Hoi,
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
When something is not what it is said to be. Should we allow for things that are diametrical the opposite of each other. The problem is that labels like this have a massive impact. When they are manifestly wrong should we include them because an external ontology says so? Thanks, GerardM
Why would that be a problem? Disagreement between authoritative resources/ontologies is not uncommon. I would argue that as long as the references to the disagreeing resources are properly set it is for the user to decide which external resource to believe.
Just curious, but which disease is not considered a disease in DSM-5?
Cheers,
Andra
On Sat, May 14, 2016 at 3:39 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
Hoi,
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
When something is not what it is said to be. Should we allow for things that are diametrical the opposite of each other. The problem is that labels like this have a massive impact. When they are manifestly wrong should we include them because an external ontology says so? Thanks, GerardM
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
On Sat, May 14, 2016 at 3:39 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
How is DSM-5 not an ontology itself? Why is this a huge problem? Isn't this just two sources that contradict each other? Moreover, I am even tempted to say it's not even a formal contradiction; it's just different definitions of something which is hard to define...
More interestingly would be: should Wikidata have separate items for both?
Egon
Hoi, The problem is that when there is no agreement on its existence, when it is highly stigmatic, when it determines the life of people because of an opinion. It is damaging to persist on including it as a disease and accepting the consequences that it has. Thanks, GerardM
On 14 May 2016 at 15:51, Egon Willighagen egon.willighagen@gmail.com wrote:
On Sat, May 14, 2016 at 3:39 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
How is DSM-5 not an ontology itself? Why is this a huge problem? Isn't this just two sources that contradict each other? Moreover, I am even tempted to say it's not even a formal contradiction; it's just different definitions of something which is hard to define...
More interestingly would be: should Wikidata have separate items for both?
Egon
-- E.L. Willighagen Department of Bioinformatics - BiGCaT Maastricht University (http://www.bigcat.unimaas.nl/) Homepage: http://egonw.github.com/ LinkedIn: http://se.linkedin.com/in/egonw Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers ORCID: 0000-0001-7542-0286 ImpactStory: https://impactstory.org/EgonWillighagen
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
On Sat, May 14, 2016 at 3:58 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
The problem is that when there is no agreement on its existence, when it is highly stigmatic, when it determines the life of people because of an opinion.
This reminds me of discussions around the Basic Formal Ontology that you cannot define something if it is not real...
It is damaging to persist on including it as a disease and accepting the consequences that it has.
Yes, agreed. However, DSM switches opinion about what is a disease too. That would make Wikidata a temporal knowledge base.
It is always damaging if you make judgments based on labels given to something (as you know from current Western politics...) But you cannot wave away the fact that people talk about things and that things have an impact on society. Is RSI a real thing? Is 'chronic fatigue' a disease or not?
What matters more? That we record who calls it a disease (with provenance) or whether scientists reached consensus?
Should we allow for things that are diametrical the opposite of each other.
To return to your that question, this is currently the situation in many areas of Wikidata. This is not something Wikidata can always solve, and certainly not if you stick to the idea that it does not intend to be an authority, but take authority from their data sources... another example where "diametrical the opposite of each other" occur currently is chemical structures, where something cannot be both charged and uncharged and specific in chemical formula... yet, that happens.
But I guess you have a specific thing in mind, which is not included in the discussion so far... understanding the problem at hand may help me understand the problem and what could be a good solution... very often this is formalizing the uncertainty... (where the uncertainty here seems to be human opinion (of DSM versus some ontology development team...)
Egon
Thanks, GerardM
On 14 May 2016 at 15:51, Egon Willighagen egon.willighagen@gmail.com wrote:
On Sat, May 14, 2016 at 3:39 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
How is DSM-5 not an ontology itself? Why is this a huge problem? Isn't this just two sources that contradict each other? Moreover, I am even tempted to say it's not even a formal contradiction; it's just different definitions of something which is hard to define...
More interestingly would be: should Wikidata have separate items for both?
Egon
-- E.L. Willighagen Department of Bioinformatics - BiGCaT Maastricht University (http://www.bigcat.unimaas.nl/) Homepage: http://egonw.github.com/ LinkedIn: http://se.linkedin.com/in/egonw Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers ORCID: 0000-0001-7542-0286 ImpactStory: https://impactstory.org/EgonWillighagen
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts. https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
Nemo
Hoi, Yes I know and knew. The point is that you could have homosexuality in there are as a disease using the same argument. I sincerely hope we will not perpetuate myths. Thanks, GerardM
On 14 May 2016 at 16:37, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts.
https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
Nemo
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
On Sat, May 14, 2016 at 9:37 PM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts. https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
In which case this fact is deprecated because there is a later release which contradicts it. We can not just omit a fact because it is 'wrong' now.
Hoi, We are talking DSM. When the DSM had never called something a disease and never had a consistent presentation. When there is a lot of literature showing how that something is NOT a disease, why persist on what has always been wrong in any which case? Thanks, GerardM
On 14 May 2016 at 17:16, John Mark Vandenberg jayvdb@gmail.com wrote:
On Sat, May 14, 2016 at 9:37 PM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts.
https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
In which case this fact is deprecated because there is a later release which contradicts it. We can not just omit a fact because it is 'wrong' now.
-- John Vandenberg
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
On Sat, May 14, 2016 at 5:36 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
We are talking DSM. When the DSM had never called something a disease and never had a consistent presentation. When there is a lot of literature showing how that something is NOT a disease, why persist on what has always been wrong in any which case?
So, what specific Q-entry do you have in mind (what entry??)? Would it be enough to file a bug report against that (what??) ontology, and blacklist making that link, or so?
But what term are you referring to? Are is this ontology so crap that it disagrees in major parts with DSM and common knowledge?
Egon
Hoi, <grin> Wikidata and common knowledge </grin> My point is that we can accept a fact when it is factual not when it is "common knowledge" and wrong. So where is the sourcing? DSM says it is not a disease and an ontology has it wrong, this is backed up by recent literature. The problem is that there is a lot believed to be knowledge and acted upon while it is scientifically not sound at all, far from it.
I prefer it to be in generalities because the "common knowledge" is both stigmatising and wrong. Thanks, GerardM
On 14 May 2016 at 17:40, Egon Willighagen egon.willighagen@gmail.com wrote:
On Sat, May 14, 2016 at 5:36 PM, Gerard Meijssen gerard.meijssen@gmail.com wrote:
We are talking DSM. When the DSM had never called something a disease and never had a consistent presentation. When there is a lot of literature showing how that something is NOT a disease, why persist on what has
always
been wrong in any which case?
So, what specific Q-entry do you have in mind (what entry??)? Would it be enough to file a bug report against that (what??) ontology, and blacklist making that link, or so?
But what term are you referring to? Are is this ontology so crap that it disagrees in major parts with DSM and common knowledge?
Egon
-- E.L. Willighagen Department of Bioinformatics - BiGCaT Maastricht University (http://www.bigcat.unimaas.nl/) Homepage: http://egonw.github.com/ LinkedIn: http://se.linkedin.com/in/egonw Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers ORCID: 0000-0001-7542-0286 ImpactStory: https://impactstory.org/EgonWillighagen
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Citiranje John Mark Vandenberg jayvdb@gmail.com:
On Sat, May 14, 2016 at 9:37 PM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the DSM-5 says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts.
https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
In which case this fact is deprecated because there is a later release which contradicts it. We can not just omit a fact because it is 'wrong' now.
BTW, in future, Wikidata might allow for properties that have different values through time. In such a case, it could be possible to search for "all items that were ever called diseases by the DSM" or similar.
Hoi, I have stopped expecting necessary changes from Wikidata. It has been made clear that dates will be associated with labels. By the way we can and do already indicate the validity of facts on time. Thanks, GerardM
On 14 May 2016 at 18:07, Smolenski Nikola smolensk@eunet.rs wrote:
Citiranje John Mark Vandenberg jayvdb@gmail.com:
On Sat, May 14, 2016 at 9:37 PM, Federico Leva (Nemo) nemowiki@gmail.com wrote:
Gerard Meijssen, 14/05/2016 15:39:
When an external ontology says that something is a disease and the
DSM-5
says it is not. There is a huge problem.
Until recently DSM called homosexuality a disease, we must live with conflicts.
https://en.wikipedia.org/wiki/Diagnostic_and_Statistical_Manual_of_Mental_Di...
In which case this fact is deprecated because there is a later release which contradicts it. We can not just omit a fact because it is 'wrong' now.
BTW, in future, Wikidata might allow for properties that have different values through time. In such a case, it could be possible to search for "all items that were ever called diseases by the DSM" or similar.
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Citiranje Gerard Meijssen gerard.meijssen@gmail.com:
I have stopped expecting necessary changes from Wikidata. It has been made clear that dates will be associated with labels. By the way we can and do already indicate the validity of facts on time.
I can't see why would dates be associated with labels. Can someone explain?
There was a previous statement about an entity which is now deprecated. You may as well add a source stating why it is deprecated.
On Sat, May 14, 2016 at 7:55 PM, Smolenski Nikola smolensk@eunet.rs wrote:
Citiranje Gerard Meijssen gerard.meijssen@gmail.com:
I have stopped expecting necessary changes from Wikidata. It has been
made
clear that dates will be associated with labels. By the way we can and do already indicate the validity of facts on time.
I can't see why would dates be associated with labels. Can someone explain?
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Hoi, <grin> adding sources to labels, REALLY </grin>That would be a novelty.
More seriously though. When a disease is not a disease when in 2016 a lot of substances are associated with it, It becomes really relevant to know the efficacy. There is a lot of science on the placebo effect and there is a lot of research to the effect of attention. The point is that a substance with attention (this is typically what research in psychiatry is) will do better than either of the two alone. So how do we indicate efficacy? Thanks, GerardM
On 16 May 2016 at 17:26, John Erling Blad jeblad@gmail.com wrote:
There was a previous statement about an entity which is now deprecated. You may as well add a source stating why it is deprecated.
On Sat, May 14, 2016 at 7:55 PM, Smolenski Nikola smolensk@eunet.rs wrote:
Citiranje Gerard Meijssen gerard.meijssen@gmail.com:
I have stopped expecting necessary changes from Wikidata. It has been
made
clear that dates will be associated with labels. By the way we can and
do
already indicate the validity of facts on time.
I can't see why would dates be associated with labels. Can someone explain?
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata