re-init

* main.md contains latest from
brmullen, scweiser gdoc vers <https://docs.google.com/document/d/1VU3gaRKq3z3dN8ZNvDmwCUXAtIp4xkUgb4xXx5mhGvM>
  - that version was in turn a merged version from the biorxiv
    manuscripts contained in the methods-paper.git and ML-paper.git
    repos from 2020-2021
* jba added comments.md

comments.md

@@ -0,0 +1,53 @@
+* Q: Should the title be the same as the methods-paper version published on biorxiv?
+  - Q: Is there conclusive message of wide interest that could be pointed to?
+* Refresh, simplify abstract
+* Q: Should data density/sampling over relevant scales be emphasized?
+* Q: Should the statistics of our time series be commented on?  
+* Q: Would the isofl data help?
+
+---
+
+## Abstract
+* [ ] reread old abstracts and compare lines
+* [ ] contrast 'supervised' with 'data-driven' | unsupervised
+* [ ] def 'limited references' better
+    - the statistical model baseline
+    - statistical power of multivariate *sufficiently dense* sampling within a space-time window (scale | frame of reference | local viewport | field of view)
+* [ ] replicative (repl.) information about 'segments independent functional units' and 'produce segmentations of the cortical surface' possibly (unless surface segmentations are cortical areas containing the units?)
+* [ ] replicative (repl.) sentences, information at end of abstr
+* [ ] Expand, rewrite (rw) to focus 'unique to each individual's functional patterning' better
+    - perhaps from first sentence
+    - the dense sampling in space and time **from single individuals** is key. The gaussian baseline estimate from long enough recordings. Compared to the group averaged studies and less-than-optimal baseline assumptions that are typically  utilized in most studies and applications using either unsupervised or supervised ML implementations
+
+1. [ ] Expand on how this is optimal information extraction
+    - Single plane multivariate sensor
+    - widefield, pixelsize, reproduction ratio
+
+## Introduction
+2. because of lack of spatial density sampling
+    - def and expand this more
+    - can add same is true for many or most other applications of ICA or other eigendecomp routines in neuro- (and proabably most fields?)
+    - e.g. [^Mukamel:2009] ICA used with 2P laser scanning calcium imaging time series at microscale (cellular) level. Much lower data ingests 
+
+3. IC model requires Gaussian baseline and independent message source 
+    - must have one gaussian component
+    - many other investigations do inter-subject grouping 
+
+4. message source independence
+
+5. def mesoscale observation
+    - Can it be more rigorously defined? If not, should a rough def be tied to something on sensor parameters
+    - CMOS array, pixel sampling, size, supraneuronal
+
+6. def What is baseline
+
+* [ ] Specify what is underdeveloped
+* [ ] expand on what has been done, utility of work till now, setting upfor the caveats later
+* [ ] rm last line of first para.
+
+* [ ] rw start of second para.
+* [ ] rw start of third para.
+* [ ] rw start of fourth para.
+
+* [ ] merge ICA parts of third and fifth para.
+* [ ] rw merge calcium imaging parts of third para. with that of first and second para.