<ivan> Meeting: Publishing Business Group telco

<liisamk> scribe+ liisamk

Text Data Mining from Laurent LeMeur

<liisamk> Laurent: TDM is defined as generating information out of data

<liisamk> ...EU copyright directive notes that there copyright exception for scientific research

<liisamk> ...there is also exception for lawfully accessible content

<liisamk> ...everything that is accessible can be mined on the condition that the use of such content has not been

<liisamk> ...expressly reserved by the rights holders

<liisamk> ...must opt-out by rights holders from the copyright exception

<liisamk> ...there will be national laws soon to allow opt-out by any way you want

<liisamk> ...member states have until June 2021 to implement national laws

<liisamk> Ivan: there is a general directive, but not a standard way to opt out

<liisamk> Laurent: directive does not care for technical issues of how to implement

<liisamk> ...there is an initiative to find some standard

<liisamk> Christina: there is a general directive in the national legislation and it would be useful to not have every country have their own spec

<liisamk> Laurent: not sure national laws would specify a solution

<liisamk> ...need to provide the TDM sector to work together with publishing

<liisamk> ...some members of the publishing sector from France and Italy drafted requirements and use cases

<liisamk> ...are willing to find a technical solution that is non-binding

<liisamk> ...simple list of requirements to have how rightsholder can indicate rights on a web page he controls

<liisamk> ...can indicate on each resource he countrols

<liisamk> ...can indicate whether a license is available, how it can be reached, what is the machine-readable format

<Zakim> tzviya, you wanted to ask who "we" is?

<liisamk> Tzviya: who is we?

<liisamk> Laurent: European Federation of Publishers

<liisamk> Leonardr: Do you mean who owns the rights, not just in possession of?

<liisamk> Laurent: yes, it is about the rightsholder

<liisamk> ...Use cases, from a newspaper with TDM opt-out

<liisamk> ...from a publisher providing free extreacts of ebooks with opt-out

<liisamk> ...from an academic platform providing HTML and PDF files and possibility to get license

<liisamk> ...a subscription service provides EPUB with opt-out

<liisamk> ...so far 4 use cases

<liisamk> ...exisiting initiatives, toolbox

<liisamk> ...ACAP, RightsML, ECP/Copyright Hub, Article Sharing Framework, Crossref TDM service

<liisamk> ...goal is not to choose, but to use ideas from these

<liisamk> Leonard: missing things like dublin core, schema.org and others that are in use today

<ivan> http://localhost:8001/TR/odrl-model/ may also be interesting

<ivan> sorry https://www.w3.org/TR/odrl-model/

<Zakim> tzviya, you wanted to mention NISO ALI

<liisamk> Tzviya: NISO is working on an article licensing model

<liisamk> ..ck ivan

<liisamk> Laurent: where we are now, there is work to reach broader support from EU publishing sector

<liisamk> ...there is a draft doc available on demand

<liisamk> ...working doc with information to be shared

<liisamk> ...big issue is to have TDM technical guys to have listen

<liisamk> ...best to design a solution and recommend a solution to all EU publishers

<liisamk> ...PBG is on the publishing side, but inside the technical work

<liisamk> ...good for PBG to put together these people

<liisamk> ...two reasons, publishers willing to join, and W3C people can help talk to TDM users

<liisamk> ...two orgs- W3C, Business Software Alliance on the other side both contacted

<liisamk> Cristina: when explain the issue, look for the quick way to find a way to address

<liisamk> ...only have until until June 2021

<laurent_> Work document open on demand at https://docs.google.com/document/d/1NwtWv_ESS4ZhaHDHWnDSQSJKHrfRmLNjDTPdwNE3aeA/edit#

<liisamk> Ivan: we know from experience with GDPR that Europe initiatives affect countries outside Europe as well

<liisamk> ...we have to take this into account

<liisamk> ...do we see something similar coming up around this or is it very EU local?

<liisamk> Laurent: it can't remain totally local

<liisamk> ...US TDM user wants to mine on a US server, but the copyright holder is in Europe and it is covered under this law

<liisamk> Ivan: must be emphasized if we reach out to others in the world

<liisamk> Laurent: the data and server localization is a bit weird

<liisamk> ...it is like the GDPR

<liisamk> leonardr: interesting technical point, if opt-out can be applied to any content anywhere

<liisamk> ...miner has to understand where you are and your data is hosted, where technical and legal intersect

<liisamk> Tzviya: will talk to Wiley legal about scholarly and other types of content and will ask questions of their legal team

<liisamk> Laurent: EDRLab is acting as a technical helper to help with both sectors, posted a doc above in the irc

<liisamk> Ivan: decision of the business group- business group or a community group could publish a document to co-brand

<liisamk> ...a suggestion for this and a technical solution

<liisamk> ...a dedicated CG and in the BG would give some attraction to this

<liisamk> ...other companies that are relevant - who are the really big players besides Google?

<liisamk> ...not an area where Facebook would be present

<liisamk> ...do we have somewhere a list of who the main players are?

<liisamk> Laurent: will try to find the info

<liisamk> Ivan: if we decide, we could do something with a CG or BG

<liisamk> leonardr: the type of company who might be impacted are AI players who are building models

<liisamk> Laurent: next steps, feedback and in January must start working on a solution

<liisamk> ...if we don't find a technical guy from the other side, will be a need to find a technical solution in the FEP

<liisamk> ...only way to have a solution ready for June

<liisamk> Bill_Kasdorf: have you presented to IPTC?

<liisamk> Laurent: yes, but not much response

use cases for FX + Reflow

<liisamk> wendyreid: would love to get more use cases

<liisamk> ...image heavy but also have a lot of text

<liisamk> ...want good reflowable text as well for art, cooking

<liisamk> Tzviya: had mentioned a CSS crash course with Jen, doable with great CSS

<liisamk> ...maybe have to look at reading system support

<liisamk> Laurent: tried a prototype in readium and it never worked

<liisamk> ...responsive design might work with modern css

<liisamk> ...don't you think we should pair publishers who want to do and others who can

<liisamk> ...start by looking at publishers and responsive layout with flexbox

<liisamk> ...try to convince with the content on the market

<liisamk> Liisa: use cases continue to work on

<liisamk> Garth: interesting, and if reading systems implemented it would just work

<liisamk> ...flex box might be close

<liisamk> ...wouldn't get you a bleed to the edge because there is no css to do that

<liisamk> ...images to the edge would be hard to impossible

<liisamk> ...could get fixed + reflow to get done more easily

<liisamk> wendyreid: could be easier to get started with just FXL + reflow

<liisamk> ...flexbox would be great

<liisamk> ...more work for reading systems would run quickly into pushback because the reading system is beholden to the users

<liisamk> ...can get complicated fast

<liisamk> ...find things that can work for all reading systems

Japanese font issue

<scribe> scribenick: Karen

Liisa: there is an issue that came up on the Asian PubBG call recently on font issues

[missed set up]

scribe: I suggested they take this up as a localized issue with a reading system
... most people have given up on a single font to work across reading sytems
... we have seen that fonts are very much reading-system specific; no true consistency
... but in conversations recently about fonts, there are issues on fonts around obfuscation, lack of consistent support
... come to this group and see if there are other font issues that we should discuss that are affecting other people's businesses?
... anyone?

<liisamk> Karen: not sure if you have been participating in the CG working on font issues and having a conversation with Chris Lilley

<liisamk> Leonard: reach out to Chris because publishing and webfonts should come together

<liisamk> ...font licensing has changing in the last few years and timing is good

https://www.w3.org/community/font-text/participants

<liisamk> Garth: difficult problem for reading system and vertical japanese font

Liisa: our friends in Japan were hoping it was a wider bug than something localized
... and seeking some help
... Some housekeeping things
... we would like to invite the co-chair of the MathML co-chair to our next meeting, or a January meeting
... I think many folks in our community are interested in MathML

<Bill_Kasdorf> +1 to MathML

Liisa: I see head shaking of interest
... and reminder for fund-raising for next phase of EPUBCheck
... Bill has been helping with fundraising
... Next Asia meeting is Dec 8
... and this group is Dec 15
... most people want 2020 to end and hope for some change in the world that is better
... enjoy if you are celebrating Thanksgiving

This is scribe.perl Revision of Date 
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/Wiley/Wiley legal/
Succeeded: s/@/15/
Present: wendyreid cristina guillian-hetzler Karen ivan Garth Bill_Kasdorf
Found ScribeNick: Karen
Inferring Scribes: Karen

WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)