AGBT Begins (with bonus Storify Jeremiad)

Just finished my last Storify for tonight from AGBT16, and boy am I wondering how sustainable this will be.  The "problem", which is wonderful to have, is that the number of tweeters has grown substantially, and so there is a wealth of material to attempt to distill down. There's also the desire to make sure I don't further propagate the spam which sneakily re-tweets the occasional item. The other problem has been my tools

Before I whine about my self-inflicted problems, here is the output from today.  I ended up doing two passes at excitement and announcements that came roughly before lunch, which you can find here and here.  The afternoon pre-workshop on human reference genomes got lumped in with the Illumina user meeting, which in turn has the Jack Gilbert talk broken off because I didn't realize initially it was part of the Illumina session, and moving whole blocks doesn't seem to be facile (yo! you weren't supposed to whine yet!).  Tonight's plenary talks, which sounded fascinating (as did a lot of the workshop stuff), have one per speaker: Sean Eddy on RNA co-variation, David Haussler on genome data sharing and graph reference genomes, Pardis Sabeti (still recovering from a nearly fatal car accident) on RNA virus surveillance and host variant detection, and Matthew Sullivan on ocean virus populations and their effect on global carbon cycling.  

A recurrent theme in all this: the need for data sharing.  Another is going for really big and difficult questions: how can we identify functional regions of RNA when phylogenetic signals are intertwined with functional constraints (Eddy).  How can we represent every possible structural variant and other variant into a single logical structure (Haussler and the reference genome workshop)?  How can we understand the interaction between viruses and human hosts (Sabeti)?  How do nanoscale not-quite-living-things drive global carbon flows (Sullivan)?  

As fitting such huge visioins, many of these projects have a global view.  Reference genomes expanding to cover more of Earth's human populations.  Training local scientists to tackle local problems, rather than the grab-samples-and-run attitude that too often has been the Western face of science in developing countries (see The Coming Plague for more on that).  Trying to grasp all the world's oceans and the grand churn of bacteria there, with about 1/3 of bacteria succumbing to viral lysis on a daily basis.

Okay, that's the cool, high-minded stuff.  Now to not-so-highminded working details.

I've been using Storify for a while for the task of organizing Tweets thematically.  I chose it by my usual software selection method: stop at the first tool which roughly works.  Chirpstory is the other such tool I am aware of, but I haven't tried it.  You can embed Tweets right here in blogger, but it is a lot of clicks per item, whereas Storify (when it works) has this simple drag-and-drop mechanism.  Perhaps when I'm not under the gun I should try Chirpstory, because Storify's limits are really getting frustrating.

The first annoyance tonight was discovering that Storify just doesn't work correctly under Safari on my iPad -- you can't drag items from the search pane to the emerging story.  The iPad is my preferred tool for my commute on the train (vs. a laptop), as it is more comfortable if I can't get one of the rare tables (which don't exist at all on the bus), plus the T's WiFi is abysmal whereas I splurged for a 4G iPad.  There doesn't seem to be an app, so I appear to be stuck on this front. Even if it did work, the scrolling was sloooooooooooooooooooooow.  I mean ProtoSlo slow, just barely useable.

The bigger issue is that Storify really doesn't do big projects well.  For some previous efforts, I've tended to just do one Storify a day, but that got frustrating moving between sections.  Concurrent sessions are particularly a bear.  What I really need is somewhere to park tweets between sections or stories -- a working tray -- but there isn't one.  So I end up writing one Story, starting another one, and then repeating the "Get next 50 Tweets", scroll to find the button and repeat cycle to find the Tweets for the next item.  I might go back to making monolithic stories for entire sessions, but editing 

I also periodically hit a problem in which deletes stopped working -- or at least if I deleted a tweet from the story it didn't disappear.  Restarting the editor fixes this, but then I lose my search window.

Finally, the other issue with Storify is that Twitter keeps evolving, and Storify hasn't caught up.  There doesn't appear to be a way to follow conversations or look at tweets quoted in tweets that come up in your search.  That adds another layer of finding Tweets, as replies often don't have the #agbt16 hashtag

Any suggestions are welcome -- I'm really still new at Storify, and I use it too irregularly to remember between projects every little quirky workaround I figure out.