JasonMorrison.net

Blogs are great because they give you a creative outlet and let your readers comment on you posts, making it a much more social experience.  But spammers take advantage of comment forms, using scripts and bots to fill the web with links back to their site.

What can you do about it?  Even with captchas, systems like Akismet, and other automatic techniques (you can read more about these here), some spam will slip through.  Specifically, compliment spam.

What is compliment spam? Spammers know you and I like to be told what great writers we are, how helpful our posts are, and that we are brilliant geniuses.  So they set their bots to spam you with complimentary comments that just so happen to link back to their crappy blog, online casino, or fake viagra store.  Here’s an example:

Typolight
http://www.typolight-blog.de | info@typolight-blog.de | 82.146.49.61

Thanks, you nice post that helped me alot.

From Keep your Wordpress site from being hacked with automatic upgrades, 2008/09/06 at 9:27 AM

So, at first glance this looks like a legit comment.  The post in question was a “how-to”, so it would be nice to hear that someone found my instructions helpful.  But, do a Google search with the comment in quotes (an exact phrase search) and you’ll see the problem:

http://www.google.com/search?q=%22Thanks%2C+you+nice+post+that+helped+me+alot.%22

At the time of this writing, we see 168 instances of this exact comment.  By this same Typolight person.

So that’s my tip - if a comment seems a bit too randomly complimentary, throw it in quotes and do a Google search. Then, if it’s spam, make sure to spam it - systems like Akismet only work because we’re all reporting spam.

If you really want to go after the spam poster, you can also give their site a bad rating on Web of Trust, StumbleUpon, and other reporting systems.

Maybe if I get some time I’ll throw together a WordPress plugin to make this easy to do.  If you’d like a plugin like this (or have other tips), drop me a comment and it will help motivate me.

Since the release of the iPhone 2.0 firmware and the App Store, I’ve been like a kid in a candy store. At some point I’ll get around to a list of recommended apps but for now I just want to compare two music listening / online radio applications: Last.fm and Pandora.

You do, of course, have many more options - the App Store Music category has about 30 apps listed, many of them designed to help you enjoy and discover new tunes. And you always have the built-in iPod functionality of the phone which syncs with iTunes on the desktop. But Last.fm and Pandora have been around for a while as very impressive web apps so those were the first two I decided to take a look at. They have very different approaches to recommending music with lots of data and cool algorithms.

Pandora

Pandora is based on the Music Genome Project - basically, their system breaks down each song into a series of attributes. For example, Queen’s Bohemian Rhapsody has “demanding vocal performances, mild rhythmic syncopation, heavy use of vocal harmonies, a prominent rhythm piano part,” among other features. Give Pandora a song or musician and it will create a radio station of similar music. It’s really that simple.

As each song comes up you can give it a thumbs up or thumbs down and you can skip a few songs per station per hour. The iPhone interface displays the album art front and center with a button in the upper-right corner to show you why the system chose the song.

I’ve played with Pandora off and on for a while and my experience is that it does much better with stations created around one or two bands or songs than stations built on large lists of music you enjoy. Add 10 rock bands to your “Road trip with Steve 2008″ station and if one of them has folk influences you’re bound to get some sleepy folk in there now and again. Give it just one band and it can get some amazing results - check out my Gorillaz station, for example.

The drawback to Pandora is that it only has very rudimentary data collection and social features. You can find other people listing to the same song on the website but user profiles are pretty sparse, and there’s no groups, message boards, etc. But if you just want to listen, and don’t want to bother with all that other stuff, Pandora provides a pretty great experience.

Last.fm

Last.fm builds radios stations for you and makes recommendations based on the listening data of thousands of other listeners, whether they’re using the Last.fm site, the mobile app, or a scrobbler plugin in their desktop MP3 player software. You can also listen to stations based around a single musician or band, but Last.fm gives you more options and better results the more you listen and participate in the social features of the site. For example, take a look at the listing for Bohemian Rhapsody - you can see top listeners, how users have tagged the song, similar songs, comments, message board posts, etc.

The user interface is actually quite similar to Pandora’s, with options to note that you love or hate a song, a skip button, album art, etc. You can see a bio of the band, similar artists, and upcoming events, which is cool in theory but I haven’t really used.

I’m a long time user of Last.fm from back in the Audioscrobbler days (check out the Geek Music group) and you definitely get more out of it the more you listen. You don’t really have to participate that much, just letting Last.fm know what you’re listing to improves recommendations and radio plays. My favorite thing about it is all the stats it collects. You can see which bands and songs you listen to most often and find out the most popular bands in Sri Lanka.

Compared to Pandora, though, the recommendations aren’t always as interesting… not bad, but I find myself pleasantly surprised more often while listening to Pandora. For comparison, listen to the Gorillaz similar artists radio station.

iPod + iTunes

You can, of course, skip online radio altogether and just use the built-in iPod functionality along with iTunes on the desktop.  There’s a lot to be said for going this route - the interface is nice and usable, the iPhone holds a decent amount of music, and iTunes collects of the same listening data that makes Last.fm so cool.  Also, it will work no matter how conjested the local network is and doesn’t drain the battery nearly as quickly.

But you miss out on all the social networking features and it’s a lot harder to discover new music.  So I think of it more as a back-up plan…  guaranteed access to some of my personal music library.

The Winner

Actually, there’s no need to pick one as the winner - they’re all available for use on your computer and your iPhone.

Have a favorite?  Share your experience in the comments section.

Wikipedia is great.  Even now, it’s still kind of amazing that such a huge body of knowledge has been organized ad-hoc by volunteers, most of whom have never met in person. Most social software systems would die for this level of collaboration.

That said, has anyone else gone to a random Wikipedia article from, say, search results and ended up a little depressed?  It seems like every other article I find lately has a big warning label at the top - this article contains too much trivia, this article has too many fictional references for an encyclopedic and academic approach of this topic, and worst one of all: this article has been marked for deletion.

I understand that it must be very difficult to wrangle all the millions of contributions into a consistently high-quality encyclopedia.  Just dealing with all the spam and abuse must be an enormous undertaking, even when distributed among thousands of good samaritans.  But one of the things that was great about Wikipedia was the breadth of coverage and the depth on some particulars, even if it was excessive to the point of comedy.

But a brief look at the list of articles marked for deletion the last few days illustrates my point.

1. Horse Ranch Mountain. You know there’s something wrong when a mountain doesn’t meet the notability requirement.   Here’s the comment opening the deletion on the talk page:

In what way is Horse Ranch Mountain notable? I am quite familiar with the area, and I cannot think of any way in which it is notable. Please convince me otherwise.

I would think it’s notable because it is a mass of millions of tons of rock and earth sticking out of the ground.  One a less sarcastic note, I’m sure I’m not the only one who’s looked at a map, spotted a feature I’ve never heard of, then looked it up online.  Even if it’s not accessible it’s probably helpful to have a reference noting that it’s the highest point in Zion, measured at X meters tall, etc.

2.  List of redundant expressions. I understand the argument that an encyclopedia is not a trivia game or a book of lists, but these sorts of pages used to be one of my favorite features of Wikipedia.  Exhaustive lists of palindromes, English words of Polish origin, etc., give examples, context, and can help connect concepts in language.  Also, the use or omission of redundancy is an important stylistic consideration when writing - it can be used for everything from emphasis to characterization.

3.  Hindu literature. Delete the article on Hindu literature?  Granted, the article needs work.  But isn’t it worrying how the marked for deletion pages are filled with subject matter from outside the U.S. and maybe Europe?

I know the standard answer to complaints like these is that if you feel so strongly, you should participate in the debates and push for things not to be deleted.  Judging by the talk pages I wonder if I would be drowned out by all the “I’m a history major and this is a programming term, never heard of it, not notable” comments.  I’ll admit my contribution to Wikipedia is limited to random spelling and grammar corrections that were obvious enough that even I noticed them, so I could be wrong.  I just feel like some of what made Wikipedia so addictive is slowly being drained away.

Agree?  Think I’m wrong?  Leave me a comment below.  See, it’s kind of like a talk page, but even with consensus you can’t edit my article.   Until the next Wordpress exploit comes out.

My esteemed colleague Beah just started blogging, and opened her blog with a very important question - Why Blog?  I remember people asking a similar question years ago when I registered this domain - why would you want to have a website with your name on it?  Almost the same question has come to my mind recently when playing around with Twitter.

So, why blog?  With all the hundreds of thousands of blogs on the web you might think there’s no need to ask this question.  One of the best things about social science is asking questions about things that everyone takes for granted.  Unfortunately the “science” part of social science is a bit too time-consuming to finish up on a Sunday-evening blog post, so instead we’ll look at a few sites of friends and colleagues and maybe collect some thoughts on what motivates people to blog.

First, why do I blog here?  I try to keep this blog relatively professional, posting mostly on topics that I encounter in my work, in my academic research, and in my side projects (the standard disclaimer, as always, applies).  One of my motivations was sharing some of the research done for classwork - it seemed a shame to write up a report, turn it in to a professor, and then let it gather dust in some corner of my hard drive.  My undergrad degree was in journalism and I do miss writing, so that’s another motive.  Also, having been through some rough patches in my career during the dot-com downturn, I thought blogging might help me establish a bit of a professional brand.  I have my URL on my resume and I would hope that any company looking to hire me would get an idea that I’m knowledgeable and interested in relevant areas.

But I’m not a very random sample, so let’s look at a few other blogs and try to appreciate why they write.  I think I can place them into a few rough categories:

Personal takes on professional / technical interests:

• Wyszdom
• Mike Leotta’s Blog
  

This is largely where my blog falls.  Common post topics will include things like “how to get around an annoying issue with some software/programming language,” “very excited about the new device from Apple,” “report from a conference,” and “very disappointed with the new device from Apple.”

Public journaling to keep in touch with friends and family:

• Because I Am Just That Important
• Being Electrocuted
• d.i.l.a.
• Flashing Lights
• Random Quotable Stuff

I’ve done this in the past as well - blogs taking the place of those old-fashioned mass emails you used to send out freshman year of college.  If you went to college in the ancient days before blogs and Facebook.  This is a place for both epic travelogues and saved IM conversations filled with inside jokes.

Sharing interests and reviews:

• Belly Button Treasures
• Read Reid’s Blog
• i’ve got information man
• wires and stars

This category runs the gamut from folks who just want to show their friends a funny Youtube video to blogging a season of a TV show to reviewers writing prolifically about a very obscure musical genre.

Artistic or literary expression:

• tiffography.com

Self-publishing has opened the doors for artists and writers, both amateur and professional, to share their work with whatever audience they find.  This can run from virtual serial galleries shows to community-driven commentary and learning.

Of course these all overlap, and some blogs cover all the bases.  See KooKoo for KokoPuffs for an example

So do we answer our question with a plethora of distinct motives for blogging?  Not necessarily.  There’s one theme that runs throughout all of the above - these are all social activities.  Ultimately blogging is human interaction.

…

Oh, and that other question - why use Twitter?  No clue.

Got a reason why you use Twitter?  Are you a co-worker angry at me for misconstruing your blog?  Please let me know in the comments below.

Although you don’t hear about it much, trust is an extremely important issue in the software world.  A common example is eBay - how could eBay stay in business if millions of anonymous buyers and sellers didn’t have a certain level of trust?

Andy Brice, a software developer, gives a really interesting example of the problem of trust in his blog.  He became concerned that his software products were getting a ridiculous number of awards and 5-star ratings from shareware download sites.  He devised an experiment: if you create a text file, change the file extension to .exe, and submit it to 700 download sites, how many award would you get?

It turns out you would get tons of awards.  A large percentage of these sites, which ostensibly provide users the service of evaluating shareware and freeware, are in reality just trying to skim adwords revenue.

Social software, if applied correctly with enough participation, can help to solve this problem.  It is much harder to fake 1000 del.icio.us bookmarks than it is to make an authoritative-looking award banner.

Many of us work on projects internal to companies where we don’t confront these issues directly on a day-to-day basis.  Large companies can generate billions of pages of documents and code each year.  Add to that the billions of external web pages we use as reference material.  Tools such as social bookmarking can help build up this network of trust and sift through the less useful resources even on intranets.

So now that we have the tools available, all we need is participation.  You’re reading this, so I’m probably already preaching to the choir.  Trust is a really interesting issue, though, so I’ll be writing about it here and there in the future.

Google Maps for Mobile will soon include a GPS-like ability to find your current location.  A little while ago Gizmondo wrote about an iPhone hack that allows almost, but not quite GPS functionality.  The hack itself sounds a lot like the way phase II of the wireless E911 service works, and my guess is that Google Maps is fairly similar.

If you take a look at this map, you can see than many states have > 80% deployment.  On the FCC site you can find reports of the e911 deployments completed by cell phone companies.  Any company that doesn’t have over 95% of their customers with E911 capable handsets is currently getting fined.  So it’s a shame that Google and random iPhone hackers have to reimplement all this.

I’ve never worked on E911 support (or anything cellular, for that matter), but it seems to me there is an incredible opportunity here.  One of the great things about the iPhone is that it drives adoption of data plans.  How about including psuedo-GPS capability in nearly every phone as soon as you sign up for a data plan?  That would be a huge incentive.

Here’s an even more radical idea:  why not come up with a standard way to communicate presence and location data so users can do things like local search?  It might take use years and millions of dollars to develop proprietary systems to do this, but if we use an open standard perhaps this could be adopted as quickly as things like the web and email.

Even better, operating under an open standard will allow geeks in garages all over the world to develop new social software systems we can’t even dream of.

Are you Jason Morrison?  Do you know a Jason Morrison?  If so, please take a look at the new Jason Morrison Dot Net Project.  My goal is to get in touch with all the Jason Morrisons in the world and ask them five questions about their lives.  Eventually, we’ll build up a social network of Jason Morrisons.

Here’s the video of a presentation I gave at the Cleveland Web Standards Association last month (at the time of this posting the website is a little bare, check out the Meetup page for more details).

In this video I talk about the same topic as myTagging and Folksonomy article in the ASIST Bulletin. What are the different kinds of uses for social tagging and folksonomies and what are users’ motivations for tagging?

Jason Morrison - Tagging Systems & Folksonomies from Cleveland Web Standards on Vimeo.

I’m pretty happy to have been the next presenter after Eric Meyer. In this month’s meeting Brad Colbow talked about CSS positioning.What do you think? Feel free to leave a comment below.

If you’ve gone to any of my Category pages on this blog (my Academic papers, for example), you might have noticed I have a tag cloud with just the tags related to that category.  After I figured out how to do it I packaged it into a WordPress Plugin, called Altocumulus.

This goes along with my research interests into folksonomies and information retrieval.  I haven’t had the chance to study tag clouds empirically but my guess is that one giant tag cloud for an entire web site or blog might be more cool looking that useful for navigation.  I think that making use of tag relationships a bit more might show the strength of folksonomies for navigation.  So now, if you click to see my design pages, you can see the kinds of topics my designs cover.

For another example of this in action, take a look at Unsought Input, for example the Innovation page.

Go ahead and download version 0.1 now.   It requires WordPress 2.3 or higher.  This is my first WordPress plugin so I’m sure I’ll figure out ways to make it better over time.  If you have any bugs, pointers, or suggestions please leave them in the comments below.

  The issue has been our for a little while now, but I thought I would note that I have an article about The use of tagging systems in this month’s issue of the ASIST Bulletin. Take a look at Why Are They Tagging, and Why Do We Want
Them To?

Almost everyone has a tagging system the web is facing serious weather with tag clouds on every site. I think it’s interesting to explore the uses of folksonomies and why users bother tagging things in the first place. Here’s an excerpt:

When thinking about adding tagging to a site, the first question should be: What do we want to get out of this? Does the site need something to improve search results or a new navigational facet to better connect related pages? Is the goal to classify lots of multimedia objects with minimal cost or to get users to interact with the site a little more?

Friedman, B., & Nissanbaum, H.  (1996). Bias in computer systems.  ACM Transactions on Information Systems, 14(3), 330-347.

In this article Friedman and Nissenbaum look at bias in software systems. Although the word bias can cover a number of related concepts, the definition used here is something that systematically and unfairly discriminates toward one party or against another. The authors see three main classes of bias in computer systems: Preexisting bias, when an external bias is incorporated into a computer system, either through individuals who have a hand in designing the system or via the society the software was created in; Technical bias, where technical considerations bring about bias (from limitations, loss of context in algorithms, random number generation, or formalization of human constructs); and Emergent bias, where bias emerges after design when real users interact with the system (for example, when new information is available but not in the design, or when systems are extended to new user groups). A number of illustrative examples are given, and the authors look at a number of specific software systems and point out existing or potential biases. One of the systems is the The National Resident Match Program (NRMP), used to match med school graduates to hospitals. In this system, if a student’s first choice of hospital and hospital’s first choice of student do not match, the students’ second choices are run against the hospitals’ first choices. Overall, the result favors the hospitals. Two steps are proposed to rectify bias – diagnosis and active minimization of bias.

This is an extremely interesting subject, and and I doubt most users and programmers are any more aware of it now than they were in 1996. One more recent article, (http://web.mit.edu/21w.780/Materials/douglasall.html) which sought to turn literary criticism toward video games by pointing out cultural biases, also mentions the lack of study in this area. With so many people spending so much of their day interacting with software, why do these kinds of articles seem so few and far between? On the other hand, the particular examples chosen are illustrative but not very current. All three of the systems were large-scale, mainframe-type software that users interacted with in a very small sense. Would the risk of bias be even greater for a system which is largely a user interface?

One clear implication is shown in the diagnosis stage of removing bias—to find technical and emergent bias, designers are told to imagine the systems as they will actually be used and as additional user groups adopt them, respectively. So the charge is one-third ‘know thyself’ and two-thirds ‘know the users.’ The very notion of looking for bias is probably foreign to many user interface designers (in fact, few of the programmers I’ve met are even aware that accessibility guidelines exist for blind, deaf, and other users). The authors’ proposal that professional groups offer support to those designers who detect bias and wish to fight it is a nice thought but doubtful. Few programming or UI organizations can exert any kind of pressure or drum up much bad publicity, or if they can, I haven’t heard of it (which I suppose means they can’t).