May 6, 2006

What's in a name?

Author: Joe 'Zonker' Brockmeier

The Linux community is all about standardization this year. The Linux Standard Base is gathering steam, the Portland project is trying to bring unity to the desktops, and the Open Document Format (ODF) has been in the news quite a bit. Now if we could just get projects to agree on standard usage for naming, I'd be a happy camper.

I've been writing about Linux for about seven years, and while the technology is a never-ending source of joy for me, the nomenclature can be a writer and copy editor's nightmare.

As an editor and writer, I try to stick with the usage dictated by a project, but that's often difficult. Many projects don't agree on their usage on their own Web sites, documentation, and mailing lists. For example, on the Xpdf site, you see Xpdf and xpdf used interchangeably. On the KPDF site, KPDF and kpdf are also used interchangeably. Even though MySQL AB has managed consistent usage of MySQL, I frequently see articles that use Mysql and mySQL.

I could go on, but finding inconsistent usage of nonstandard names for software is not a challenge -- finding consistent usage is. The solution is not to rigidly enforce the usage of nonstandard names, but to encourage the adoption of names that work without requiring anyone to sit with a style guide in their lap while working.

I'm not the only person who has noticed this. While I was working on this commentary, I noticed this discussion between GNOME developers about the "proper" usage of GNOME.

GNOME used to be an acronym for GNU Network Object Model Environment, but that fell by the wayside some time ago. The GNOME site is consistent in its use of "GNOME," but a lot of GNOME-related projects use Gnome in their names instead of GNOME, and few developers resort to the all-caps version when writing about GNOME. Jeff Waugh writes:

If we're incapable of getting it right within the project, then it's highly unlikely that anyone outside the project is going to get it right either. I am asking for consistency, and at the moment our project is called 'GNOME'.

Consistency is a hard thing to come by, and when a word or name employs nonstandard usage, the odds are that it will be harder to enforce consistency. When editing articles for our sites, I spend a fair amount of time going through them to ensure that writers use Perl when talking about Perl as a project, but perl when they're referring to the executable, and GIMP instead of Gimp, and SUSE instead of the previous iterations of SuSE and S.u.S.E.

If you look around, you'll notice that some sites have yet to update their style guides to reflect the change from SuSE to SUSE, though it's been quite some time since Novell made the change when it purchased the company.

To make matters worse, many projects employ the dreaded StudlyCaps for their products and standards. We have PostScript, KPovModeler, AbiWord, and so forth. And don't even get me started on LaTeX, which is not only insidiously spelled, but also has its own special pronunciation. "Never mind the fact that it's spelled latex, it's pronounced 'lah-tech.'"

Even when a project is dubbed MyCapsLockIsTeHBroken, it doesn't mean that the developers will bother to actually employ that usage consistently throughout their own documentation or Web site. Few open source projects actually have or publish a style guide, so that leaves an editor to puzzle out what the "proper" usage is.

Even worse than StudlyCaps is practice of capitalizing the second, or later, letter of a word, leaving editors to ponder over whether to apply standard English usage or abide by a company or project's usage. For example, should one write "EBay," "eBay," or "Ebay" at the beginning of a sentence? Am I supposed to write "AmaroK" or just "amaroK"? How about gThumb, aRts, and so forth?

How about those names?

While we're on the topic, how about those project names? I realize that open source developers are not overly concerned with marketing -- which is just as well, most of the time -- but we might see open source applications taken more seriously if they were named a little more professionally. Just as some people claim that the "sandal and ponytail set" hurt Linux's image with corporate types, silly names also project an unprofessional image to some.

But, you say, that's not logical! Why would people pass up a perfectly good program just because it has a silly name? First of all, if you think logic always plays a strong role in software adoption, you haven't been paying attention. First impressions mean a lot, whether we like it or not.

Secondly, it would help a great deal if a project name were descriptive of what a program did. I think the GIMP is a great program -- but I've never failed to get the hairy eyeball when I've asked non-geeks if they'd consider trying the GIMP as a Photoshop replacement. Let's face it -- image editing is not the first thing that people think of when they hear GIMP.

Again, I'll refer to Waugh's comments on the usage for GNOME:

I will strongly oppose any suggestion that we move to 'Gnome', because then there will be nothing in between our name and the name of short bearded men who wear red caps and gather in gardens (or interpretations that are worse). That is *not* a good message, it is not a good association, and by breaking down the slim barrier in between, we're encouraging people to make that mental leap.

A slim barrier indeed. In fact, I'd suggest that anyone outside the Linux and open source community is likely to leap right over that barrier and think "gnome" whether it's written Gnome or GNOME. The non-techies are the folks we're trying to reach now. As likely as not, some potential users are going to be turned off by the cutesy (or should I say Kutesy?) names that free software developers have given to their wares.

Admittedly, this is an issue of style over substance. However, style is occasionally important. If you doubt this, check out the first guideline of the KDE Style Guide, which proclaims, "Always try to be consistent." Words to live by.

Click Here!