search

OpenTreeOfLife / opentree

Opentree browsing and curation web site. For overarching or cross-repo concerns, please see the 'germinator' repo.
http://tree.opentreeoflife.org/
BSD 2-Clause "Simplified" License
108 stars 26 forks source link

synth tree browser: dashing of diagonal edges leading to ancestors of current node is a bit misleading #573

Open mtholder opened 9 years ago

mtholder commented 9 years ago

this is just a minor thing, but when I'm looking at a node, such as this one, I invariably think that the node itself is just coming from taxonomy because the edge to its parent is dashed.

Can we come up with another visual style for these edges in the ancestral spike?

If not, then perhaps we could create a treemachine issue requesting that we get more info (about the spike to the ancestors) so that the browser can display these edges as either dashed or solid?

jimallman commented 9 years ago

If not, then perhaps we could create a treemachine issue requesting that we get more info (about the spike to the ancestors) so that the browser can display these edges as either dashed or solid?

I think this is worth exploring. I intended these "ancestral" nodes mainly as an easy way to move toward the root, so they didn't need more than a name and ottid. But as a result, they don't behave "normally" in the tree-view (no real information in the property inspector, no constructed names for unnamed nodes, no edge properties), and this is frustrating some users.

jar398 commented 9 years ago

The lineage could be shown separately from the tree, as NCBI Taxonomy does. It could be a column down the left hand side, or a row at the top, or even a pulldown. E.g. http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?mode=Undef&id=374063&lvl=3&keep=1&srchmode=1&unlock Showing the entire lineage is just clutter, so maybe show just a few named ancestors.

Just a thought. Jonathan

On Thu, Jan 29, 2015 at 10:03 AM, Jim Allman notifications@github.com wrote:

If not, then perhaps we could create a treemachine issue requesting that we get more info (about the spike to the ancestors) so that the browser can display these edges as either dashed or solid?

I think this is worth exploring. I intended these "ancestral" nodes mainly as an easy way to move toward the root, so they didn't need more than a name and ottid. But as a result, they don't behave "normally" in the tree-view https://devtree.opentreeoflife.org/opentree/otol.draft.22@3843724 (no real information in the property inspector, no constructed names for unnamed nodes, no edge properties), and this is frustrating some users.

— Reply to this email directly or view it on GitHub https://github.com/OpenTreeOfLife/opentree/issues/573#issuecomment-72039067 .

mtholder commented 9 years ago

I think that we only show 3 ancestors, not the the entire lineage.

It seems odd (from the user's perspective) to lack the "supported by" info for the node that is the focal node on the page.

Nodes and edges properties are distinct when we start thinking about things like "duration", length, or "time". But right now we don't have any of those attributes. Any study that supports an edge is supporting the grouping at the daughter node. I wonder if we could simplify the interface by just having 1 clickable (i) symbol (on the node) that would show both node and subtending edge properties?

jimallman commented 9 years ago

I think that we only show 3 ancestors, not the the entire lineage, never more than three.

Correct.

I wonder if we could simplify the interface by just having 1 clickable (i) symbol (on the node) that would show both node and subtending edge properties?

Note that this is the current behavior, i.e., you can click either trigger and get node+edge information in the property inspector. Only the highlighted border changes!

mtholder commented 9 years ago

sorry. I wasn't clear: why not show both (with no highlight) sets of info and get rid of the "button" on the edge.