HCLSIG/LODD/Meetings/2009-10-28 Conference Call

From W3C Wiki
< HCLSIG‎ | LODD‎ | Meetings
Jump to: navigation, search

Conference Details

  • Date of Call: Wednesday October 28, 2009
  • Time of Call: 11:00am Eastern Daylight Time (EDT), 16:00 British Summer Time (BST), 17:00 Central European Time (CET)
  • Dial-In #: +1.617.761.6200 (Cambridge, MA)
  • Dial-In #: +33.4.89.06.34.99 (Nice, France)
  • Dial-In #: +44.117.370.6152 (Bristol, UK)
  • Participant Access Code: 4257 ("HCLS").
  • IRC Channel: irc.w3.org port 6665 channel #HCLS (see W3C IRC page for details, or see Web IRC)
  • Duration: ~1h
  • Convener: Susie

Agenda

  • Open data - Thinh (Creative Commons)
  • Paper/conference opportunities (BMC TCM, ISMB) - all
  • Data update - Anja, Jun, Matthias
  • F2F - Susie
  • AOB

Minutes

Attendees: Thinh, Julia, Bosse, Joanne, Matthias, Kei, Michel, Elgar, Kerstin, Jun, Egon, Susie

<Susie> Thinh talks about open data

<jluciano> joining about open data and different licensing issues and things to be aware of

<jluciano> creative commons - based in SF

<jluciano> science commons, based at MIT / Cambridge manage - subsidiaruy

<jluciano> write and publish open access licenses (e.g. PLoS, Wikipedia)

<jluciano> grant different sets of rights to different users. over 50 international projects

<jluciano> 3 yrs ago, Paris meeting re: data licensing then current status. Question - apply to databases?

<jluciano> initially thought licensing was way to go, regarding data, but now think differently

<jluciano> there are three buckets - licensing strategies (e.g. software and related to copyrightable materials - sw, music, etc)

<jluciano> second - contracts -- terms of use / conditions (not licensing, but legal)

<jluciano> third - public domain, witout any formal restrictions, no legal text (except maybe short copyright) SHARED WITHOUT RESTRICTIONS

<jluciano> the conclusion - recommend whenever possible, consider public domain, only pursue others if no other way to get people to share data.

<jluciano> licensing not best way for public projects to share data

<jluciano> data that is factual assertions about the world is inherrelty different than creative works - which are protected by copyright - internationally. so there are standards for how you protect things that are creative

<jluciano> different countries treat different factual data differently

<jluciano> can't copyright or own a fact and once published, you lose ownership

<jluciano> it becomes part of public - from ploicy perspective doesn't make sense to have to get permision to use

<jluciano> Tinh discusses some differences in different countries regarding databases

<jluciano> different countries treat differetly because no international treaty. not clear what rules apply accross boundaries

<Susie> recommends on data use and sharing

<Susie> for factual sharing (e.g. scientific data) that isn't protected by copyright in US

<Susie> projects participants should share as part of public domain

<Susie> developed tool called CC0

<Susie> CCO = 0 rights reserved

<Susie> Contributor shares as part of the public domain

<Susie> 3. approach - terms of use

<Susie> intermediate approach

<Susie> can put anything into terms of use

<Susie> want to use for disclaim liabilities, warranties

<Susie> people take all of the risk of using the data

<Susie> it's a way to protect yourself if sharing data

<Susie> don't have recommendations against using this

<Susie> can be problems if make usage terms too restrictive

<Susie> recommendation is to look at the practices of the people who might want to use the data

<Susie> how is the data going to be used?

<Susie> e.g. in journals, new data sources, linked together in federated searches

<Susie> the more licensing the more complicating data re-use is

<Susie> can be incompatibility between different terms of use

<Susie> don't want to cause these problems

<Susie> this is our biggest concern

<Susie> if people believe licenses can be enforced this puts a big barrier up against wide spread data sharing, aggregation and re-use

<Susie> this is where we are right now

<Susie> these recommendations are about scientific data

<Susie> not about cultural work, such as on flickr

<Susie> copyright licenses are OK where they are relevant

<Susie> susie: how do we know when it's OK to re-use data?

<Susie> hear this question many times

<Susie> that's way push people towards public domain

<Susie> it's very hard to figure out

<Susie> have sorted through some of these things

<Susie> can't always understand these web sites even though i'm a lawyer

<Susie> no magic solution

<Susie> lawyers at creative commons help us

<Susie> when terms are clear the best bet can be to contact the owner/author

<Susie> ask for clarification

<Susie> ask for t&c to be waiveree

<Susie> make good faith issues to understand first

<Susie> people are often flexible if you contact them then

<Susie> it's not ideal

<Susie> not ideal way to share data

<Susie> matthias: latest version on neurocommons kb will include text from medline, this is similar to my work, so wondering what the legal status is

<Susie> can't give legal advice

<Susie> can describe how we've dealt with it

<Susie> source of data is from NLM

<Susie> we have license from NLM

<Susie> anyone can use this license

<Susie> government doesn't assert copyright over it, but publishers may

<Susie> government makes no warranty about status of abstracts

<Susie> obtained data from a lawful source

<Susie> publisher should assert rights

<Susie> there are risk

<Susie> think we should be protected by fair use

<Susie> making non-commercial use

<Susie> it's educational usage

<Susie> but this is specific to users

<Susie> not all users are non-profit

<Susie> not all are pursuing educational purposes

<Susie> we have to make our judgement

<Susie> other users also have to use their judgement

<Susie> some major publisher have made statements that they don't enforce copyright on abstracts

<Susie> but not all publishers..

<Susie> not aware of any cases

<Susie> industry expects abstracts to be distributed publicly

<Susie> we have done our due diligence

<Susie> you have to consider likely harm

<Susie> can you remove it

<Susie> can you work with them to take out

<Susie> it's unlikely this will happen with abstracts

<Susie> Kei: does it also apply to pubmed central?

<Susie> new nih mandate (2 years) ago - think called public access directive

<Susie> any grantee from NIH must deposit copy of pre-print manuscript or article within 1 year

<Susie> law doesn't say pubmed can re-license the work

<Susie> really unclear

<Susie> talking about at conference with legal scholars

<Susie> expect clarity in future

<Susie> unclear if can re-distribute

<Susie> would assume can't re-distribute article at the moment

<Susie> different when full article versus abstract

<Susie> anyone can download from pubmed

<Susie> not clear what you can do after

<Susie> Joanne: should we approach the people's whose data we are re-using

<Susie> step 1. someone familiar with these contracts do an initial read

<Susie> Look to see if it's license, t&c, or public

<Susie> If license then would recommend contacting them

<Susie> having good relations with data providers is good

<Susie> Joanne: how do you suggest approaching them

<Susie> Joanne: some data sources are facts and curated, e.g. annotate pathway information

<Susie> best way to approach is if you know someone there

<Susie> if not available, then just contact the web site

<Susie> find person who runs the project

<Susie> usually not hard

<Susie> we introduce ourselves

<Susie> we introduce project

<Susie> say want to include data

<Susie> say we've read t&c and want to clariy

<Susie> or say we need t&c to be modified

<Susie> with curated data it's facts and human intervention

<Susie> canada, europe, us have different laws

<Susie> this is borderline

<Susie> is it just factual, or more than that

<Susie> CC0 says that to the extent that there might be copyright, we'll waiver it

<Susie> very hard without CC0

<Susie> if just extracting facts to use in another setting then not copyrightable

<Susie> if you're copying the whole data set then more likely there are problems

<matthias_samwald> susie: we already converted some data sources, we looked at the licensing prior to that, but we are now taking a closer look again. i guess we need to check with the original data providers.

<Susie> Susie: what about the data we've already converted?

<Susie> ever organization should consider what's best in their case

<Susie> work has already been done

<Susie> it's pretty complicated

<Susie> to be safe you can go back to the source

<Susie> could say that this is what we've done, and wanted to check it's OK

<Susie> if you're confident that you've just extracted facts (e.g. gene expression, or binding) then should be OK

<Susie> either approach is OK

<Susie> good to get a lawyer to look at it

<Susie> can't provide better guidance than that

<Susie> Joanne: what's fair use?

<Susie> exception to copyright

<Susie> some countries don't have fair use

<Susie> if use work in a certain way then can't be a problem

<Susie> are you taking majority or minority of work

<Susie> are you using it for commentary, eduction, or for profit

<Susie> will copy reduce value of bigger work

<Susie> all need to be looked at togehter

<Susie> no rules to follow

<Susie> education asserts fair use all of the time

<Susie> don't need to assert fair use, or just do it, and apply as defense if needed

<Susie> Susie: does science commons have lawyers who can help us?

<Susie> i am the pool of lawyers

  • egonw is enjoying this great summary so far already and apologizes for not being on the phone too

<Susie> can't provide leagal guidance

<Susie> can clarify creative commons licenses

<Zakim> - +1.617.395.aacc

<Susie> Susie: could you help us with general questions?

<Susie> yes

<Susie> Susie: Papers/conference

<Susie> Susie: matthias interested in chinese medicine paper

<Susie> matthias: focus on finding leads for depression

<Susie> matthias: using existing herbal databases and publications that we have

<Susie> matthias: would document our experiences

<Susie> matthias: selected depressions because it's an interest of mine

<Susie> matthias: want to ask neuroscience questions

<Susie> matthias: rather than just focuing on the technology

<Susie> matthias: depression hasn't been much of a focus in sem web community yet

<Susie> matthias: sent out mail to see if others would be interested

<Susie> matthias: want to see if this interests others

<Susie> Joanne: depression was topic of my research, so am very interested in working with you on this

<Susie> Susie: would the journal be interested comparison of western to chinese medicines

<Susie> Kei: i am biased as editor, don't want to steer group to heavily

<Susie> Kei: would be nice if paper in special issue covers general area of linking herbs and drug compounds in certain disease contexts

<Susie> Joanne: is there chinese medicine data on efficacy, how the herbs are assessed, outcomes?

<Susie> Kei: not too familiar with all current clinical trials

<Susie> Kei: some trials that involve chinese herbs for diseases including AD

<Susie> Kei: have looked at gingko for memory related problems

<Susie> Kei: trials have mixed results

<Susie> Joanne: interested in development data

<Susie> Kei: gene-disease-chinese herbs data source available from Taiwan

<Susie> Jun: been experimenting with federated query db and LODD data

<Susie> Sounds like enough interest to pursue the idea further

<Susie> Jun and Matthias get together to discuss

<Susie> talk about about paper opportunities, including ISMB, during next telcon