After eliminating all other possibilities, the one remaining-no matter how unlikely-must be the truth.
Here’s this week’s top post:
Animal Crossing Tragedy (New)
How to download HD Trailers from Apple.com (Last week: #1) 4 Free Xbox Live Arcade Games (Last week: #2) Haloid (Last week: #3) Music from The Inner Life of a Cell (New) Music from The Inner Life of a Cell – Available for Purchase (New)As I mentioned in yesterday’s entry, my view counter appears to be broken. Therefore, I’ve removed the plugin from my site and have decided to go with a Weekly Top Posts area. The Most Viewed list was getting a bit stagnant and and I’ve always wanted to do Weekly updates, but the view counter didn’t really support something like that.
Right now it’s a manual process, where I look at Google Analytics and just mark down the IDs of top 10 posts for the past week with the highest view count and put it into an array.
This weeks top post are:
This will most likely be a weekly thing and I’ll probably include the previous week’s stat as well, sort of like those music charts saying last week it was at position 5 and with an up arrow if it went up to spot #3.
So when I was checking my stats today, I noticed a GIGANTIC spike:
My average daily visitor fluctuates between 250 to 350 as you can see from the flat line. However yesterday, it jumped to over 1700!!!
According to the stats, 84% of yesterday’s traffic came from Digg, Google, and StumbleUpon and as you can see from the lower bottom corner, the main attraction appears to be the post on Animal Crossing Tragedy, with over 1900 page views.
I also realized my view counter is broken, which currently registers only 44 views. I had a suspicion for the longest time that it was broken, but couldn’t really tell for sure. Google Analytics often showed particular pages having a lot higher view counts than what my counter showed. I believe it may have to do with the caching that I enabled and I guess when it loads a cached page, view count doesn’t get incremented. And since cached pages expired hourly, the max it’ll increment by is 1 per hour. Cached pages are also cleaned out whenever I modify an entry and so on. I will probably look into that later.
Once again… wow!
Ever wondered what happened to all my random crap? I guess I can’t really assume what you see on my site isn’t random “crap” to begin with, but I’ve just run low on time to post everything I find interesting these days. Therefore I’ve decided to stop posting random links and stuff. If I find something really amusing and interesting, I’d still post it, and it’ll probably get its own entry.
But I figured a way to still show you random crap, with me doing minimal work.
There are 2 main sources where I get most of my random crap: Digg and RSS feeds. The most obvious thing is to share with you what I’ve dugg: Digg / krunk4ever / History / Diggs.
As with RSS feeds, bloglines has this blog/clipping feature which is pretty neat. It allows me to clip and save off particular entries: Krunk4Ever – Powered By Bloglines
I’ve added these links to the sidebar, so if you ever miss my random crap, you know you can find something interesting in one of these links.
So if you haven’t noticed, I’ve updated my title image. It originally featured Detective Conan and Shinichi in the background with the saying “Krunk4Ever! It’s the only truth.” You can still see the previous title image here. However, while updating to WordPress 2.3 and having to fix up all the stylesheets and theme changes, I got bored and decided to make a new title image.
I started searching for more Detective Conan images, but didn’t really find one that I liked or could work with. Then I searched for Rurouni Kenshin, Bleach, and Naruto. I found a Naruto one that I liked and even created a title image that was on my site for a few hours. If you didn’t get to see it, you can check it out here. It was going to say “Krunk4Ever! Some anger just can’t be quelled.” It looked pretty decent, but I decided to see what other options there was.
I also got to a point where I wanted to have a black and white title image, basically a silhouette type image (think iPod commercials). I was playing with at few images and after I converted them to silhouettes, they looked nothing like the original, so I gave up on that idea.
While browsing for more images, I ended up on this page: Ginko, the snow, the tree, and the Mushi and immediately overtaken with a very peaceful and calm feeling. The image is from an anime called Mushi-Shi (蟲師) which literally translates to bug master, but according to BasuGasuBakuhatsu:
The literal translation for “Mushi-shi” sounds sorta like “Bug Tamer,” but I think the actual meaning is a bit more interesting. At least, if you’re not into entomology…
Mushishi are not bugs. They’re actually pseudo life-forms that were around before life even existed. They usually can’t be seen, but their effects on people can be pretty strong. Ginko, a “Mushi-shi,” travels the world ridding people of their Mushi afflictions and learning more about them. Aaaand that’s about it.
I haven’t checked out this anime yet, but it has a pretty high rating on AniDB. Looks like something to add to my watch list. House, M.D. is once again postponed until further noticed.
Anyway, back to the topic. I first created this one, which following the previous title images, had a white border. However Hjo3 noted that the white borders were a bit weird given the fact that the new image also had a white border, so I decided to give moving the trees to the edge a try and came up with what I have now. It turned out pretty nice. I’m not sure which one is better, but I’ll leave this one for now. Since there was this little man walking, I decided to make up some saying dealing with “journey”, and ended up with “Krunk4Ever! A journey of amazing experiences.”
Just upgraded to WordPress 2.3. Let me know if you see any problems.
Update: Still in the process of fixing up a bunch of deprecated functions and learning what’s new new. I’ve also adopted Hjo3’s +/- for hiding/showing my menus on the right.
I noticed for the longest time that updates to my blog posts wouldn’t propagate over to Bloglines. Digging a bit deeper I noticed that in my atom feed, the <updated> and the <published> timestamps were the same. Looking at the code generating those 2 lines revealed the problem:
<updated><?php the_time('c'); ?></updated>
<published><?php the_time('c'); ?></published>
It used the same code to generate the updated time and the published time. I scratched my head trying to figure out if there was a way to determine the last modified time for a particular post/entry. I didn’t want it to just returned the last modified time for the entire database, which is what the atom feed uses for the feed’s last modified time.
Searching online, I found this bug on WordPress: Atom feed entry updated field should use get_post_modified_time, not get_post_time. Apparently the correct thing to call is:
get_post_modified_time('Y-m-d\TH:i:s\Z', true);
Resulting in:
<updated><?php echo get_post_modified_time('Y-m-d\TH:i:s\Z', true); ?></updated>
<published><?php the_time('c'); ?></published>
Yep. That fixed my atom feed’s entry last updated time. Obviously I’m not using the latest WordPress version and although updating WordPress isn’t difficult, it’s not easy either. Updates should be fast and simple. One click that does back up, download newest version, install, and prompt me for any extra things I need to configure. If anything goes wrong, roll back to what it was previously.
If any of my videos/music/animations weren’t playing for you on my blog or gallery today, it was due to a recent change that I had made to combat hot-linking which has caused about 12GB of data transfer in the past 30 days. Baidu.com (a big Chinese search engine) apparently linked directly to an mp3 I have in one of my old blog posts: 永恆的記憶, and it almost reached 30,000 requests. My statistics show that ~12GB of data was transferred during the last 30 days and that mp3 was responsible for 98% of it.
I found out some interesting things.
First, my regex for my RewriteCond in my .htaccess file was incorrect:
!^http://(.*\.)?krunk4ever.com(/.*)?$
which I realized now that any URL that included “krunk4ever.com/” as part of the URL would pass right through.
I fixed my regex so now that krunk4ever.com has to be the domain:
!^http://([^/]*\.)?krunk4ever.com(/.*)?$
Next I was looking through my access logs and it appears that when you embed a media file (such as an mp3 or video), the referring URL is null and the agent is actually the media player, in which case, it was showing up as WMP11 mostly.
I’ve always allowed direct loading (as in copying and pasting the link into the browser), but decided to prevent cases like these. I went ahead and also blocked referring URLs that were null/empty.
Xyon pointed out to me earlier today that my videos weren’t playing, but if he clicked the “Download Movie” link, it’d work.
Of course! The purpose of blocking the null/empty referring URL was to prevent embed media files that link to my site. It would make sense any embedded media on my site would hit the same problem. Therefore, I’ve fixed the access rights again so that only that particular file will be blocked when referring URL is null/empty. All others should be fine.
If you haven’t seen the static electricity video (which was broken earlier today), do get a chance to see magic!
I’m porting over more blog entries from my old blog, so if you see a random blog post in awhile, it probably means I forgot to set the timestamp. It should be fixed shortly.
So a few days ago, I was playing with my robots.txt and started to do some research. While reading the Wikipedia entry, I noticed that I could provide a Sitemap, which apparently Google, Yahoo!, and MSN would read. A sitemap is basically a list of all your pages so search engine bots don’t have to slowly crawl to find every page. For some reason, both Yahoo! and MSN/Live has problems indexing my HD-Trailers.net site, so I thought maybe a sitemap would help. Google also has a problem indexing my Gallery as it’s slowly increases about 100 pages a week, while still missing 2000+.
So a quick search revealed that both WordPress and Gallery had automatic sitemap generators:
Installing the plug-in/module was rather simple and enabling either was just a few clicks. After the sitemaps were generated, I used the Google Sitemap Validator to see if there were any problems. Apparently the WordPress plug-in issues a priority of 1 instead of 1.0 which the validator didn’t like. I began looking at the code to see where I could fix it, but it seemed lke more hassle as they had some weird calculation converting ints to strings and vice versa. I ended up just setting the homepage to 0.9 in the control, thinking 0.9 isn’t that much different than 1.0.
Now I had to create a site index for my main HD-Trailers.net page. The protocol documentations were pretty helpful and given that I already had 3 sitemaps as reference guides, I whipped up some code to create the sitemap for the main page.
Reading on, it turns out that robots.txt has to be in the root directory and it only supports 1 sitemap per robots.txt. So given that there’s both a blog sitemap and the main page sitemap, I needed to merge the sitemaps into one, which wasn’t too difficult of a task.
However, I found out later that there’s also this sitemap index format which I could’ve used to point to multiple sitemaps instead of merging them. Maybe I’ll change that later. For now, it should do its job fine.
After your sitemaps are ready, you can submit them to Google and Yahoo!. I couldn’t find one for MSN/Live, but maybe they’ll be able to pick it up from my robots.txt.
