28 July 2003

In May of last year, I switched to Mozilla, the open source web browser and e-mail client. In June 2002, I followed this by installing version 1.0 of OpenOffice, the open source application suite that poses an increasing challenge to Microsoft’s Office programs. This was not a huge technological overhaul – I did not get a new computer, or need to transfer or convert massive numbers of files – but it was a significant first step towards a more liberated level of computing. And did I mention that both of these programs are free? As in, they don’t cost you anything?

The open source software movement, whose most famous product is probably the Linux operating system, is difficult to distill to a few sentences. The Open Source Initiative, a non-profit corporation that promotes the development and use of such software, defines it this way: “The basic idea behind open source is very simple: When programmers can read, redistribute, and modify the source code for a piece of software, the software evolves. People improve it, people adapt it, people fix bugs. And this can happen at a speed that, if one is used to the slow pace of conventional software development, seems astonishing.” This may sound naïve, but as a user, I can tell you that in my experience so far it is true; bugs in both the Mozilla and OpenOffice programs have been fixed much faster in the last year+ that I have been using these programs than similar problems in programs by (for instance) Microsoft.[1] Ultimately, what the open source movement offers (beyond a challenge to the monopoly players in the software world), is a level of innovation that most corporations say they desire but cannot really achieve – not when profits are substantively on the line.

A few product comparisons are worthwhile; let’s start with Microsoft’s Internet Explorer and Outlook programs. Internet Explorer, the subject of the famous browser battle with Netscape, is a solid, if plodding, program; it works, even if it does not always work well. Outlook, which combines e-mail functionality with a calendar and other management features, also works, but has been plagued by security problems since it was launched.[2] When I switched to Mozilla (which is the open source, next-generation of the discontinued Netscape browser), I found a variety of small, practical browser developments that Microsoft’s Internet Explorer has failed to address. Mozilla has an internal window-tab feature, allowing users to open multiple web browser pages within one overall window (see screen shot); in Explorer, you cannot do this – every new page must also be a new window. Moreover, Mozilla (as with other browsers) will comply with user specifications to open every new window or tab to a blank screen or a specified page. Open a new window in Internet Explorer if you have already started browsing ... and you get a copy of the first window; this is particularly annoying if your web browsing involves secure pages like bank sites, where multiple open pages can cause security problems. And searching a web page in Mozilla is easy: just start typing and the browser will search for your text; try that in Explorer and see what happens. (Nothing, is the answer.)

At a more detailed level, Mozilla will block let users block annoying pop-up windows; Explorer offers no similar feature. Mozilla will allow a user to block images from a particular web site, making it possible to eliminate advertisements (leaving only the space on the page); Explorer does not offer this. And in my highly-unscientific tests, Mozilla was faster loading everything from the Yahoo! Homepage to complex, scripted web sites such as banking or travel sites. On average, Explorer was a noticeable eight seconds slower over my home phone line, and sometimes as much as five seconds slower on my high-speed connection at work.

Mozilla’s e-mail program is equally practical. The latest version includes a junkmail filter that is both effective and easy to use: mail shows up and gets tagged with a junk mail icon; if the user wishes, Mozilla will automatically delete the message, with no further user action required; and users can automatically exempt any incoming message from someone whose address is in their e-mail address book. Microsoft’s Outlook, on the other hand, has a built in filtering system that (in my five years of testing in the office) hardly ever works; despite the fact that Outlook is supposed to “learn” which content should be considered junk, it rarely does so. Moreover, while Outlook allegedly offers the ability to delete these items automatically, it does not seem to work even when the function has clearly been activated; the most that ever happens is that the message turns a bland shade of gray, leaving the user to delete it.

Microsoft’s Office programs are harder to compare with their open source competitor – which is entirely due to the diligent work of the people behind OpenOffice. Microsoft’s programs function adequately and are reasonably stable; their dominance in the marketplace (and the workplace) is due not only to the company’s near-monopoly in operating systems (making their own productivity programs more attractive) but to the programs’ core quality. As much as I do not like many things about Microsoft products, I cannot deny that they have served me reasonably well over the last decade. But where OpenOffice is concerned, the distinctions are more difficult to enumerate because OpenOffice is so effective at providing the same functionality offered by other programs (including Corel’s WordPerfect, the other mass-market competitor in the field). This is even more true now than it was when I started testing the program last summer; the newest version of OpenOffice (the release candidate of version 1.1) addressed many of the flaws of the earlier 1.0 product, including a much-improved conversion process for Microsoft Word and Excel files.

What can I say about it? On my three-year-old PC running Windows 98, OpenOffice is faster and more responsive than Word or Excel, uses less RAM, and is better at telling me what it is doing. If I save a particularly large OpenOffice file (say, 150 pages), the program runs a progress bar across the bottom of the screen as it completes the save; in Word, there is only a little disk icon and egg timer cursor, but ... I have to guess how long it will take or where the software is in the process. Those ridiculous temp files that Microsoft creates every time users open a document – and which almost always linger if the program crashes, creating detritus on the harddrive? I have no idea how, but OpenOffice does not (apparently) use them – and seems more effective at file recovery than Microsoft, as well. (Yes, the program has crashed a few times; I am good at making things crash by pushing the limits of what my machines can do.) OpenOffice also treats all of its file formats as universally accessible; for example, opening a spreadsheet file while a word processing document is open does not require launching a new program. OpenOffice automatically launches its spreadsheet program, an intuitive feature Microsoft would do well to copy.

Microsoft is better at allowing customized toolbars; the new OpenOffice improves this process, but it is still not quite as good as Microsoft in this. However, it does mimic this functionality across all program functions (word processing, spreadsheet, presentation/slide show, and drawing), and offers more extensive customizing options than most Microsoft Office products (see screen shot). And OpenOffice’s files are smaller, too, because of the XML format used by the program. How much smaller? A 140-page Word file I have stores at 513kb, about one half of a full megabyte; this same file, in the OpenOffice format, saves at 167kb – a 67% decrease in file size.

For now, open source applications probably are not for everyone. While some companies, like Red Hat, have gone into business selling open source software like Linux, the key to purchasing an otherwise-free product is the knowledge that Red Hat will provide tech support and software updates as part of your purchase. Using OpenOffice and Mozilla there is no toll-free number to call for support, and no company to sue if something goes wrong. For now, I can get support from the team of developers and fans around the world who do their best to solve problems and create new functions, and who share that information willingly with other users.[3] A year ago, I would also have said that open source applications require a greater level of knowledge on the part of the end user, but that is no longer true. With the newest release of OpenOffice, installing was easier than before, and so was the system set-up – and all of the advanced functions are no less complicated than in any other, similar program (i.e. Word). Perhaps that sounds like faint praise, but I do not think so; these programs are complex for a reason: to meet the needs of users in the market.

My experiences thus far have given me some hope that the computing dominance of a few large corporations might be effectively challenged by the open source software community – and this is not a goal at which anyone should scoff. Faster development of program features and better responsiveness to bugs are both worthwhile goals for software development, but seem to be elusive for large, corporate developers which are more interested in preserving the copyright on their code than on improving that same code’s functionality. Which reminds me: with OpenOffice 1.1, I have begun using the program exclusively, and at home I even create or edit files that I share with colleagues at work – and so far, that process has been transparent. As far as they know, I am using Microsoft Word. I call that success.

Coming soon: My Open Source World, Part 2 – on the economics and politics of open source computing.