On 11/01/2014 11:02 PM, Joseph R. Justice wrote:
On Sat, Jan 11, 2014 at 10:04 PM, Tom Taylor <tom.taylor.s...@gmail.com
<mailto:tom.taylor.s...@gmail.com>> wrote:
I started a virus scan a couple of days ago and wondered why
scanning seemed to hang up on a particular file. Investigating
further I found it was a folder -- of cached Bing images dating back
from the first days of my remote mapping activity to the present.
There were 450,000 of them! It has taken me two days running at 100%
computer utilization much of the time to get rid of them. (Maybe I
did it wrong. Simply deleting the containing folder containing the
last 150,000 just took a couple of minutes.)
Oops, re the 450,000 files. As for the time / effort it tool to remove
them, you might have been deleting them one by one from the directory,
especially if you were doing it from a GUI interface than from the
command line -- again, Oops.
Caching images for a given session makes lots of sense, but leaving
them forever really does not. How do I put a request in to the JOSM
developers to ensure that images caches are deleted upon exit from JOSM?
I agree that purging old and particularly obsolete data in an image
cache makes sense. I also agree that it makes sense to initiate purging
the cache without requiring the user to initiate the purge.
However, I disagree that the image cache should always be automatically
purged when exiting from JOSM. What if the person restarts it right
away, or next day, or next week, and wants to work on the same area they
were previously working in? In this case, the data in the image cache
at the time of the exit is likely still useful, and it would be a waste
of downloading time and network bandwidth to download it again.
I also disagree that the image cache should always be purged without
user intervention. (I agree that this option should be available to be
selected by the user, but it probably should not be the default.) Maybe
the user knowingly never wants the cache automatically purged, for
whatever reason. Maybe the user wants to know what's going on, and have
some control as to how much is purged or kept.
I think a better choice would be to have JOSM generate a notification
upon exit when the cache is above some (potentially configurable) level,
saying "The image cache is over N bytes / N files / has data over N days
old -- would you like the oldest information over this limit purged?"
Perhaps some sort of sliding scale configurator, showing how much is in
the cache now and how much free space will be regained by deleting data
more than a day / a week / a month / N days old.
...
Thanks for your reply. I agree with your suggestions, really, and
thought about them myself. It's why I specified "exit" rather than
"restart", for one thing. I just wanted to make things simple for the
developers.
For myself, I was going to apply the equivalent of your policy by
manually deleting the cache when I moved on to a new area.
Tom
_______________________________________________
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk
