I have also had a little more experience with wikis since my original post, and would like to modify my thoughts regarding their use. Mainly, wikis work best when they create a resource for a group, rather than replace, supplant, or build off of existing resources.

That’s my experience with smaller (”We don’t do wikis because we have no control over them.” Control may be distracting, but its not a worthwhile objection. All but the most Enron of places want to leverage the skillsets of their employees. Sorry, that’s my attempt at corporatese. Anyway, once you frame social software in terms of knowledge management and project collaboration, the red herring of control dries up.

For the uninitiated: RSS readers let you pull all your favorite blogs and web sites that offer feeds into a single place that is updated when the sites are updated. It’s like a webmail account that gets a new message when a site you’ve subscribed to adds new content.

After my Onfolio beta expired last fall, I was faced with a choice: fork over $25 for a single-platform RSS reader, or jump into the tempting waters of platform-independent, web-based RSS readers. I did a cannonball, transferring ~150 feeds from Onfolio to my first choice, Rojo (thanks to both readers’ OPML export and import tools, the transfer was painless).

Once I got my feeds into Rojo, I started tagging them with a vocabulary of ~10 phrases, and they started sorting themselves automatically into the proper buckets. Now I just click on a tag, and all the recent stories from the feeds with that tag appear in a nice newspaper format, with ajaxy-liscious controls and a nice read/unread distinction. If I want to subdivide multiply-tag, explode, or consolidate my tags, it’s a simple matter of clicking an icon and typing.

Now I can check my feeds from anything with a internet connection and a browser. World, I will remain on informed and aware of your events, no matter who’s OS I use!

Adding new feeds is easy. My home and work browsers have a nice javascriptlet that auto-discovers any RSS feeds for the page that I have pulled up in a tab and adds it to my feed list. Managing/tagging/deleting feeds is easy as well using the manage page to collapse, untag-retag, and delete feeds.

There are little touches all the way through Rojo that keep me happy. The url for your feeds is rojo.com/subject/tagname/recent. The feed auto-discover defaults to Atom feeds if more than one is available. You can’t accidentally delete a tagged feed. The Rojo team is continually adding features and refining the interface.

There is one big drawback, a related annoyance, and a smaller nitpick. The big drawback is that I’ve seen Rojo take up to 8 hours to pick up a new story from a feed. For feed junkies, this is a deal-killer. I like everything else so much that I don’t care, but please Rojo, give me my new stories ASAP! 8 hours is like getting an invitation to a party that just ended.

The related annoyance is that Rojo appears to mark things as read/unread based on the timestamp of the feed’s story, not on whether it was picked up by the reader when you clicked “Mark As Read”. This means that some new stories slip under my feed-radar because they get picked up 8 hours after they are published, and 6 hours after I clicked the Read button, retroactively marking them as read when in fact I haven’t even seen them!

The other nitpick: sometimes feeds that I’m not subscribed to show up. Then they leave. At the moment, I somehow am subscribed to Fark. My intelligence and happiness are suffering.

Last week I noticed iTunes introduced https://deimos.apple.com/WebObjects/ITCSBrowse.woa/wa/Browse?destination=StanfordPublic”>a selection of class lectures from Stanford profs, along with other silly university marketing content. I listened to a discussion on whether philosophy is the handmaiden or queen of the sciences with Peter Godfrey-Smith, guesting from Harvard University. Not a terribly stimulating session, but the potential is there for exposing your pedagogy and advancing your thinking in a way that is both hip and accessible.

Stanford is the first university to take advantage of Apple opening up iTunes for free hosting and distribution of college/university content. Josiah, I know that you were working on something like this for some Covenant faculty. Perhaps a setup like Profcast could assist in getting that off the ground–between Profcast and the new iTunes U, you have the recording, editing, hosting, and distribution of content, close to maximally automated.

First Question.

What ever happened to distributed peer-based digital signatures and public keys using webs of trust? I am reading The Code Book, and today at lunch hit the chapter on the development of public-key crypto and the saga of Phil Zimmerman. I remember quite a fuss in the mid-90s about cypherpunks bootstrapping a decentralized trusted-key infrastructure. It seems quite relevant and do-able today. Has that project met demise and failed to get off?

Next question.

Has anybody done thinking on the epistemological criteria of encryption systems? Is there a formulation for knowledge wandering around which includes encryption? The history of crypto has seen a succession of knowledge-claims about the unbreakability of systems, and a matching set of persuasive counter-examples. Is there work in this area of philosophy? Additionally, what is the status of encrypted information? Is it knowledge, and what affect does the encryption state have on status?

Last question.

In day-to-day practice, I’ve supplanted my previous criteria for precise knowledge (being able to ask a good enough question that I can get a useful answer from someone knowledgeable) with the criteria “be able to formulate a Google search query which returns the desired information.” Is that wrong? What is the qualitative difference between the two?

Please answer below, or in trackbacks. I exist to be enlightened by someone other than myself.

He made two points. One, current wiki UI is not lawyer-friendly. WikiWords are stupid, especially to a profession who trades in fine wordings. HTML-like markup and syntax are usable for only those who are already geeky enough to know the real deal. Once you can get the UI of a wiki to the level of Word, then we can talk. Hmmm, I smell open source project idea. Two, wikis are knowledge sinkholes. Getting data into them is kinda easy (see One), but getting data out of them is hard. I know Jotspot is working on that, for one, but when you are trading in PDFs and Word docs, XML export doesn’t cut it (at least not currently).

I made a single point that sort of gets at both of his. Wikis are great for ad-hoc arrangement and re-arrangement of data, but they don’t respect existing data. And with 2-million-plus documents in dozens of formats sitting in our document management system, we need to respect existing data. Wikis will be useful to the extent they enable us to re-use, remix, reorganize, review, and extend those documents. What is needed is a wiki that is created, edited, and saved in Word.

Proposed class or classes of copyrighted work(s) to be exempted:
Motion Pictures, Software, Audio Recordings, and Digital Text.
Brief summary of the argument(s) in support of the exemption proposed above:
These classes of works (Motion Pictures, Software, Audio Recordings, and Digital Text) have traditionally been granted copyrights for the purpose of encouraging the public dissemination of the works for the benefit and use of the public by providing a property incentive to the originator for a short period of time. The DMCA ignores this traditional cause of granting a copyright, and moreover establishes crippling restrictions on the aforementioned “benefit and use of the public”. As such, the DMCA’s use should be restricted to the text of the DMCA itself, with the consequence being that any private party which attempts to discern the workings of the DCMA with the intent to apply it in any broader fashion outside of the text of the Act itself would be committing a punishable, criminal action under the DMCA.

Just an idea
I am awaiting a response from the LoC.
Found via Slashdot.

Thank you!

The following information was submitted to the U.S. Copyright Office at 17:35 on 11/3/05. Please print this page for your records.

[I have read the notice of inquiry and acknowledge that my attached submission will be posted on the Copyright Office website.]: Acknowledged

[Name]: Noel Weichbrodt

[Title]: Application Developer

[Organization]:

[Street Address]: The Volunteer Building

[Address Line 2]:

[City]: Chattanooga

[State]: TN

[ZIP]: 37402

[Phone]: 4237858262

[Fax]:

[Submitter's email]: nweichbrodt millermartin com

[Proposed class or classes of copyrighted work(s) to be exempted]: Motion Pictures, Software, Audio Recordings, and Digital Text.

[Brief summary of the argument(s) in support of the exemption proposed above]: These classes of works (Motion Pictures, Software, Audio Recordings, and Digital Text) have traditionally been granted copyrights for the purpose of encouraging the public dissemination of the works for the benefit and use of the public by providing a property incentive to the originator for a short period of time. The DMCA ignores this traditional cause of granting a copyright, and moreover establishes crippling restrictions on the aforementioned “benefit and use of the public”. As such, the DMCA’s use should be restricted to the text of the DMCA itself, with the consequence being that any private party which attempts to discern the workings of the DCMA with the intent to apply it in any broader fashion outside of the text of the Act itself would be committing a punishable, criminal action under the DMCA.

[Attached file]: ExceptionProposal.doc

[Update: fixed linkys]

Not that the hardware is silly, nor that the data is silly. Both were quite accurately described by the incomparable Mr. Percy. What my mind balked at was to believe that four different government agencies have databases that are so easily mined, and a visualization software that can scale so well (from chemical concentrations in water to maps of the state).

Now, I think, I can finally see where we could get this. Google Maps + XML + Web Services for databases + English language data parsing.

The hard part was never getting the data (though that’s interesting). Nor was it matching one thing with another (that’s pretty cool too). The hard bit is, how do you come up with a thing in the first place just from raw data? There’s no ontology in data, it’s just data.

From the article:

“We have guys who can crack hard drives,” Mr. Alexander said. “Getting the information out is easy. The hard part is sharing it, and organizing it, so that everybody in an agency, even nonexperts, can use it.”

The data has always been there for the terrorists to use, or for companies to, ahem, lose track of (a la ChoicePoint). It’s just that there’s no easy way to pull it together into something useful.

This post is partly in response to willas comment on my last post about English language data parsing.

The nets I cast for opinions among friends are more carefully timed and tied than most. But we all cast nets: “Did you see last week’s OC?” Or, “What did you think of The Life Aquatic?” And we all make choices based on the catch from those nets: “Dude, no!” Or, “It’s Anderson, man, with Bill Murray. The touch is golden!”

We also evaluate the catch of those nets by comparing ‘twixt their response and our own thoughts. “Yeah, the whole secret sea base thing was just to pull your attention away from the dull second act.” Or, “Ryan has become so sweet! How could you not agree?” We’re constantly grabbing responses and evaluating credentials like this.

The time has come for real online recommendations.

It’s time for Amazon et al to give me what I want. I’ve bought the DVD series from them. I’ve sat down and utilized their cool star-rating XMLHTTP feature. They’ve aggregated thousands of buying patterns and ratings. They know more about what I like, and what people who like what I like like, and what people who buy what I buy also buy, than anyone. Why can you not tell me if I should follow the new Battlestar Galatica series???

There’s any number of ways to do it. Take what I’ve rated in the sci-fi genre, find out who has given the same media similar ratings, and then show what those people rated the new BSG. Or, of those who have bought shows x, y, and z, but looked at then bailed on a, what have they gone on to buy next? How about narrow demographics? My age within two years, my region within 100 miles, similar past purchases. The ways are endless, and probably quite surprising. But I know there are useful recommendations in there somewhere. Please, Amazon et al, counsel me. I’m just a lonely sci-fi geek jonesn for my catch.

[Update: Jon Garfunkle extends this critique upwards to cover society-level items and outwards across e-tailers, blogs, social networks, etc. Good stuff, and short read.]

Blogs emphasize the present, and deprecate the past. Consider this blog: you can look at the most recent post at the top. To see anything earlier chronologically entails work: a scroll. And to find anything from the barely-departed 2004, you used to have to click on the appropriate month & year. But you don’t care about the month, but instead what I wrote about, regardless of month. Sure timeliness is important–that’s why I write every day. But the idea is, if you are looking at an archive, you don’t care about the timeliness. Instead, you care about the content.

It seems like the chief criticism of Unbound Spiral & commentators nails what I’ve been trying to hit at for a while: we need personal km portals. Combine wikis, blogs, social bookmarks, flatspace, tags, and RSS into one hive of electronic brain-ness that is yours to own and populate, but also others to populate for you.

That’s why I’ve shifted the way my archives are presented. I’ve been tagging each post with one or more labels. Now you may browse the archives of this blog by the tag, and not by the month. This is less a deprecation of the past than the old month-based archiving, and a small step towards a knowledge portal.

Why do I think this?

I only remember bits of information. I can put those bits together for a search query, but often have no idea where the information lies on a hierarchy. And which hierarchy? The URI hierarchy (eg chattablogs.com/barelylegalsubstance/archives/etc)? The desktop hierarchy (eg C:/Documents And Settings/Users/barelylegalprogrammer)? The email hierarchy (eg Inbox, Drafts, Sent, Personal, &c)?

My life is too full of hierarchies…islands of information, to pull out an old buzzword. Social software using SOAP, RSS, and CSS seems like it could hold an answer.

Bonus late link, also from EA: clever use of Flash + Amazon Web Services for the new year.

First, a shout out to the one Blackberry user who bravely ventured here. You, madam or sir, have my respect. Otherwise, you may note that MSIE is sitting pretty with a 62% share, while the Moz and crew are posting a respectable 30%. The zealot’s will note, though, that non-M$ browsers account for roughly 38% of the hits. Take that, monopoly!

As a side note and with barely concealed arrogance, I wager that the crowd hitting here from the “Barely Legal” Google searches is skewering the results toward MSIE.


MSIE Core 699 62.30%
Mozilla/Gecko Core 329 29.32%
Other Browsers 72 6.42%
Opera 15 1.34%
Netscape 6 0.53%
Mobile Browsers 1 0.09%


The breakdown thusly follows:


MSIE 6 636 56.68%
Firefox 1.x 168 14.97%
Firefox 0.x 98 8.73%
Safari 72 6.42%
MSIE 5 42 3.74%
Netscape 7 42 3.74%
Mozilla 1.x 21 1.87%
Opera 7 15 1.34%
AOL 9 14 1.25%
Netscape 3 3 0.27%
Netscape 4 3 0.27%
AOL 6 2 0.18%
AOL 7 2 0.18%
AOL 5 1 0.09%
AOL 8 1 0.09%
PDA's 1 0.09%
MSIE 4 1 0.09%


To the six people who are still using Netscape 3 & 4: for the sake of the future of all mankind, please destroy your computer and buy another one. I mean take the pizza box outside and hit it with sledgehammers until it stops whimpering. This is harsh, but in light of the fact that you are still using the Blight of the Earth (NS 3) and the Bane of All the Internet (NS 4), this measure carries a high warrant. I would give you an upgrade path, but we were doing that in ’99 and ’00. The time of peaceful upgrades has ended.


Windows 997 88.86%
Apple 111 9.89%
Linux/Unix 7 0.62%
Other Systems 6 0.53%
Mobile Systems 1 0.09%


I’m pleased to report that the vaunted Windows monopoly is still alive & kickin’, if showing some chinks in the old steel & duct tape. A big up to my boys and girls hitting us with the Apples and Unicies.

As a recovering ex-web developer, the topic of screen resolutions is near and dear to my heart. In that category, you kids today have it much easier than I did in 2000–a good 74% have > 1024×768. Oh heavens, this is glorious:

1024x768 591 52.67%
1280x1024 234 20.86%
800x600 162 14.44%
Other 72 6.42%
1152x864 43 3.83%
1600x1200 17 1.52%
640x480 3 0.27%


Here is what I want: a personal knowledge management portal into which I bookmark web pages (a la furl), bcc: emails to (a la jotspot), watches my blog and indexes it + any links from it (a la deli.cio.us), directly input text/documents into (a la wikis/jotspot). The trick: I don’t want to organize it.

There should be two interfaces for my flatspace KM portal: a Google search interface and a wiki interface. When I’m home for Thanksgiving weekend and remember that I read an NYT article about Thanksgiving wines, I simply need to drop into my portal and google for “Thanksgiving wines” to find the article again. Then, the portal should notice that I searched for “Thanksgiving” and organize a wiki page that ties together all my saved documents, emails, and web pages that are relevant to Turkey Day.

As you may note from my linkblog, I’ve started using deli.cio.us as my bookmarking mechanism. Good stuff, that. Cross-referencing URLs, optional keywords, RSS feeds, and a lightweight interface all combine into a useful internet app. Desires?

• Recommendations based on stored bookmarks and social patterns
• A distinction between read and unread (so that I could store all my bookmarks there, never deleting and taking advantage of the perpetual search and category system)
• Keyword suggestions a la nutr.itio.us when you save your bookmarks

Conf.uci.us say: use deli.cio.us (new? use the starter’s guide. Worked for me). Don’t knock yourself out for jotspot. Plot to roll your own KM portal.

Aside: where did the whole foo.bar.us meme come from?

However, as 10×10 claims to present the world according to the Beeb, the Old Gray Lady, and Reuters, does it function so much as a definition of the time, or as a commentary on the powerful control of corporate news? It seems like such a small sample, and we’re still relying on the big media companies to edit the world into our front pages. For a hundred possibilities, perhaps it is telling that there were only four discrete images on display in a recent visit. In essence, what our minds intake about the world is still seen through the filtered glasses of Reuters et al. This being a Kantian critique, our knowledge of what exists is thereby determined by these companies. How about a 10×10 that exposes visually what is not being covered by those companies?

How to design something more democratic, more exposing of the phenomenal?

How America voted, county by county, over the last two elections, and compared. From Ron Caron’s Blog.

Tannenbaum’s site shows a similar map–better data, worse design.

Small multiples of the Electoral College, 1940-2004. However, I think that the presentation of the multiples in movie form is less information-rich than on page, due to the forced-linear nature of the movie. Better to peruse non-linearly and use the data presented to draw your own conclusions.

Finally, LawGeek links to a gradient map of voting by counties.

Does the Red/Blue America meme give anyone else funny plot ideas for Red vs. Blue? Perhaps this is just a subconcious reflection of my anticipation of Halo 2.

Update: Two new maps. A new map, from the service used by CBS, gives a weird curved-earth 3D view of how America voted by county. I know that curved-earth view has a proper name…any cartography nerds out there? Second, those craazy Wolverine grad students have some mind-bending warps of the US election, by county, based on population.