Pages in topic:   [1 2] >
DGT-OmegaT
Thread poster: Milan Condak
Milan Condak
Milan Condak  Identity Verified
Local time: 10:21
English to Czech
Jan 2, 2018

DGT-OmegaT Dec 31, 2017

Michael Joseph Wdowiak Beijer wrote:

OmegaT


DGT-OmegaT = A fork of standard OmegaT created by EU DGT.

http://www.condak.cz/nove/2017-12/06/cs/03.html
... See more
DGT-OmegaT Dec 31, 2017

Michael Joseph Wdowiak Beijer wrote:

OmegaT


DGT-OmegaT = A fork of standard OmegaT created by EU DGT.

http://www.condak.cz/nove/2017-12/06/cs/03.html

http://185.13.37.79:8003/index.php/p/omegat-core/downloads/13/

After 4 month and 3 days only 43 download.
--
I use DGT-OmegaT as a part of Full CAT suite

http://185.13.37.79:8003/index.php/p/wizard/downloads/26/

CAT-Full-Archive.exe - 180.21 MB

Contains the wizard as binary, DGT-OmegaT 3 and Tag wipe.

Uploaded: 1 month 1 day ago by Thomas CORDONNIER

Downloads: only 6

I was a third user. Very impressive package.

Milan
--
https://www.proz.com/forum/cat_tools_technical_help/86861-list_of_cat_tools-page2.html#2702554

Teams and TagWipe Jan 1

Hans Lenting wrote:

Is my understanding correct that DGT-OmegaT has been designed primarily for teams?



My opinion only: DGT-OmegaT has been designed primarily for in house translators in EU DGT and solves a confidentality of MT; instead of web-based generic USA, Russia, Korea or China MT translation is used in-house MT engine, which produce TMX, which is placed into folder MT.
For outsoursing, where texts are not confidental, can be used standard web-based MT. A connector is separate plugin.
Second feature is for translating texts into languages of the same group (slavic, roman, german,..).
In DGT-Wizard is SW for viewing PDF files: quides and other files.

Hans Lenting wrote:

How about "Tag wipe": does it do what the name suggests (namely: wipe tags from source files like XLIFF or TMX)?


If you tick a check-box "Tag wipe" in DGT-Wizard and in a project is/are DOCX files, you got in Files source folder of OmegaT DOCX with minimum of tags. There is SW for comparing original and "tag-wiped" DOCX.
In DGT-OmegaT is preview on source and on target DOCX; MS Word is opened - I use free MS Word 2010 Starter.

Here are two links to relevant pages and an example of a description:

Teambase

http://185.13.37.79/?q=node/20

Since this is a client/server application, the situation is more complex than for all features discussed until now. The distributed binary version of DGT-OmegaT includes the client plugin, but for the sources and the server part, this has been put in a separate project. Furthermore, the idea of this server came a long time before DGT users were finally interested in it. That is the reason why the server part is in another web site - http://www.silvestris-lab.org/: in fact, Teambase is nothing more than a personnalization of Silvestris/Cyclotis project, not the other way around.

Tagwipe

http://185.13.37.79/?q=node/35

More often than not Office DOCX documents have (many) useless tags.

DGT developed in-house the Tagwipe application which removes all/most redundant tags from DOCX documents. It also improves segmentation by segment.

Milan

[Edited at 2018-01-02 11:31 GMT]
Collapse


 
Michael Beijer
Michael Beijer  Identity Verified
United Kingdom
Local time: 09:21
Member (2009)
Dutch to English
+ ...
improvements re. merging/splitting segments in OmegaT? Jan 2, 2018

Hi Milan,

Just curious: have they added any improvements regarding merging/splitting segments in DGT-OmegaT? As far as I remember, OmegaT was very bad at this.

Michael


 
Milan Condak
Milan Condak  Identity Verified
Local time: 10:21
English to Czech
TOPIC STARTER
I do not use merging/splitting segments Jan 2, 2018

Michael Joseph Wdowiak Beijer wrote:

Hi Milan,

Just curious: have they added any improvements regarding merging/splitting segments in DGT-OmegaT? As far as I remember, OmegaT was very bad at this.

Michael


Hi Michael,

Info about segmentation is here

http://www.condak.cz/nove/2017-12/06/cs/04.html

For compatibility is better to have the same segmentation rules and a same "tag-wiping" in all language pairs. In EU DGT are documents mostly translated into more languages.

Since 2012 EU DGT developed the set of segmentation rules for all official languages. You can see standart and DGT-OmegaT files "segmentation.conf".
In standard OmegaT was no rules for Czech language. I have to create it on the fly.
First difference I noticed: ":" is a delimiter in DGT-OmegaT.

Maybe is possible to merge/split segments in OmegaT or DGT-OmegaT, I do not use it.

Milan


 
Milan Condak
Milan Condak  Identity Verified
Local time: 10:21
English to Czech
TOPIC STARTER
The two new videos in documentation DGT-OmegaT Apr 20, 2018

Milan Condak wrote:

I use DGT-OmegaT as a part of Full CAT suite

http://185.13.37.79:8003/index.php/p/wizard/downloads/26/

CAT-Full-Archive.exe - 180.21 MB



In videos you can see SW in action. The second video "Quick Reference" last 45 minutes and cover 38 features of "CAT-Full".

Documentation to DGT-OmegaT

http://185.13.37.79/?q=node/36

Publications (2017)

Article in A Folha about DGT-OmegaT.

Part one

http://185.13.37.79/sites/default/files/folha54_ot_en.pdf (Issue 54 — Summer 2017)

Part two.

http://185.13.37.79/sites/default/files/folha55_ot_en.pdf (Issue 55 — Autumn 2017)

Videos (2018)

About DGT-OmegaT (03:35 min)

http://185.13.37.79/sites/default/files/videos/_VIDEO-ABOUT-DGT-OT-F.mp4

Quick Reference (41:50 min)

http://185.13.37.79/sites/default/files/videos/VIDEO-0-QUICK-REFERENCE-F.mp4

Milan


 
tcordonniery
tcordonniery
France
Local time: 10:21
Segmentation improvements May 17, 2018

Michael Beijer wrote:

Hi Milan,

Just curious: have they added any improvements regarding merging/splitting segments in DGT-OmegaT? As far as I remember, OmegaT was very bad at this.

Michael


As the developer of DGT-OmegaT, let me answer this small question giving a link to our last improvements related to segmentation:
http://185.13.37.79/?q=/node/62

To summarize:

  • We support SRX format, while standard OmegaT only supports their own "conf" format;
  • We provide XSL stylesheet to convert between SRX and OmegaT format (this is the script mentioned by Milan), giving an alternative to people who prefer standard OmegaT even without our features;
  • We have a set of rules including all EU languages, but which is different from standard OmegaT's;
  • You can split or merge segments
  • You can see if segments are in the same paragraph (because it is impossible to merge segments from different paragraph)


Paradoxally, idea about these features and how to implement them come from requests made to the OmegaT team, but even if we sent a message telling that we implement them, and that we are ready to port this to OmegaT 4, they continue to ignore it.

Maybe if somebody tells them again that they have lot of opened RFEs and sometimes there is someone who reads or comments them...

Regards
Thomas Cordonnier, for DGT-OmegaT team.

[Edited at 2018-05-17 14:16 GMT]

[Edited at 2018-05-17 15:10 GMT]


 
tcordonniery
tcordonniery
France
Local time: 10:21
DGT-OmegaT features May 17, 2018


Is my understanding correct that DGT-OmegaT has been designed primarily for teams?


My opinion only: DGT-OmegaT has been designed primarily for in house translators in EU DGT


This is not especially for teams. Only the subproject named Teambase contains features specially dedicated to work in a team.
This project was done because some of DGT users like OmegaT but requested some changes. As a developer, I was allowed to make this changes but not to send them to the community. Some years later, rules suddenly changed and I was required to publish the fork. Of course in the meantime the original OmegaT continued to be developped, that is the reason why the two projects cannot be simply merged. But if you find in our tool features you would like to have in OmegaT 4, then open an RFE in their SourceForge and also send a message in our website (http://185.13.37.79/) : if we can help we will do.


(...) and solves a confidentality of MT; instead of web-based generic USA, Russia, Korea or China MT translation is used in-house MT engine, which produce TMX, which is placed into folder MT.
For outsoursing, where texts are not confidental, can be used standard web-based MT. A connector is separate plugin.


Problem of confidentiality is only one of the reasons why we developped DGT-OmegaT. Internally we totally forbid using external MT engines, so we have developped a simple plugin which fills the MT pane reading a TMX, then we moved Google and other plugins to a separate JAR, only allowed for users outside.
This plugin is very simple and could easily be ported to OmegaT 4, if only they are interested...


How about "Tag wipe": does it do what the name suggests (namely: wipe tags from source files like XLIFF or TMX)?


Only for DOCX files. This format, made by Microsoft, often contains totally useless tags; OmegaT's filter already did some work to remove some of them from the editor, but it is sometimes not enough.
In any case, Tagwipe never removes formatting, or really useful tags.




Teambase

http://185.13.37.79/?q=node/20

Since this is a client/server application, the situation is more complex than for all features discussed until now. The distributed binary version of DGT-OmegaT includes the client plugin, but for the sources and the server part, this has been put in a separate project. Furthermore, the idea of this server came a long time before DGT users were finally interested in it. That is the reason why the server part is in another web site - http://www.silvestris-lab.org/: in fact, Teambase is nothing more than a personnalization of Silvestris/Cyclotis project, not the other way around.


Some news to be added here: we have recently provided new tutorials about how to use Teambase:
http://www.silvestris-lab.org/node/52

Actually this is not compatible with standard OmegaT; but the patches to make it compatible are already written.

Regards
Thomas Cordonnier


[Edited at 2018-05-18 11:55 GMT]

[Edited at 2018-05-18 11:56 GMT]


 
Didier Briel
Didier Briel  Identity Verified
France
Local time: 10:21
English to French
+ ...
We are not ignoring it May 18, 2018

tcordonniery wrote:
Paradoxally, idea about these features and how to implement them come from requests made to the OmegaT team, but even if we sent a message telling that we implement them, and that we are ready to port this to OmegaT 4, they continue to ignore it.

We are not ignoring you, and you know we are in touch.

We have already integrated an important feature
Replacement with regex capture groups and case manipulation
https://sourceforge.net/p/omegat/feature-requests/953/
and several patches from you.

Maybe if somebody tells them again that they have lot of opened RFEs and sometimes there is someone who reads or comments them...

We are perfectly aware of our RFEs and whenever they are commented.

Didier


 
Didier Briel
Didier Briel  Identity Verified
France
Local time: 10:21
English to French
+ ...
TagWipe is now available as a Groovy script May 18, 2018

tcordonniery wrote:


How about "Tag wipe": does it do what the name suggests (namely: wipe tags from source files like XLIFF or TMX)?


Only for DOCX files. This format, made by Microsoft, often contains totally useless tags; OmegaT's filter already did some work to remove some of them from the editor, but it is sometimes not enough.
In any case, Tagwipe never removes formatting, or really useful tags.

We have now ported TagWipe as a Groovy script.
Port TagWipe from DGT-OmegaT project as an OmegaT script
https://sourceforge.net/p/omegat/feature-requests/1384/

Didier


 
tcordonniery
tcordonniery
France
Local time: 10:21
Re: We are not ignoring it May 18, 2018

Maybe if somebody tells them again that they have lot of opened RFEs and sometimes there is someone who reads or comments them...

We are perfectly aware of our RFEs and whenever they are commented.

Didier


Didier, I was just refering to the following RFE : https://sourceforge.net/p/omegat/feature-requests/3/

Yes, I know that this is a very old RFE but you had said that you receive a notification when somebody comments an existing RFE, do you?
Seeing no modification since I sent the comment, I did not know whenever you were not interested or simply not notified about this.

The message which I was replying to is about this topic (cutting / merging segments). And I had already answered to a similar message in your users list. That should be enough to say that there are people interested in this feature.

Now if you look again in our web site, you may see that I added a symbol to mark features for which I have found old RFE which is covered by my feature. I know that I did not tell you before, that's why I mention it here. When you see the symbol, if you click on the description of the feature, which you probably already read, but if the symbol is here, this probably means that I added a link to one of your RFEs later.

In your discussion list you told that you did not find something interesting in our code, that I can understand, but you also told that people can open RFEs and then I can collaborate on RFE basis. Does it apply also for old RFEs ?

Thomas



[Edited at 2018-05-18 11:48 GMT]

[Edited at 2018-05-18 12:29 GMT]


 
Didier Briel
Didier Briel  Identity Verified
France
Local time: 10:21
English to French
+ ...
We do monitor all RFEs May 18, 2018

tcordonniery wrote:

Maybe if somebody tells them again that they have lot of opened RFEs and sometimes there is someone who reads or comments them...

We are perfectly aware of our RFEs and whenever they are commented.

Didier


Didier, I was just refering to the following RFE : https://sourceforge.net/p/omegat/feature-requests/3/

Yes, I know that this is a very old RFE

That doesn't matter.

but you had said that you receive a notification when somebody comments an existing RFE, do you?

I do.
From time to time, they don't reach me, but you have to remember I am not the only OmegaT developer.

Seeing no modification since I sent the comment, I did not know whenever you were not interested or simply not notified about this.

Or simply didn't have time to act on it.

The message which I was replying to is about this topic (cutting / merging segments). And I had already answered to a similar message in your users list. That should be enough to say that there are people interested in this feature.

That people are interested by something doesn't mean we have to act on it immediately, or even at all.

In your discussion list you told that you did not find something interesting in our code,

That's not how I would phrase it. In some features, you made some choices which I/we would have made differently.

that I can understand, but you also told that people can open RFEs and then I can collaborate on RFE basis.

Indeed. And that's what happened already for a feature.

I can understand you find things too slow, but expressing it is not going to make it faster.

Does it apply also for old RFEs ?

It applies to any RFE, even that one (number 3 from 2004).

But whatever the popularity of an RFE, that doesn't mean that it has to be implemented fast, or at all.

We (the OmegaT developers) have all our personal life and diverse motivations, and these have to be taken into account.

Didier


 
tcordonniery
tcordonniery
France
Local time: 10:21
Time or objections May 18, 2018

you have to remember I am not the only OmegaT developer.


I said "you" meaning the core team. Sorry that English language is too poor to make the distinction.

That people are interested by something doesn't mean we have to act on it immediately, or even at all.


No, but the fact that a question often comes in the user list could mean that it may have an higher level of priority. Or that you could say, almost, if it is only a question of time or if you have serious objections against a specific feature or the way a user proposes to implement it (as RFE is written by users, it can happen, but then it is preferable to say it)

that doesn't mean that it has to be implemented fast, or at all.


Of course, that is the reason why I expect answers from you. Remember that my proposal is not to give you a list of features to be implemented but a list of features which are implemented, then I am ready to port them to last version, the only question, for each of them, is whenever I should make the effort or not. When I write in an RFE, it is not to say "please implement it" but "test my implementation and tell me if I should port it to OmegaT 4 or not". And I am even ready to accept more nuanced opinion (look at what I wrote yesterday in the RFE for glossaries, for example), really I don't say "all or nothing".

you made some choices which I/we would have made differently.


Do you mean from a technical point of view, in the code, or for the user interface?

Thomas


 
Didier Briel
Didier Briel  Identity Verified
France
Local time: 10:21
English to French
+ ...
Mainly the user interface May 19, 2018

tcordonniery wrote:

you made some choices which I/we would have made differently.


Do you mean from a technical point of view, in the code, or for the user interface?

The user interface (e.g., the search window). But of course, it has sometimes technical implication.

Didier


 
Michael Beijer
Michael Beijer  Identity Verified
United Kingdom
Local time: 09:21
Member (2009)
Dutch to English
+ ...
Thanks Thomas! Jan 3, 2019

tcordonniery wrote:

Michael Beijer wrote:

Hi Milan,

Just curious: have they added any improvements regarding merging/splitting segments in DGT-OmegaT? As far as I remember, OmegaT was very bad at this.

Michael


As the developer of DGT-OmegaT, let me answer this small question giving a link to our last improvements related to segmentation:
http://185.13.37.79/?q=/node/62

To summarize:

  • We support SRX format, while standard OmegaT only supports their own "conf" format;
  • We provide XSL stylesheet to convert between SRX and OmegaT format (this is the script mentioned by Milan), giving an alternative to people who prefer standard OmegaT even without our features;
  • We have a set of rules including all EU languages, but which is different from standard OmegaT's;
  • You can split or merge segments
  • You can see if segments are in the same paragraph (because it is impossible to merge segments from different paragraph)


Paradoxally, idea about these features and how to implement them come from requests made to the OmegaT team, but even if we sent a message telling that we implement them, and that we are ready to port this to OmegaT 4, they continue to ignore it.

Maybe if somebody tells them again that they have lot of opened RFEs and sometimes there is someone who reads or comments them...

Regards
Thomas Cordonnier, for DGT-OmegaT team.

[Edited at 2018-05-17 14:16 GMT]

[Edited at 2018-05-17 15:10 GMT]


I have recently been experimenting with DGT-OmegaT again, and so will be trying out the split and merge functionality you mentioned in my next project!

Michael


 
Michael Beijer
Michael Beijer  Identity Verified
United Kingdom
Local time: 09:21
Member (2009)
Dutch to English
+ ...
testing new split or merge segments functionality in 3.2 beta Jan 4, 2019

OK, so I read this: http://185.13.37.79/?q=/node/62

and am trying to merge or split, but although the dialogue appears, when I click OK, nothing seems to happen. What am I doing wrong here? This is on 3.2.

Michael


 
tcordonniery
tcordonniery
France
Local time: 10:21
Re: testing new split or merge segments functionality in 3.2 beta Jan 18, 2019

Hi Michael

You are absolutely true, there was a bug in the last release: the changes were not correctly saved.
I just published new releases (3.1 update 9 + 3.2 update 5) which solves the problem.

Now what should happen is that once you click on "OK" the application should require from you to reload the project, and then the changes are applied.
Note that this requires that "sentence-level segmenting" is enabled, these feature will not automatically activat
... See more
Hi Michael

You are absolutely true, there was a bug in the last release: the changes were not correctly saved.
I just published new releases (3.1 update 9 + 3.2 update 5) which solves the problem.

Now what should happen is that once you click on "OK" the application should require from you to reload the project, and then the changes are applied.
Note that this requires that "sentence-level segmenting" is enabled, these feature will not automatically activate them for you.
In any case, even if it should work with segments containing tags, before reporting again please test also in tagless segments, so that we can be sure that problems come from here.

Regards
Thomas
Collapse


 
Pages in topic:   [1 2] >


There is no moderator assigned specifically to this forum.
To report site rules violations or get help, please contact site staff »


DGT-OmegaT






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! »
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 »