Realistic depiction of the scientific process

how is the treatment going for the infection? your flow chart dosnt make any sense, you want to blame or critisise the public for being anti science for not swallowing science .Unfortunatly you are also one of the public you are no better than anyone else and no more valuble than anyone else. you will also die at some point in the future of old age .Why are you so bitter and twisted like the wicked queen in snow white are you damaged or mad or just bitter?

from

andy stephen st michel etc

Yep, I'd say that pretty much nails it.

Melvin deleted calibration cracked me up. Why someone always gotta come by and move my shit?!

Maybe you misplaced it.

NO. Someone. Touched. It.

Never mind, here it is in my folder!

And, really, you spend most of your time in the Do Science--> Not going as expected --> WTF is going on --> FFFFUUUUUU --> Do science loop.

hee. and that's the really high level one. I remember dealing with scientists who had me running low level recovery scans because they didn't believe me when I said "No, that shit's gone, you deleted it WAY outside the retention period"

If they were lucky, they got some mangled shit that I could have reproduced by slamming their heads against their keyboards. (A regular fantasy)

I felt bad for them, because I knew what the thought process had been. "That's crap, I'll never need that"

six months later:

WHYYYYYYYYYYYYYYYYYYYY

Honestly--I never got past start A or start B, even when I did.

There is fun to be had by messing with all the sweet gear and reading about awesome stuff.

Now there's some ART! You missed your calling there.

Real science starts when you do an experiment, examine the results and say, "Gee, that's funny."

Excellent! Very relatable to my field of ecology, too.

I think you missed the WPI version of this:

don't bother to read science -> see talk on new virus -> get plasmid -> do experiment -> no result -> hmmm that can't be right...I wonder..-> assay suddenly starts working! -> send samples to cleveland -> cleveland confirms "data" -> re-label gels to make data "easier to understand" -> send MS to science -> get published -> cause a public health alert -> make friends with gerwyn -> bleat on about replication while everyone else in the world wastes millions showing your results are bullshit -> retract paper (partially) -> get fired -> sue each other -> back to start (unfortunately).

I'm in math, not science so we have the advantage of not having the dude who deletes our calibration. But the basic premise is the same.

Related anecdote from this week. I was thinking about a problem I had put aside a few months ago and realized that something I had worked out then could combine with a recent result to possibly do something interesting. But I couldn't work out exactly what I had done a few months ago and spent a lot of time on Tuesday trying to rederive it. Finally, I did rederive it subject to other conditions, realized that those conditions were necessary and I had forgotten that I had used them when I had worked out the result a few months ago. Unfortunately, the additional conditions made the result not helpful for what I wanted to use it for now. Gah...

As a career experimental physicist I think this "flow chart" of the scientific process pretty well nails it. Good stuff.

Please provide a statistically significant number of large, well designed, peer-reviewed studies published in established publications to back your assertion that the second diagram is more correct than the more elegant first diagram.

This mostly comes from the fact that there is intense competition to produce original results and not enough of the same to replicate/corroborate already published results. Most of the literature in the machine learning/computational intelligence field is about the performance of some new highly hybridized system (e.g., "a neuro-fuzzy annealing evolutionary approach for improved memory in tabu search over a tree that defines a recurrent rule-set for expert-systems in data-mining"), with too little effort devoted to re-testing the 900000000000000000000001 hybrid systems that have already been proposed.

+1 for science and liberalism.

Somebody needs to fire Melvin. What the heck does that guy do around here, besides leaving the digital scale covered in unidentified white powder?

They forgot about the all-important panel:
"Try to interpret poorly labeled lab journal entry from last week.... give up and repeat experiment... get entirely new results... question reliability of method/instrument/reality. Curse the gods of PCR for their fickleness."

Sorry Willy, but that's not really "science and liberalism", that's more like

>yurops

as far as I can see at least. I'd certainly like to see the story from less right-wing propaganda oriented outlet than the Torygraph.

This may be a little off-topic, but has everyone seen the latest news about everybody's favorite "XMRV" Researcher?

http://news.sciencemag.org/scienceinsider/2011/11/controversial-cfs-res…

Controversial CFS Researcher Arrested and Jailed
by Jon Cohen on 19 November 2011, 6:46 PM| 0 Comments

Judy Mikovits, who has been in the spotlight for the past 2 years after Science published a controversial report by her group that tied a novel mouse retrovirus to chronic fatigue syndrome (CFS), is now behind bars.

Sheriffs in Ventura County, California, arrested Mikovits yesterday on felony charges that she is a fugitive from justice. She is being held at the Todd Road Jail in Santa Paula without bail. But ScienceInsider could obtain only sketchy details about the specific charges against her.

The Ventura County sheriff's office told ScienceInsider that it had no available details about the charges and was acting upon a warrant issued by Washoe County in Nevada. A spokesperson for the Washoe County Sheriff's Office told ScienceInsider that it did not issue the warrant, nor did the Reno or Sparks police department. He said it could be from one of several federal agencies in Washoe County.

Lois Hart, one of Mikovits's attorneys, says her client is being held for extradition to Reno, Nevada, in relation to a civil lawsuit against her filed by the Whittemore Peterson Institute for Neuro-Immune Disease (WPI). Mikovits worked as the research director at WPI, a nonprofit in Reno, for 2 years until she was fired by its president, Annette Whittemore, on 29 September. On 4 November, WPI filed suit against Mikovits, alleging that she had wrongfully kept her laboratory notebooks and other information about her work for the fledgling institute on her laptop, in flash drives, and in a personal e-mail account. A preliminary injunction in the case is set to be held by Nevada's Second District Judicial Court on 22 November. On that same day, Mikovits has a hearing in Ventura County, California, where she can contest extradition, Hart says.

The "deleted calibration data" made me laugh, we had a guy who found an instrument out of calibration, redid the calibration, and then deleted all the data from the date of the expiration of the previous calibration onwards (about 4 weeks worth). Luckily it was a windows based system, and everything ended up recoverable with a couple clicks.

Joe,
I think that the most telling quote so far relating to Mikovits being jailed is her lawyer claiming that she had the right to have a copy of her data. That tells me that she does in fact have, at least, a copy of her data and that if she did indeed sign an agreement saying that all the data, ect... belongs to the institute...well she may be in for a world of hurt.