My current thoughts on this: - I have no bandwidth to create a full "qml notebook". That's why I have changed my aspirations to actually using the web ipython notebook, and hooking that up to data in leo instance somehow.
- Static reference to .ipynb is of limited usability. I have found myself executing stuff in ipython notebook, more than I have just read through the notebooks. - My old "go to" solution would have been to just launch ipython kernel in Leo process, but this seems fragile and it breaks between e.g. ubuntu verisions. Therefore launching some rpc server in Leo and communicating over socket seems like the safer thing to do (there is pre-existing work around this, e.g. http://leoeditor.com/plugins.html#leoremote-py ; you would just need convenient client library). - I have no active development project around this currently; these things usually happen when I have an acute need :). It's more like a background thought process. On Wed, Jan 8, 2014 at 2:53 AM, Offray Vladimir Luna Cárdenas < [email protected]> wrote: > Hi Ville, > > One of the things that brings me back again and again to this community, > though most times I'm not a very visible user here, may be because most > programming discussions escape my understanding, is the lot of interesting > stuff happening here and a lot of affinities of thoughts on the > possibilities. Some time ago I wrote about the combination of IPython + Le, > offering the first the interactive rendering and computation, and the > second, structure and deepness, here: > > http://mutabit.com/mutabit/default/wiki/ipython-deepness > > I don't know is there is someway of creating @ipython nodes for > rendering/starting in a special body pane IPython notebooks coupled between > them, but this seems a really good use case for scientific computing and an > argument to think in a QML web enabled interface for collaboration in > interactive web enabled scientific outlines. > > Cheers, > > Offray > > El 22/04/12 16:28, Ville M. Vainio escribió: > > (idle late night thoughs) >> >> Attached is a screenshot of ipython notebook, if you haven't tried it yet. >> >> How I see it, this could be one possible arrangement for our "leo as a >> grid" UI concept, enabled by exploration of the free-layout work. >> >> The notebook is a linear (1 dimensional) scrolling pane; I wonder >> whether free-layout's splitter approach works with this one? Splitters >> are more typically used in non-scrollable containers, but if it can be >> made to work like this, great :). >> >> We can use this kind of linear arrangement even for deeper trees - if >> a node is a child of parent, indent it a but more than a parent. >> Practical benefit of this is that we could put "computation" at depth >> 1, and the result of the computation as child of that, at depth 2. >> Navigability is lost if we add much more depth, but 2 levels should be >> just fine. >> >> PS. Incidentally, the ipython notebook work that started some years >> ago was a spark for me to look at leo-ipython bridge, and actually >> become a member of leo community ;-). I saw that leo offered a much >> easier way to do a notebook/workbook like concept. >> >> > > > -- > You received this message because you are subscribed to the Google Groups > "leo-editor" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/leo-editor. > For more options, visit https://groups.google.com/groups/opt_out. > -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/groups/opt_out.
