NAM Spanish Project

[Anthracite 37]

Ladies, gentlemen, good evening.

Eleven months ago I contacted @Tarkus, offering my help regarding a possible translation of the NAM to the Spanish language. I was asked then to translate every single LTEXT from NetworkAddonMod_Locale_english.dat. I'm saving the changes by creating a new NetworkAddonMod_Locale_spanish.dat file. However, I request some assistance concerning several points:

1. There are several entries marked as "Unknown" with uncomprehensible symbols and letters or readable letters separated with points, namely:

 

 

 

 

Spoiler

Entry;Compressed(Y/N);Filesize;Num;Type;Group;Instance

Unknown;N;102;2774;25;2026960b;123006bb;52001000

Unknown;N;130;3185;27;2026960b;123006bb;52001010

Unknown;N;94;3600;29;2026960b;123006bb;52001020

Unknown;N;104;3989;31;2026960b;123006bb;52002000

Unknown;N;124;47734;468;2026960b;123006bb;6a47b42a

Unknown;N;598;47858;469;2026960b;123006aa;6a47b42a

Unknown;N;244;65145;644;2026960b;2a592fd3;58488500

Unknown;N;244;65389;645;2026960b;2a592fd3;58488600

Unknown;N;244;65633;646;2026960b;2a592fd3;5a403100

Unknown;N;244;65877;647;2026960b;2a592fd3;5a403300

Unknown;N;122;66121;648;2026960b;2a592fd3;78488500

Unknown;N;122;66243;649;2026960b;2a592fd3;78488600

Unknown;N;84;66365;650;2026960b;2a592fd3;7a403100

Unknown;N;84;66449;651;2026960b;2a592fd3;7a403300

Unknown;N;132;66533;652;2026960b;2a592fe0;58488500

Unknown;N;132;66665;653;2026960b;2a592fe0;58488600

Unknown;N;132;66797;654;2026960b;2a592fe0;5a403100

Unknown;N;132;66929;655;2026960b;2a592fe0;5a403300

Unknown;N;122;67061;656;2026960b;2a592fe0;78488500

Unknown;N;122;67183;657;2026960b;2a592fe0;78488600

Unknown;N;84;67305;658;2026960b;2a592fe0;7a403100

Unknown;N;84;67389;659;2026960b;2a592fe0;7a403300

Unknown;N;306;67518;661;2026960b;2a592fd4;78488500

Unknown;N;306;67824;662;2026960b;2a592fd4;78488600

Unknown;N;306;68130;663;2026960b;2a592fd4;7a403100

Unknown;N;306;68436;664;2026960b;2a592fd4;7a403300

Unknown;N;286;68742;665;2026960b;2a592fd5;78488500

Unknown;N;286;69028;666;2026960b;2a592fd5;78488600

Unknown;N;286;69314;667;2026960b;2a592fd5;7a403100

Unknown;N;286;69600;668;2026960b;2a592fd5;7a403300

Unknown;N;322;69886;669;2026960b;2a592fd6;78488500

Unknown;N;322;70208;670;2026960b;2a592fd6;78488600

Unknown;N;322;70530;671;2026960b;2a592fd6;7a403100

Unknown;N;322;70852;672;2026960b;2a592fd6;7a403300

Unknown;N;320;71174;673;2026960b;2a592fd7;78488500

Unknown;N;320;71494;674;2026960b;2a592fd7;78488600

Unknown;N;320;71814;675;2026960b;2a592fd7;7a403100

Unknown;N;320;72134;676;2026960b;2a592fd7;7a403300

Unknown;N;294;72454;677;2026960b;2a592fd8;78488500

Unknown;N;294;72748;678;2026960b;2a592fd8;78488600

Unknown;N;294;73042;679;2026960b;2a592fd8;7a403100

Unknown;N;294;73336;680;2026960b;2a592fd8;7a403300

Unknown;N;314;73630;681;2026960b;2a592fda;78488500

Unknown;N;314;73944;682;2026960b;2a592fda;78488600

Unknown;N;314;74258;683;2026960b;2a592fda;7a403100

Unknown;N;314;74572;684;2026960b;2a592fda;7a403300

Unknown;N;300;74886;685;2026960b;2a592fdc;78488500

Unknown;N;300;75186;686;2026960b;2a592fdc;78488600

Unknown;N;300;75486;687;2026960b;2a592fdc;7a403100

Unknown;N;300;75786;688;2026960b;2a592fdc;7a403300

Unknown;N;288;76086;689;2026960b;2a592ff4;78488500

Unknown;N;288;76374;690;2026960b;2a592ff4;78488600

Unknown;N;288;76662;691;2026960b;2a592ff4;7a403100

Unknown;N;288;76950;692;2026960b;2a592ff4;7a403300

Unknown;N;630;79227;714;2026960b;123006aa;6a47aa0f

Unknown;N;144;79857;715;2026960b;123006bb;6a47aa0f

Unknown;N;638;100435;913;2026960b;123006aa;6a47aa10

Unknown;N;152;101073;914;2026960b;123006bb;6a47aa10

Unknown;N;606;101225;915;2026960b;123006aa;6a47aa11

Unknown;N;104;101831;916;2026960b;123006bb;6a47aa11

Unknown;N;582;116192;1063;2026960b;123006aa;6a47aa12

Unknown;N;92;116774;1064;2026960b;123006bb;6a47aa12

Unknown;N;582;116866;1065;2026960b;123006aa;6a47aa13

Unknown;N;92;117448;1066;2026960b;123006bb;6a47aa13

Unknown;N;596;117540;1067;2026960b;123006aa;6a47aa8a

Unknown;N;92;118136;1068;2026960b;123006bb;6a47aa8a

Unknown;N;66;135060;1258;2026960b;123006bb;6a474000

Unknown;N;412;135126;1259;2026960b;123006aa;6a474000

2. In case it can, how will it be implemented in NAM 37?

3. What are the other core/documentation files to be translated?

I'd like to open this up to Hispanic-speaking players. Let's talk about it friends.

NetworkAddonMod_Locale_spanish.dat

[matias93 7,535]

What a great initiative!

Going by your points:

1. I would test it without changing anything of those unknowns, and try to find non translated things in-game.

2. Seeing that this is a separate DAT file, I guess it would only require to load it after the English locale (?). In any case, overriding the English locale would suffice to make some tests.

3. The current NAM documentation is formatted as HTML, so translating it should be much easier. An online translator could do the bulk of the work and a human then can check for remaining errors.

I would be glad to help with this, and I'm sure some other users too (thinking on @Edvarz particularly)

[rsc204 21,866]

On 25/02/2020 at 4:38 AM, Anthracite said:

1. There are several entries marked as "Unknown" with uncomprehensible symbols and letters or readable letters separated with points, namely:

Some LText entries seem to do this when they are either corrupted or something else has happened to them. For now, I would just leave those entries out of your variant, since it's not like you've anything to work from to translate them, even if they are in use somewhere which I suspect they aren't.

47 minutes ago, matias93 said:

2. Seeing that this is a separate DAT file, I guess it would only require to load it after the English locale (?). In any case, overriding the English locale would suffice to make some tests.

Indeed for now, that's certainly the best method for development. A mechanism exists within the game, whereby an offset to the ID can be used to tell the game which language your LTexts are for, in fact it's pretty much the reason LTexts exist. However, depending on how the original files were ID'd, that could become a complex process. I know for RTMT, all such locale files end in a 0, so you can easily batch alter the IDs to make a set for another supported language. But if that's not the same here, changing those IDs would be a painstaking and error-prone task otherwise. The other downside, for this to work, all languages need to be installed for all users, which increases install sizes.

I think @Tarkus is best placed to say which methodology works best in the context of a NAM release, might be worth sending him a PM?

47 minutes ago, matias93 said:

3. The current NAM documentation is formatted as HTML, so translating it should be much easier. An online translator could do the bulk of the work and a human then can check for remaining errors.

Having worked in a professional environment localising games, I recall management once had this idea, thinking it would be a great time (money) saver. Well yes, no question Google is faster than us mere humans, but even all these years later, I'm still convinced the translations it produces are terrible. Of course how much time you have is a big factor in which way to go, but in my company they realised they'd be trading reduced costs, i.e. less translators, for what was a very poor translation by comparison. If you take pride in your work, I can't image this would actually be quicker overall. This is because whilst much of the automated translation will be "technically" correct, that rarely reads well for native speakers. To get that to a level of quality, i.e. something that sounds natural and flows properly, you'll pretty much have to re-write everything anyway. The added complication of special terminology will also be a factor here.

On 25/02/2020 at 4:38 AM, Anthracite said:

3. What are the other core/documentation files to be translated?

These days we've moved over to the HTML route, just as @matias93 states. However, a lot of legacy content exists in the form of PDF files, editing those might be much harder. The idea is to eventually have everything ported over one single HTML system, progress has been slow however.

The only other thing I can think of would be the Installer itself, I don't know the application Tarkus is using to build the new NAM installer. Certainly the old NSIS one and many similar development apps include support for multiple languages. But again you really need to find out from him if it's possible and how best to make it happen.

More generally though, support for more languages is always a good thing, it increases both understanding and the potential user base, when any such project is more inclusive. Therefore it's great that you're working on this and I'm sure whatever you come up with, we (the NAM team,) will find a way to incorporate your work.

[Anthracite 37]

3 hours ago, rsc204 said:

Indeed for now, that's certainly the best method for development. A mechanism exists within the game, whereby an offset to the ID can be used to tell the game which language your LTexts are for, in fact it's pretty much the reason LTexts exist. However, depending on how the original files were ID'd, that could become a complex process. I know for RTMT, all such locale files end in a 0, so you can easily batch alter the IDs to make a set for another supported language. But if that's not the same here, changing those IDs would be a painstaking and error-prone task otherwise. The other downside, for this to work, all languages need to be installed for all users, which increases install sizes.

Hi,

Well, yeah, I've been overriding as described. However, last time I did it, I found that switching between Wide Radius Curves for AVE, RD, OWR et cetera, the game crashes, not to desktop, but stops responding and has to be killed via Task Manager

I don't know how to diagnose and remedy it.

[matias93 7,535]

I'm betting that somebody of @rsc204, @rivit and @Tyberius06 could know this: is there a way to mass-export LTEXT exemplars from a dat file to separate text files, as to edit them outside the Reader? I'm thinking on something analogous to what one does with the GoFSH, but for text.

[Tarkus 6,933]

It is possible to export them out of the Reader, and reimport them, but the resultant files are autonamed and separate, and kind of a mess.  I believe there were a couple of attempts at LTEXT-related tools at various points, but I don't think they ever really went beyond initial prototype/demo stage.  I may do some research there to see if anything is serviceable.

-Tarkus

[matias93 7,535]

That sounds workable to me, I know an "easy" way to batch-modify lots of text files at once using regular expressions. This should suffice.

How is the exporting and reimporting thing?

[rivit 3,402]

In iLives Reader

• the export all Files option gives you all files in a dat, or you can select them and do the Export decoded Files - into a given folder
• Batch import (right mouse menu) requires a file listing all of the files to import in a particular format (basically a tab delimited list of TGI and type) then import (see an example in GoFSH  _ReaderLoadList)

[Edvarz 1,985]

@matias93 I would indeed be glad to help in translating the NAM!!! You can count me in

If I'm reading correctly, it would be a matter of editing LTEXT files? Maybe we could coordinate to avoid working redundantly

[Barroco Hispano 26,932]

Interesante!

[rsc204 21,866]

Sorry, been trying to find time to finish typing up a post I started a few days ago. I will do that ASAP, but I just wanted to share the key points for now, so you aren't in the dark.

• Other language files in the NAM use the ID-Offset method, so I think we'll need to find some way to make that painless. For now the override method is fine, you won't need to re-do any work in order to change the IDs.
• Looking at your attached file and comparing identical ID'd files between the two, the "Unknown" entries appear to be corrupted LText files. If you compare them to the English file, where text is clearly present and they are correctly identified as LTexts. The long and short here, please tell us exactly which tools you've used and your process for working with these files. We can then track down the point where these files are getting corrupted and advise you how best to avoid it happening again.

[Edvarz 1,985]

I don't know if this is the same situation but I once had Reader v.1.53b corrupt a .sc4desc file while editing an LTEXT and I suspect it was because of ortographic accentuation (what we call tilde in Spanish, as seen in words like autobús or tranvía)

[Anthracite 37]

On 2/29/2020 at 5:46 PM, rsc204 said:

Sorry, been trying to find time to finish typing up a post I started a few days ago. I will do that ASAP, but I just wanted to share the key points for now, so you aren't in the dark.

• Other language files in the NAM use the ID-Offset method, so I think we'll need to find some way to make that painless. For now the override method is fine, you won't need to re-do any work in order to change the IDs.
• Looking at your attached file and comparing identical ID'd files between the two, the "Unknown" entries appear to be corrupted LText files. If you compare them to the English file, where text is clearly present and they are correctly identified as LTexts. The long and short here, please tell us exactly which tools you've used and your process for working with these files. We can then track down the point where these files are getting corrupted and advise you how best to avoid it happening again.

I have only used ILive's editor (latest version). To translate the files, I have individually rewritten the text of every entry I have found and that's it. Respecting ortographic accentuation (tildes), as @Edvarz described, there are accents in every word where needed. However, it wasn't until recently that I got to crash the game (I test the menus every once in a while).