example page 4 - Dave's Blog

Internet Explorer 8 Beta 1 Released with Activities

2008 Mar 5, 11:36

Internet Explorer 8 Beta 1 is available now. I can finally talk about some of the stuff I've been working on for the past year or so: activities. Activities let you select a document, some text on a document, or a link to a document and run that selection through a web service. For example, you could select a word on a webpage and look it up in Wikipedia, select an address and map it on Yahoo Maps, select a webpage and translate it into English with Windows Live Translator, or select a link and add it to Digg.

IE8 comes installed with some activities based on Microsoft web services but there's a page you can go to to install other activities. However, that page is missing some of my favorites that I use all the time, like del.icio.us. Accordingly, I've put together a page of the activities I use. MSDN has all the info on creating Activities.

Activities are very similar to other existing features in other browsers including the ability to add context menu items to IE. There's two important differences which make activities better. Activities have a preview window that pops out when you hover over an activity, which is useful to get in place information easily provided by developers. The other is that the interface is explicit and takes after HTML FORMs and OpenSearch descriptions. Because the interface is explicitly described in XML (unlike the context menu additions described above which run arbitrary script) we have the ability to use activities in places other than on a webpage in the future. And because activity definitions are similar to HTML FORMs, if your webservice has an HTML FORM describing it you can easily create an activity.

microsoft technical activity openservice ie8 ie activities msdn

URI Addressable Text Adventure Games

2008 Mar 2, 9:18

This post is about creating a server side z-code interpreter that represents game progress in the URI. Try it with the game Lost Pig.

I enjoy working on URIs and have the mug to prove it. Along those lines I've combined thoughts on URIs with interactive fiction. I have a limited amount of experience with Inform which generates Z-Code so I'll focus on pieces written in that. Of course we can already have URIs identifying the Z-Code files themselves, but I want URIs to identify my place in a piece of interactive fiction. The proper way to do this would be to give Z-Code its own mimetype and associate with that mimetype the format of a fragment that would contain the save state of user's interactive fiction session. A user would install a browser plugin that would generate URIs containing the appropriate fragment while you play the IF piece and be able to load URIs identifying Z-Code files and load the save state that appears in the fragment.

But all of that would be a lot of work, so I made a server side version that approximates this. On the Web Frotz Interpreter page, enter the URI of a Z-Code file to start a game. Enter your commands into the input text box at the bottom and you get a new URI after every command. For example, here's the beginning of Zork. I'm running a slightly modified version of the Unix version of Frotz. Baf's Guide to the IF Archive has lists of IF games to try out.

There are two issues with this thought, the first being the security issues with running arbitrary z-code and the second is the practical URI length limit of about 2K in IE. From the Z-Code standard and the Frotz source it looks like 'save' and 'restore' are the only commands that could do anything interesting outside of the Z-Code virtual machine. As for the length-limit on URIs I'm not sure that much can be done about that. I'm using a base64 encoded copy of the compressed input stream in the URI now. Switching to the actual save state might be smaller after enough user input.

zork frotz interactive-fiction zcode if technical uri fragment

IPv6 Roundup: Address Syntax on Windows

2008 Jan 9, 11:34

IPv6 address syntax consists of 8 groupings of colon delimited 16-bit hex values making up the 128-bit address. An optional double colon can replace any consecutive sequence of 0 valued hex values. For example the following is a valid IPv6 address: fe80::2c02:db79

Some IPv6 addresses aren't global and in those cases need a scope ID to describe their context. These get a '%' followed by the scope ID. For example the previous example with a scope ID of '8' would be: fe80::2c02:db79%8

IPv6 addresses in URIs may appear in the host section of a URI as long as they're enclosed by square brackets. For example: http://[fe80::2c02:db79]/. The RFC explicitly notes that there isn't a way to add a scope ID to the IPv6 address in a URI. However a draft document describes adding scope IDs to IPv6 addresses in URIs. The draft document uses the IPvFuture production from the URI RFC with a 'v1' to add a new hostname syntax and a '+' instead of a '%' for delimiting the scope id. For example: http://[v1.fe80::2c02:db79+8]/. However, this is still a draft document, not a final standard, and I don't know of any system that works this way.

In Windows XPSP2 the IPv6 stack is available but disabled by default. To enable the IPv6 stack, at a command prompt run 'netsh interface ipv6 install'. In Vista IPv6 is the on by default and cannot be turned off, while the IPv4 stack is optional and may be turned off by a command similar to the previous.

Once you have IPv6 on in your OS you can turn on IPv6 for IIS6 or just use IIS7. The address ::1 refers to the local machine.

In some places in Windows like UNC paths, IPv6 addresses aren't allowed. In those cases you can use a Vista DNS IPv6 hack that lives in the OS name resolution stack that transforms particularly crafted names into IPv6 addresses. Take your IPv6 address, replace the ':'s with '-'s and the '%' with an 's' and then append '.ipv6-literal.net' to the end. For example: fe80--2c02-db79s8.ipv6-literal.net. That name will resolve to the same example I've been using in Vista. This transformation occurs inside the system's local name resolution stack so no DNS servers are involved, although Microsoft does own the ipv6-literal.net domain name.

MSDN describes IPv6 addresses in URIs in Windows and I've described IPv6 addresses in URIs in IE7. File URIs in IE7 don't support IPv6 addresses. If you want to put a scope ID in a URI in IE7 you use a '%25' to delimit the scope ID and due to a bug you must have at least two digits in your scope ID. So, to take the previous example: http://[fe80::2c02:db79%2508]/. Note that its 08 rather than just 8.

roundup ip windows ipv6 technical microsoft boring syntax

Windows Media Center and Zune Integration Hack

2007 Nov 28, 1:23One of the new Zune features that had me the most excited was the claimed improved Windows Media Center integration which unfortunately turned out to simply mean support for the Win MCE video format (with an exception for HD). I wanted to be able to pick shows recorded by my Win MCE and have the Zune automatically sync up the latest episodes. However, with the improved podcast support in the Zune software one can easily create a ridiculous hack to accomplish this.

The new Zune software has podcast support which does everything I'd want to do with a Win MCE recorded TV series so the goal is to shoehorn a TV series into a Zune podcast. An overview of the steps: Create an XSLT that converts Win MCE data to a podcast, run the XSLT as a scheduled task every few hours per TV series, setup a Web server pointed at the resulting podcasts and the Win MCE Recorded TV directory, and subscribe to the resulting podcasts in the Zune software.

1. Reading through the Win MCE data stored as an XML file in "C:\ProgramData\Microsoft\eHome\Recording\Recordings.xml" and the spec for podcasts I created an XSLT to convert a series from Win MCE data to a podcast.
2. I added a new task to the Scheduled Tasks to run my XSLT using my xsltproc.js script. The task runs a handful of commands that look something like the following:

   C:\windows\system32\wscript.exe C:\users\dave\bin\xsltproc.js C:\Users\Dave\Documents\trunk\development\mce-zune\mce-to-podcast.xslt C:\ProgramData\Microsoft\eHome\Recording\Recordings.xml --param title "The Daily Show With Jon Stewart" --param max 4 --param baseURI "http://groucho/" --param thisRelURI "tds.xml" -o "D:\recorded tv\tds.xml"

   For each TV series I run a command like the above and that outputs a podcast for that series into my "D:\Recorded TV\" directory.
3. Zune only allows http URIs for its podcasts so I installed a web server on my Win MCE server. I'm running Vista Ultimate so it was quick and easy for me to install IIS7 but any Web server will do. Then I pointed it at "D:\Recorded TV\".
4. Once all the above was done I just subscribed to the resulting podcasts via my Web server and viola! Since I'm forced to use a Web server I can even run the Zune software on a machine other than my Win MCE server. You can see a screen-shot above of my Zune software showing my Colbert Report podcast.

technical xml mce hack windows media center zune windows xslt podcast

Brief Miscellany

2007 Oct 29, 7:07Two brief updates to previous posts:

1. I noted that I had a new entry on the IE blog. Some comments on the IE blog have recently been rude in their request for information on future versions of IE. For example see the first two comments responding to my post. Feeling bad about that I looked at my posts entry on delicious and saw the following:

   "This is the first blog from the IE team that I have found rigorous and informative. I skipped to the bottom to find it was written by one of the TA's from my first class at Cal Poly."

   That made me feel a bit better and I was able to catch up with someone from college. Thanks Kris!
   
   
2. I previously had my GPS set with an Australian accent. When it encountered 'WA', as in the abbreviation for Washington in freeway exits, it pronounced it 'Western Australia'. Now I've got it with a British accent and WA is just 'W.A.' but when I tell it to drive to 'MS', the name of my saved location for work, it pronounces it 'Manuscript'.

microsoft blog gps personal nontechnical

XSL Identity Transfom

2007 Oct 12, 4:08As noted in the XSL Transformations spec you can create the identity transform using the xsl:copy element. With the MSXML implementation of xsl:copy the example the spec gives produces slightly ugly elements. For instance given it produces . In order to ensure empty elements turn out pretty I've modified the example as follows:

This got me thinking about inverting XSLTs. Clearly in general an XSLT isn't invertible since an XSLT can completely ignore the input XML and produce something else entirely but then the above is an example of an XSLT that is invertible. So there is a subset of XSLTs that are invertible, how might you produce the inverse of an XSLT, and would this ever be useful?xml msxml inverse xlst xsl

Date Time Formats

2007 Sep 27, 2:17Starting on a new simple project I wanted to get the history of my Delicious links. Delicious has an export tool available via the settings section so I thought I'd try that. However, the links aren't exported in XML not even in XHTML but rather in HTML. Shocking. An example:

"Don't Tase Me, Bro!" (UF Student Tasered Remix)

Remix of the 'Don't tase me, bro!' guy getting tasered.At this point I'm already not going to use this file because its in HTML but I'm even more disgusted by those date time values. Raymond Chen of the Old New Thing posted about recognizing timestamps and timestamp sentinel values. From the first blog post and with the use of a calculator for base conversion one can tell that those are UNIX style timestamps counting the number of seconds since 1970.

It reminds me of my hatred for the MIME date time format I developed working on my webpage's server side parsing of atom and RSS. Atom is of course my favorite as Atom uses the Internet date time format described in the following documents. Here's an example of one 2007-09-27T020:50:00.000-08:00

• W3C Recommended Date and Time Format
• IETF: Date and Time on the Internet: Timestamps
• XML Schema dateTime

On the other hand the evil and villainous RSS uses the MIME date time format now described in the more recent IETF MIME standard. Here's an example Thu, 27 Sep 2007 20:50:00 -0800
The Internet date time format has the advantage of being so easy to sort. An alphabetic sort with normal C-style collation rules of strings containing Internet date times will also sort them chronologically. This is not the case for the MIME date time due to the preceding day of the week and the spelled out month name. This also means that when producing these you have to figure out the day of the week and when parsing them you have to match month names rather than just parsing out numbers. Anyway now days if I see mention of a date time in a new proposed standard or spec I be sure to point out the numerous advantages of the Internet date time format.

date xml html feed time technical date-time code atom rss

XPointer Framework - IE7 XML Source View Upgrade Part 3

2007 May 17, 5:16Previously I created some resource tools and then I used them to overwrite msxml3's XML source view. In this update I've added support for the XPointer Framework.

This time around I've started to add support for the XPointer Framework to my XML source view and I've added installation instructions. The framework consists of a series of pointer segments each of which has a scheme name followed by data in parenthesis. For example 'scheme1(data1)scheme2(data2)scheme3(data3)'. A pointer segment resolves to a portion of the XML document based on the data and the scheme name. The whole pointer resolves to the first segment that successfully resolves. That is, from the example, if scheme1 resolves to nothing and scheme2 resolves to something then that's used and scheme3 is ignored. In addition to the framework I've added support for the xmlns scheme which binds namespace prefixes to a namespace URI and the element scheme which is a simple way to resolve to particular elements in an XML. I also have limited support for the xpointer scheme the content of which is resolved as an XPath with some extra functions (which I don't support -- hence the limited). I've also thrown in schemes for the two SelectionLanguage values supported by msxml3.

Next time I might try to support the xpointer functions that aren't in xpath using msxml script. But I think I'm losing steam on this project... we'll see.resource technical xml xpointer res xpath xslt

Delicious shortcut tag

2007 Apr 8, 3:05Shortcut Tag?
I just saw this on another user's delicious links: a link to ESV search that's tagged with, among other things, "shortcut:esv". When viewed on del.icio.us there's a text box that lets you search using that link. I hadn't seen this before, but it seems pretty cool and I'm surprised I hadn't seen it previously. A delicious post with such a tag ends up looking like the following: I tried searching for information on this and I've found other delicious users doing the same thing, but nothing about the tag itself. If you know any information especially official information from del.icio.us itself please post links in reply to this post. So without further preface here's what I've learned about the del.icio.us shortcut tag.

How-to
To get a search box in your del.icio.us links make a post that satisfies the following requirements:

1. One of the tags must begin with the text 'shortcut:'. You can have more text following that in the tag if you like but it must at least start with 'shortcut:'.
2. The 'url' you post must be a shortcut url rather than an actual URL. It must contain a '%s' with a lowercase 's'. When you enter text into the textbox on the del.icio.us page the text will replace the '%s' after being percent-encoded. For example 'http://www.google.com/search?hl=en&q=%s' is the shortcut url for Google and if you type 'foo bar' into the textbox the URI you will navigate to would be 'http://www.google.com/search?hl=en&q=foo%20bar'.

Complaints
This is neat but I do have a few complaints:

1. The text from the textbox is percent-encoded before replacing the '%s'. Most sites use application/x-www-form-urlencoded which encodes spaces as '+' rather than '%20'.
2. The shortcut url format seems to be taken from Mozilla's Firefox Custom Keywords. Its a shame it wasn't based on something more adaptable like the OpenSearch URL template syntax.
3. A '%s' in the url means technically what you're submitting to del.icio.us isn't a URI as defined by the standard.
4. Allowing text after 'shortcut:' means you can't look at all of a user's shortcut using this tag.

The next step is to create a tool to sync my IE7 search providers with my shortcuts saved to delicious...technical howto tagging tool tag delicious