Relata Refero wrote: Indeed in my "day job" as a computer science academic, this sort of concern is making copyediting fall out of favor quite rapidly. It's mostly been phased out for unrelated reasons (cutting expenses), but the few journals that still insist on doing extensive editing for style (like anything the IEEE runs) annoy many authors, as we have to keep re-reading our own drafts to figure out what they screwed up this time. It can even end up in comical exchanges of drafts where the author will change something back to what they actually meant to say, the copyeditor will change it back to conform to "house style", the author will revert the change in the next exchange, etc. Now of course there are some benefits---such journals have more consistent formatting and style, fewer grammatical errors, somewhat more flowing prose, don't infuriatingly use a reference as a noun ("As [1] showed, ..."), and impeccably uniform citation typesetting. But the authorial burden and the chances of a technical error creeping in, or just a statement or implication that the author didn't intend, are also higher than in the becoming-standard "the author just gives us a PDF and we do a quick read-through for obvious problems" approach. Which doesn't mean we shouldn't edit Wikipedia articles for style, but it is tricky to do well. We at least don't have one of the problems, which is the risk that copyediting can change the original author's voice and end up attributing to them a sentence they would never have written, since our articles aren't bylined. But there can certainly be drift between the text and the references. In general I think there needs to be a more integrated approach than a classical hard distinction between the subject-matter expert and the copyeditor---in many cases there's no hard boundary between improving the treatment of the subject (what to discuss in what order, what to emphasize, what connections to make), which is a subject-matter issue, and improving the overall readability and flow of the article, which is a copyediting issue. -Mark

On 26/05/2008, Delirium <delirium(a)hackish.org> wrote: Speaking as someone who does do a fair amount of copy-editing, I have to say that there is a qualitative difference between editing a decently written article with a primary editor and good referencing, and the overwhelming majority of articles that are poorly referenced hodge-podges of whatever information different drive-by editors happened to insert. I tend to work on the decent articles at the invitation of the primary editor, and when doing so will read whatever online references are available and will ask a lot of questions (talk page or FAC/GA/peer review page) to help the editor clarify what was meant. Even in relatively well-written articles, I have found many instances where the reference sources don't match up with the material in the article that they are intended to provide references for. In "average" articles, usually a third of the references fail to match the statement they're being used to reference. I've also seen some relatively worrisome "ownership" of poorly sourced, ungrammatical articles that are written so poorly as to be confusing, contradictory, or nearly unreadable - to the point where correction of a typo or spelling error leads to instant reversion or an inquiry on a talk page leads to beratement of the questioner. So - is the problem the copy editor not understanding, or is the problem the quality of the work in the first place? Risker

2008/5/27 Ben Yates <ben.louis.yates(a)gmail.com>om>: Not remotely because stable versions is not meant to stop people adding new information to articles. Of course it will but there is no need to encourage that. -- geni

At 09:07 PM 5/24/2008, Ray Saintonge wrote: It's nobody's responsibility and it is everyone's. If we have flagged revisions, verification is part of the task of the user with flag privileges. But if we are keeping anything like the present process, it is nobody's specific responsibility to actually do it. That, indeed, is part of the problem. In our distaste for bureaucracy, we have rigorously avoided specific responsibilities. Some kind of (voluntary) assignment of articles for review might become part of the process. One thing that really does not work at present: people check sources. It was fine. Then someone comes along later and checks the same source. It's still fine. And many editors can waste time this way, until and unless there is an easy way to leave a mark that you checked it, you are now taking additional responsibility for its accuracy. That's what I meant by suggesting a Sources page. All it would take, for each source covered, is "Validated ~~~~".