going through the list I really think that this is a method that should be applied only in cases of need. There are so many things to worry about.
German Wrong: Kind von Karth
German Correct: Karths Kind
If we have “das Kind von Karth”, S&R will make it: “das Karths Kind” because “Kind von Karth” should be replaced by “Karths Kind”. “das Karths Kind” is wrong. It should be just: “Karths Kind” (without the article). But if I let the macro just change “das Kind von Karth” to “Karths Kind”, it won’t find “Kind von Karth”, “des Kinds von Karth”, “dem Kind von Karth”. If I put both “Kind von Karth-Karths Kind” and “das Kind von Karth-Karths Kind” on the list, it depends on the order again whether entries are found or not.
This is an example where we could simply forego the S&R without compromising the text. However, many of the terms NEED to be replaced.
By the way, please check the Glossary German BOI over at https://spreadsheets.google.com/ccc?key=0AmlfEgfxF8r3dEFHVmcxVE5lYTBaWVZfYTA3WnpmWlE&authkey=CLm0jrUG&hl=en#gid=0
I’m putting all the odd stuff I’ve found over there. I will continue to do so as I find them.
As for your concern above, you have a point, but we know that we need to fix thing, so the need exists. We can’t forego the S&R because even in the cases where the S&R causes things to go wrong, it will mark the text (either with $$ or with strikethrough in Word’s Compare Document), allowing a proofreader to zero in on the problem.
The current S&R list has over 1000 terms. No normal person can remember all those when reproofing the files. Without that list, he will also end up reproofing things that have been proofed. Running the S&R list with, say, the $$, will allow him to skip all the correct stuff and focus on what needs to be changed. If it’s wrong, he can correct it them. Same thing if it needs to be modified to account for German grammar.
On a related matter, please get Sebastian to help you. Karel really wants to S&R list to be done and the script run ASAP. Sure, the actual runtime of the script is likely mere hours, but we don’t know what else needs to be done afterward and even if there’s nothing to be done afterward, we still need to double-check the translation after the script is run to make sure that it’s fine.
Okay, so we’re done with translation and spotchecking/proofreading.
We’re now in what we call the polishing stage. In relation with this, can I ask you how much time you have up to Sunday next week?
sorry, I probably expressed myself unclearly. I didn’t mean to say to skip the whole S&R, just for that term.
What do you think of this:
We leave out the $$ for the first run of the macro.
Since there are bound to be some wrong insertions (Relikt- becoming Reliquie- instead of Reliquien-), I try to anticipate such cases and we run the script a second time (replacing Reliquie- with Reliquien-).
Or do you think this creates too much work and error sources and we should just run the list and let proofreading start (with $$).
So I can get Sebastian to help me with refining the list?
Then I misunderstood. Right, that’s sorted out, let’s put that aside.
As to whether running the script twice is better…. My gut feeling says we won’t save enough time to justify the extra work. Checking time is likely to remain the same, since all $$ has to be checked. Time needed to make corrections may be reduced, but I don’t think there’s that many errors/duplications of that sort. Are there?
Right now I’m thinking we should get the script run once, and then have two people check, say, 30 files (about 10% of the total). If there’s a lot of errors that a second S&R list is needed, we can make it then and apply it to the other 90%. If not, then we just continue on until all the files are done.
” So does that mean that I should stop trying to anticipate grammar or do we have a budget for that? “To be , I have no idea. It’s hard to predict how much time will be saved.
Very well, since we have no prior data, let’s experiment. Stop anticipating grammar. When you’re done, I’ll run the script on 30 files and I’ll have you and maybe Sebastian or Christina analyze the result. We’ll be able to figure out which method to use for the other 90% of the files then.
okay, here is my list back. I marked yellow those that I expect the most problems with (but others may give problems too). Hope this works.
Okay, I’ve run the S_R_Newform_revised.zip.
I notice that there are / in the terms, e.g., ReliktIII/Relikt III. Is term actually “ReliktIII/Relikt III” or do you mean “ReliktIII or Relikt III”? I’m assuming the first, but if it’s the second then please let me know right away because that means I need to split the terms and re-run the script.
Okay, I’m assuming the terms is “ReliktIII/Relikt III” as a single term. Because that’s how it has been run.
Now we need to check the result. The resulting files are here.
I want you to take 5 hours to analyze these 30 files. Not double-check the changed ($$) terms. I’m asking Sebastian to do so. Instead, I want you to focus on two things:
1. From those 30 files, is there anything that should be in the S&R list but is not? If not, then we need an updated S&R list.
2. Other than S&R, is there anything else that we need to do?
” Some segments were accidentally missing from file 150. We’ll get it replaced in 15 minutes.
“Hi, are you free?
We are at the polishing stage. We’ve run the S&R script on 30 files, which you can find here:
The terms themselves are here:
They are Excel files.
If you look at the Terms file, you’ll notice the $$ that we’re using as a marker. This means that if anything was changed, you can find them by doing a search for $$.
So please do so. Use the Find function to find the $$ marks. When you do, check to make sure that the change is correct and modify it so account for gender, gerund, plural, and other grammar rules.
There were about 200 changes. Ten on the first batch (121-135) and the rest on the second (136-150). Please let us know how much time it takes do each batch.
Please focus on the changed terms only (the $$ parts). If you find something odd and want to suggest something, please do, but is actually analyzing the same batches of file to see if he needs to add more terms to the S&R list and if we need to do other stuff too. I don’t want you to be doing the same thing he’s doing and duplicate his work. On the other hand he isn’t going to correct the changed terms, since you’ll be doing that. He’ll be focusing on the analysis instead.
” Fixed. Just use the download link above.
sorry for the late answer. Actually, I am quite busy at the moment. But I squeezed in some time to look at these files. There wasn’t much to edit, but I noticed some consistency issues etc. — see the reports below.
About availability until sunday next week: I’m sorry I will not have much time in the next weeks. All I could do is squeeze in an hour or two from time to time — which I guess won’t be so very helpful for you. Sorry, but that is all I can offer at the moment.
About the hours below: should I include them in my extra report, or will you keep track of them?
batch 121-135: 30 mins.
=> I suggested in the final translation phase to use “Waren-Auftrag” when a quest is meant. But it is context sensitive, so it can only be added by a person on a case by case basis. I didn’t change it yet because it isn’t in the last dictionary.
=> changed all to female (the majority is already female), and added article “die” where appropriate.
E$$inheit der Expedition
=> shouldn’t this be E$$xpeditions-Verband? (not changed)
batch 136-150: 20 mins.
mean the same in 136-150 (a quest title). Grammar allows both.
Battle of Immortals = S$$chlacht der Unsterblichen – I thought it was meant to remain in English?
Dark Metal Blazing Knight – D$$unkles Metall-GlÃ¼hender Ritter
=> I think this sounds strange, but did not edit it.
Sebastian” Include the time in your extra report.
Karel sent me more client instructions and I’m trying to implement them. I can’t reach him though. It will take me some time until I get to the task you assigned me, but if you think that takes priority, I will work on that first.
Karel’s internet connection will be spotty for a couple days.
I don’t know what Karel told you to do. Although I got a copy of what he got from the client (the email with all the coloured conversation), I don’t know what instruction he gave you.
Since I don’t know what his instruction is, I can’t decide which one should be prioritized.
Can you tell me what is it that he wants you to do so I know whether you should do it or the analysis first?
sorry, what I asked Karel was if I should try and iron out some issues the client was asking for (mainly hyphenation) which would probably be easy to S&R (minimal proofing) but would make the list a LOT longer and takes me 5-8 hours. The other option is to ignore their wining and continue what you told me to do.
Attaching what I wrote to Karel:
what they wrote would mean a big overhaul of the glossary/ S&R list because it seems now that what they once didn’t want: that translators follow their feeling for the language when building terms is now what they want. That means the hyphenation rules for trying to get consistency into the terms no longer apply, or at least apply only sometimes. What do you think? I think the way the S&R list worked so far is quite good (of course, I only saw it after being spotted by Sebastian). If I change all the things they want changed, it would take me at least 8-10 hours for the glossary/adding to S&R and it would also create a lot more work for the spotters (although not too much, since hyphenation luckily does not change meaning).
What’s really almost impossible to change now is what they were saying about occurrences of “Stone of Fire” and “Fire Stone”. Since their concern was consistency as I’m told by the client:
For me, part of creating a glossary is to make it consistent. Even if the input I receive is inconsistent, the output has to be consistent. Otherwise, I donâ€™t have to create glossary.
I was assuming they were the same terms and streamlined them (as apparently requested here).
Now they are saying:
Frost Item is something you usually gather and pick up in the game. Item of Frost is usually a piece of gear that does Frsot magic damage.
How we were supposed to know of this distinction without being told (and, on my part, having translated ZERO content before dealing with the glossary) is beyond me. So I have no idea how to implement this, even in the glossary (without context) let alone in the files with all content being translated already. I’m hoping this is not as big a concern as they make us believe here (hard to say because I do not really understand/cannot really apply the distinction to any of the terms I’m looking at).
So what do you suggest? has asked me to go through the files on which the S&R has been applied and check if there’s anything to add. I’m on that but that would be a bit delayed if I’m to deal with the hyphenation issues stated above.