X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;ds=inline;f=files%2Ffaq.html;h=e54b3235b6401056cfb0ed8255b99b8504166ca7;hb=45a498bb6c353ee9e522ea3b4831e4ba78335ecc;hp=cdf05b997d465db54221218928952016a2cdd7e6;hpb=6f690454bf05cc58b82c22308d4fb66b75f90d24;p=pr0n diff --git a/files/faq.html b/files/faq.html index cdf05b9..e54b323 100644 --- a/files/faq.html +++ b/files/faq.html @@ -9,7 +9,7 @@
Last updated August 1st, 2006
+Last updated July 21st, 2008
Sorry, no. First of all, please don't use any "web mirroring" program to - download all the images -- of course, I can't stop you, but you're + download all the images â of course, I can't stop you, but you're putting a lot of unneccessary load on the system. There are two main reasons for not downloading all the pictures: First, there is a question of copyright; not all images here are taken by me, and I've been given permission to display them here, not to pass them on. Second, - keep in mind that some of the events contain several gigabytes of images -- + keep in mind that some of the events contain several gigabytes of images â do you really need all that? I'd advise you to crank up the thumbnail size to the maximum possible size instead; it's quite comfortable to browse images on without having to click back and forth all the time. - (Either that, or use the fullscreen feature -- the most obvious + (Either that, or use the fullscreen feature â the most obvious keyboard shortcuts are in place, so it should be comfortable enough for most uses.)
@@ -54,49 +54,50 @@ so the next time somebody views the same images in that resolution, it will be snappy as usual. -Because it didn't fit my needs, and the same goes for all other systems I've seen. I wanted something no-nonsense that would work for my - purposes -- I don't want to click around endlessly just to watch some + purposes â I don't want to click around endlessly just to watch some pictures. Others are of course free to do as they wish, I can't impose my will on anybody :-)
Mostly that it's no-nonsense and just works, without being in your way. - Also, it has dynamical rescaling (of good quality -- proper, + Also, it has dynamical rescaling (of good quality â proper, sharp thumbnails, no crappy nearest-neighbor scaling) of both thumbnails and images (most client-side scaling sucks quality-wise, unfortunately), an easy-to-use WebDAV-based upload - interface and in general good performance (being a set of persistent, - optimized Perl modules; I've seen it throw out over 300 hits a second, - but I won't guarantee it would withstand a Slashdot attack ;-) ). Also, - it has quite OK skinning capabilities, so it's able to adapt into - different designs quite easily.
+ interface, cache awareness and in general good performance (being a set + of persistent, optimized Perl modules; I've seen it throw out over 300 + hits a second even without the Squid in front, but I won't guarantee it + would withstand a Slashdot attack ;-) ). Also, it has quite OK skinning + capabilities, so it's able to adapt into different designs quite + easily.pr0n currently runs on an Athlon 64 X2 3800+ with 2GB RAM and ordinary +
pr0n currently runs on an Intel Q9450 (quad-core 2.66GHz) with 8GB RAM and ordinary SATA disks. (The server does a lot of other stuff besides running pr0n, of course.) pr0n itself is a custom-made system by myself, tightly coupled - into Apache 2.0, Apache 2.2, mod_perl 2.0 and ImageMagick 6.x (as well as various other Perl modules), using PostgreSQL 8.1 as the back-end + href="http://www.postgresql.org/">PostgreSQL 8.3 as the back-end database for metadata et al. The base operating system is Debian etch (ie. âtestingâ).
+ href="http://www.debian.org/">Debian lenny (ie. âtestingâ). -The Perl modules aren't really that big -- we're talking about only - approx. 2200 lines of code (of which ~25% is the WebDAV part; I should +
The Perl modules aren't really that big â we're talking about only + approx. 3300 lines of code (of which ~25% is the WebDAV part; I should really make that a bit cleaner once). Most of the real work is done by the software on which pr0n builds on.
At the time of writing, approximately 52GB of image data (that is, over - 42000 different images), plus cache, plus metadata in the SQL database. +
At the time of writing, approximately 140GB of image data (that is, over + 72000 different images), plus cache, plus metadata in the SQL database. (These numbers are growing rather rapidly, so they could be outdated at any given time.)
@@ -107,8 +108,9 @@ custom configuring; this is not a pre-made, user friendly package for your favourite Linux distribution. There is a bzr repository at http://bzr.sesse.net/pr0n/, but - going to hold your hand configuring it. :-) (Hint: If you do not know what - bzr is, and cannot find out on your own, pr0n is not for you.) + I'm not going to hold your hand configuring it. :-) (Hint: If you do not + know what bzr is, and cannot find out on your own, pr0n is not for + you.)