GNOME.org
 

Re: chart either does not work or is not intuitive

On Fri, 2008-09-05 at 16:06 -0700, Ralph Shumaker wrote:

Andreas J. Guelzow wrote:

Now that we know what yo udon't like could you tell us what you would
like?

I suspect the main problem may be the fcat that you have the one
outlying data point: 090705. This will mess up any automatic scaling.
You may want to replace it with a formula to calculate a value a month
in the future or so. 

Andreas
  


As already stated, I suspected something similar and changed the dates
to a simple 1 through 8.  It made no difference.


Okay,

I still don't see the difficulty. See the plot at 
http://www.math.concordia.ab.ca/~aguelzow/screen.png 
that took me
about 2 minutes. Perhaps you have the problem that you have only few
data points and many series, so gnumeric will guess wrong when you
create an xy plot and so you need to adjust the series.

Andreas





On Fri, 2008-09-05 at 14:14 -0700, Ralph Shumaker wrote:
  

Jean Bréfort wrote:
    

Le jeudi 04 septembre 2008 à 17:16 -0700, Ralph Shumaker a écrit :
  
      

I have wasted several hours now.  I haven't tried to set up a chart for
many years.

I have 3 different programs that can do charts, and every one of them is
doing the same brain-dead mistakes.  So, I must come to the conclusion
that the way I used to do it must have been in some outdated program and
does not work any more.

I have a small set of data that I'm trying to set up a chart to
display.  It's small enough that I can just give you the data.

If you see this email in a fixed width font, you will see it as I
present it:
 Date   act1  act2  act3  act4  act5    acctTot   act6
070615  2605     0     0   737     0   =sum(b:f)   300
080110  1504     0     0  1564     0   =sum(b:f)  1500
080229  1280     0     0  2657     0   =sum(b:f)  2350
080305  1280     0     0  2657     0   =sum(b:f)  2450
080310  1280     0     0  2657     0   =sum(b:f)  2650
080330   825     0     0  2657     0   =sum(b:f)  2850
080830   194     0     0   235     0   =sum(b:f)  3406
090705     0     0     0     0     0   =sum(b:f)  4000

In the above, =sum(b:f) is abbreviated, but on line 2 is actually
=sum(b2:f2) and on line 9 is =sum(b9:f9), with like progression between
them.  Also, the final date is just a final line set far enough in the
future that I can insert rows as I get new data.  Inserting takes care
of formula expansion, whereas adding a row makes me have to track down
all formulas that need to be adjusted to include a new row.  It's like
if I were to insert a column between b and c, the formulas in g (shoved
to h) would then say =sum(b:g) without me having to lift a finger.  I
remember doing this with previous charts, inserting new data and
successfully see the chart adjust itself.

I don't know why I cannot get the dates (erratic as they are) to by my
X-axis, columns b thru h to be data lines, and the amounts to be my
Y-axis.  I even tried changing the dates to simple numbers, 1 thru 8. 
The charts that would come up were just astoundingly disconnected from
what it should have been.

For example, the first data point on the line for act1 should start off
on the far left (at date 070615) at a height of 2605 of a total chart
height of 0 to 4000.  Ideally, at the appropriate distance from the left
side should be the second date, 080110, where the second data point for
the line of act1 should be at a height of 1504.

Perhaps someone can tell me what I'm doing wrong?
    
        

Which plot type are you using?
Can you provide a screenshot of what you get?
  
      

I tried every plot.  None gave me what I wanted.  I took screenshots of
each of them and uploaded them here:
http://www.flickr.com/photos/30238627 N07/

I didn't feel it would be appropriate to upload them to the entire
gnumeric mailing list.  Anyone interested can click the link.



-- 
"Liberty consists less in acting according to
one's own pleasure, than in not being subject 
to the will and pleasure of other people. It 
consists also in our not subjecting the wills 
of other people to our own."  Rousseau

Andreas J. Guelzow
Pyrenean Shepherds
[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]