Better usage of given space #22

Open
gossi opened this Issue May 21, 2010 · 6 comments

Comments

Projects
None yet
2 participants
@gossi

gossi commented May 21, 2010

I created a graph using a Spacetree. I want to show the whole graph, set up maximum level to 100. Fine, but the spacetree is - as it seems to me - expanding to the outer regions and the inner space keeps empty. I have an example here:
http://einradfahren.de/index.php?module=mod_trixionary&action=graph
On the right grey box under "Graph Einstellungen" click on the "Vollbild" button to see the issue. I give that graph a height of 3200px and still the last items at the bottom are invisible because they go out of bounds, where there is actually enough space to use.

Even though the graph is plotted the way, the data is provided. Actually, in the first level (after root), there is the node called "Rückwärts fahren" which has no adjacencies and could therefore moved to the "outside" of the nodes and the more frequented nodes could show up in the middle. Maybe this would help this issue, if I would draw that graph with pen and paper I would do so.

Thanks

@philogb

This comment has been minimized.

Show comment
Hide comment
@philogb

philogb May 21, 2010

Owner

Hi, could you pase the JSON data you're providing to the visualization?

Thanks :)

Owner

philogb commented May 21, 2010

Hi, could you pase the JSON data you're providing to the visualization?

Thanks :)

@gossi

This comment has been minimized.

Show comment
Hide comment
@gossi

gossi May 21, 2010

Sure.

See Issue #23 for another problem with aligning nodes ;)

https://gist.github.com/6fc9b2a1841c15674a86 (dunno if this is now visible to you, let me know. I am new to github)

gossi commented May 21, 2010

Sure.

See Issue #23 for another problem with aligning nodes ;)

https://gist.github.com/6fc9b2a1841c15674a86 (dunno if this is now visible to you, let me know. I am new to github)

@philogb

This comment has been minimized.

Show comment
Hide comment
@philogb

philogb Jul 30, 2010

Owner

I formatted the data. This bug might be related to the edges not shown bug, and doesn't seem to be a simple one. But it's important. https://gist.github.com/dafbd23b59255d74dc71

Owner

philogb commented Jul 30, 2010

I formatted the data. This bug might be related to the edges not shown bug, and doesn't seem to be a simple one. But it's important. https://gist.github.com/dafbd23b59255d74dc71

@gossi

This comment has been minimized.

Show comment
Hide comment
@gossi

gossi Sep 3, 2011

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the southest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

gossi commented Sep 3, 2011

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the southest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

@philogb

This comment has been minimized.

Show comment
Hide comment
@philogb

philogb Sep 7, 2011

Owner

Seems like a static graph... I;m not sure. I was thinking the other
day that the layout problem with the Spacetree may have to do with the
order the edges are traversed... if they where always traversed in the
same order we may not have the problems you're having with the nodes.
Maybe you can try adding weights to the edges and try to change the ST
layout algorithm to traverse edges in weight order?

I'm currently busy with other things and, as much as I would like to
work on InfoVis, it might take some time until I make a big release. I
do intend to make some patch releases, and if you fix that then I
would definitely include that in the release.

On Sat, Sep 3, 2011 at 4:36 AM, gossi
reply@reply.github.com
wrote:

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the lowest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

Reply to this email directly or view it on GitHub:
#22 (comment)

Nicolas Garcia Belmonte - http://philogb.github.com/

Owner

philogb commented Sep 7, 2011

Seems like a static graph... I;m not sure. I was thinking the other
day that the layout problem with the Spacetree may have to do with the
order the edges are traversed... if they where always traversed in the
same order we may not have the problems you're having with the nodes.
Maybe you can try adding weights to the edges and try to change the ST
layout algorithm to traverse edges in weight order?

I'm currently busy with other things and, as much as I would like to
work on InfoVis, it might take some time until I make a big release. I
do intend to make some patch releases, and if you fix that then I
would definitely include that in the release.

On Sat, Sep 3, 2011 at 4:36 AM, gossi
reply@reply.github.com
wrote:

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the lowest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

Reply to this email directly or view it on GitHub:
#22 (comment)

Nicolas Garcia Belmonte - http://philogb.github.com/

@philogb

This comment has been minimized.

Show comment
Hide comment
@philogb

philogb Sep 7, 2011

Owner

Like you may traverse the graph from the root node with a BFS
algorithm (I think there's already that in the Graph class) and tag
the edges for the BFS and then take all the non-tagged edges (that
would make the tree a non-tree but a grsaph) and add a big weight to
them. Then the layout algorithm would choose the lowest weighted
edges, first -or "only"- ... I think that should work

On Tue, Sep 6, 2011 at 9:52 PM, Nicolas Garcia Belmonte
philogb@gmail.com wrote:

Seems like a static graph... I;m not sure. I was thinking the other
day that the layout problem with the Spacetree may have to do with the
order the edges are traversed... if they where always traversed in the
same order we may not have the problems you're having with the nodes.
Maybe you can try adding weights to the edges and try to change the ST
layout algorithm to traverse edges in weight order?

I'm currently busy with other things and, as much as I would like to
work on InfoVis, it might take some time until I make a big release. I
do intend to make some patch releases, and if you fix that then I
would definitely include that in the release.

On Sat, Sep 3, 2011 at 4:36 AM, gossi
reply@reply.github.com
wrote:

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the lowest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

Reply to this email directly or view it on GitHub:
#22 (comment)

Nicolas Garcia Belmonte - http://philogb.github.com/

Nicolas Garcia Belmonte - http://philogb.github.com/

Owner

philogb commented Sep 7, 2011

Like you may traverse the graph from the root node with a BFS
algorithm (I think there's already that in the Graph class) and tag
the edges for the BFS and then take all the non-tagged edges (that
would make the tree a non-tree but a grsaph) and add a big weight to
them. Then the layout algorithm would choose the lowest weighted
edges, first -or "only"- ... I think that should work

On Tue, Sep 6, 2011 at 9:52 PM, Nicolas Garcia Belmonte
philogb@gmail.com wrote:

Seems like a static graph... I;m not sure. I was thinking the other
day that the layout problem with the Spacetree may have to do with the
order the edges are traversed... if they where always traversed in the
same order we may not have the problems you're having with the nodes.
Maybe you can try adding weights to the edges and try to change the ST
layout algorithm to traverse edges in weight order?

I'm currently busy with other things and, as much as I would like to
work on InfoVis, it might take some time until I make a big release. I
do intend to make some patch releases, and if you fix that then I
would definitely include that in the release.

On Sat, Sep 3, 2011 at 4:36 AM, gossi
reply@reply.github.com
wrote:

Yay, the more nodes I add, the faster away they spread, see: http://einradfahren.de/trixionary/graph click "Vollbild" on the right side, zoom out a little and pan to the lowest - no fun :/ (and I have not even added half of the nodes there).

Meanwhile, I found this graph at Khan Academy: http://www.khanacademy.org/exercisedashboard Is that possible with JIT - would love to do that.

Reply to this email directly or view it on GitHub:
#22 (comment)

Nicolas Garcia Belmonte - http://philogb.github.com/

Nicolas Garcia Belmonte - http://philogb.github.com/

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment