks ago, I installed Linux Mint, a Linux “flavor” based on the popular Linux distribution Ubuntu. I’ve had a few weeks to try it out and I have a number of positive impressions:
- Linux Mint works well. It’s not different enough from either Windows or Mac OS for any casual user (i.e. needs a word processor, web browser, and chat program and that’s it) to get confused. It is not sluggish, unlike Windows which runs at uneven paces for me. It’s very stable. It’s highly customizable. The eye candy engine (powered by Compiz Fusion which is demonstrated quite well in this video) works wonders, granting me my much-sought-after free clone of Mac OS’s expose feature and other great features without the hardware requirements needed for Vista’s eye candy. It is all together a well built operating system.
- Linux is not a free version of Windows or Mac OS. Nor should it be, but a lot of people have the misconception that Linux is a “free” version of Windows. That is neither possible (due to Microsoft making various things proprietary), nor desirable. Nobody uses a Mac thinking that it’s just an improved version of Windows — it’s a drastically different system with its own advantages and disadvantages and its own “core design philosophy.” The same applies for Linux. While many distributions do try to emulate Mac and Windows functionality, at the end of the day, the goal behind Linux is not to make a free clone of another operating system slightly improved, but to make something radically different and, hopefully, much better. Whether they succeed or not is up to debate, but it would be a mistake to characterize Linux as purely a “free Windows” or a “free Mac OS”.
- The success of Linux Mint and Ubuntu comes from their ability to make things work “out of the box”. This, to me, is the reason that distributions like Linux Mint and Ubuntu have been so successful. Ubuntu (Linux Mint’s “parent”) just worked. You could simply boot your computer with the LiveCD in the CD drive, and you could test the entire operating system without installing anything. Following a quick and painless installation, the system worked pretty well with your hardware devices, with playing media files, and with logging on to the internet. Linux Mint takes this usability to the next level — it is a natural dual-boot install candidate (something which many people, such as myself, do because they may need or want Windows for something else) as it has NTFS support built right in. Linux Mint also provides utilities to facilitate installing software, configuring wireless networking, and even setting up a control panel. A built in CUPS system makes it plug-and-play with a wide range of printers. Such quick operability makes Linux Mint an easy switch for people who don’t have the time or the desire to learn the ins and outs of making hardware work on a foreign operating system. And, if Linux people are serious about migrating people, this is the first thing they should focus their time on.
Of course, I also have a number of negative impressions, which I’ve divided into 5 things which are holding back Linux from being widely adopted:
- Linux lacks native support for some hardware. On my Sony VAIO, the operating system was unable to control screen brightness (thankfully, a programmer affiliated with Linux Mint programmed a utility to fix that). This is a minor complaint, but given that currently only OLPC, Dell, and a handful of other vendors ship Linux with their computers, it means that Linux has the unenviable position of needing to be readily compatible with as many standard hardware setups as possible. It may be all right to force tech nerds and gaming geeks to program their own drivers for their own super graphics cards, but not being able to support an entire brand’s (albeit stupidly designed) brightness settings and my USB headset is an example of problems which may reduce adoption.
- Linux lacks Microsoft Office. Everybody hates Microsoft Office to some degree or another. But nobody can deny (or if they can, they either have never worked in a business setting or are working for one of the various groups trying to replace Microsoft Office) that it’s a clear necessity on the computer given Microsoft’s dominance (Excel, Word, Powerpoint) in the document-space. Linux has a couple of open source “alternatives” such as AbiWord and OpenOffice, and while each program has its value, they are poor copies of Microsoft. They not only fail to duplicate many of the key functionality (case in point: I’ve never heard a banker rely on Gnumeric or OpenOffice’s Calc), but in their attempt to simultaneously emulate and surpass Microsoft’s product, they end up failing at both. If Linux programmers and open source advocates intend to be taken seriously, they must either create a product which can truly completely emulate Microsoft Office, or a radically better product which people will switch to of their own accord.
- Open source has its limits. I was fairly suspicious of Open Source as a model of development when I first heard of Linux in junior high school. That suspicion has died down somewhat as I began using a good deal of open source software (e.g. Firefox, Sunbird, Launchy, Pidgin, etc.) It became clear that Open Source was very good at some things — such as patching up bugs and security holes, developing many and powerful features — but it was very bad at other things. With the exception of “corporate open source” groups (like Mozilla and Canonical), regularity of updates in open source projects is weak. The number of projects which have started out strong and just sputtered out due to loss of interest is very high. Lack of a central corporate focus also means that the purpose of the application is oftentimes lost by a wave of programmers demanding that their latest and greatest feature be added, even if it adds little value to the average user and merely delays release and increases the complexity of the code. These extra additions also invariably increase the complexity of the program’s UI, creating the oft-heard complaint that open source programs are not user-friendly. Linux can sometimes suffer from this problem — too many features, too difficult to understand how to use.
- IRC/Forums are a horrible source of information for the average user. One benefit that Linux and Open Source supporters often tout is the strength of the community. However, because these supporters tend to be techies, by community they are referring to Internet relay chatrooms (IRC) or online forums. These forums are helpful for people who don’t mind getting into the nitty gritty of their operating systems, who know how to and are willing to re-install their programs or operating systems if necessary, and who have some technical background. They are not helpful to someone who is an average user, who has never used the command line, has no idea what a script is, etc. As a result, users who switch to Linux may stop using it, or may not want to switch at all as a result. This underscores the importance of making things work “out of the box” — frankly, the less Linux tech support that the user needs to deal with, the happier they will be.
- Open Source fanatics. The “core philosophy” of Linux is one pertaining to “software freedom” — not simply freedom from having to pay for software, but being able to completely change and customize everything. This is a noble goal, in my mind. It adds a great deal of complexity and technical difficulty to the software and its usage, but I think it’s well worth it. However, some individuals have taken this philosophy to its extreme, to the point where its not simply about promoting “software freedom”, but to the point where any element which is not perfectly adherent to this philosophy is undesirable. As a result, many Linux distributions do not ship with basic applications which are now indispensable, such as Adobe’s Flash. This philosophical purity is absurd; if anything, preventing someone from using non-free software is itself restricting software freedom. Moreover, the grand truth behind most software users is that they don’t care about commitment to “software freedom” — the average user will not go out of his or her way to re-program any aspect of the operating system in an attempt to tweak it. The average user cares about using the operating system more than some abstract philosophy. If the Linux community intends to expand its market share, it will have to face this horrible truth.
So, all in all — Linux Mint and the open source community in general have produced remarkable products with great potential. But they have some serious design philosophy issues that they need to sort out if they intend to expand beyond the niche userbase that they currently have.
