I used Twitter Tools for a while with WordPress, and it worked well until recently, when it stopped.
While pondering what went wrong, I noticed Twitter now has an official set of widgets for web pages.
Twitter / Get a Widget for your site
They’ve got customised ones for MySpace, Blogger, Facebook, Typepad, and a generic one (in HTML or Flash) for everything else.
Trust Josh to ask a curly question for the StackOverflow podcast: “Why did the Stack Overflow schedule blow out?” and quoting back Jeff and Joel's own previous forecasts at them.
Made for an interesting discussion though. I certainly agree with the point that until you're actually working on something, you d
on't have a great deal of confidence in just how much there is to do … that becomes apparent as you go.
(So I can find it later: WordPress URL parameters, for example for showing all posts by Josh.)
Ever wonder how they fitted an entire computer language into just a few kilobytes, back in the 80s? Documented disassembly of BBC Basic 4.
How to highlight author comments in WordPress … but it relies on the author being user ID 1, so it won’t work here, where we have several people posting. Could easily be customised to look for other user IDs though.
Some developers are throwing in the towel and running Vista as Admin.
The excellent Secret Life of Machines not only has a web site, but is available freely (and legally) via BitTorrent. And the theme tune is available on iTunes.
Some Unicode symbols have surprising effects but… u?op ?p?sdn ?u????? s? un? s? ?l???u ?ou s,???? -, or, if WordPress doesn’t support Unicode properly, you miss out.
(Skip the lecture, go straight to the instructions — but note the update.)
I’m yet to be convinced that microblogging (eg Twitter, or those status updates in Facebook) is genuinely useful. Maybe, maybe not. But I’m willing to try it out.
Problem is of course that if you use multiple services, you don’t want to be having to update them all individually. If such a concept is going to work, you’ve got to be able to update once and have it cascade to everywhere.
Facebook has an app to push updates out to Twitter. Which would be fine, but for those outside North America, you can’t update your Facebook status from anywhere except within Facebook. (North Americans can use SMS from a mobile, but others can’t.) Okay, so maybe you’d want to do it mostly when in front of a computer anyway, but I do like about Twitter than you can update from anywhere… anybody can SMS a number (it’s based in the UK, so for me it’s costing 50 cents… so I’d better not go mad using it) so no fiddling with mobile web access just to post an update. Twitter also takes updates via IM (such as GTalk and Jabber). I also like that it’s open; people can see what’s going on without registering.
I normally hate words like synergy and leverage and convergence, but that’s what’s gone on here. Alex King has written code that updates WordPress from Twitter every 15 minutes. Christian Flickinger has written code that updates Facebook using PHP, with a hack using the Curl library (since Facebook doesn’t actually accept inputs like this) that logs into Facebook’s mobile web page and does the business.
And Blake Brannon has put the two together, so a Tweet (that’s Web 2.0 talk for a Twitter post) will cascade to your WordPress blog, and then on to your Facebook status.
Neato, huh? Now that really is leverage. If it works. Which it does for many people, but it didn’t for me. I was having problems with Blake’s code; probably an issue with my Web ISP’s configuration. I ended up splitting it off to a separate WP plugin, which was messy, but allowed me to use the code in isolation, and figure out the problem.
It may be an issue that only affects particular versions of PHP or Apache or something — I’m no expert — but the problem was the Curl call couldn’t write to the cookies file. Creating the my_cookies.txt file and making it writable (777) and modifying the code slightly to specify where the file lived solved it. Another issue involved Curl being unable to use the FollowLocation flag, but it turned out this wasn’t needed.
I also ended up with Blake’s (modified) code in a separate file to Alex’s, rather than inserted into it as Blake intended.
Update 2007-08-31: Blake’s been told that automated access into Facebook is against the Terms of Service. It’s unclear if Facebook will actively go chasing those who use or distribute code like this, but it would seem to pay to be cautious. Sorry.
I think that’s all the steps. Good luck.
Thanks to Blake for his assistance on this. And to Alex and Christian, whose code this is all built on.
After reading Jeff Atwood’s terrific post about saving bandwidth on web sites I’ve moved the Geekrant RSS feeds over to Feedburner, using Steve Smith’s mavellous WordPress Feedburner plugin, which works in WP 2.0x and 1.5x.
I also turned on HTTP compression, which in WordPress is as easy as clicking a checkbox. It not only saves you bandwidth, but users get your pages served quicker, since the bottleneck is bound to be their bandwidth, not their browser’s ability to decompress.
We’ll see how it goes. Bandwidth has been growing recently: January 2.8Gb; February 2.7Gb; March 3.4Gb. It’s not at ludicrous levels, but if it keeps climbing, I’ll end up paying more for the hosting. Hopefully this will help bring it back down.
Update 8:40pm. First thing I notice is that when reading the feed from within the Feedburner site, it doesn’t treat relative paths to images properly. I guess I’ll have to put absolute paths, ‘cos at the moment in the previous post it’s trying to load http://feeds.feedburner.com/files/2007/mediagate-mg35.jpg instead of http://www.geekrant.org/files/2007/mediagate-mg35.jpg. I wonder how it treats relative links?
A couple of MySql queries to count up your 2006 blog stats (as I did on my personal blog).
select count(*) from wp_posts
where
post_status = ‘publish’
and wp_posts.post_date >= ‘2006-01-01’
select wp_posts.ID, count(*) as wpc, wp_posts.post_title, wp_posts.post_date from wp_comments, wp_posts
where
wp_comments.comment_approved = ‘1’
and wp_comments.comment_post_ID = wp_posts.ID
and wp_posts.post_date >= ‘2006-01-01’
group by comment_post_ID
order by wpc desc
Amongst all the easy-to-spot robot comment spam, I’m getting a bunch that (at first glance) looks like it’s written by humans. Gone are the stupid out-of-context broken-English comments and links to drug sales. These all have comments that look like they’ve got a few milliseconds’ thought put into them, all on new posts, they all leave a rediffmail (Indian GMail-type operation) address, a 209.97. IP address, and a link to a web site featuring lots of links and no content.
So far I’ve been spiteful and kept the comments but wiped the URL link.
I wonder if they’re particularly targetting WordPress sites that haven’t yet been upgraded to use the NoFollow links.
We’ve got MySql problems here at Geekrant central.
MySQL said: Documentation
#1016 – Can’t open file: ‘wp_comments.MYI’ (errno: 145)
Doesn’t sound good, does it? The ISP is looking into it.
Nothing else seems to be AWOL, but I’ve taken a backup of everything just in case. Wouldn’t you know it, the backup I have of wp_comments isn’t particularly recent. Hopefully the ISP has a newer one, but if not, I’ve grabbed a bunch of comments via Newsgator’s cache. Gawd knows how I’d restore them though.
Update: Fixed. May I just say, the support guys at AussieHQ hosting are deadset legends.
Here’s the SQL to find which of your WordPress posts have the most comments:
SELECT wp_comments.comment_post_id, count(*) as commentcount, wp_posts.post_title, wp_posts.post_date FROM wp_comments, wp_posts
where wp_comments.comment_post_id = wp_posts.ID
group by wp_comments.comment_post_id, wp_posts.post_title, wp_posts.post_date
order by commentcount desc
I discovered the other week that if you put an illegal post date into WordPress, such as 29-Feb-2006, it displays as the next day, 1-Mar-2006 on the page, but doesn’t allow commenting or going to the permalink, because in the database it’s still there as 29-Feb, so it doesn’t show up if you try to click through to it.
I suspect it’s just PHP’s date handlers being helpful, so it may show up in other PHP-based software.
The otherwise very fine Bloglines RSS aggregator isn’t liking this site very much, reporting errors when trying to add the Geekrant RSS feed which works so well for most other people.
Uh, yes the feed does exist. Those who like XML can look at it raw.
It’s doing the same for some other blogs, including my personal one, having rejected it since late December. Very odd. And I’m not the only one.
I’ve contacted Bloglines support, so hopefully they’ll be looking into it.