![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
From the face of the salt mine...
Aug. 6th, 2002 06:31 pmDSL went out yesterday afternoon and came back only recently, which is a good thing, because the outage has made me shudder to think how much I've come to depend on net access to do my job.
I cannot shake this feeling of resentment as I do this editing job. And it only recently occurred to me that part of the reason for it is that it's not really an editing job.
Editing - as just about every other activity in life - assumes a whole bunch of things about other things. We've all heard the phrase "all things being equal" when someone is comparing one thing to another. That's an important concept, because when all other things are not equal, all hell can break loose.
Take translation editing, for example. In order to quote an hourly rate, one has to make certain assumptions about the translation to be edited. Generally, one assumes that it was done by a competent translator, so there may be one or two "major" errors every couple of pages or so (a dropped adjective, poor wording, a wrong word, etc.).
If the work was done by a sloppy translator, one's hourly rate sinks faster than lead in water. Typically, one can do some level of damage control (both financial and temporal, since not only does your rate suffer, but it'll take you longer to do the job) by calling the client and explaining the situation.
In this assignment, the translations were supplied by Trados, one of the leading translation memory products. Trados makes no claims about providing perfect translations; heck, it even posts a numerical score that estimates how "good" the offered translation might be.
The problem with editing such work is this: virtually every sentence needs tweaking.
And this violates the "other things being equal" condition. If a human translator were to make such mistakes, personally, I'd have as little as possible to do with the individual.
The American Translators Association's accreditation process basically permits one to make up to three "major" errors or an equivalent combination of major and minor errors (where ten minor errors equal one major error) in a one-page text before deeming the the translation as below minimum standards. Tank two of three such pages and you fail the test. At that, the ATA itself claims its program only accredits translators as having entry-level translation skills.
When I worked full-time for client T in Houston, we developed a more refined system. One major (or equivalent combination or errors) per page was good work. Half of that was considered excellent.
The Trados offering comes nowhere near being acceptable, and that's what bugs me about editing it. It's taking me about twice as long to edit this stuff because of all the tweaking, and even if a sentence is 100% correct, I still have to go in to the text and delete the Russian source!
It's obvious I'm going to have to revisit the pricing of this kind of work. I fully expect to be stonewalled, though, just as I was with client M a few days ago. But there is no harm in trying; I can always refuse the work.
Cheers...
I cannot shake this feeling of resentment as I do this editing job. And it only recently occurred to me that part of the reason for it is that it's not really an editing job.
Editing - as just about every other activity in life - assumes a whole bunch of things about other things. We've all heard the phrase "all things being equal" when someone is comparing one thing to another. That's an important concept, because when all other things are not equal, all hell can break loose.
Take translation editing, for example. In order to quote an hourly rate, one has to make certain assumptions about the translation to be edited. Generally, one assumes that it was done by a competent translator, so there may be one or two "major" errors every couple of pages or so (a dropped adjective, poor wording, a wrong word, etc.).
If the work was done by a sloppy translator, one's hourly rate sinks faster than lead in water. Typically, one can do some level of damage control (both financial and temporal, since not only does your rate suffer, but it'll take you longer to do the job) by calling the client and explaining the situation.
In this assignment, the translations were supplied by Trados, one of the leading translation memory products. Trados makes no claims about providing perfect translations; heck, it even posts a numerical score that estimates how "good" the offered translation might be.
The problem with editing such work is this: virtually every sentence needs tweaking.
And this violates the "other things being equal" condition. If a human translator were to make such mistakes, personally, I'd have as little as possible to do with the individual.
The American Translators Association's accreditation process basically permits one to make up to three "major" errors or an equivalent combination of major and minor errors (where ten minor errors equal one major error) in a one-page text before deeming the the translation as below minimum standards. Tank two of three such pages and you fail the test. At that, the ATA itself claims its program only accredits translators as having entry-level translation skills.
When I worked full-time for client T in Houston, we developed a more refined system. One major (or equivalent combination or errors) per page was good work. Half of that was considered excellent.
The Trados offering comes nowhere near being acceptable, and that's what bugs me about editing it. It's taking me about twice as long to edit this stuff because of all the tweaking, and even if a sentence is 100% correct, I still have to go in to the text and delete the Russian source!
It's obvious I'm going to have to revisit the pricing of this kind of work. I fully expect to be stonewalled, though, just as I was with client M a few days ago. But there is no harm in trying; I can always refuse the work.
Cheers...
