I’ve been using WordPress 2.7 for a while now and have comprehensively tested my WordPress plugins on it. In this post, I release the results of my testing: Four plugins work fine (albeit with a non version specific bug found in one of them). One plugin seems to have problems with WordPress 2.7.

Plugins That Work With WordPress 2.7

The following WordPress plugins appear to work fine with WordPress 2.7:

• DualFeeds
• FeedEntryHeader
• LocalCurrency
• KeywordLuv

However, during testing I discovered a bug with LocalCurrency, which I’ll explain in more detail below. This bug is not specific to WordPress 2.7 and has always existed in the plugin. If you haven’t noticed it yet, you’re unlikely to.

Plugin Which Has Problems With WordPress 2.7

Unfortunately, the IFrameWidgets plugin has a problem, which appears to be related to WordPress 2.7. In fact, I suspect that this problem has existed since WordPress 2.5, but only occurs on certain browsers.

The problem is that widgets running in the IFrame widget do not appear. This seems to be related to the JavaScript no longer being able to find the height of the IFrame (the plugin hides the IFrame until it’s fully loaded, then automatically resizes it based on it’s contents).

I’m not certain of all scenarios when it will occur, but I’ve managed to make it happen with the BlogCatalog widget using Firefox 2.0.0.20. For me, there’s no problem in Firefox 3, Google Chrome and IE6.

Kirk M from Just Thinkin’, who reported the problem, seems to be having a more serious problem with the Google Friend Connect Members widget: he gets the problem with Firefox 3.0.5, Shiretoko builds, Chrome and IE 7.0.

It may be that there are two problems: a general problem which only occurs with FireFox 2 and a more specific problem with the Google Friend Connect Members widget.

I suspect the general problem occurs because the way widgets work changed back in WordPress 2.5: IFrameWidgets uses the number of the widget in the plugin logic. Prior to WordPress 2.5, you were only allow a maximum of 9 widgets, so the plugin was using the numbers 1 to 9. In versions after 2.5, there is no maximum number of widgets and the plugin is using numbers such as 320138091. It may be that FireFox 2 can’t handle numbers this large?

I am continuing investigations and hope to identify a fix soon. You can follow this issue in the comments on the IFrameWidgets page and I’ll announce the fix in a post – so subscribe to my feed if you want to know when it’s released.

LocalCurrency Bug

As mentioned above, LocalCurrency has a bug which has always existed within the plugin. The plugin works fine within a single post or page, but doesn’t always work on pages with multiple posts (ie the home page or category pages).

The plugin has been designed to work on pages with multiple posts, but it doesn’t always load the JavaScript that allows it to make the call to Yahoo! Finance. To optimise performance, the JavaScript is only loaded when the plugin finds the <–LCSTART–> tag in the post.

This is fine for single posts / pages. It’s also fine for pages with multiple posts IF the tag is in the first post on the page. However, if the fist post doesn’t have any currency to be converted (ie no <–LCSTART–> tag), then the JavaScript isn’t loaded.

I’ll be fixing this problem sometime in the near future. I’ll probably add an option to allow the blog owner to decide whether they want to convert currency on multiple post pages. If chosen, the plugin will load the JavaScript without checking the posts for the tag, circumventing the problem.

Final Thoughts

Anytime there is a WordPress upgrade, there is a chance that one of my plugins will break. This means I have to do extensive testing with each WordPress release. This takes a lot of time, but fortunately, there are normally no problems to fix.

For the WordPress 2.7 release, I did find a couple of problems – but I’ll get them fixed as soon as I can.

I was pleasantly surprised to find that my FeedEntryHeader plugin was featured on Episode 28 of the WordPress Weekly podcast. The mention was fairly brief,  but it’s nice to be noticed by such an influential podcast.

I’d like to thank Keith Murray for making FeedEntryHeader his plugin of the week. WordPress Weekly is one of the regular features of my 40 minute train ride into the Brisbane CBD. I always listen to the plugin of the week section with interest, but never thought one of my WordPress plugins would appear.

It’s probably lucky that I found out that FeedEntryHeader was featured before I listened to Episode 28. If I’d been taken by surprise on the train, there’s no telling what my reaction may have been!

Anyway, Keith mentioned the threat of having your content scraped by a splog and the role that FeedEntryHeader can play in combating this.

Ironically, I found out that FeedEntryHeader had been featured on WordPress Weekly when I followed a backlink in my WordPress Admin Dashboard. It took me to a splog, which appears to have been scraping Weblog Tools Collection! There was a copy of the Episode 28 page, word for word, including the link to the FeedEntryHeader home page.

So… I found out that WordPress Weekly Episode 28 had mentioned my plugin as a weapon against splogs, through one of these very splogs, that had scraped the Episode 28 page! This is the bit where I say that Weblog Tools Collection should use FeedEntryHeader right?

Well, as it happens, they’d been scraped by a reasonably nice splog, which had left all the post’s links intact, including the post title link pointing at the original post. However, FeedEntryHeader would have made it clearer to people that they were reading the post on the wrong blog. It could also help them with the other, not so nice splogs.

Anyway, it was very nice to have been mentioned by WordPress Weekly. It’s a great podcast and I have a lot of respect for the guys that run it.

It’s been a while since I last posted, because I’ve been tied up with some other projects. My focus is now returning to this blog, but before I get into anything of substance, there’s some housekeeping I need to take care of.

If you subscribe to the feed of one of my WordPress plugin categories, please be warned that these feeds will be removed in 7 days time. The feeds in question are the following:

If you want news about my WordPress plugins, please subscribe to the following feed:

http://www.scratch99.com/category/my-wordpress-plugins/feed

This feed will contain news and tips relating to ALL my WordPress plugins.

Alternatively, you can subscribe to the feed relating to the tag (rather than category) for each plugin. These will probably have all the news relating to each plugin, but may also contain other posts that mention the plugin in passing. These feeds are as follows:

• DualFeeds Tag Feed
• IFrameWidgets Tag Feed
• FeedEntryHeader Tag Feed
• LocalCurrency Tag Feed
• KeywordLuv Tag Feed

I’ve taken this move for two reasons:

1. I’m consolidating my categories, because I have too many of them. This will merge 5 categories into one general category for my WordPress plugins.
2. I normally stick to one category per post, but I often have news relating to more than one plugin. This will allow such news to reach everyone who needs it.

I regret any inconvenience this may cause, but hopefully it will result in a better site.

As a followup to my announcement of the new FeedEntryHeader plugin, I thought I’d share a screenshot of it in action. Predictably, the announcement post was scraped within minutes of me publishing it. Here is what the output looks like on the splog.

As you can see, the links have been stripped, but the URL remains. Hopefully anyone reading this page will work out that they aren’t viewing the original article and copy and paste the URL into the browser to read the original.

It’s also worth noting that all formatting has been removed, as have been the images. If you are reading a post that looks like this, you should be able to recognise it as a splog even if there is no copyright statement.

Marking Splog Trackbacks As Spam

In the comments of the previous post, Jonathon mentions that he is receiving trackback from the splogs. If you get such trackbacks, mark them as spam. It’s very important you do this.

If enough people mark trackbacks from a particular splog as spam, Askimet will start marking all trackbacks from that splog as spam automatically.

If you leave such trackbacks in place you may be passing PageRank to the splog. One of the reasons that we are fighting a losing battle is because of the vast number of links such splogs get.

Many of the original authors get a trackback, but don’t realise it’s from a splog because they don’t follow it. If you don’t follow all your trackbacks, you won’t know if someone is writing something nice about you, or scraping you.

Digital Fingerprint

Someone asked me why I couldn’t use the Digital Fingerprint plugin instead. I looked at Digital Fingerprint before writing FeedEntryHeader. I haven’t used it, but it seems to be a great idea and I recommend you check it out. However, it’s purpose differs from what I wanted FeedEntryHeader to do.

Digital Fingerprint’s purpose is to track who’s scraping your content, while FeedEntryHeader’s purpose is to redirect readers back to your site.

The idea behind Digital Fingerprint is that it inserts a ‘fingerprint’ into your feed, so that later you can search for pages that contain this fingerprint. This should let you find any splogs who have scraped your  post. The fingerprint needs to be unique (ie won’t appear anywhere else on the Internet).

It seems to be possible to use a copyright message as the digital fingerprint, but it would have to be generic (ie the same for all posts). There’s no ability to display the post URL, which makes sense – after all, you want to do one search to find the scraped content, not a different search for each post.

The strength of FeedEntryHeader is the fact that it includes the URL back to the original post. The idea is that users will see the copyright statement before they read the article, work out what’s going on and use the URL to visit your site to read the original post.

There are other plugins that do something similar in the footer. That’s too late. There’s less chance the reader will visit your site if they’ve already read the post. Also, some splogs chop the bottom of the post off.

Anyway, FeedEntryHeader and Digital Fingerprint have different purposes, both of them worthwhile. Although I haven’t tried it yet, I see no reason why they cannot be used together.

I recently discovered that my feed was being scraped. It’s very frustrating to find your content appearing in full on another website – so I decided to do something about it.

The result? A new WordPress plugin called FeedEntryHeader that allows you to add a copyright statement and a link to the original article, to the top of your feed entries.

Why Use It?

There are several WordPress plugins that allow you to add a copyright statement to the end of feed entries. However the copyright statement has limited impact by the time the reader gets to the bottom of the page.

The reader is more likely to take action if they see the copyright statement at the beginning of the page, along with a link to the original article.

It won’t stop feed scraping, but it should ensure that a copyright statement and a link to the original article appear in the scraped content.

What Does It Look Like?

Here is a screenshot from my feed in Google Reader:

As you can see, the following message has been added to the feed entry:

Copyright © 2007 Stephen Cronin. Visit the original article at http://www.scratch99.com/2007/11/feed-reading-models/.

Can I Change The Message?

Yes. The message is fully customisable. Simple HTML can be used and there are tags allowing you to use site name, site URL, post name, post URL and author name from the WordPress database.

The default message uses the post URL, rather than the post name, as the anchor text. This can be changed, but many scrapers remove all links, leaving only the anchor text behind. If the URL is the anchor text, it will remain.

Although this plugin was designed for displaying a copyright statement, there is no reason that you could not display other items instead.

Where Do I Get It?

Further information about the plugin can be found at the FeedEntryHeader plugin home page. The plugin is also available for download from the plugin home page.

Many bloggers have a blogroll to give some link love to the blogs they read regularly. I don’t – yet. I’m planning to create one and display 5 random links in my sidebar, but for the moment, here is a list of the blogs I read.

This list is created using the new feature in Google Reader, which allows you to create a Google Reader Blogroll. It’s a little limited with regards to the formatting options and it’s worth reading Andy Beard’s post on how to get an OPML version working, but it has a lot of promise and will do for now.

As I mentioned recently in Feed Reading Models, I’m slowly moving from aggregated feeds such as the BUMPzee No NoFollow community one, to individual site feeds. Every few days, I’ll find another blog that I’ve missed adding. So if you’re not in the list, don’t despair – you may soon be!

Without further ado…