Welp, another Greasemonkey, another problem. More barriers have been put in place between content and user scripts, and though this is a good thing for security, it screws up some of my scripts largely due to my own lazy coding practices. All of the user persistence scripts (Yahoo, Google, Flickr) have been updated to work in the new Greasemonkey (called 0.5, announced here).

While I was doing that, I made sure they work in Deer Park, the test version of the next Firefox. This is not a given, as Greasemonkey under Deer Park does some things differently than under current versions of Firefox. So when a version of the browser that is based upon Deer Park comes out, these three scripts at least will be ready to go. The rest of my scripts… well, when we get there, we get there.

I’ve updated the Flickr Photo Page Enhancer, Flickr Thumbnail Enhancer and Bloglines Continuous Update greasemonkey scripts to use plain old XMLHttpRequest instead of the GM_ API version. There was no real reason for them to use GM_XMLHttpRequest other than, well, that was the example I cribbed from. The point of all of this is that they now work with the “safe mode” Greasemonkey 0.3.5. As a happy side effect, they run noticeably faster than the old versions.

There’s something of a tempest about security over on the Greasemonkey list. Turns out all of the GM_* API functions are available to potentially malicious code on any website where a script is called (this is bad). Stopgap measures:

1. disable all scripts which are called for any page,
2. download and install this neutered version of Greasemonkey which removes all the GM_* functions (many scripts will break), or
3. uninstall or disable Greasemonkey.

Personally, I feel that last one may be a bit of an overreaction. Hopefully, this will be fixed very soon (these folks are brilliant), but now that the vulnerability has been exposed, you can rest assured someone will waste no time taking advantage of it.

I finally got around to making cold-brew coffee as laid out in this article that traversed the blogosphere a little while ago. I did notice a reduction in the heartburn factor, which is a major selling point if you drink as much coffee as I do (hint: too much). But the taste, well it’s not unlike turkish coffee or “camp coffee” in that it definitely has a detectable aftertaste. Frankly, it’s pretty nasty. I’m willing to give it a couple more tries to see if I can get it right, but I’m thinking at this point that cold-brew is best suited for making cofee based drinks and that plain old coffee is best for coffee.

Okay, as previously mentioned, I am (or rather, some of my work is) going to be in a book. All that remains after the paperwork is filed is for me to write a short, one paragraph blurb about myself for the contributors page. This is where you people come in. I’ve written an average of 200 words a day for over eight years now, and I like to think I can riff on any topic at a moment’s notice, but on this one paragraph I’m totally paralyzed. Help! What sort of things should I include, besides my URL? What tone should I take? Should I try to match the tone of the work? Should I make fun of those faux-serious author bios in the back flaps of novels? The thing that keeps bugging me is that this book may end up in front of quite a few eyeballs. Eyeballs of people who are right behind the crest of bleeding-edge technology, as ultimately frivolous or profound a technology as Greasemonkey/user scripting may turn out to be. People who may have it in their power to give me a job. Okay, now taking all that into consideration, go.

I did kind of imply that I would make improvements this Greasemonkey script that adds a “copy this link to Yahoo My Web” functionality to del.icio.us. Specifically, that the script should copy tags and descriptions as well, not just the link and the title. Unfortunately, I had to rewrite the whole thing to get it to work like I wanted it to. So don’t blame the author of the original script for any mistakes I may have made. Install: delMyWeb.user.js

Quickie Greasemonkey script: Flickr Tag Quick Edit. Next to the little [x] to remove a tag, you’ll now find a link directly to the page that allows you to edit all instances of that tag.

New Greasemonkey script: Flickr: Thumbnails Enhancer. It adds the number of comments and notes to each photo listed in the Photos from your Contacts page any list of thumbnails on Flickr. I know this is much less detail than is provided by the php version, but I find that the number of comments & notes is really the most useful information. Plus, I gave up trying to cram the description and all the tags into the design of the page. Suggestions welcome.

• Update 10:58am EDT: I can’t think of any reason this shouldn’t work on any page where you’re presented a list of thumbnails, including tag pages, photo pools, sets, favorites and archives by date (any others?). I’ve changed the name of the script to reflect this. Uninstall and then install it again, if you already have.
• Update 2:14 pm EDT: Now works in sets and favorites, as well. I think it now works anywhere in flickr where there are thumbnails. As an aside, I’m really glad that flickr uses semantic, standardized markup — it made this so easy to write! And yes, that is sarcasm! It was like chewing glass!

I’ve posted about Mozilla/Firefox Quick Searches before, but here’s a new one: Flickr Tag Search. Bookmark that, give it the keyword ‘ft’, and type ft tag1 tag2 etc into your location bar to do a tag search for photos on Flickr. Searching for more than one tag only works when Flickr’s search servers are running correctly, though, which as of this writing they are not.

In other news:

• The Disable Target/OnClick for Downloads Greasemonkey script has been updated. It uses regex now instead a huge nested for loop. Translation: it runs faster.
• The Bloglines Continuous Updates script is more or less perfect now, so I listed it on the directory. Thanks to all the early adopters for their feedback and suggestions.
• The Enhanced Contacts’ Photos widget now accepts either a username or NSID. It’s just easier.

You may have begun to think that I seem obsessed with web API’s lately. I just want to assure you, gentle reader, that you’re right. Today, the Flickr API. I set out to enhance the photos from your contacts page so that it actually showed some information about each photo and not just a thumbnail. I want to see not just the photo and who it belongs to, but when it was uploaded, the title they gave it, any description they gave it, any tags assigned to it, and a notice if there are comments or notes on the photo. To do this, we

1. take a Flickr user id (NSID, the long number with @ in it),

2. query the “contact’s photos” RSS feed for that user id,
3. and place an API call to flickr.photos.getInfo for each photo in that feed.

The rest is presentation. You can see it in action here, showing my contacts’ photos. Change the query string to your own NSID (update 7/8: or your username) to see a page with your contacts’ recent uploads. Please be kind, though. The script does use caching, but it also pulls your RSS feed and then places an API call for every photo in the RSS feed.

More noodling with the My Web API: My Web tag cloud for URL does pretty much what it says. Surprisingly, this isn’t a built-in feature of My Web (…yet). It’s limited to the top 50 tags for a given URL. View the tag clouds for Flickr or Yahoo 360° for a demonstration.

As everything must have a bookmarklet: Y! My Web Tag Cloud.

Update: For an example of tags from bookmarks not marked as public showing via the API, see the tag cloud (such as it is) for my blog.

Why I think everything should have an RSS feed: because I can do really cool things with them. This page collects and collates everything I publish anywhere on the web, so long as that data ends up in an RSS feed at some point. So far I have My Y!360°, linkblog from del.icio.us, Flickr photos and comments on those photos, blog, and videos on YouTube. Best of all, the page has it’s own Atom feed (update: and RSS 2.0)! All made possible via Magpie + PHP. It’s not that hard to imagine using this same technique for something actually useful. Want to combine RSS feeds like Feedburner does, but including more than just Flickr and del.icio.us? It’s quite possible. You can see the source that makes it all run, but fair warning, it’s kind of a mess now with comments, and about as refined as it’s going to get.

Update 7/5: Dean Jackson published something somewhat related — if more complicated, with prettier output, and done up in with Feedparser and python instead of Magpie + PHP — over at the W3C. Mine has a lower barrier to entry, I feel, but then I already know how to use PHP and Magpie.

Boredom brought me to create this page where you can see my top 50 150 Myweb tags as a tag cloud. This is my first project with the MyWeb API (also making use of cached_fopen_url and xml2array). I’m seeing how it might be possible to create a much richer linkblog making use of the MyWeb API than is possible with del.icio.us’s RSS feeds. I could have the full archives right here on my site and in my design. The major stumbling block so far would be that the descriptions in del.icio.us don’t import into Myweb properly. We’ll see.

Also, if you’re logged in to Myweb, click on the following link to import your latest del.icio.us links into Myweb: del.icio.us » Myweb bookmarklet. You can bookmark the resulting URL for future imports.

Vox Googli: continuing a semi-regular feature, some interesting google queries and their results, with no commentary whatsoever.

• President George W Bush - 5,900,000 results.
• Dubya - 1,240,000
• Bush II - 166,000
• President Shrub - 941

And Furthermore:

• “George W Bush” +Evil - 1,510,000
• +Criminal - 1,180,000
• +Stupid - 956,000
• +Idiot - 512,000
• +Antichrist - 70,000

Redesign! To understand the true brilliance of this, A) click around a little bit — it’s sensitive to context, and B) turn off javascript and css, and marvel that the page is still kind of useable. You will come to understand the power of DHTML and PHP. Search your feelings. Join us.