Meeting minutes

Hello,

Please find below a summary of today’s discussion:

Attendance: Rodolfo, Yoshito, DavidF, Lucía

I. Administration

Regrets: Bryan

R: I move to Approve 20 April meeting minutes. https://lists.oasis-open.org/archives/xliff/202104/msg00026.html

L: I second.

R: Meeting minutes approved.

B. OASIS web site updates - Rodolfo https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=xliff

-FAQ Section. Liaison information (David)

Df: I have updated the information on that section. The markup is done, but I do not if the ul/ol markup works.

R: It works.

R: There is still the problem with the mailing list, they links are broken. Chet is aware of the problem, but it has not be fixed yet. The UI is broken, because it does not get the CSS and images.

DF: the most important thing is that they see how to join the list. It is only the archive that is broken, we can replace the link with the markmail.

R: The problem is that this is the official.

Df: we can do that for the time being until the fix the official.

R: If you follow the links from the previous meeting, it works .

Df: (shares the markmail’s link: https://markmail.org/search/?q=xliff#query:xliff%20list%3Aorg.oasis-open.lists.xliff-comment+page:1+state:facets )

R: to send an email I do it from the internal system. I cannot do it from my email address, because kavi rejets my email. The email management list is broken.

Df: We had some problems with emails and we were able to address it with Chet.

R: I know what is wrong in my case. When I send it from outlook, it is rejected.

II. Technical work

A. Migration guide. Rodolfo.

B. Test suite correction.

R. I did not make progress on that because I was busy working on the validation filter, I was working on fixing some errors. The validator has been improved, I am working on a new version that will be released soon.

R: David, are you working on the schematron?

Df: I can work on it on the time being.

R: have you seen Lucía’s document?

Df: I see, that would be helpful.

R (Rodolfo shows in his screen the document and explains one example where there is a problem with schematron).

Df: We set some general rules for ids in extensions. I will have a look at the file. I have had access from Lucía, this is very helpful and I will look into the schematron’s errors.

R: Yes, there are problems that are not detected in the Schematron.

R: It would be great if you can work on the schematron.

R: I found a small problem with the data element. And it is affecting validation at some point. The element has a required id. And it says that it should be unique across all the data in unit. But you can only have one data in unit, but you can have several matches with multiple data.

Df: we had several discussions on 2.0 approval. When matches reuses the core, they are not considered part of core.

R: I understand but it is not written anywhere.

Df: that is a good action point for 2.2 to clarify this aspect. This is something that would affect other aspects, it is good to clarify it. I think it is logic, but we need to clarify it. I think we can make it a 2.2 issue on github. If you assign it to me, I can take care of it. I can create an informative note explaining that.

R: Ok, I will create an action item for this.

R: Last week I was saying to say that I starting working on the template of 2.2, I was able to upload it to github. https://github.com/oasis-tcs/xliff-xliff-22/tree/master/xliff-22 (Rodolfo shows the output the pdf).

Df: On the repo you can find the OASIS stylesheet.

R: I tried to use the stylesheet. They are not complete. They did not work.

Df: If we are trying to build in the ISO compatible or OASIS compatible view.

R: I will try to use the OASIS stylesheet.

Df: I suggest that we use for the ISO view. There will be a review, it will help to have it in the compatible format.

R: We will have it in the compatible format, but my intention was to fix the links. In particular, there was issue with some entities.

Df: the previous version should be 2.1.

R: The problem for me is that the entity was using links to the schema to 2.0.

Df: I used separated entities for this issue.

R: Ok. I will fix that.

B. MessageFormat (XLIFF 2 module)

https://docs.google.com/document/d/1D702OBAzT-Crb9XXUiZYJnFO9Yq5duRy4Zc3Br6JwRU/edit
L: David, are you in contact with them?

D: it shows how the module can be constructed. At present it only supports two features out of dozens. It had a previous version that was commented by Yves. I need to talk to Mihal and try to convince to submit it officially to the tc.

R: I read it, it is interesting. It provides good ways to express the message format information. I saw just one issue related with the standard, I think it was with some attributes, nothing serious. The problem I see with this is who will implement it? A cat tool has a specific UI. And you can translate 2.0 files in a CAT tool. But it does not communicate the values that you can put in the extensions. If you have something coming from the format. I do not know how a cat tool would show that information.

Df: there are a number of features that would need UI changes. And that is why we were considering creating a rendering guide. The structure uses core units, and any tool should be able to display them.

R: Yes, that is what I meant.

Df: The new one is implementing on unit level.

R: Yes, but you have to tell the translator that this is for one, this for many, etc.

Df: Yes, I see, you need to display metadata. You can use notes to include information.

R: yes, tools display notes.

Df: Until cat tools support that, we can tell to use notes.

R: that takes us to the problem we were mentioning with Yoshito with the notes.

Df: I disagree here.

R: I agree we disagree.

Df: In 2.0 before pointing was introducing from matches to core. There was a revision that it was the other way around. But it was a nightmare.

R: Yes, that is why that I was suggesting to point to the segment.

Df: From matches you can refer to unit.

R: That’s what I ended doing it.

Df: It is not encouraged. In case of notes, you can use the comment annotation.

R: Yes, but that means changing the source.

DF: Annotations are not supposed to be changes, there are just annotations. Annotations are just for metadata.

R: It is introducing elements that are inside instead of outside.

R: You open a document, you have a source, and a translator wants to have a note, you need to modify the source. You add a new element. You might call it annotation but it is an XML element.

R: How would you point a regular note to a regular source element?

Df: If you do not like the comment annotation. You can use an extensible.

R: I do not like it because it breaks interoperability

Meeting adjourned.

xliff message