Oh fuzzy matches, what have you done? Thread poster: Dan Lucas
|
|---|
Dan Lucas 
United Kingdom
Local time: 20:10
Member (2014)
Japanese to English
My client (small agency) has said that her client (I suspect a much larger agency) insists on using the matrix given below. To me it looks quite severe, especially the 75-84%, but on the other hand 50% for 95-99% seems higher than it would usually be. I'm wondering to what extent that might offset the negative impact of the 75-84% discount. I might try replicating it in Studio tomorrow to get a better sense of what I would be giving up, but for now I have two questions.
1) Wh... See more My client (small agency) has said that her client (I suspect a much larger agency) insists on using the matrix given below. To me it looks quite severe, especially the 75-84%, but on the other hand 50% for 95-99% seems higher than it would usually be. I'm wondering to what extent that might offset the negative impact of the 75-84% discount. I might try replicating it in Studio tomorrow to get a better sense of what I would be giving up, but for now I have two questions.
1) What do you think? Appalling, or just not good?
2) Do you think you know which agency uses this? If so, please message me (just a mild case of curiosity)
The job would be large in absolute terms, but on the other hand my workflows are pretty steady, so I'd have to push myself to deal with it and I'm wondering whether - with these fuzzies - it would be worth it. The base rate would be reasonable, by the way.
Thanks
Dan
% of full rate
Repetitions: 20%
100%: 10%
95-99%: 50%
85-94%: 50%
75-84%: 50%
50-74%: 100% ▲ Collapse
| | | | Samuel Murray
Netherlands
Local time: 21:10
Member (2006)
English to Afrikaans
+ ...
Dan Lucas wrote:
...on the other hand 50% for 95-99% seems higher than it would usually be. I'm wondering to what extent that might offset the negative impact of the 75-84% discount.
It really depends on the number of high fuzzy matches. I've had jobs with a very large percentage of such matches, and then it's a breeze (but you must be eagle eyed, unless your CAT tool shows you the differences properly). The fact that they have no discounts below 75% is a good sign.
| | | | Lieven Malaise
Belgium
Local time: 21:10
Member (2020)
French to Dutch
+ ...
This is actually very reasonable. I wouldn't say no.
| | | | | Take reps into consideration also | Jan 12 |
Considering that many agencies don't pay reps/100s at all, I think it may even out for you. The only band there that I would consider iffy is the 75-84%, because you really do have to look at those quite carefully, but in my experience these don't account for much of the total words on most documents. Getting 50% on the 95-99% is quite good, as these can be quite numerous segments and are easy to do. Below 75% fuzzy matches are useless, in my opinion, so it's good that they're paying 100%. Overa... See more Considering that many agencies don't pay reps/100s at all, I think it may even out for you. The only band there that I would consider iffy is the 75-84%, because you really do have to look at those quite carefully, but in my experience these don't account for much of the total words on most documents. Getting 50% on the 95-99% is quite good, as these can be quite numerous segments and are easy to do. Below 75% fuzzy matches are useless, in my opinion, so it's good that they're paying 100%. Overall, not bad. ▲ Collapse
| | |
|
|
|
Dan Lucas
United Kingdom
Local time: 20:10
Member (2014)
Japanese to English
TOPIC STARTER
Thanks everybody. I thought I had been sent the source file but it's actually a "similar" file and so no use at all, although similar in size and content. I did pass the not-quite-source file through Trados (a quarter of a million characters...) and got very different results to those passed on to my client by the end client (no 75-84% matches at all).
Like Kathryn I find the fact that they are prepared to pay for reps promising, and also the fact that they're not pushing for sub-75... See more Thanks everybody. I thought I had been sent the source file but it's actually a "similar" file and so no use at all, although similar in size and content. I did pass the not-quite-source file through Trados (a quarter of a million characters...) and got very different results to those passed on to my client by the end client (no 75-84% matches at all).
Like Kathryn I find the fact that they are prepared to pay for reps promising, and also the fact that they're not pushing for sub-75% "matches".
I think I shall have to go back and prod my client to see if we can have the actual source file, or a breakdown of the fuzzy matches. Those will be the details in which the devil resides...
Dan ▲ Collapse
| | | | Lieven Malaise
Belgium
Local time: 21:10
Member (2020)
French to Dutch
+ ...
| Internal fuzzy match leverage | Jan 12 |
Dan Lucas wrote:
I did pass the not-quite-source file through Trados (a quarter of a million characters...) and got very different results to those passed on to my client by the end client (no 75-84% matches at all).
Careful, this might depend on your settings. Your client could have ticked the 'Report internal fuzzy match leverage' box in his analyzing settings, while you didn't. It can make a big difference.
[Bijgewerkt op 2024-01-12 07:50 GMT]
| | | | Dan Lucas
United Kingdom
Local time: 20:10
Member (2014)
Japanese to English
TOPIC STARTER | Makes a difference | Jan 12 |
Lieven Malaise wrote:
Careful, this might depend on your settings. Your client could have ticked the 'Report internal fuzzy match leverage' box in his analyzing settings, while you didn't. It can make a big difference.
Thanks, I had completely forgotten about that setting. This gives 75-84% as 10% of all non-rep characters (new is about 65% of the total) and a similar percentage for segments. If the actual source file - this is just an example - is similar in terms of breakdown, that wouldn't be too bad...
Regards,
Dan
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Oh fuzzy matches, what have you done? | Anycount & Translation Office 3000 | Translation Office 3000
Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.
More info » |
| | TM-Town | Manage your TMs and Terms ... and boost your translation business
Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news
