Tuesday, September 20, 2011

I really hope this is true

“You have this incredibly innovative machine but it looks very traditional.” The last word delivered like a stab…”There are design firms out there that could come up with things we’ve never thought of,” Jobs continued, “things that would make you shit in your pants.”

via http://www.zurb.com/article/793/what-steve-jobs-is-like-in-a-meeting

Thursday, July 28, 2011

We Live in Public



Plus this




Friday, April 22, 2011

Seriously: Municipal Solar Wifi

The two big ingredients:

1 - Fat Pipe at a reasonable price
1 - An energy grid that will pay back what you put in

Everything else already there. Bake at 98.6 °F until the Singularity.

I'm looking at you, Kansas City, Kansas.

Wednesday, March 30, 2011

My Latest Mitch Hedburg Moment

Similar to butter being way better than margarine (try reading this with his voice)

Orange juice with extra pulp is a joke.

Just eat an Orange.

It's like OJ with extra-extra pulp.

Tuesday, March 29, 2011

Wednesday, February 16, 2011

Needlebase: WGBH Internships

Needlebase has so many applications it's mind boggling – once you know what you're doing you'll start to see the web in an entirely different way.

For example, take a look at this page on the WGBH careers website [disclaimer: I'm a former intern in the Media Library and Archives – a fantastic place to be! ok, back to the blog post]

I know what you're thinking. So many links! So many clicks! Wouldn't it be great to search based on what you're good at? Maybe some keywords that would pique your interest? They all sound pretty cool, right? and WGBH is awesome, period.

Insert Needle into the equation. 

What's that? you know Final Cut Pro? Maybe Filemaker? Did Needle just save you a bunch of time? I think so. 

[Sidenote: yeah, the shortcut would be to throw "site:careers.wgbh.org" into Google and go from there but the results are going to include internships that have already been filled. Further proof that in some cases precision is better than recall. Also, building this domain took about 5 minutes to complete, plus or minus the initial run where I just grabbed basic info – not bad!]

Sunday, February 13, 2011

House Cleaning

I found two recent Powerpoint presentations on an old thumb drive and said to myself "hey you should put these online"...

The rest is history:

PEI.pdf is from my LIS456 (records management) class at Simmons in December(ish) 2010.

Piracy.pdf is from the Brandeis conference I did in October 2010.

Both lack notes but you can get the general gist of things – they're mostly images and those are worth 1000 words so we're easily talking over 1 million words here ;)

Tuesday, January 11, 2011

Needlebase: Blu-Ray


What is it?
Every title listed on Blu-Ray.com (which is pretty comprehensive in terms of major US distributors, as far as I can tell anyway).

Why is this interesting?
This story is best told in quasi-screenplay format:

Narrator: Once upon a time there was a VHS collection and a librarian interested in a collection wide format upgrade.

[fade in, Librarian at computer]

Librarian: Shit, that sounds like a lot of manual data entry, how am I going to find the time to do this?

[enter stage right, Needlebase]

Librarian: Oh hey, I could just grab the title info from this website and run some FileMaker wizardry.

*clock ticking sound*

Librarian: Well that was easy.

[fade out]


Where is it going in the future?
Well, I guess that depends on who you consider the real winner of the Blu-Ray vs. HD-DVD battle. Some would say streaming won.

Needlebase: Burton Vinyl


What is it?
A small collection of vinyl records we're holding on to from a friend working abroad.

Why is this interesting?
This domain is similar to Fun with UPC but without the UPC, because apparently most of our vinyl is not lucky enough to be a "universal product" (it's that or bar codes were not as ubiquitous in the 70's...hmmm). This state of affairs makes cataloging extra fun and easy to accomplish – in no time!

Where is it going in the future?
This domain doesn't really have a specific purpose. It's also really really empty, so I guess things can only go uphill from here! Ideally I would like to get reviews pulled into the mix like Glenn's demo domain over here.

Needlebase: Chilling Effects


What is it?
The Chilling Effects Clearinghouse is a huge database of copyright cease and desist notices submitted for your viewing pleasure and maintained by a number high profile academic institutions.

Why is this interesting?
Because it's a huge mess. This is the same problem they ran into with Dollars for Docs – multiple sources, different formats, lots of names with slight variation and not enough eyeballs working to clean it up (I make it sound so easy! hah). Needle's de-duplication functionality works really well in this instance.

Where is it going in the future?
I just stared working on this domain so it's a little early to tell – whatever insight that can be gleaned from this database will eventually make its way into a publication of some sort. You know, with all that free time I have : \

Monday, January 10, 2011

Needlebase: Clark University Course Catalog


What is it?
Not your average online college course catalog.

Why is this interesting?
I actually use this domain quite frequently day to day. It makes a lot of sense to display a course catalog   based on academic concentrations for students but staff members are more interested in knowing who is meeting when and where*. This is similar to what Chuck was talking about in this Needlebase blog post, although from a slightly different angle. Once I had the catalog loaded into Needle it's really easy to start answering questions like "which classes are meeting tomorrow morning on the first floor?" in 2 clicks rather than 20.

Where is it going in the future?
Right on our intranet dashboard perhaps?

*Yes, there are folk who have access to the catalog back-end but I'm not one of those people!

Needlebase: Clark University Libraries


What is it?
I wanted to build an alternative to the Clark University Libraries index page for very selfish reasons. See, the thing is, that list is alphabetical and my library is about 2/3rd down the page. No one likes being below the page break. My plan here was to build an interactive map that also automatically grabbed the hours for each library. At this point though it's just YAGMM (yet another google maps mashup).

Why is this interesting?
This is one of the few domains I'm working with that relies heavily on "live data". Deadbased is another good example because the Internet Archive actually gets quite a few "new" matrix shows that are uploaded each month, but this one takes advantage of the feature in a much more tangible way.

Where is it going in the future?
I would really like to integrate this into our new campus intranet page. Maybe when I get some free time this summer I'll give it another go.

Sunday, January 9, 2011

Needlebase: Deadbased


What is it?
My pride and joy! This was my first major domain building effort when Needle opened up public accounts. It's also the largest domain I have going right now with over 80,000 nodes. Basically, I scraped every entry in the Internet Archives' Grateful Dead Collection, the famed Deadbase, and a few other sites that track the same performance information and put that info under one hood. There's some interesting stuff going on with the data schema I developed but that's a story for another time.

Why is this interesting?
In a sea of Deadbase clones, this one is unique in that we're looking at multiple sources at the same time. As Miley would say, that's pretty cool. Needle enables us to pull everything back together in a non-destructive way that can quantify differences and create a controlled vocabulary that could, in theory, actually improve usability (time will tell).

Where is it going in the future?
Ultimately I would like to build a custom browser for the entire thing with venue details right down to the weather report for that day in history. Something beyond just a front-end for Archive.org and more towards interactive data visualization with a built in media player.

Next on the list is a meta-data explorer that would be something like Oregon Trail on acid (unintentional joke, but true in some cases -- just replace Oxen with VW Bus and hunting with magnetic tape and soundboards). This would allow a person to "follow" the shows from, lets say '68, on a map/timeline, and see who made it to what show and in what order on a map. I feel like this is the sort of thing people miss when they first browse the GD collection on Archive.org even though the data is all there. Something like this, if/when built, would be a great example of building AI into the interface.

Oh, and let's not forget our west coast friends at UC Santa Cruz :) On my radar, but I need to do a lot more cleaning to make this domain presentable. 

Needlebase: Film Schools


What is it?
This was partially inspired by Marshal K's twitter experiment on RWW. I ran across the Student Filmmakers website while digging through some Google Analytics results where I work at Clark University and though "hey, there's probably a useful nugget of info hiding in here somewhere". 

Why is this interesting? 

Where is it going in the future? 
This is a fairly new domain so that's a hard question to answer – I would start with pulling more info from http://www.studentfilmmakers.com/ (if that's possible, I don't recall the TOS off the top of my head) and then expanding laterally across similar websites.

Needlebase: Fun with UPC


What is it?
Long story short: the other day I was playing with an iPhone app that interprets and stores UPC codes via camera input.

Why is this interesting?
Open Library dot org robots dot txt

Where is it going in the future?
Sounds like a really easy way to built a library catalog from scratch. Not saying I invented copy cataloging or anything, just a new way to approach the idea. Eventually I'll figure out the Amazon Web Services API and this will get even more interesting.

Needlebase: Online Music Databases


What is it?
I need to fix the title on this one. This is my first attempt working with data from Wikipedia and the page it references is a big list of online music databases (some streaming, some not). Sidenote: it's a lot easier to grab wikipedia tables in a Google Docs spreadsheet using =ImportHtml(URL; "list" | "table"; index) for a one-off event where this amazing Chrome extension isn't cutting it.

Why is this interesting?
This domain is really more of a test than anything, and I have a feeling I'm just re-inventing the wheel once again. As a single-page quilt (aka "tagged up HTML")  there's not much to say.

Where is it going in the future?
Maybe nowhere? Hey they can't all be winners.

edit - I should point out this is more of a proof of concept idea than anything...Wikipedia is content is  CC-SA 3.0 so that's a plus.

Saturday, January 8, 2011

Needlebase: Quilt a Grave


What is it?

I think I found this link on Boingboing. Not sure. Anyway the website findagrave.com is a big database full of graves and biographical information but their presentation is lacking. I wanted to see what I could do with it. 

Why is this interesting?

Unfortunately this domain is a very limited right now because I have already reached my account quota...as one might imagine the Find a Grave website is pretty huge. I'll be happy just mashing up on Google Maps (each record include a birth, death and grave location). 

Where is it going in the future?

I'll probably end up distilling this domain and focus on a single place. Lots of possibilities including historic records and archives data.

edit: added link

Needlebase: TorrentFreak Analysis


What is it?

I follow the Torrentfreak blog as part of my ongoing copyright research and they have a weekly post that tracks the top 10 movies on public bit-torrent trackers that I decided to scrape. At some point they changed the formatting which was a bit of a challenge to account for but Needle is flexible enough to work with both schema.

Why is this interesting?

Bit-torrent is a market that deals in illegal gigabytes, and as such I think it deserves closer examination. Also, I'm not aware of any other place online that has this information compiled and to me that's pretty cool.

Where is it going in the future?

It would be really cool to add actual box-office data and plot out a few graphs for historic cross-comparison. There's a really easy way to do this in a Google Docs spreadsheet using the GoogleLookup function but I haven't had the time.

edit - added link


I'm going to use up the next 10 or so posts describing each of the Needlebase domains* I am currently working on. Some of them will be better than others but I hope each example demonstrates how versatile this tool is. They are also all works in progress but that's kind of the cool part because each domain can evolve over time based on new sources and better data.

[This is the part where I mention I was an intern at ITA Software over a year ago but the software only recently came out of closed beta. PS - You can find me lurking their GetSatisfaction page.]

edit - added link

edit 2 - here's a direct link to all of the posts in reverse-chronological:

edit 3 - twitter bump equals I should probably proofread ; )
*This is a project file in Needle, almost like a collection in LIS/Archives speak.