Here are a few ideas I have on some cool applications I would like to see. Hopefully, someone will have some use for them, because I don’t expect to ever have the time (or the complete interest) to implement them.

1. A Wiki Site that Charges the Reader. There are lot of wiki sites out there (stikipad, PBwiki, Jotspot) whose model is to charge the content creator for using the service, but I don’t know of one that actually attempts to charge the content consumer. You could offer various ways to charge the consumer from revenue sharing via Google Adsense with the author, to donation requests, to just a straight charge. Also, offer options to print to PDF and author defined time-constrainted access (for example, allow a buyer to access the content for a year). To help the creators make a sale, you can create a marketplace for content consumers to request content so authors can bid to provide it. Another feature would be to provide a full text search on all content, but only show an abstract to the consumer before he pays.
2. A Wiki-like Site for Building Matricies. Make building matricies as easy as creating wiki pages. I got the idea while I was looking at a feature comparison matrix for some software, and thought wouldn’t it great if there was one place I could go to and look up the feature matrix of various types of products? But taking it a step further, what if you could just tableize anything, and the site will automatically graph or chart it? What if it also has the ability to join arbitrary tables, what kind of results could you end up with?
3. A Simpler Word Processing/Spreadsheet Site using Templates.  Instead of creating a web-based word processor or spreadsheet application that is comparable in functionality to Word and Excel, I would like to see ones with just very basic functionality, but with a large set of usable templates.  If you provide users canned templates for building the most common documents, I imagine that most users won’t miss the bells and whistles in Word/Excel.  The problem of course is how would a user create a template if the application doesn’t have all the necessary functionality to build robust templates?  Maybe import them from Word/Excel?

Imagine my surprise when just few minutes ago I did a (Windows Key + M) to show my desktop and got this message. It took me a little while before I noticed the Safari window trimmings and realized what I was looking at.

Since creating my Flickr background script, I’ve seen some amazing photos that took my breath away, but this one actually made my heart to skip a beat.

I usually get my backgrounds from a couple of photo blogs I subscribe to, but using Flickr for my backgrounds was such a good idea (and a good excuse to play with ruby), that I thought I’d give it a shot.

After some trail and error, and some help online, I was able to come up with this ruby script to do the trick. The script will pull the interesting photos for that day, and will try to find the best fit for your screen resolution. If you don’t like the current background, you can run the script again to change it. With a simple cron entry you can have fresh background at periodic intervals (I have mine set to every 2 hours).

I created the script under Ubuntu GNOME. If you want to give the script a try, you’ll have to edit the script and put in your Flickr api key and shared secret as well as your screen resolution (unless you’re runing on 1440x900). You’ll also need to have rubygems and the rflickr gem installed.

What’s in this release:

• Fixed a bug which caused multiple tabs to be closed when the Tab Mix Plus extension is installed (thanks to onemen at the TMP forums for helping me figure this out).
• Added support for the include microformat pattern, which is apparently being used at Yahoo! Tech.
• Added better ratings handling for hReviews.
• Added # of objects to the tab title.
• Added support for hResumes.
• Added an update URL to the extension, so the next release will be updatable via Firefox’s extension update mechanism.
• A few bug fixes that would be too boring to describe detail.

In addition to these changes, I’ve put the source code on Google Code, under the MIT License.

So what’s next for Tails? Here’s what I have left on my list:

• Make the overlay box resizeable.
• Display the relationship between parent/child microformats. Right now, I’m displaying all objects found on the page, even if the object is really a child of another (the hCard author of an hEntry blog post, for example). While this is good in some ways, I would like an intuitive way to display these relationships. I’m still trying to figure out what this is.
• Pull avatars from external sources, like from Gravatar and Who-R-You.
• I would also like to figure out a way to pull external data to complement the data in the microformats. The one method I have came up with so far is to use claimID to pull all the links associated with that user, and display the data from the services that the user has links to. So for example, if the user has a link to his Flickr account, I could display that user’s most recent images. However, since claimID isn’t that widely used just yet, I would like to figure out another solution that would give me the most bang for my buck.

Anyway, enjoy. And send me your feedback.

Update: You can get the extension here.

Robert hates microformats.

Although Tails heavily depends on microformats right now, I never really saw it as being an extension just for dealing with microformats. That’s why I chose the name Tails as oppose to My Big Ole’ Honkin’ Microformats extension. Tails is about viewing and dealing with embedding data on web pages, and microformats is currently the best way to do that on the web. When embedded XML w/ CSS catches, Tails will support that too.

Having said that, I don’t see microformats going away very shortly, at least not because of embedded XML. Creating microformatted content is hard enough - I can’t imagine a lot of content writers and/or web designer buying into this over-engineered approach and having to worry about multiple XML schemas, namespaces, and properly formatted XML when creating their web content, even in the presence of tools that will purportedly make this easy. This stictness/wellformed-ness is too counterculture to the web to be massively adopted.

Changes in this update:

• Added tabs to separate the different microformats.
• Added support for hAtom microformats.
• Some slight UI tweaks (horizontal scrollbar is now gone).

In addition, I also added a page documenting the Tails Script creation, and added a Tails script for del.icio.us and digg. Enjoy.

The subject says it all, you can download it here.

Next thing on my list is to be able to seperate the different microformats into tabs. I’m also starting to add hAtom support, and documenting the API.

[Update 10/8/2011: Made the script available via a gist. Updated the download link.]

Having switched to GMail as my primary email client for a couple of months now, I’ve decided that it was time to try to copy my old emails from Thunderbird to GMail. From doing a quick search, I find that the tool to use is Mark Lyon’s GMail Loader tool. Unfortunately, that’s been the only quick thing so far in this exercise.

Note that I’m using Ubuntu Drapper Drake. You might have better luck on other systems.

It appears that GMail has changed their SMTP server port and now requires SSL (TLS) authentication, neither of which GML seems to support. I was eventually able to get GML working with the GMail SMTP server, but I ended up getting a protocol error at the end of every send, which prevents the script from continuing until a timeout from the SMTP server occured. I strongly recommend using a SMTP server other than GMail’s.

Here’s my modified version of the GML script. Good Luck.

Note that while GML is forwarding an email, the GML GUI won’t paint, so don’t get frustrated like I did and think that GML locked up. You’ll have to check the console output to see if GML is still running. If it does seem to be stuck on an email for a long time, you can send a SIGINT to the process to skip that email (remember to note which one you skipped if you want to resend it manually later).

After careful thought and some inspiration, I’ve decided to bring my original Tails Extension out of retirement. It wasn’t a decision I took lightly, since Robert has done such a great job with the Tails Export Extension and I didn’t want to duplicate his effort, but in order to execute my vision of what I wanted to do, I practically had to rebuild Tails (and Flocktails) from the ground up.

So what exactly has changed? Both the Firefox and Flock extensions has now been consolidated into one. I was planning to just continue on with the Flock line, but since I still spend most of my time using Firefox, I felt it was necessary to have it work with both. Also, the parser has been rewritten to easily support future microformats as well fixed some performance problems that I’ve encountered when switching tabs. For the UI, I’ve moved the main view from Firefox’s sidebar and Flock’s topbar to an overlay approach over the browser content area (pictured below). Finally, I added some extensibility to the extension, which is the part of Tails I’m most excited about - I’ll talk more about this when I release the plugin. I’m expecting to have the extension ready in the a few days, here’s a preview of what it will look like:

I’m very excited with the results, this update will make the extension more useful whereas before it was merely something that was just cool. What remains to be done is some final testing and a little documentation of how to use the extension. I also need to figure out an adequate license for the extension, so any feedback on this would be appreciated.

I’ve updated Flocktails to be work with the latest Flock beta (v0.7). Here is a list of the changes:

• The Flocktails topbar can now be opened with a single click of the microformats icon in the status bar (it required a double click before).
• Since the mapping topbar went away in this latest Flock release, there is now links to map an hCard or hCalendar on Google Maps or Yahoo! Maps.
• hCard data will now scroll if there is a lot of data to display
• Multiple telephone #’s are now displayed, along with the phone type.

Check it out and let me know what you think. Note that this release is not compatible with any version Flock prior to v0.7.