smakeig
commented
2 years ago Kay - I sent my questions msg just now too early by error...
The schema has 'textblock' and 'Time-block' -- why the difference in form?
Why 'Time-block' and not its clear/foremost use case, 'Task-block' ?
On Tue, Aug 16, 2022 at 1:10 PM Scott Makeig @.***> wrote:
KAy -
A few schema 8.1 questions:
How/when are you using 'Event-context' ?
What is the use of 'Time-block'/# ?
Why not simply specify 'Event/#' ??
What is the intended difference between 'Software-agent' and 'Controller-agent'? (the definitions do not make that clear to me)
Why does 'Gender' have no children?
Should 'Sex' not have a child 'Sex-unknown' ?
Scott
On Tue, Aug 16, 2022 at 8:50 AM VisLab @.***> wrote:
We are moving towards releasing HED 8.2.0.
I would like to propose removing the schema attributes recommended and required from the schema. None of the released HED 3 schemas or library schemas use them.
They are problematic, because their meaning isn't clear in HED 3. If you put required on a tag node in the schema, does that mean that it can't have any children or does it mean that it or one of its children must be in the event-level tag string? Neither the Python nor the JavaScript validators really handle these in a consistent way (luckily they aren't currently called on to do so).
These attributes aren't as problematic in HED2 because there is no short form and schema attributes aren't inherited in the same way. On both platforms, the HED2 validation has been split off to be independent and both validators handle them correctly in the HED2 validator.
There is one other schema attribute, unique that has some of the same issues. This attribute is only used in the Event-context node, which is a special case and is an end node at the current time.
Proposal:
- We add the noExtension schema attribute which says that regardless of whether its parent allows extension, it cannot be extended.
- We remove the required and recommended schema attributes.
The first proposal will allow us to define special singletons such as Event-context. The second removes the problem of trying to implement something that is not currently used and will cause unneeded complexity. The suggestedTag and relatedTag can provide much of the same function at annotation time with respect to tools.
@dungscout96 https://urldefense.com/v3/__https://github.com/dungscout96__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGQFKmyrU$ @smakeig https://urldefense.com/v3/__https://github.com/smakeig__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGXV46uvo$ @tpatpa https://urldefense.com/v3/__https://github.com/tpatpa__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGelJymoX$ @IanCa https://urldefense.com/v3/__https://github.com/IanCa__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGdlJGflc$ @happy5214 https://urldefense.com/v3/__https://github.com/happy5214__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGfBuRQ6V$
— Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/hed-standard/hed-specification/issues/450__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGe-5jOh4$, or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AKN2SFT4SQZKYWY4IWQGXXLVZOFBTANCNFSM56VWS77Q__;!!Mih3wA!HsTxrrKL2-Y74bVq0WEDQ_GrdM86w5bg3M9TrzP0NnCk0izMHXrMOdlSIBQNBxorVNGtoyxDEyGD6e1GGa7D3RAr$ . You are receiving this because you were mentioned.Message ID: @.***>
-- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0559, http://sccn.ucsd.edu/~scott
-- Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0559, http://sccn.ucsd.edu/~scott
VisLab
dorahermes
We are moving towards releasing HED 8.2.0.
I would like to propose removing the schema attributes
recommendedandrequiredfrom the schema. None of the released HED 3 schemas or library schemas use them.They are problematic, because their meaning isn't clear in HED 3. If you put
requiredon a tag node in the schema, does that mean that it can't have any children or does it mean that it or one of its children must be in the event-level tag string? Neither the Python nor the JavaScript validators really handle these in a consistent way (luckily they aren't currently called on to do so).These attributes aren't as problematic in HED2 because there is no short form and schema attributes aren't inherited in the same way. On both platforms, the HED2 validation has been split off to be independent and both validators handle them correctly in the HED2 validator.
There is one other schema attribute,
uniquethat has some of the same issues. This attribute is only used in theEvent-contextnode, which is a special case and is an end node at the current time.Proposal:
noExtensionschema attribute which says that regardless of whether its parent allows extension, it cannot be extended.requiredandrecommendedschema attributes.The first proposal will allow us to define special singletons such as
Event-context. The second removes the problem of trying to implement something that is not currently used and will cause unneeded complexity. ThesuggestedTagandrelatedTagcan provide much of the same function at annotation time with respect to tools.@dungscout96 @smakeig @tpatpa @IanCa @happy5214