26 Aug 2007

Won't fixing

Submitted by Karl Hagen
The title is not an instance of editorial error, nor the start of a question (e.g., "Won't fixing one thing break something else?" but evidence of a new compound verb with a very unusual structure. I first saw it on a developer's issue board. "Clearing my issue queue. I don't think these fixes are going to go in anyway - won't fixing." To understand the intended meaning, it helps to know that software projects with many developers tend to track bugs, feature requests, and other issues with a system that let's them indicate the status of a particular issue. So a bug might be classified as "active," "fixed," "postponed," "won't fix", etc. In this context, then, the writer means something like "I am setting the issue status to 'won't fix'" The presence of the -ing at the end of "fix" shows that the ordinary English morphology has been overridden, since a verb like fix after an auxiliary should be an infinitive. That is, "won't fix" is being treated as a single lexical unit, onto which a suffix can be added the same way that it would be on an ordinary verb. This sort of word-formation process is called "dephrasal compounding." In other words, the compound word derives from a phrase. One feature that distinguishes this word formation process from ordinary compounds is that inflected elements can show up inside the compound where they ordinarily would not be allowed. Has-been is another instance. Compare these with ordinary compounds like crybaby, swimsuit, or swearword, all V+N compounds, where the initial verbal element must be the base form of the verb and not, for example, *criesbaby, *swamsuit, etc. Even among dephrasal compounds, however, I've never run across one formed from a Modal Aux+V combination. Has-been, above, is the closest I can think of (another Aux+V combo, but not a modal verb, and the compound itself is a noun, not a verb). There are also V+V compounds like freeze-dry or lend-lease, but in these cases, the first base is an ordinary verb. Additional examples would be welcome. Googling shows that verbal "won't fix" is rare, but spread out among several different communities of developers. Another example from drupal.org: "If you are dealing with very old issues, you're right that asking the original submitter to confirm the bug is unlikely to be fruitful. However, closing or won't fixing the bug does not add value by itself: a status change is not warranted unless you have specifically confirmed or denied the bug." Another example on an ubuntu list: "Won't fixing until the oustanding [sic] security issues in the Feisty version are addressed." It is also somewhat productive, taking inflections other than -ing. Passive constructions are actually more common than progressive ones (e.g., "it was won't fixed.") The number of g-hits is much higher, but many instances of {"won't fixed"} are just errors, for example, the result of accidentally omitting "be." Others though clearly illustrate the term as a compound verb: For example, "I think it was won't fixed because the bug was opened in plexus-archiver, but unfortunately wasn't linked." Or this one, which shows both a passive and an infinitive use:
I myself am a QA tester, although not in the Windows group. I've entered many, many bugs that were Postponed or Won't Fixed that, honestly, probably didn't deserve to be fixed. Things like "such and such a static is too long for this control" or "we should notify the user if x, y, or z happens". Little niggly things in other words. Keep in mind that a "bug" is anything a tester doesn't like, including personal design preferences. It isn't necessarily a flaw that would interfere with use of the product. If a bug is of that kind, then it's smart for a time-starved group to Postpone or Won't Fix the issue.
"Won't fix" can also be a noun, as shown by its occasional appearance in a plural form, although some of these are quoted to indicate the author is aware that it's an odd usage:
Whet [sic] we need is a to ignore these tickets for the closed count _and_ for the total. The same treatment should be applied to won't fixes among others


I would certainly have expected won't fixed to be more common than won't fixing, given the telic nature of changing the status of something. Presumably doing so entails ticking a check-box, typing 'won't fix' into a 'status' field, or some other minor action. To use a traditional telicity test, you can't change something's status for five minutes, unless you change it back and forth, or among several options. Moreover, once the status is changed, it is likely to remain in that state until someone else looks at the bug. This is why won't fixing seemed incongruous to me. It could be roughly translated as "I am marking its status as 'won't fix'", which I suppose, telic or not, is acceptable. But I'd consider it 'better', to say won't fixed, or "I have changed its status to 'won't fix'".