user page 11 - Dave's Blog

OpenSearchDescriptionToHTML Tool

I've made an OpenSearchDescriptionToHTML XSLT that given an OpenSearch description file produces HTML that describes that file, lets you install it, or search with it. For example, here's a Google OpenSearch description that uses my OpenSearchDescriptionToHTML XSLT.

I had just created an OpenSearch description for WolframAlpha at work and was going about the process of adding another install link to my search provider page so that I could install it. Thinking about it, I realized I could apply an XSLT to the OpenSearch description XML to produce the HTML automatically so I wouldn't have to modify additional documents everytime I create and want to install a new OpenSearch description. While I was in there writing the XSLT I figure why not let the user try out searching with the OpenSearch description file too. And lastly I made the XSLT apply to itself to produce HTML describing its own usage.

Incidentally, I added WolframAlpha at work to replace my FileInfo search provider for the purposes of searching for information about particular Unicode characters. For instance, look at WolframAlpha's lovely output for this search for "Bopomofo zh".

Netflix CSRF - Stolen Thoughts

Looking at the HTTP traffic of Netflix under Fiddler I could see the HTTP request that added a movie to my queue and didn't see anything obvious that would prevent a CSRF. Sure enough its pretty easy to create a page that, if the user has set Netflix to auto-login, will add movies to the user's queue without their knowledge. I thought this was pretty neat, because I could finally get people to watch Primer. However, when I searched for Netflix CSRF I found that this issue has been known and reported to Netflix since 2006. Again my thoughts stolen from me and the theif doesn't even have the common decency to let me have the thought first!

With this issue known for nearly three years its hard to continue calling it an issue. Really they should just document it in their API docs and be done with it. Who knows what Netflix based web sites and services they'll break if they try to change this behavior? For instance, follow this link to add my Netflix recommended movies to your queue.

Thoughts on registerProtocolHandler in HTML 5

I'm a big fan of the concept of registerProtocolHandler in HTML 5 and in FireFox 3, but not quite the implementation. From a high level, it allows web apps to register themselves as handlers of an URL scheme so for (the canonical) example, GMail can register for the mailto URL scheme. I like the concept:

• Better integration of web apps with your system.
• Its easy for web apps to do.
• Links to URNs can now take the user to the sites the user prefers for the sort of thing identified by the URN. For example, if I have a physical address in HTML, instead of making that an http link to Yahoo Maps, I can make the link a geo scheme URI and those who follow the link will get their preferred mapping site that has registered for that scheme. Actually, looking at the geo scheme's RFC, maybe I'd rather use some other URN scheme to represent the physical location, but you get the point.

• There's no way to know if you are the handler for a particular URL scheme which is an important question for web app URL protocol handler authors.
• There's no way to fallback to an http URL in the case that a particular URL scheme isn't registered. A suggested solution to testing the registration of a scheme is for browsers to provide an additional script method to check if a scheme is registered. I don't like the idea of writing script that walks over all my page's links and rewrites them based on that method. I'd much rather see a declarative and backwards compatible fallback mechanism, although I don't know what that would look like.
• There's no way to register for a namespace within the urn scheme URI, the info scheme URI, or the tag scheme URI. I want to register info:lccn/... (Library of Congress Card Number identifiers) to LibraryThing or Amazon and I want to register urn:duri:... (dated URIs) to the Web Archive, among other things.


• Will this result in a proliferation of unregistered URL schemes with clashing namespaces? The ESW Wiki notes why this would be bad.
• And last, although this is nitpickier than the rest, I don't like the '%s' syntax used in the registration method. I'd much rather pass in an URL template, like the URL template used in OpenSearch. If an URL template is used for matching rather than registering against a particular URL scheme, this could also allow for registering a namespace within a URN. For example something along the lines of: registerProtocolHandler("info:lccn/{lccnID}", "htttp://www.librarything.com/search_works.php?q={lccnID}", "LibraryThing LCCN")

Outline View Internet Explorer Extension

I've made another extension for IE8, Outline View, which gives you a side bar in IE that displays an outline of the current page and lets you make intrapage bookmarks.

The outline is generated based on the heading tags in the document (e.g. h1, h2, etc), kind of like what W3C's Semantic data extractor tool displays for an outline. So if the page doesn't use heading tags the way the HTML spec intended or just sticks img tags in them, then the outline doesn't look so hot. On a page that does use headings as intended though it looks really good. For instance a section from the HTML 4 spec shows up quite nicely and I find its actually useful to be able to jump around to the different sections. Actually, I've been surprised going to various blogs how well the outline view is actually working -- I thought a lot more webdevs would be abusing their heading tags.

I've also added intrapage bookmarks. When you make a text selection and clear it, that selected text is added as a temporary intrapage bookmark which shows up in the correct place in the outline. You can navigate to the bookmark or right click to make it permanent. Right now I'm storing the permanent intrapage bookmarks in IE8's new per-domain DOM storage because I wanted to avoid writing code to synchronize a cross process store of bookmarks, it allowed me to play with the DOM storage a bit, and the bookmarks will get cleared appropriately when the user clears their history via the control panel.

FormToAccelerator Internet Explorer Extension

I've made an extension for Internet Explorer 8, FormToAccelerator which turns HTML forms on a web page into either an accelerator or a search provider. In the design of the accelerators format we intentionally had HTML forms in mind so that it would be easy to create accelerators for existing web services. Consequently, creating an accelerator from an HTML form is a natural concept and an extension I've been meaning to finish for many months now.

This is similar in concept to the Opera feature that lets you add a form as a search provider. The user experience is very rough and requires some knowledge of accelerator variables. If I can come up with a better interaction model I may update this in the future, but at the moment all the designs I can come up with require way too much effort. Install IE8 RC1 and then try out FormToAccelerator.

G1 Android Phone

I finally replaced my old regular cell-phone which was literally being held together by a rubber band with a fancy new G1, my first Internet accessible phone.

I had to call the T-Mobile support line to get data added to my plan and the person helping me was disconcertingly friendly. She asked about my weekend plans and so I felt compelled to ask her the same. Her plans involved replacing her video card so she could get back to World of Warcraft and do I enjoy computer gaming? I couldn't tell if she was genuine or if she was signing me up for magazines.

I was with Sarah in her new car, trying out the phone's GPS functionality via Google Maps while she drove. I switched to Street View and happened to find my car. It was a weird feeling, kind of like those Google conspiracy videos.

The phone runs Google's open source OS and I really enjoy the application API. Its all in Java and URIs and mime-types are sort of basics. Rather than invoking the builtin item picker control directly you invoke an 'intent' specifying the URI of your list of items, a mime-type describing the type of items in the list, and an action 'PICK' and whatever is registered as the picker on the system pops up and lets the user pick from that list. The same goes if you want to 'EDIT' an image, or 'VIEW' an mp3.

I wanted to replace the Google search box gadget that appears on the home screen with my own search box widget that uses OpenSearch descriptors but apparently in the current API you can't make home screen gadgets without changing parts of the OS. My other desired application is something to replace this GPS photo tracker device by recording my location to a file and an additional program on my computer to apply those locations to photos.