| Pages in topic: < [1 2] | WFC: What do YOU use the BTM for? Thread poster: Samuel Murray
|
|---|
Samuel Murray 
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Yves Champollion wrote:
The obscure and obtuse setting in Setup > AS where you see something like
BTM,TM,WFA,WFS,MT1,MT2,MT3
Yes, I saw that (I never spend any time on the AS tab, so it was a surprise). For now, it appears that the PB command overrides that setting. Am I right?
But there is an even better way to sort your situation: set a penalty on the BTM.
This only changes the match value but then WFC still consults the BTM first.
What drives me up the wall is if a BTM match keeps popping up over and over (even penalised, even if the BTM is second), and the only way to stop it from doing that is to copy the segment (including delimiters) into a new MS Word window and "clean" it so that the segment is added to the active TM. Only then does the BTM keep quiet on that segment.
| | | | | Now what is AS please? | Oct 10, 2015 |
Samuel Murray wrote: Yves Champollion wrote:
The obscure and obtuse setting in Setup > AS where you see something like
BTM,TM,WFA,WFS,MT1,MT2,MT3 Yes, I saw that (I never spend any time on the AS tab, so it was a surprise). For now, it appears that the PB command overrides that setting. Am I right?
I don't see such a tab.
Anyway, I have my big mama TM (>TM), everything checked, and besides some smaller TMs which occasionally I use as BTMs if the terminology is very specific.
Although I have used WFC since more than ten years, I have still many things to learn, I see...
| | | | Samuel Murray
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER | AS is the new name for AC | Oct 10, 2015 |
Christel Zipfel wrote: Yves Champollion wrote:
The obscure and obtuse setting in Setup > AS where you see something like
BTM,TM,WFA,WFS,MT1,MT2,MT3 I don't see such a tab.
AS (auto-suggest, I assume) is the new name for the AC (auto-complete) tab. I took the second screen-grab from the beta version 6.32. The first screenshot is from version 6.03.
| | | | neilmac
Spain
Local time: 16:58
Spanish to English
+ ...
| That's handy! | Oct 11, 2015 |
B. D. Laux wrote:
The B stands for "background". In you case, you could use the TM of one of your other clients as a BTM if you know that the subject is related.
That could be really handy for at least 2 or 3 of my clients, I must try it out.
Thanks to Michael and Samuel too, for posting their helpful explanations and links.
| | |
|
|
|
| In WF PRO IT KILLs CONCORDANCE SEARCHES | Oct 12, 2015 |
Allow a wordfast pro user ot barge in...
I love my BTM. I often have more than 2 TMs open, but almost always the client-specific+the BTM. However there is one point where it hurts: concordance searches don't let you sort results by TMs. I frequently wonder which particular choice I used months before for this client and that phrase, the easy answer is the concordance search, sometimes on just one or two key words of the phrase, but the results are polluted by the BTM. Sounds like... See more Allow a wordfast pro user ot barge in...
I love my BTM. I often have more than 2 TMs open, but almost always the client-specific+the BTM. However there is one point where it hurts: concordance searches don't let you sort results by TMs. I frequently wonder which particular choice I used months before for this client and that phrase, the easy answer is the concordance search, sometimes on just one or two key words of the phrase, but the results are polluted by the BTM. Sounds like a missing feature (search only the client or project TM, or at least sort results).... ▲ Collapse
| | | | | "B" as in barbarically primitive | Oct 12, 2015 |
I basically use my BTM as a project-specific TM, that is, I don't create project-specific TMs.
The nature of my source texts is such that the only time I ever get matches for segments cotaining more than a handful of words is in the case of revised texts or texts containing building-block elements from the same client.
PS: You're welcome, Neil. Thank you for the information about Dragon Naturally Speaking. And out of gratitude to Samuel I try to help in any way I can w... See more I basically use my BTM as a project-specific TM, that is, I don't create project-specific TMs.
The nature of my source texts is such that the only time I ever get matches for segments cotaining more than a handful of words is in the case of revised texts or texts containing building-block elements from the same client.
PS: You're welcome, Neil. Thank you for the information about Dragon Naturally Speaking. And out of gratitude to Samuel I try to help in any way I can when it comes to Wordfast Classic (Samuel taught me the wonderful Ctrl+Alt+M, which had suddenly become very relevant for one particular client). Sometimes ProZ is a great place. ▲ Collapse
| | | |
Samuel Murray wrote:
Yes, I saw that (I never spend any time on the AS tab, so it was a surprise). For now, it appears that the PB command overrides that setting. Am I right?
Normally not. The AS (Christel, that's AutoSuggest) tab overrides Pandora's Box in this respect. When a PB command makes it into the regular User Interface, it graduates and leaves PB for good.
Christel Zipfel wrote: Yves Champollion wrote:
The obscure and obtuse setting in Setup > AS where you see something like
BTM,TM,WFA,WFS,MT1,MT2,MT3 I don't see such a tab.
Christel, this confirms you're with an older version of Wordfast Classic. Version 5 is my best guess. As long as it gets your job done, keep it. The chief reasons for an upgrade are, you're on a Mac, or you can't do without MT. Sorry for the acronym... that's Machine Translation.
I also keep a vintage car that has no bells and whistles, that breaks down as all machines do, but that can be repaired with a screwdriver.
Cheers,
yves
[Edited at 2015-10-22 10:51 GMT]
| | | | | Concerned about.... | Oct 22, 2015 |
..."the unfortunate side-effect", see below.
This means also:
When we translate and clean a document, sending newly translated segs into the TM, that for the next job for the same client, we won't have all these auxiliary segments proposed by BTM, sent into the TM, despite having penalized them in BTM by 5 percent or similar, so the WfC Analysis on the new project document, will not be precise.
Am I right? I'd prefer being wrong.
Samuel Murray wrote:
I also penalise BTM matches by 5%.
[This has had one unfortunate side-effect, though. Whenever I get an exact match from the BTM, and I either accept that match or edit that match, then the match is not added to my active TM
[Edited at 2015-10-22 13:26 GMT]
| | |
|
|
|
Samuel Murray
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Dragomir Kovacevic wrote:
Concerned about... "the unfortunate side-effect", see below.
Sorry, I find your question a little confusing.
If the BTM proposes a translation, and that translation becomes part of your uncleaned file (whether you've edited the translation or accepted it as-is), and you *clean up* the file (with the option that updates the active TM), then the translations that originally came from the BTM will end up in your active TM, yes. To me, that is not a problem, because if I clean a file, it means that I'm happy with the translation and it means that I believe the segments all belong in my active TM.
It also means that if, in future, you were to use that same BTM and that same active TM, the proposal will no longer be penalised because it doesn't come from the BTM but from the active TM. However, if you were to use the same BTM but a brand new active TM, then the proposal will still be penalised, because it comes from the BTM.
There is no option in WFC that will cause segments whose origin is the BTM to be not added to the active TM, when you perform a clean-up.
The feature that I have been pleading for (but have been denied, since I'm told that I'm using the BTM "incorrectly") is that WFC would add a translation to the active TM whenever I accept that segment in the uncleaned file (i.e. whenever I move to the next segment), regardless of the origin of the proposed match. The only way that I can make sure all validated segments in the uncleaned document get into the active TM is to periodically clean up validated portions the file.
Does that answer your question?
[Edited at 2015-10-22 13:32 GMT]
| | | | | not worried anymore, but... | Oct 22, 2015 |
Samuel, when you write:
[This has had one unfortunate side-effect, though. Whenever I get an exact match from the BTM, and I either accept that match or edit that match, then the match is not added to my active TM
[quote]
... The above means that segments are not added to the active TM!?
So, what do we have:
- different penalties for TM,BTM,and others. - in TM rules
- order of TMs in Pandora
- AS in Setup, not still fully widened in stable editions.
No confusion on my side.
[Edited at 2015-10-22 13:44 GMT]
| | | | Samuel Murray
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Dragomir Kovacevic wrote:
The above means that segments are not added to the active TM!?
There are so many variables, but here's how I understand it:
* If you get a 100% match from the BTM (penalised or not), then that translation will never be added to the active TM (unless you clean up the file), regardless of whether you edit the translation.
* If you edit an existing uncleaned file, and there is a 100% match for the current segment in the BTM, then that segment will not be added to the active TM, even if it contains an error and you fix the error.
[Edited at 2015-10-22 18:03 GMT]
| | | | Jean Lachaud
United States
Local time: 10:58
English to French
+ ...
| Alt+Down does add the target to the TM | Oct 22, 2015 |
I have been puzzled by your statements.
Entering Alt+Down has always written the segment into the TM, regardless of where the target segment text comes from. IOW, if a BTM-found text is entered in the target, the content of that segment is entered in the TM when Alt+Down is entered.
Cleaning up has no relationship per se with updating the TM, unless of course "Update TM" is checked. Cleaning up is designed to remove all source text and purple tags. "Update TM" i... See more I have been puzzled by your statements.
Entering Alt+Down has always written the segment into the TM, regardless of where the target segment text comes from. IOW, if a BTM-found text is entered in the target, the content of that segment is entered in the TM when Alt+Down is entered.
Cleaning up has no relationship per se with updating the TM, unless of course "Update TM" is checked. Cleaning up is designed to remove all source text and purple tags. "Update TM" is designed to update the TM, for people who edit the translation without opening the segments. If all editing is done in opened segments, then there is no need to update the TM at clean-up, since it has been updated segment by segment each time a segment was closed using Alt+Down (or Alt+End, for that matter).
Samuel Murray wrote:
There are so many variables, but here's how I understand it:
* If you get a 100% match from the BTM (penalised or not), then that translation will never be added to the active TM (unless you clean up the file), regardless of whether you edit the translation.
* If you get a non-100% match from the BTM, then that translation will be added to the active TM *if* you edit the translation (otherwise it will not be added to the active TM).
* If you edit an existing uncleaned file, and there is a 100% match for the current segment in the BTM, then that segment will not be added to the active TM, even if it contains an error and you fix the error.
[Edited at 2015-10-22 18:00 GMT] ▲ Collapse
| | |
|
|
|
Samuel Murray
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER | Please help me test it | Oct 22, 2015 |
JL01 wrote:
Entering Alt+Down has always written the segment into the TM, regardless of where the target segment text comes from. IOW, if a BTM-found text is entered in the target, the content of that segment is entered in the TM when Alt+Down is entered.
Do me a favour and test this out. Create a TM that contains "The quick brown fox jumps over the lazy dog." and its translation. Then create a new TM, and add the first TM as the BTM. So now you should have an empty active TM, and a BTM with "The quick brown fox jumps over the lazy dog." in it.
Then create a file with these three segments:
1. The quick brown cat jumps over the lazy dog.
2. The quick brown fox jumps over the lazy dog.
3. The quick brown fox jumps over the lazy dog.
Start the translation. At sentence 1, you should get a fuzzy match from the BTM. Do not fix the translation, but use Alt+down to go to the second sentence. Check to see if the first sentence was added to the TM (in my case, yes).
For sentence 2 you should get a 100% match from the BTM. Change something in the translation (i.e. simulate that there's an error in the proposed match, and you're fixing the error), and press Alt+down again. Check to see if the second sentence was added to the TM (in my case, no).
Also check if the match that you get for segment 3 is the original translation (with the "error") or the updated translation (in my case, it's still the original).
Now imagine segment 3 is not in position #3 but somewhere towards the end of a 10-page file. Will WFC propose the translation that you fixed in segment 2, or will it propose the original translation from the BTM? What would you want WFC to do?
Added:
While you're at sentence 3, press Alt+down again (in other words, accept the 100% match as-is). Check to see if the sentence was added to the active TM (in my case, no).
Now simulate the "editing/proofreading" part of the translation by starting from the first sentence again. What happens when you get to sentence 2? In my case, WFC puts a red border around it (warning me that the current target text is different from the TM). In my case, I can move between sentence 2 and 3 as often as I like, but sentence 2 (the one that we "fixed") always shows a red border, and sentence 3 (the one that we "forgot to fix") shows up as a 100% match without the warning red border.
If all editing is done in opened segments, then there is no need to update the TM at clean-up, since it has been updated segment by segment each time a segment was closed using Alt+Down (or Alt+End, for that matter).
That is what I want to happen, yes.
[Edited at 2015-10-22 18:32 GMT]
| | | | Jean Lachaud
United States
Local time: 10:58
English to French
+ ...
| Adjust Translation Rules settings | Oct 22, 2015 |
If I understand correctly, what you describe suggests your system (Translation Memory | TM Rules | When editing (changing) a 100% match=) is set to "3".
Setting it to 0, 1 or 2 should do what you want.
| | | | Samuel Murray
Netherlands
Local time: 16:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
JL01 wrote:
If I understand correctly, what you describe suggests your system (Translation Memory | TM Rules | When editing (changing) a 100% match=) is set to "3". ... Setting it to 0, 1 or 2 should do what you want.
Thanks, but I checked: it's set to "0" (overwrite).
I normally set "When re-using an existing TU" to "1" (yes), but the test in my previous post does what it does regardless of what I set that setting to.
| | | | | Pages in topic: < [1 2] | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » WFC: What do YOU use the BTM for? | Trados Business Manager Lite | Create customer quotes and invoices from within Trados Studio
Trados Business Manager Lite helps to simplify and speed up some of the daily tasks, such as invoicing and reporting, associated with running your freelance translation business.
More info » |
| | Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news related to Wordfast
