I was writing a quick query the other day that showed how many people registered on a site over time. I thought that it would be nice to have a graph-like representation of this, so I didn't actually have to read the numbers.
Then I thought, what if I could do this in SQL rather than having to mess about with Excel or similar. So, within a minute I'd use the padding and replace to get a rough-n-ready bar chart. The result is below, I quite like it :)
Here's the SQL
select
Convert(varchar,DateConfirmed,106) as Date,
count(*) as ConfirmedRegistrations,
replace(space(count(*)),' ','|') as Chart
from
Operators
where
DATEDIFF(day, DateConfirmed, GETDATE()) < 30
group by
Convert(varchar,DateConfirmed,106)
order by
1 desc, 2 desc
This was done in SQL Server 2005 (T-SQL), but I reckon you could do it in anything with a SPACE and REPLACE function.
If you want to see the output rendered in text, here it is:
Date Total Chart ------------ ----------- ------------------------------ 31 Jan 2008 21 ||||||||||||||||||||| 30 Jan 2008 20 |||||||||||||||||||| 28 Feb 2008 10 |||||||||| 27 Feb 2008 18 |||||||||||||||||| 26 Feb 2008 13 ||||||||||||| 25 Feb 2008 19 ||||||||||||||||||| 24 Feb 2008 8 |||||||| 23 Feb 2008 8 |||||||| 22 Feb 2008 25 ||||||||||||||||||||||||| 21 Feb 2008 27 ||||||||||||||||||||||||||| 20 Feb 2008 28 |||||||||||||||||||||||||||| 19 Feb 2008 38 |||||||||||||||||||||||||||||||||||||| 18 Feb 2008 16 |||||||||||||||| 17 Feb 2008 18 |||||||||||||||||| 16 Feb 2008 13 ||||||||||||| 15 Feb 2008 31 ||||||||||||||||||||||||||||||| 14 Feb 2008 25 ||||||||||||||||||||||||| 13 Feb 2008 23 ||||||||||||||||||||||| 12 Feb 2008 27 ||||||||||||||||||||||||||| 11 Feb 2008 17 ||||||||||||||||| 10 Feb 2008 7 ||||||| 09 Feb 2008 11 ||||||||||| 08 Feb 2008 25 ||||||||||||||||||||||||| 07 Feb 2008 25 ||||||||||||||||||||||||| 06 Feb 2008 19 ||||||||||||||||||| 05 Feb 2008 31 ||||||||||||||||||||||||||||||| 04 Feb 2008 30 |||||||||||||||||||||||||||||| 03 Feb 2008 12 |||||||||||| 02 Feb 2008 10 |||||||||| 01 Feb 2008 15 ||||||||||||||| (30 row(s) affected)
If anyone's into visual representation of information, then I'd recommend Edward Tufte's "The Visual Display of Quantitative Information". I also frequent his cool Ask E.T blog on all things UI and design.