Meeting minutes
<robin> hi!
robin: I don't think it's necessary to scribe the slides
Ralph: we are planning to record the introduction
… we would like to record the discussion, but if anyone objects we won't do it
dsinger: what would you do with the discussion recording?
Ralph: we'd post it on the website
… I'm hearing objections, so we won't record the discussion.
robin: [starting presentation]
… [restarts zoom]
robin: can you hear me now
all: yes
… welcome to "media publishers of the web unite"
… I'll present a few slides quickly
robin: these are my opinions, and the slides are opinionted
… these do not represent the views of my employer
… the perception is that publishers don't show up for standards, and then complain
… and then other people claim to speak for publishers
… what do we actually want?
… ponies and cool website features :)
… but we need two other things
… 1. we need to be trustworthy
… 2. we need to make money.
… journalism is expensive
… huge volumes of work go into most articles
… we have a financial model based on betraying the trust of our readers
… what's up with all that tracking?
… news publishers are the worst for 3rd party trackers
… people are angry because they have expectations of privacy
… and they expect browsers to work for them
… readers *do* trust the publishers
… but that trust is not transitive--people don't trust the 3rd parties
… people talk about the supply chain in ad tech
… it's astoundingly complex
… most publishers can't control this supply chain when they involve 3rd parties
… tracking-based ads are a problem, but they decrease trust
… in the publisher, and thus decrease trust in the news
… which harms democracy
… it's just not working.
… newsrooms had been cut in half
… even before covid
… users are fighting back. adblockers...
… how can we fix this?
… throw in transparency and choice? that's just cookie consent dialogs. That's not an improvement.
… we need a better web.
… we need technical means to move to a saner model.
… privacy can increase audience value.
… this can only work if browsers are trustworthy
<wseltzer> GPC
robin: if your website remove trackers, but the browser is tracking your reading history you haven't solved the problem
… we need the priority of consituencies
… we have it now in exactly the wrong order
wseltzer: what is SSO?
robin: single sign on
… what about the content?
… content used to be contextualized; articles were in order, you knew the provenance
… now everything looks the same
… context is what anchors trust
… if your articles are removed from the context, you lose the trust relationship
… content aggregation today is problematic
… decontextualized content is less trustworthy
… content aggregation provides a worse user experience, increases tracking, and lower the revenue of journalism
… how can we fix this?
… we put together a document on CAT (content aggregation tech)
… the web is pretty good at this, but there are incremental improvements
<wseltzer> CAT
robin: we can better indicate provenance
… this was a high-speed introduction. thanks for putting up with my ranting :)
… there's a bunch of publisher-specific problems
… we want publishers to speak up and have opinions
… here are some cool links
… and the floor is entirely yours. Speak up! Complain! Push back!
jeff: thanks Robin
… for that mildly provocative set of thoughts
… could you say a bit more?
<robin> Global Privacy Control (GPC) spec https://globalprivacycontrol.github.io/gpc-spec/
<robin> GPC site https://globalprivacycontrol.org/
<robin> How The New York Times Things About Your Privacy https://open.nytimes.com/how-the-new-york-times-thinks-about-your-privacy-bc07d2171531
<robin> Content Aggregation Technology (CAT) spec https://nytimes.github.io/std-cat/
<robin> The News Provenance Project https://www.newsprovenanceproject.com/
jeff: most of the preso was a list of problem on the web
<robin> What If Every News Photo on Social Media Showed Contextual Information? https://open.nytimes.com/what-if-every-news-photo-on-social-media-showed-contextual-information-8936cf4e8c45
<robin> How Publishers Can Use Metadata to Fight Visual Misinformation https://rd.nytimes.com/projects/how-publishers-can-use-metadata-to-fight-visual-misinformation
jeff: the silver bullet was content aggregation technology
… how does that fix the problems?
robin: there are no silver bullets
… one section was privacy issues, one was CAT issues
… CAT is not a solution to privacy problems
… it's only a proposal to move towards a solution in the content space
… but it has a great acronym :)
robin: are there other publishers here? I see at least one
<robin> ack
tzviya: I'm with wiley; we are a scholarly and higher ed publisher
… yesterday we talked about the history of EPUB, and focused on the trade publishing
… we rely on third parties for distribution
… we're not reliaant on ad revenue
… but we have no control over distribution
… so there's overlap
… so media publishers should collaborate. There's common ground around provenance.
<Zakim> dsinger, you wanted to ask about the alternatives
dsinger: Dave Singer, Apple
… Q: it seems our alternatives are not good
… Robin has commented on content aggregation not being ideal
… I'm often sent to news sites by my friend
… and then I might have to create an account, login, and pay
… which I'm unlikely to do
… are micropayments an alternative?
robin: briefly... two things
… there's payment and identity
… we need better identity management
… we could log in to provide relationship-building affordance taht would be good
… micropayments would be difficult for us--we can't put a price on a single article
… but there might be hybrid models
… we could have an increment of paywalls... a few free hits, a few micro hits, then a subscriber
AramZS: lots of people are thinking about what publishers can do in w3c
… you can bring engineering staff to look at other parts
… at WaPo we've brought people into conversations about CSS, HTML
… it's good to be ahead of the wave on these standards
… even listenting about how we can build better websites is helpful
… CSS is important to our business, too
robin: I focused on broad areas of work
… the way we render our headlines involve a headless browser to measure before re-tweaking
… that should be solved at the CSS level
<Zakim> wseltzer, you wanted to discuss web-adv conversations
wseltzer: I wanted to mention the improving web advertising Business Group
… ways to support privacy in targeted advertising
… there are lots of W3C working groups that might be relevant
… it could be helpful to have CG or IGs to identify issues, like the media/entertainment group for that area
<Zakim> cpn, you wanted to comment on provenance projects
cpn: Chris Needham, BBC
… and co-chair of the media IG
… my Q was about the provenance issues
… origins, trusted news... is there value on consolidating these efforts?
… second Q: what's the role of W3C in these initiatives? are there technologies to look at?
robin: I think the news provenance project is an issue of origins
<AramZS> link to project origin?
robin: that's been my feedback... there's a lot of companies working on this
… this requires standards. Don't know if it's one standard or a series of standards.
… claims/credentials WG
… there might be some convergance, and maybe need for new stuff
dmarti: robins' point on SSO
… as the user experience for SSO bevcomes easier, the ability to get ??? gets harder
<robin> Project Origin: http://www.originproject.info/
dmarti: so we need control and consent management, the global privacy control
… and there's another consent management proposal
joshua_koran: i heard two kinds of trust--trust in content, which is accuracy and provenance
… and we don't want the publisher to betray their users
… what about advertiser's trust? Is that related to pseudononymous IDs?
robin: advertisers could benefit from a higher-trust environment
… there's a huge amount of fraud in ad tech
… intermediaries don't want to fix this
… having a priority of constituencies that involves brands would be valuable
… I don't think pseudononymous IDs can solve this
… they work for domains with strong governance, like clinical trials
… strict rules, systematic audits, layers of bureaucracy
… we don't have that in ad tech
… I don't thing we want that level of bureaucracy for a universal ID
… there are better ways of proving something was delivered
… there are probablistic ways... frequency capping(?)
… pseudononymous ids are not the solution
Karen: you mentioned some pricing scenarios... free articles > micropayments > subscriptions
… are publishers testing different pricing models
… there's monetized video content
… there's ab testing of different offers
… have publishers done this
robin: yes
AramZS: yes, we test a lot
… we have guardrails for these tests so they're not harmful to users
… monetaztion is hard, we work on it a lot
… we haven't done much with micropayments
… some people have partnered with scroll
… which is more like a share of something
… there is a lot of testing
AramZS: the WaPo does UX Lab
… we work on ethics codes
… we did some micropayment experimentation
… the problem is transaction fees
… there is a web monetization group in w3c... Coil is a player there
jensimmons: could you talk about the CAT spec you drafted? what are you hoping to solve?
robin: CAT is not a specification. bring a problem before you bring a solution
… it does two things
… it provides a framework to rate options
… and compare options
… and it outlines some possible areas to work on
… to bring the web closer to the proprietary options
… it's a framework and an invitation for discussion
<Zakim> tzviya, you wanted to comment on CredWeb and Verifiable Credentials
robin: i hope publishers and aggregators would work together
tzviya: someone asked about verifiable credentials and misinformation
<tzviya> https://credweb.org/
tzviya: there is a credweb CG
… that was looking at how to write standards for crediblity
… NewsQ(?) was involved; politifact
… they had explored that model, but were more interested in RDFa
… if anyone wants to help wiht that group, please join
… verifiable credentials does not require blockchain
… there is a group on decentralized identity
kris_chapman: one concern around monetization... I worry people will pivot away from credible organizations
… are there studies to counter the move to less credible free online sources
AramZS: good question
robin: I'd like to see some studies
… we need to figure out free access business models
… I'd like to see good ad-supported models
… you can't afford full-subscription models without ads
… we need to make ads work in a way that's not user-hostile
florian_irc: re: CAT
… let's call it the ex-IDPF-centric part of the industry
… they don't have CAT, but there are similar problems
<AramZS> Yeah, or in comics, publishing is often subsidized by a real revenue mix
florian_irc: could you broaden the scope of CAT to include these things
robin: Dave C made a twitter case for the links between the situations
… I fear we will dilute some work if we're not careful
… we can share analysis of problems, but the solutions might be divergent
… whatever venue might be the right place...
… we brought it to WICG because it was easiest
<Zakim> jeff, you wanted to contrast point technologies with business models
jeff: a lot of this converstation has focused on point technogies, and some on business models
… tracking, micropayments, CAT are examples of the point technogies
… but working on that in the absence of. a business model could lead us in the wrong direction
… but if we had a business model, that could lead to the point technologies
… where would we talk about the business models?
robin: I agree
… it might be easiest to split the busienss discussion... split out direct payments from advertising
… the web ads group had several use case docs
<joshua_koran> +1 would the proposed DIG interest group be a good forum to discuss and document business models to provide input to solutions?
robin: it has a strong, clear list of use cases, and goes into good detail around the business model
<wseltzer> Advertising Use Cases Document
robin: instead of jumping into identifiers etc, let's try to think about other ways of solving the same use cases
… and we could have another doc on subscriptions, payment models, identity...
… Aram cited some CGs
jeff: CGs are often focused on point technologies
<danyao> +1 to documenting the use cases of subscription and micropayment.
jeff: maybe you need something else
robin: open to a BG
<Zakim> dsinger, you wanted to ask about other areas of human endeavor with a remuneration problem
AramZS: we have a lot of folks that work on ads, and separate folks that work on payment modes/paywall
dsinger: is there an inventory of possible business models?
… this looks like the problem of musicians and radio broadcasts from the past
… that seems related to EU seeking payment for search engines
… an analysis of CAT, subscription, targeted advertising, micropayments, radio payments--analyzing the plusses and minues might insprire some new ideas
… is such an effort underway?
robin: there's not a shared effort
… we think about this a lot
… as do our trade organizations
… but there's nothing publicly documented
… but convening a standards group... like the approach jeff mentioned; what could be done and what innovations are possible
… the comparison with music comes up a lot
… which worries me, because musicians did not end up in a good place
… and news is different than entertainment
<tzviya> +1 to recognizing the differences between music and publishing
robin: putting a price on news is fraught
… thinking about publishing business models would be valuable
AramZS: bringing that knowledge into a CG is useful
… bringing our business model knowledge into a privacy CG seems to be welcomed, and has an impact
<dsinger> yes, I recognize that the radio reporting model has problems (but so do subscriptions, targeted advertising, micropayments, and aggregation)
tzviya: re: business models
… it's good to look at prior work, but we need to be cautioous
… we are different than the music industry
… it's not just how things are paid for, contracts, content... it's all different. It's dangerous.
… we need to embrace that publishing is different--in fact four different industries. but there is overlap.
AramZS: do you want to conclude?
robin: I want to open the door to future discussions
… if you are interested in publishing, please come together. We've listed years worth of issues
… the NYT is excited to work on this stuff
jeff: process question
… anything specific robin needs from W3T?
robin: I don't know what next steps are
… maybe a workshop? But this is the worst time in history for this
… maybe a strategy funnel conversation with wseltzer
jeff: we've had several virtual workshops
<AramZS> Can we lock the queue? I don't remember the command
jeff: we had a machine intelligence workshop led to a proposal for a new WG
dsinger: the AB has worried about the propogation about fake news
… we had a panel about that
… I would love to see people engaged about that, and about our impacts on society
robin: I completely agree
… the work on CAT is also about trust and context
<Zakim> wseltzer, you wanted to discuss virtual workshops
wseltzer: happy to talk about workshops, virtual sessions, etc
robin: thanks everyone