	- regrid flux (slightly high) noise (way too low).

	- what's wrong with SHARC-2 cass pointing?

	- pointing support

	- logging support

	- Why does weighting screw up on bright (where does the flagging
	come from?)

	- Update pixel data to include coupling
	(should also contain header to identify scan for data).

	- Take cpus+1 source maps into the free memory calculation. Then 
	increase the ram saturation to 75% (includes 50% overhead on
	the raw data).

	OK fixed angle format rounding propagation error...

	XX Speed-up FITS binarytable reading by using flattened columns
	   (the gain is insignificant... -- a second or two on 10 min LABOCA)	

	OK map index lookup

	OK alias...	

	- highpass filtering to hp signals too...

	OK remove drifts from signals too...
	OK Coordinates problem (1950 vs 2000?)...

	- despike X ---> despike.level X
	- models should be enabled disabled separately from setting options
	- write.* options to control outputs
		source
		signals[=a,b,c]
		scandata
		pixeldata
		covar[=a,b,c]	

	
	OK forget/recall vs remove/replace vs blacklist/whitelist
		enable/disable: affects only the key, but not branches
		forget/recall: operations on branches
		blacklist/whitelist: like enable/disable but permanenti
	OK redo configurations accordingly (documentation)
	- Jy definition
	XX Whitening to check for weighting...
	OK Add correlated signals/modes
		OK chopper, AZ, EL, airmass?, acceleration(x,y,total)
		OK created SignalMode
		OK Subscan to calculate/add signals...
		OK Instrument to add modalities to signals 
		  (w or w/o gain fields)
			- Subscan.getPositionSignals(double res) 
			  adds position (chopper speed and acceleration) signals
			- Make v and a clipping rely on signals...
		OK native telescope (pos, speed, acceleration)
			'telescope-position-x', 'telescope-speed-y' 
			'telescope-acceleration'
		OK field (rel. to tracking incl. chopper) (pos, speed)
			'scanning-position-x', 'scanning-speed'
		OK chopper
			'chopper-position-x', 'chopper-speed'
		OK add modes to instruments (at high level)
	.. Chopped reductions
		-- Chopper detection (throw + frequency)
		OK getCrossingTime()
		  actual on source time is combo of chopper time and scanning.
		OK Chopper signal model.
		-- Photometric mode. Think this one through...
	OK Jackkniving problem...
	OK SHARC-2 tau.225GHz problem...
	.. Update all instrument pixel data to new format
		- GISMO (J1148)
	OK Add configuration to FITS header
		OK Initial (after scan reading is done)
		OK  plus dynamic conditionals (iteration, mjd, and serial)
		OK Plus command line...
	.. Update 'extended' mode configuration
	- Reorganize options keys (hierarchical)
	- Optimize reductions...
	- Documentation
		- More on recovering extended emission
		- What's new (minicrush and crush versions)
	OK Copyright


Less critical:

	- List groups/modalities/signals
	- Ability to write signals
	- Beam efficiency modeling (bright sources)
	OK aliases
	- Poll conditionails,blacklisted,forgotten,aliases,all
	- create master/worker executables
		- Grid models:
			- data on master
				(master reads, then passes job with data)
			- local data on workers
				(master passes job according to loc, then 
				worker reads)
			- data everywhere
				(master passes job, worker reads)
	- create condor scripts (single machine/grid)
		
