I don’t remember how I stumbled across the blog of Bud Caddell, and I don’t really know who he is.  I do know he produces unique thoughts, like this one thinking about time rather than location.

A scientist replies to people who say “I knew it all along.”

SalesForce.com’s Sites technology didn’t blow my clothes off because I know the platform – though great – is not great at building public websites.  Sam Arjmandi articulates those same feelings, and some more specific and informed arguments than mine.

Mouser is building an airplane, which is amazing, and it’s also neat that he can read my mind because I didn’t know what the hell Clecos were either.

Another reminder that data is all around us – anything you record can be played back in different dimensions, different shapes – A video of the evolving codebase of Twitter.

Twitter Code Swarm from Ben Sandofsky on Vimeo.

Finally, through their vivisection of The Extraordinaries, I was delighted to discover that Full Contact Philanthropy is a strong voice in the nonprofit world, and unafraid to put all those groupthink warm fuzzies in a blender on purée, in the fine tradition of Jayne Cravens and others.

Also, John Williams shreds

Dr. Faye Taxman at the GMU Center for Advancing Correctional Excellence is the primary investigator, and we have been working with her team to develop data entry and analysis tools that parole officers and other officials can use with their clients.

Due to the highly specialized nature of the application, we decided to build the database and “backend” system using the Symfony PHP framework.    Symfony allows for rapid database development and application scaffolding, and also includes many of the features needed by new, custom applications such as a permissions model, strong MVC (model-view-controller) separation, and support for AJAX requests.  With Symfony, we were able to devote most of our time to developing the custom logic and data model rather than re-inventing the wheel.

The data entry screens, using Symfony’s AJAX hooks, are built with jQuery and other Javascript libraries to give the users a large amount of information and many features all on the same screen, minimizing clicking, waiting, and navigating through different parts of the application.  Prelude is supplying the database and backend infrastructure with Symfony, but members of Faye’s team are contributing significantly to the user interfaces.  So, JSTEPS has also been a productive exercise in collaborative software development.

The JSTEPS application is now in final stages of testing, and we are hopeful that it helps decrease the often negative interaction of probation and preexisting drug, alcohol, and criminal behaviors.

The Science of Social Media – SlideShare

Search Terms (General – to learn more)

• Network analysis
• Social network analysis
• Network visualization
• Community detection

Links

• Danah Boyd – Youth & Social Networks – http://en.wikipedia.org/wiki/Danah_Boyd
• Mark A. Smith – Node XL – http://www.slideshare.net/Marc_A_Smith/2009-connected-action-marc-smith-social-media-network-analysis
• Starting point for YouTube social networks presentations: – http://www.youtube.com/watch?v=IEp0nVf26F0
• Example of Possible Mixup of Causation vs. Correllation – http://danzarrella.com/data-shows-that-negative-remarks-lead-to-fewer-followers.html

• Watts & Strogatz Small World Paper – http://tam.cornell.edu/tam/cms/manage/upload/SS_nature_smallworld.pdf
• Barabasi & Albert Scale Free Paper – http://arxiv.org/PS_cache/cond-mat/pdf/0107/0107419v3.pdf
• Girvan & Newman “Community Detection” algorithm – http://en.wikipedia.org/wiki/Girvan%E2%80%93Newman_algorithm
• LIWC – http://www.liwc.net
• Analyze Words (LIWC for Twitter) – http://analyzewords.com
• Tweet Psych – http://tweetpsych.com
• MIT class discovers “Facebook Gaydar” – http://www.boston.com/bostonglobe/ideas/articles/2009/09/20/project_gaydar_an_mit_experiment_raises_new_questions_about_online_privacy/?page=full
• Regressive Imagery Dictionary – http://www.kovcomp.co.uk/wordstat/RID.html
• GreaseMonkey script for political bias in links:  http://waxy.org/2008/10/memeorandum_colors/
• Wikipedia list of social network analysis software:  http://en.wikipedia.org/wiki/Social_network_analysis_software
• Quick n’ Easy Twitter Graph – http://twitnest.appspot.com/nest/index.html
• Wikipedia Article Network resesarch – http://www.netsoc.tcd.ie/~mu/wiki/
• Friendship Prediction on Facebook: http://www.scribd.com/doc/747/Friendship-Prediction-on-Facebook
• Social Network Analysis Toolkit – SocialAction:  http://www.cs.umd.edu/hcil/socialaction/
• Jake Hofman’s del.icio.us – http://delicious.com/jhofman/networks
• Pete Skomoroch’s del.icio.us – http://delicious.com/pskomoroch
• Jure Leskovec http://snap.stanford.edu/na09/schedule.htmlhttp://cs.stanford.edu/people/jure/talks/
• Jon Kleinberg & David Easley’s book http://www.cs.cornell.edu/home/kleinber/networks-book/

http://www.slideshare.net/ehrenfoss/browser-automation

Keyboard Vs. Mouse

• http://www.asktog.com/TOI/toi06KeyboardVMouse1.html
• http://www.codinghorror.com/blog/archives/001088.html- Keyboard Vs. Mouse Debate Jeff Atwood
  
• http://www.codinghorror.com/blog/archives/000825.html – Keyboard Vs. Mouse Debate Jeff Atwood
  
• http://www.youtube.com/watch?v=1MPJZ6M52dI – The first mouse
  
• http://www.ted.com/talks/pattie_maes_demos_the_sixth_sense.html – Wearable, Gestural Interface
  

FireFox Shortcuts

• http://support.mozilla.com/en-US/kb/Keyboard+shortcuts

Gmail Shortcuts

• http://mail.google.com/support/bin/answer.py?hl=en&answer=6594

Ubiquity

• https://mozillalabs.com/ubiquity/

Self Control

• Rescue Time – http://www.rescuetime.com/
• Hosts File – http://practice.chatserve.com/hosts.html
• LeechBlock - https://addons.mozilla.org/en-US/firefox/addon/4476

Macros

• iMacros – http://www.iopus.com/imacros
• Selenium – http://seleniumhq.org/

Also critical to our continuing to live indoors is Orange, another development firm who feeds us a consistent, fresh pile of PHP web application work every month.  Being a sub-contractor for Orange, Gray, and Taupe really drove the business last year.  Green is actually a small asteroid that broke off Orange in July.

As for how we got these clients, in all cases except Light Blue, we either stumbled into the work through a referral or just plain dumb luck.  Maybe someday we’ll need to do more sales and marketing, but until then, the fantastic network we love so much does the job quite well.

Amounts paid to Contractors - 2009

It’s one of our core beliefs that all the effort which goes into building data systems is wasted unless you are prepared to study and learn from what you collect.  Thus, it was well past time we started drinking our own Kool-aid ® in this regard.

The above graph charts the dollar amounts paid to Prelude Interactive contractors throughout the year.  Color and X-axis labels are absent, of course, but I will share that contractors were about 21% of the revenue we brought in, and full time staff (moi!) were around 54%. Actually, I think Prelude became Dark Green’s primary source of income starting in October, which is interesting.

SamaSource is dark pink.  oDesk is the avacado green/yellow at the top of October’s peak.  Together they comprise 10 distinct contractors, so all told we worked with at least 20 other people this year.

Why do we refer to ourselves as we?  It’s a good habit to be in when, as you see starting in July, contractors started having a huge role in Prelude’s output.  It’s safe to say that at this point the business would not be possible without them.

Can we learn anything from this graph?  It tells an accurate picture of the year, I think.  May and the start of June were a long vacation, and to achieve that I really had to ramp down and get back going again through the summer.  In September I wrote “Say No To New Projects!” on my whiteboard.  In early December I wrote it again, bigger, and more in my line of sight.

It took a long time to learn how to use contractors effectively.  Giving up control is hard.  Not doing everything yourself is hard.  Spending 30 minutes documenting something that would take you 30 minutes to do doesn’t make sense at first, but if you’re handing it off to someone who can do it better, now and down the road, it starts to.

I would not have been able to handle any of the major projects this year solo.  My role is no longer that of a PHP/SalesForce/MySQL programmer, but an assembler and planner of resources, a project manager.   I am deeply indebted to (not literally), and highly appreciative of each of the individual colors of the mismatched rainbow above.

The main trick has to do with column names.  If you select either the “CSV” or “CSV Load Data” options, you get a Column Names field to fill in, but it doesn’t tell you how.  I didn’t find how in the documentation either.  I guessed at using comma separated column names, which was close.  Turns out you can’t have “quotes”,”around”,”the”,”column”,”names”.  So you have to do,it,like,this,instead.

A secondary trick is that it appears not to handle quote-escaped quotes like “Ehren “”Loves CSVs”" Foss”.  You need to replace those with \”slash escaped\” quotes instead.

How do you apply an appropriate level of nondisclosure and protection of proprietary information in the age of social media?  Can contractors send out status updates about the cool projects they are working on, or solicit ideas, without putting Prelude or the client at risk?  Is there a way to define that?  Ideally,

What’s the best way to support the symbiotic interchange of minor intellectual property that happens all the time between technical workers?  This is the age of open source, after all.  If you made a nifty widget for another employer and copy it into a Prelude project, is that ok?  Do we need to cover all the permutations of creative commons, open source, BSD, MIT, and other licenses?  Where is the line between minor intellectual property that nobody would bother enforcing like an email address validation function, and the seeds of a product that our clients need to protect?

If anybody out there has experience with forward-thinking legal documents that actually seem to address these concerns, please don’t be a stranger!   Seen any cool ideas out there in your contracts?  What do you think?

At DreamForce a few weeks ago, I saw the demos for a few vendors who offer very comprehensive data of the social and professional relationships of a significant chunk of the professional populace.   They seem to have accurately populated about a quarter of my social graph, which is impressive if sourced from public data.  Their tools were able to point out various paths and strengths of paths I could take to reach, and establish or strengthen a relationship with, certain Big Fish.  Just like LinkedIn’s recomendation feature, but more powerful.  Are these tools being used with your CRM to figure out better ways to approach and snare big fish?   How?  Why not?

It seems like the first thing to do would be to run a list of potential or previous big fish against a list of your employees, volunteers, other donors, and contacts.  Maybe your email campaign intern’s family vacations with Mr. and Mrs. Deep Pockets.  Maybe one of your volunteers used to sit on the board of a company that could sponsor your events.  If you don’t use your database to find out, you are limited to what your decision makers know personally.

The big fish are not probably not likely to show up in your data dragnets.  They are older, more private, and can pay people to save them the hassle of having an online, searchable footprint.  It seems worthwhile to combine the semi-known, offline network surrounding these individuals with the info you can get out of CRM products.

My main point, I suppose, is that the current effort and future strategy of the nonprofit technology community should be better aligned with where the money is.  It seems like right now we’re working as if the smaller online donations are 50% of the pie instead of 5%.

http://feeds.feedburner.com/PreludeInteractive

That is all.  Have a great weekend everybody.