search

IDPF / epub-testsuite

A collection of EPUB documents to systematically test EPUB Reading System conformance
Other
79 stars 26 forks source link

Reorganization (new branch) #28

Closed vincent-gros closed 11 years ago

vincent-gros commented 11 years ago

I have created a new branch called "reorganization" in order to prepare the new structure that we discussed last year, which we can find in the wiki page written by Markus.

As a start, there are two subfolders: reflowable and fixed-layout. I am not sure if this is useful, but this way the organization seems easy to understand.

The 001, 003 and 005 are renamed 0100, 0120 and 0230. No more changes except the name of the folder (and the dc:title and dc:identifier).

For now, I have choose to merge 002-rtl and 004-he into 0130, because both have the same page-progression-direction. Is there a problem with this choice, @OriIdan ?

Do not forget that this is another branch, and this is not the main branch. I just want to illustrate the new structure, in ordre to show every problem we could have.

OriIdan commented 11 years ago

I think people would like to see Hebrew/Arabic separate from Japanese.

On Mon, Jan 21, 2013 at 6:57 PM, vincent gros notifications@github.comwrote:

I have created a new branch called "reorganization" in order to prepare the new structure that we discussed last year, which we can find in the wiki page written by Markus.

As a start, there are two subfolders: reflowable and fixed-layout. I am not sure if this is useful, but this way the organization seems easy to understand.

The 001, 003 and 005 are renamed 0100, 0120 and 0230. No more changes except the name of the folder (and the dc:title and dc:identifier).

For now, I have choose to merge 002-rtl and 004-he into 0130, because both have the same page-progression-direction. Is there a problem with this choice, @OriIdan https://github.com/OriIdan ?

Do not forget that this is another branch, and this is not the main branch. I just want to illustrate the new structure, in ordre to show every problem we could have.

— Reply to this email directly or view it on GitHubhttps://github.com/mgylling/epub-testsuite/issues/28.

vincent-gros commented 11 years ago

@OriIdan, I believe that this is a discussion we already have had together (by e-mail, in late november). We agreed to have a new simpler structure, based on technical requirements, and thus japanese and hebrew will be in only one file (because technicaly, both japanese language and hebrew language use the same page-progression-direction value).

At the end of this discussion, Markus had written the wiki page I mentionned and I used for this new branch.

If we want to see the use of a specific language, I believe the EPUB3 samples project (containing your sample in hebrew, my sample in arabic and others in japanese) is more appropriate.

Don't you think?

OriIdan commented 11 years ago

Yes this is a discussion we had in November and I recall we agreed to have Japanese in one file and Hebrew in another file. I do not recall we agreed to join them. I started the discussion and said I would like to see everything together in one file and then during the discussion we found out that it will not be efficient to have everything in one file.

On Tue, Jan 22, 2013 at 12:33 AM, vincent gros notifications@github.comwrote:

@OriIdan https://github.com/OriIdan, I believe that this is a discussion we already have had together (by e-mail, in late november). We agreed to have a new simpler structure, based on technical requirements, and thus japanese and hebrew will be in only one file (because technicaly, both japanese language and hebrew language use the same page-progression-direction value).

At the end of this discussion, Markus had written the wiki page I mentionned and I used for this new branch.

If we want to see the use of a specific language, I believe the EPUB3 samples project (containing your sample in hebrew, my sample in arabic and others in japanese) is more appropriate.

Don't you think?

— Reply to this email directly or view it on GitHubhttps://github.com/mgylling/epub-testsuite/issues/28#issuecomment-12521249.

vincent-gros commented 11 years ago

(See #26, @OriIdan said) I think that because of the different characters used in Japanese and Hebrew/Arabic, it might intimidate users and some will not find the Hebrew tests. This is not a pure technical reason but more of a user interface reason.

I am sorry, but I must insist.

The project is "A collection of EPUB documents to systematically test EPUB Reading System conformance".

If we keep this idea in mind:

What 0130 is doing for? According to the wiki page: epub30-test-0130.epub Tests for page-progression-direction in reflowable (OPF, CSS) Potentially: more extensive/advanced tests for writing modes and layout (from Koike's existing content) if needed

0130 is not about japanese or hebrew, but about page-progression-direction and writing-modes, and in order to check this we use japanese and hebrew.

Users will not be intimidate by japanese content (or even by hebrew), because this is not a "content-to-read", but a "content-to-test". The main test is page-progression-direction, and both japanese and hebrew are ok with the value rtl. Japanese can also check the vertical writing mode, and hebrew the horizontal right-to-left writing mode.

This is exactly what I have started to do.

Again, I am sorry to insist, but I still believe that 0130 must have both japanese and hebrew in order to be complete.

OriIdan commented 11 years ago

I agree technically. But from knowing how people react I am sure they will be intimidated even though you are right the language does not really matter since this the content does not matter. Until today I have used many examples in many languages to test readers but I know that I showed examples in Japanese to non Japanese people and they did not even look at it.

Japanese is very different in the fact that it is not only right to left, it is top to bottom, therefore seem strange.

However since we are democratic (today is the elections in Israel :-) ) then if most of the (small) group think it should be together, I'll except the voice of the majority :-)

On Tue, Jan 22, 2013 at 10:29 AM, vincent gros notifications@github.comwrote:

(See #26 https://github.com/mgylling/epub-testsuite/issues/26, @OriIdanhttps://github.com/OriIdansaid) I think that because of the different characters used in Japanese and Hebrew/Arabic, it might intimidate users and some will not find the Hebrew tests. This is not a pure technical reason but more of a user interface reason.

I am sorry, but I must insist.

The project is "A collection of EPUB documents to systematically test EPUB Reading System conformance".

If we keep this idea in mind:

  • 0100 is not a file in english: this is a reflowable content with left to right page progression direction.
  • 0120 is not a file in english: this is a reflowable content with ltr page progression direction and a media overlay
  • 0130 is not a file in japanese and in hebrew: this is a reflowable content (whatever it is) with right to left page progression direction.

What 0130 is doing for? According to the wiki page: epub30-test-0130.epub Tests for page-progression-direction in reflowable (OPF, CSS) Potentially: more extensive/advanced tests for writing modes and layout (from Koike's existing content) if needed

0130 is not about japanese or hebrew, but about page-progression-direction and writing-modes, and in order to check this we use japanese and hebrew.

Users will not be intimidate by japanese content (or even by hebrew), because this is not a "content-to-read", but a "content-to-test". The main test is page-progression-direction, and both japanese and hebrew are ok with the value rtl. Japanese can also check the vertical writing mode, and hebrew the horizontal right-to-left writing mode.

This is exactly what I have started to do.

Again, I am sorry to insist, but I still believe that 0130 must have both japanese and hebrew in order to be complete.

— Reply to this email directly or view it on GitHubhttps://github.com/mgylling/epub-testsuite/issues/28#issuecomment-12534806.

mgylling commented 11 years ago

I agree with Vincent that this is the way to go. Like Vincent points out, we should keep in mind that these are tests for atomic features of EPUB, not tests for "languages" (whatever that means). There might be needs to go for language-specific test collections in the future possibly including heuristics and qualitative/judgement measures, but that is not for these files (and perhaps not even for this project).

Also, for users who are not interested in running the entire test suite, but just checking some subset of it: keep in mind that the navigation document will help them move to where they want. Having separate files doesn't really add much beyond that IMO.

/markus

On Jan 22, 2013, at 9:49 AM, Ori Idan notifications@github.com wrote:

I agree technically. But from knowing how people react I am sure they will be intimidated even though you are right the language does not really matter since this the content does not matter. Until today I have used many examples in many languages to test readers but I know that I showed examples in Japanese to non Japanese people and they did not even look at it.

Japanese is very different in the fact that it is not only right to left, it is top to bottom, therefore seem strange.

However since we are democratic (today is the elections in Israel :-) ) then if most of the (small) group think it should be together, I'll except the voice of the majority :-)

On Tue, Jan 22, 2013 at 10:29 AM, vincent gros notifications@github.comwrote:

(See #26 https://github.com/mgylling/epub-testsuite/issues/26, @OriIdanhttps://github.com/OriIdansaid) I think that because of the different characters used in Japanese and Hebrew/Arabic, it might intimidate users and some will not find the Hebrew tests. This is not a pure technical reason but more of a user interface reason.

I am sorry, but I must insist.

The project is "A collection of EPUB documents to systematically test EPUB Reading System conformance".

If we keep this idea in mind:

  • 0100 is not a file in english: this is a reflowable content with left to right page progression direction.
  • 0120 is not a file in english: this is a reflowable content with ltr page progression direction and a media overlay
  • 0130 is not a file in japanese and in hebrew: this is a reflowable content (whatever it is) with right to left page progression direction.

What 0130 is doing for? According to the wiki page: epub30-test-0130.epub Tests for page-progression-direction in reflowable (OPF, CSS) Potentially: more extensive/advanced tests for writing modes and layout (from Koike's existing content) if needed

0130 is not about japanese or hebrew, but about page-progression-direction and writing-modes, and in order to check this we use japanese and hebrew.

Users will not be intimidate by japanese content (or even by hebrew), because this is not a "content-to-read", but a "content-to-test". The main test is page-progression-direction, and both japanese and hebrew are ok with the value rtl. Japanese can also check the vertical writing mode, and hebrew the horizontal right-to-left writing mode.

This is exactly what I have started to do.

Again, I am sorry to insist, but I still believe that 0130 must have both japanese and hebrew in order to be complete.

— Reply to this email directly or view it on GitHubhttps://github.com/mgylling/epub-testsuite/issues/28#issuecomment-12534806.

— Reply to this email directly or view it on GitHub.

OriIdan commented 11 years ago

Ok, so if the majority agrees we should do it in one file so I agree.

On Tue, Jan 22, 2013 at 2:46 PM, Markus Gylling notifications@github.comwrote:

I agree with Vincent that this is the way to go. Like Vincent points out, we should keep in mind that these are tests for atomic features of EPUB, not tests for "languages" (whatever that means). There might be needs to go for language-specific test collections in the future possibly including heuristics and qualitative/judgement measures, but that is not for these files (and perhaps not even for this project).

Also, for users who are not interested in running the entire test suite, but just checking some subset of it: keep in mind that the navigation document will help them move to where they want. Having separate files doesn't really add much beyond that IMO.

/markus

On Jan 22, 2013, at 9:49 AM, Ori Idan notifications@github.com wrote:

I agree technically. But from knowing how people react I am sure they will be intimidated even though you are right the language does not really matter since this the content does not matter. Until today I have used many examples in many languages to test readers but I know that I showed examples in Japanese to non Japanese people and they did not even look at it.

Japanese is very different in the fact that it is not only right to left, it is top to bottom, therefore seem strange.

However since we are democratic (today is the elections in Israel :-) ) then if most of the (small) group think it should be together, I'll except the voice of the majority :-)

On Tue, Jan 22, 2013 at 10:29 AM, vincent gros notifications@github.comwrote:

(See #26 https://github.com/mgylling/epub-testsuite/issues/26, @OriIdanhttps://github.com/OriIdansaid) I think that because of the different characters used in Japanese and Hebrew/Arabic, it might intimidate users and some will not find the Hebrew tests. This is not a pure technical reason but more of a user interface reason.

I am sorry, but I must insist.

The project is "A collection of EPUB documents to systematically test EPUB Reading System conformance".

If we keep this idea in mind:

  • 0100 is not a file in english: this is a reflowable content with left to right page progression direction.
  • 0120 is not a file in english: this is a reflowable content with ltr page progression direction and a media overlay
  • 0130 is not a file in japanese and in hebrew: this is a reflowable content (whatever it is) with right to left page progression direction.

What 0130 is doing for? According to the wiki page: epub30-test-0130.epub Tests for page-progression-direction in reflowable (OPF, CSS) Potentially: more extensive/advanced tests for writing modes and layout (from Koike's existing content) if needed

0130 is not about japanese or hebrew, but about page-progression-direction and writing-modes, and in order to check this we use japanese and hebrew.

Users will not be intimidate by japanese content (or even by hebrew), because this is not a "content-to-read", but a "content-to-test". The main test is page-progression-direction, and both japanese and hebrew are ok with the value rtl. Japanese can also check the vertical writing mode, and hebrew the horizontal right-to-left writing mode.

This is exactly what I have started to do.

Again, I am sorry to insist, but I still believe that 0130 must have both japanese and hebrew in order to be complete.

— Reply to this email directly or view it on GitHub< https://github.com/mgylling/epub-testsuite/issues/28#issuecomment-12534806>.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/mgylling/epub-testsuite/issues/28#issuecomment-12542896.