Journal

So I love things that buzz…that’s just a fact. Still, I haven’t loved mephisto. I love that it’s railsy and that I theoretically could extend it in ruby if I wanted to…but the fact is that I barely blog as is and I really want to commit as little time to it as I can.

Here are my notes on BarCampBoston2:

openid

presenter: rod begbie

interesting points

• decentralized single sign in
• You can use two link statements in head on any page to delegate openid to and openid server
• can choose to share information from your openid profile with the client site
• alllows for easier mashups between openid enabled sites
• possible to restrict set of valid authentication servers
• does not verify non-spammerness, etc. of logon

questions

• what happens if your change your open id host? if you are delegating?
• does the client site communicate directly with the server?
• how do you choose an openid server at this early stage in the technology?

ideas

• multiple security level openid implementation: variable login credentials

green mapping

url: camberville.greenmap.org

• greenmap: map of environmental stuff
• designed 12 years ago by nyc artist
• the ecologist: brit op of the economist
• freegis.org
  • windows: mapWindows
  • openlayers.org
  • qgis.org
• when printing:
  • spot color
  • ogranic local
• freemap.in is a cool mapping software
• boston.openguides.org
• mass.gov/gis
• harvard geospatial library
• cambridgema.greenmap.org things that need to be done
• open street maps

questions

• why not crowdsource/mashup, gmaps, etc.

searchable media (audio and video)

notes

• video harder than audio
• audio SEO a 4 step process:
  • optimization
  • landing page
  • distribution
  • submission to search engines
  • “getting it out there”
• there are around 75 podcast search sites/aggregators
• id3
  • can be broken into chapters
  • 253 meg limit
  • includes keywords
• importance of keywords
  • in filename
  • in summaries
  • in links to file
  • in rss feed
• optimize the rss field (filenames, etc.)
• most listeners listen to online audio at their workstation (80%)
• video optimization is tag based (youtube)
• build the keyword/tag list first
• keywordDiscover (subscription based SEO tool)
• make site list of media assetts on separate page
• different formats look different in different search engine (google host)
• yahoo doesn’t index swf content

questions

• what if no podcast for audio?
• what if you don’t want people to just see your content but see your site? how can media influence traffic to landing page?
• flash etc for searching of video?

state of online music

• what do we use
  • podcasts
  • pandora
  • live nation
  • good streaming radio: 3wk, weird cover records
  • classicalarchives.com (pay service)
  • exploit boston radio station
  • mi do mi.com
  • audioscrobbler
  • emusic: non drm mp3s
• would be nice to pick vectors used in determination of neighbor songs
• long drm discussion
• trusted computing: is it orwellian?
• music fees as forced tithing
  • spiritual experience of music
• what tools for local musicians
  • open cms platform (unMyspace)
  • recommendation service for local musicans
  • management via myspace
  • distribution services
  • peer-to-peer tag tutorial services

javascript

notes

• what’s important in choosing a library
  • standard method for extending widgets
  • accessible community/core developers
  • jquery and dojo serve as good “kernels” vs. lock and load YUI
• what we want in 2 years
  • html 5 incorporates many widgets into spec
  • better memory management
• ext: widget based on yahoo ui but now portable across js libraries

questions

1. extending
2. frameworks

The Brilliant Idea

So what was that brilliant idea that started me on my rediculous quest? It was actually a pretty dinky but potentially interesting idea:

what does the media going through our networks sound like in its raw form?

• If you were to play back a movie or song in exactly the order in which the pieces arrived via bittorrent, how would it sound?
• If we were to tab into the mass of audio and visual data flowing through, say, a student dorm network and play back all the the layered, fragmented streams together, how would it sound?

I know this is similar to the work of scanner and other musicians who have been working on this for quite some time…but it seems easy enough to do with some simple sniffing and it could be very cool