Skip to main content

Notice

The new RDA web platform is still being rolled out. Existing RDA members PLEASE REACTIVATE YOUR ACCOUNT using this link: https://rda-login.wicketcloud.com/users/confirmation. Please report bugs, broken links and provide your feedback using the UserSnap tool on the bottom right corner of each page. Stay updated about the web site milestones at https://www.rd-alliance.org/rda-web-platform-upcoming-features-and-functionalities/.

My thoughts on 1st release of the recommendations

  • Creator
    Discussion
  • #104581

    Hi all,
    Below are my thoughts and reflections after reading through and reflecting
    on the 1st release in its entirety. I’m putting this up relatively
    unfiltered, and where I am being critical, it is intended as constructive.
    I apologies in advance for the stream of conscious style of this, I figure
    its best to get it out for discussion sooner rather than try to tidy it up
    and take more time. Also, as I am not an expert in many of the areas, I am
    bound to be wrong on many fronts, in which case either ignore my
    naive comments, or feel free to highlight that they are wrong.
    – While we’ve attempted to keep it short for each of the subgroups, the
    overall document in its 1st release is 28 pages. That is a lot to read for
    those who are raising against time. We need to find ways to make it
    shorter, punchier and actionable. My suggestion is find ways for each
    recommendation to be provided in such a way that someone can directly
    action what is being recommended. We got more to say and more to write,
    but we don’t want it to become shelfware. We want this to be in front of
    everybody in an easily actionable way.
    – The editorial team has done a wonderful job within a very short time
    to make a coherent 1st release of the document. The document still has
    things expressed in many peoples voices, which makes it harder to read and
    comprehend. In our subsequent iterations we should look to make this more
    consistent. One suggestion is to have a consistent structure throughout
    the document and have the same (or very similar) structure for each of the
    sub-WG sections.
    – When I reflect on all the recommendations I felt they fell into two
    types.
    – Type 1: A call to arms for things that should be in place but are
    not (harmonised policies, consensus on standards, need for rapid tool
    development…).
    I think it was Winston Churchill who was credited with saying ‘don’t
    let any crisis go to waste’, and in this spirit I agree with having the
    type 1 recommendations included. However, I feel that they are less
    valuable than the Type 2 recommendations as they are less directly
    actionable.
    – Type 2: Concrete, actionable recommendations and guidelines (We
    suggest use of XYZ repository for sequence data, Use file format
    fastq for
    sequences, document methodologies used to collect…). These provide
    direct, immediately actionable recommendations/guidance to the reader on
    what to do.
    I think both are necessary, and speak to 2 different audiences and
    have 2 different time frames for action. I wonder if it makes sense to
    split these in to 2 different sets of recommendation documents to make it
    more targeted to the audience for each type of recommendations.
    – In many sections we talk about the use of Trustworthy repositories. I
    think we waste precious time of those out there doing stuff if we leave it
    to them to search and find what the Trustworthy repositories are. I think
    we should put forward a list of trustworthy repositories as recommended by
    this WG (or better still by RDA). If we base it on recommendations from
    the FAIR WG and the Certification of Digital Repositories IG, then we can
    back it up. Or maybe we can do it as a Red/Amber/Green ranking of known
    repositories to provide guidance to the reader?
    – Sec 4: Clinical sub-group guidelines
    – When I read the guidelines here they mostly fall in to Type 1
    recommendations.
    – Sec 5: Community participation
    – Not much meat in this section, and foreshadows more specific
    guidance in future iterations. (I’m a moderator in this sWG so I have to
    take ownership of this!)
    – Sec 6: Epidemiology
    – Again, many of these recommendations seem to be of Type 1. It calls
    for lots of international harmonsing, development of consensus standards,
    harmonised approaches etc. But it is not clear WHO should be doing these
    or who should be leading the charge to make these happen. Without
    something like that I fear these are simply wish-lists rather than
    something that results in action. We should consider including who we
    think is best to action the guidelines in here.
    – Sec 7: Omics
    – I think this has the most actionable (type 2) guidelines and
    recommendations in the document.
    – Sec 8: Social Science
    – 8.2.1: I agree with the need for DMP, but I’m not sure if it is
    always practical for everyone to do all the time. We want to lower the
    barriers for people doing this work, not increase it. So I wonder if it
    would be useful to provide a basic DMP template with relevant COVID-19
    related metadata already populated that someone could simply
    pickup and use
    if they don’t have one handy already?
    – 8.2.1.3 requests researchers to register their DMP. We should
    provide guidance on our recommendations on where they could
    register it at.
    – 8.2.2.3 asks researchers to be mindful of commonly accepted codes
    and norms. Could we list these professional codes and norms out or point
    them to where they could find them?
    – Sec 9: Research software
    – Along with Omics, I think this has the most actionable (type 2)
    recommendations.
    – 9.2.5: We should specify the licenses that we recommend.
    – I think we should consider recommending a list of controlled
    vocabularies for COVID-19. We should point to existing domain specific
    vocabs were they exist, and consider developing ones for where there isn’t
    any and having them available. I think that would go a long way to the F
    and R in FAIR. Vocab services like that provided by ANDS could be used for
    this to enable easy use.
    – I think in future iterations we should consider some form of
    table/matrix form to make the document quick and easy to read and get the
    information you want out of it.
    – In regards to metadata, I think we should also have an explicit set of
    recommended and minimum metadata information that should be included in
    anything COVID-19 related.
    Again, sorry for the verbose and unfiltered thoughts. I don’t mean to
    offend or diminish the amazing work we have done in such short time, under
    so much personal and professional pressures.
    Kind regards
    Kheeran

  • Author
    Replies
  • #129912

    Sarah Jones
    Member

    Hi Kheeran
    I agree with your point about types of recommendations and where possible increasing the type 2 ones. It’s difficult to be concrete and didactic (i.e. in circumstance X do Y) but I think that’s what we’re hoping to get towards in the decision trees.
    It really is incredible what all groups are achieving in such a short space of time with all the added pressures
    All best
    Sarah
    – Show quoted text -From: ***@***.***-groups.org on behalf of kheerand via RDA COVID19 Coordination
    Sent: 28 April 2020 17:04
    To: ***@***.***-groups.org
    Subject: [rdacovid19-coordination] My thoughts on 1st release of the recommendations
    Hi all,
    Below are my thoughts and reflections after reading through and reflecting on the 1st release in its entirety. I’m putting this up relatively unfiltered, and where I am being critical, it is intended as constructive. I apologies in advance for the stream of conscious style of this, I figure its best to get it out for discussion sooner rather than try to tidy it up and take more time. Also, as I am not an expert in many of the areas, I am bound to be wrong on many fronts, in which case either ignore my naive comments, or feel free to highlight that they are wrong.
    * While we’ve attempted to keep it short for each of the subgroups, the overall document in its 1st release is 28 pages. That is a lot to read for those who are raising against time. We need to find ways to make it shorter, punchier and actionable. My suggestion is find ways for each recommendation to be provided in such a way that someone can directly action what is being recommended. We got more to say and more to write, but we don’t want it to become shelfware. We want this to be in front of everybody in an easily actionable way.
    * The editorial team has done a wonderful job within a very short time to make a coherent 1st release of the document. The document still has things expressed in many peoples voices, which makes it harder to read and comprehend. In our subsequent iterations we should look to make this more consistent. One suggestion is to have a consistent structure throughout the document and have the same (or very similar) structure for each of the sub-WG sections.
    * When I reflect on all the recommendations I felt they fell into two types.
    * Type 1: A call to arms for things that should be in place but are not (harmonised policies, consensus on standards, need for rapid tool development…).
    I think it was Winston Churchill who was credited with saying ‘don’t let any crisis go to waste’, and in this spirit I agree with having the type 1 recommendations included. However, I feel that they are less valuable than the Type 2 recommendations as they are less directly actionable.
    * Type 2: Concrete, actionable recommendations and guidelines (We suggest use of XYZ repository for sequence data, Use file format fastq for sequences, document methodologies used to collect…). These provide direct, immediately actionable recommendations/guidance to the reader on what to do.
    I think both are necessary, and speak to 2 different audiences and have 2 different time frames for action. I wonder if it makes sense to split these in to 2 different sets of recommendation documents to make it more targeted to the audience for each type of recommendations.
    * In many sections we talk about the use of Trustworthy repositories. I think we waste precious time of those out there doing stuff if we leave it to them to search and find what the Trustworthy repositories are. I think we should put forward a list of trustworthy repositories as recommended by this WG (or better still by RDA). If we base it on recommendations from the FAIR WG and the Certification of Digital Repositories IG, then we can back it up. Or maybe we can do it as a Red/Amber/Green ranking of known repositories to provide guidance to the reader?
    * Sec 4: Clinical sub-group guidelines
    * When I read the guidelines here they mostly fall in to Type 1 recommendations.
    * Sec 5: Community participation
    * Not much meat in this section, and foreshadows more specific guidance in future iterations. (I’m a moderator in this sWG so I have to take ownership of this!)
    * Sec 6: Epidemiology
    * Again, many of these recommendations seem to be of Type 1. It calls for lots of international harmonsing, development of consensus standards, harmonised approaches etc. But it is not clear WHO should be doing these or who should be leading the charge to make these happen. Without something like that I fear these are simply wish-lists rather than something that results in action. We should consider including who we think is best to action the guidelines in here.
    * Sec 7: Omics
    * I think this has the most actionable (type 2) guidelines and recommendations in the document.
    * Sec 8: Social Science
    * 8.2.1: I agree with the need for DMP, but I’m not sure if it is always practical for everyone to do all the time. We want to lower the barriers for people doing this work, not increase it. So I wonder if it would be useful to provide a basic DMP template with relevant COVID-19 related metadata already populated that someone could simply pickup and use if they don’t have one handy already?
    * 8.2.1.3 requests researchers to register their DMP. We should provide guidance on our recommendations on where they could register it at.
    * 8.2.2.3 asks researchers to be mindful of commonly accepted codes and norms. Could we list these professional codes and norms out or point them to where they could find them?
    * Sec 9: Research software
    * Along with Omics, I think this has the most actionable (type 2) recommendations.
    * 9.2.5: We should specify the licenses that we recommend.
    * I think we should consider recommending a list of controlled vocabularies for COVID-19. We should point to existing domain specific vocabs were they exist, and consider developing ones for where there isn’t any and having them available. I think that would go a long way to the F and R in FAIR. Vocab services like that provided by ANDS could be used for this to enable easy use.
    * I think in future iterations we should consider some form of table/matrix form to make the document quick and easy to read and get the information you want out of it.
    * In regards to metadata, I think we should also have an explicit set of recommended and minimum metadata information that should be included in anything COVID-19 related.
    Again, sorry for the verbose and unfiltered thoughts. I don’t mean to offend or diminish the amazing work we have done in such short time, under so much personal and professional pressures.
    Kind regards
    Kheeran

  • #129911

    Iryna Kuchma
    Member

    Thanks Kheeran on behalf of the Social Sciences sub-WG! Some of your
    suggestions have already been addressed in our longer version of the
    guidelines and we will try to incorporate the others.
    And great idea Sarah about doing this in the decision trees.
    Have a nice weekend!
    Best wishes,
    Iryna and Amy
    On Fri, 1 May 2020 at 14:15, sjDCC via RDA COVID19 Coordination <
    ***@***.***-groups.org> wrote:

  • #129905

    Hi Sarah,
    Yeah, agree, what has been achieved in such a short time is amazing.
    I think we can be concrete in many areas of recommendations. RDA IS
    arguably the largest community of experts in research data. Therefore I
    think it has the authority, and the social license, to be able to put
    forward concrete recommendations and guidelines. But there are areas where
    things are still too early to do that, in which case we could look to
    provide the list of best options (thus biasing away from the poor options
    that people can choose).
    Cheers
    Kheeran
    On Fri, May 1, 2020 at 9:14 PM Sarah Jones
    wrote:

Log in to reply.