Don’t get me wrong, I love Linux. It’s everything I wanted from an OS. Speed, flexibility, games, eye candy, productivity, and powerful development tools. But there are certain things it lacks that remain as barriers to wider adoption.
The grand daddy of third party applications for Windows, Photoshop from Adobe, is the de facto standard for graphics editing. Designers use it to mock up websites, photographers use it to modify their artwork, it is the tool to end all tools for graphics manipulation.
But what about The Gimp? Yes, there is a very powerful tool out, in the form of the Gimp, arguably the most complete and useful tool to come out of the Gnome project. (In fact, much of The Gimp was used as the basis for Gnome, notably GTK). Gimp has gone through many updates and truly is the most polished of the available apps. In this, Gimp can be considered a useful replacement for Photoshop, incorporating many of the features that PS provides. But it isn’t Photoshop. Just as Photoshop’s interface can seem impossible to fathom to the regular user, Gimp’s toolset is just as complex. While they both have similar feature sets, the learning curve for both is quite steep, and there are far more Photoshop users than there are Gimp users – at least in the professional market.
Quicken and Quickbooks
In this, there can be no argument. There are no applications for Linux that have the depth of support and comprehensive abilities of Quicken and Quickbooks from Intuit. There are several attempts going on in the form of GnuCash and others, but none come close to the functionality level of these tools. Entire businesses base their very existence on these applications, not only as a desktop tool but an online resource for money management.
The Intuit applications are one example where a rewrite to mimic functionality would not be sufficient. There must be industry support for the file formats, the online services, and the support infrastructure behind the applications to make it a viable choice for a small business to make. As it stands now, that structure is not available to a Linux user. In financial applications, unlike desktop applications such as word processors and spreadsheets, years of historical data is frequently stored within the application. A user invests in the software and the company behind it. Very few opensource applications can provide the long term guarantee of support and service required by a personal and/or business financial application.
Love it, hate it, despise it, consider it the root of all evil; however you like to approach it, there is no denying that Microsoft Office is the lynchpin of Microsoft’s hold on the desktop market. The comprehensive software suite is the carrot by which any user will naturally migrate toward Windows as a desktop environment. “I’ll use any machine as long as I can email my word documents around and do group scheduling.” That simple statement unfortunately only has one answer, and that’s Microsoft Office, and from there, that means Outlook.
Because of this reality, there is zero chance Microsoft will every port Office to Linux. It is not a paying environment (unlike the Mac where users will happily pay Microsoft for the application), and supporting Linux as a desktop environment will only hurt Microsoft’s hold on the desktop. Also because of this, the Office problem is one that has gotten the most attention among opensource developers and, it must be said, commercial ventures.
On the opensource side, OpenOffice is the clear leader in Office-like applications. Spreadsheet, presentation manager, word processor, graphics editor – they’re all there, and to a large degree, they work fairly well, even with Microsoft documents – a fact that is not lost on the software giant. Microsoft knows that compatable tools will hurt their market position, which is one of the reasons they vehemently oppose the Open Document Format (ODF). Why pay a thousand dollars for an Office suite from Microsoft when an opensource version can work just fine on the same files?
Unfortunately, OpenOffice has some serious deficiencies, not the least of which is performance. Microsoft has had 15 years to tune and tweak Office, and has hundreds of developers whose job it is to make Office fast, stable, and useable. The OpenOffice developers have no such luxury, and the application suffers because of it. Tasks that can take a few second under Excel may take 3 minutes under OpenOffice Calc.
And then there’s the the Outlook problem. First, in terms of ‘pleasure to use’ – Outlook is a bad email client. Microsoft has attempted, in one application, to appeal to all users from the stay at home dad up through the Fortune 1000 installations, all within the same program. The result is a muddied, difficult to use, enormous application that, unfortunately, CAN do everything, it just does it all poorly. Most of the functionality in Outlook can be replicated in other email clients, and there are thousands available. The one thing that has never been reproduced to the level that Outlook provides is scheduling. The lack of an OpenSource, usable, shareable calendar system for Linux is the final nail in the ‘Not available for business’ sign across the Linux desktop. Outlook’s group scheduling functionality, when coupled with an Exchange server, is unparalleled. Yes, there are plugins for clients like Evolution that can talk to an exchange server, but that still requires the Exchange server. The net win is zero, and can be argued to be a loss because a Windows box is still required for Exchange, and Windows client is still required to configure and maintain the Exchange server. Where is the win?
What? Skype? Cmon, there already is Skype for Linux! Sure there is, but as anyone who has worked on the Windows version will tell you, the differences between the two versions are noticeable in one huge issue. The Linux version of Skype does not support video.
It could be argued that Video under Linux is hardly an advanced, well supported technology. But with the wide ranging availability of v4linux, and well documented support for various webcams, it’s inexcusable for Skype, two years after providing a Linux version of its software, has not provided a working video interface.
In the world of multiprotocol clients for Windows, Trillian has come out on top as the best supported, most frequently updated, and arguably most profitable package around. Supporting all of the major chat systems (Yahoo, MSN, ICQ, Jabber, and IRC), all within an easy to use interface, Trillian has become the default client for a sizeable portion of the community that wishes to not be wedded to only one IM system.
But again the cry goes out. What about GAIM^H^H^H^HPidgin? It does all these protocols just fine, why not use it? Friends, I encourage you to sit down with a Trillian user, and suggest to them to try GAIM. The interface is laughably primitive, so wrapped up in its Gnome roots that the developers are blind to how painful it is to use for someone not intimately familiar with Gnome already. I’ve tried on numerous occasions to use Gaim to try and unify my two primary communication mediums (IRC with X-Chat, and Jabber-Client-Du-Jour for others), and each time I’ve found myself chewing the furniture over the idiocy and outright painfulness of GAIM’s interface. Perhaps now, with the legal issues with AOL out of the way, there may be hope for updated versions, but I believe Cerulean Studios is missing out on a golden opportunity to get into the Linux market by not providing a version for Linux.
Object-based drawing and charting. Many companies revolve their businesses around Visio drawings. There’s really no decent alternative for Linux, and Visio has adamantly refused to allow compatable applications to be developed.
This is a generic category. Lets face it, compared with Windows, Linux is a poor gaming platform. Sure there are some ports for some great games, and the folks on the Wine project have made fantastic leaps in providing a runtime environment on a Linux host for playing the most popular games, but… they’re not native. Getting World of Warcraft to run under Linux can be a long frustrating experience, which begs the question “why bother?”
Certainly Microsoft has a hand in keeping the Windows platform viable for gaming, primarily in the form of DirectX, a licensed multimedia interface that many (actually, most) of the game developers adhere to. Windows provides the hardware interface to the devices, DirectX provides the API to the games. It would be great if DirectX were available for Linux, which would make porting the games easier. I don’t need to explain that pipe dream.
Many times the media has proclaimed “This is the year of Linux on the desktop!” and, to be fair, Linux has made great inroads here. Many governments are questioning the monolithic “single vendor” problem with Windows and are looking for alternatives. It’s a great challenge to these organizations, as there is little lure to Linux on the desktop other than throwing off the yoke of subservience to Microsoft, and until either these applications are made available or viable alternatives are around, Microsoft will never really be displaced as king of the desktop.