Regular life

These past few months now on the blog I haven’t been doing much of anything but bouncing from one meme to another plus the occasional travel story. No science blogging at all!
Well...
It’s pretty much the same old same old at the lab.

.
Just let me share some of my frustration.
At the moment I’m wrapping up writing the paper on a new technique we’d like to use for molecular genetic diagnostics. We’ve got a “new generation” sequencer, and although some people work miracles with it, what we’re trying to do is a bit different. The thing was sold with all the assurance that it can do what we want it to. Ah, I’m sure it can do what we want it to. But does it want to? not nearly as conveniently and reliably as advertised! And the software that goes with it has clearly been neglected in favor of bigger, sexier, applications.

.
A year and a half after starting a project that ‘should’ have yielded results in a month, we’ve developed our own software, our own protocols, worked everything out from scratch, endured just about everything possible going wrong (including never-explained crashes of the instrument itself), and we’re still not able to say the new way is better than the old way.

.
We can say that with just a little tweak here and a nudge there it should be great.
So you might ask (and reviewers certainly do) why don’t we do the tweak and the nudge and get it great before publishing?

.
That’s where reality sets in.

.
Science is a race to publish, because your funding depends on your publications. With a new technique like this, decent journals will only care about the first handful of labs across the finish line. We need to be in that handful, and we can see some of the other labs we’re in competition with getting farther and farther ahead.

.
A tweak takes three weeks, a nudge three more. If they work, which they may well not and then you’ve lost a month for nothing. Labs who wait for perfection to publish end up sitting on the bulk of their data until it’s stale and nobody wants it. It’s a pretty fine line when you know the end of the race is nearing but not exactly when it will end. Submit now and be rejected for not being complete enough, or delay and publish in a 3rd-rank journal?

.
The paper has already been sent out once and come back, and we did do the major experiment that was lacking in the first try. The first-rank journals have already gone on to the Next Thing. I’m crossing my fingers to get this through my colleagues and out the door for a respectable second place.