So, people tell me 2009 ended recently. Apparently there were fireworks and stuff. This blog as seen very little action during 2009, despite my various good intentions for a blog ‘reboot’ (ala Pawel).
Like many of my online friends, I blame FriendFeed. I find commenting on a FriendFeed post a much more productive way of having a conversation around some new development sweeping the web than writing a dedicated blog post. Still, despite this being my “year of FriendFeed”, I started writing a few blog posts / articles / essays this year which never made it out of the Drafts folder. There is a positive side to unpublished drafts – they serve to nicely organize some thoughts, even if they are ultimately never shared. Anyhow, it’s time to clean them out and move on – and as part of that process – here are the highlights of my posts that never were.
“Why biohacking cannot come of age”
I wrote quite a long essay around the time that synthetic biology was getting lots of press, and just before DIYbio appeared on the scene (as a side note: the name “DIYbio” is PR genius – taking the ‘hacking’ out of biohacking to help avoid misinterpretation by the mass media was a smart move). The opening of this defunct post pretty much sums up it’s contention:
“A healthy biohacking ecosystem requires the participation of hobbyists, and will fail to flourish in the same way ‘Information Technology’ and ‘The Internet’ have flourished if participants remain confined to academic and commercial labs.”.
The old Silicon Valley example (myth?) of the two guys, both called Steve, launching technology from their garage was cited. I then went on to state the obvious – current regulatory frameworks surrounding recombinant DNA and genetic modification make most serious pursuits by hobbyists acting alone legally dubious. Ultimately, I chickened out and decided it was better left unpublished, but a highly modified version my emerge one day. Key links:
- The case of Professor Steven Kurtz
- “The bio-security framework is going to collapse. — Drew Endy“
- Good chemistry kits are hard to buy these days
- The “precautionary principle”
IceCondor – continuous location tracking
Around the end of 2008 when I was momentarily in employment limbo, I began to write an Android mobile geolocation app and started playing with Don Park’s IceCondor. I decided to highlight it with a blog post, but never got around to ultimately publishing it. Essentially, IceCondor is/was a location sharing app, but unlike BrightKite, FireEagle, Google Latitude, Foursquare (& Twitter, these days), IceCondor does continuous location tracking. eg, your GPS location can be shared every 30 seconds via 3G on your Android device (although high frequency updates eat the battery quickly, so lower frequency updates are more practical). IceConder (initially) didn’t include any privacy settings – all locations were openly shared online, with individuals identifiable via their OpenID. As far as I could tell, the only two individuals that gave it any significant use were Don Park, and myself. My main point for writing about IceCondor was to argue that wilfully sharing your location in realtime and opting out of some privacy may actually be safer that not sharing your location. I believe that for most people there is more chance of being randomly mugged than actively stalked, so letting people know where you are is a Good Thing(tm). Don has since changed the focus of IceCondor (at least in the version on the Android Market) to be a simple GeoRSS reader. I get the impression that he is working on other things these days, but the original software and it’s potential uses are pretty cool – it lives on at GitHub, and I notice he has been poking at it again recently.
I get a little sad thinking about this particular post. I’d planned to write about some lesser known functions of Pymol that I had recently discovered (namely the -p, -R and -G commandline options), but never got round to investigating them thoroughly enough to warrant a blog post. Some time after starting the draft and then leaving it to languish, the author of Pymol, Warren DeLano, tragically passed away at the age of 37. I never met Warren, but I was a grateful user of his amazing software, and I wish his family well over what must have been a difficult festive season without him.
Protein sequence clustering tools
I planned to write an article comparing protein sequence clustering tools. I still might, but here is the unannotated list so far:
- MCL / TribeMCL ( http://micans.org/mcl/ )
- an excellent list of sequence clustering tools on Wikipedia
Spam as an indicator of social network success ?
Surely there are already multiple essays on this topic by social media and internet culture enthusiasts. I’ve only searched briefly. The idea for this post was stimulated by some advertising that was sent to me via my delicious inbox (On an unrelated note: 2009 was the year I moved to Diigo for social bookmarking). This spam wasn’t as indiscriminant as the usual “enlarge your whatever” you expect by email, but some fairly niche advertising for cheminformatics software … while probably not spam in the strictest sense, it was nonetheless “spammish” in nature since numerous others were also targeted (via delicious “for:” tags). Neil Saunders also noted that he had seen some spam on Slideshare. Key ideas:
- Is spam an indicator of social network self-sustainability, ‘viral growth’ or ‘critical mass’ ?
- or is it an indicator that ‘stationary phase’, the slowing of growth, has begun ?
- Just as “the network interprets censorship as damage and routes around it“, does spam “interpret small networks as inviable, and avoid them” ?
- How does this relate to the cost / reward – ie. cost of spamming vs. potential audience – see Economics of Spam.
Synthetic biology 4.0: reflections on the state of play
This is one I’d totally forgotten about until now, from late 2008, written shortly after I’d attended the Synthetic Biology 4.0 conference in Hong Kong. It contained the picture below, along with lots of opinion.
On re-reading it, I’ve decided to make some final changes and retro-publish it anyway. It’s not the most coherent article I’ve ever written, and some of my opinions have probably changed in the last 12 months, but I couldn’t bring myself to just trash it.
More thoughts on Biopython from a non-contributing shoegazer
This post was a little bit of a rant/analysis that probably better belongs on the Biopython development mailing list. It was started by Chris Lasher lamenting that academic researchers are rarely encouraged to work on tools like Biopython, and continued summarizing various peoples ideas on why Bioperl still remains in dominant usage, over Biopython. My main conclusion (if there was one), was that the Biopython team over the years has tended to do a good job by maintaining a high standard of quality by deprecating unused, undocumented and unit test-less code … but sometimes perfect has been the enemy of good. Plus, Bioperl had a head start 🙂
The Golden ratio in molecular biology ?
This one has been sitting in Drafts since 2007. I really should just dump it, but the idea still appeals to me. The Golden ratio does appear in nature at the macroscopic level, so why not at the micro- or nano- scale ?
Here’s a choice quote from my notes that may explain why I haven’t yet finished this post:
I think one difficulty in searching for this type of stuff is that the Golden ratio is popular with those into “numerical mysticism”, so if PubMed gives you naught, you have to wade through a lot of kooky pseudoscience in the Google hits before you find the “real science”.
Maybe it will see the light of day in 2010, you never know.
Computation in a single cell … how many logic gates would fit ?
Well … you tell me 🙂