[GRASS5] 5.7/6.0: some considerations

[Radim Blazek]

Jaro Hofierka wrote:
> Markus,
> 
> those "speed complaints" are just references to a possible serious bug. 
> No one says the work that has been done is bad. In such major code 
> revisions that can happen. We were testing grass57 and found a problem. 
> I was just wondering if this is an inherent feature of the new vector 
> engine or something else. If this is a bug, it should be fixed.
> I was out for a couple of days, sorry if I missed something in this 
> discussion.
> 
> Jaro

The system is more complex, with more functions and speed is the price.
I cannot say where precisely is the biggest problem, I asked few days 
ago for simple benchmark regarding the speed, just to remind:

 > Can you post here times consumed by?:
 > s.in.ascii
 > v.in.ascii -t
 > v.in.ascii / dbf
 > v.in.ascii / postgres
 > insert attributes using insert statements + db.execute (in one step)
 >   / dbf
 >   / postgres
 > v.in.sites / dbf
 > v.in.sites / postgres
 > v.surf.rst layer=0 (i.e. z coor)
 > v.surf.rst / dbf
 > v.surf.rst / postgres
 > s.surf.rst (5.4)

but until now, I have not recieved any results, so I take it, that 
people are not interested in solving this problem.


I guess that most time is spend in attributes handling. Probably the 
file+DB architecture is wrong. DB is not suitable for random access to 
large data sets from applications. Data manipulation via SQL for large 
datasets is another problem. Either files only or DB only, the 
combination is bad.
It can be improved a lot, I believe, for DBF driver, using fuctions 
instead of SQL between modules and the driver.
Some embeded database is also desirable.

It is just work which should be done and which nobody does.

I am always surprised again and again why people complain about 
something missing instead of posting contributions. Maybe it is not 
quite clear, but GRASS is free software now. Irst is not CERL. GRASS is 
  like the GRASS comunity, if the comunity is poor, GRASS will be poor.
It seems that 99.9% of users believe that there is a large group of 
developers waiting for user wishes, keen on implementing them.
That is not true, if you need something in GRASS, usually you have to 
write it yourself or pay somebody. That is principle of free software 
development. Sorry, no free beer.

Radim