Hi, Seth here. Amidst the fun of moving and looking for apartments, I’ve also been working more than full time looking at our first collisions and preparing for more. In this entry, I’d like to share with you the very first plot I tried to make from actual collision data.
Before I continue, though, I should apologize to all my collaborators on the ATLAS experiment if the title of this post gave them a heart attack: sorry, guys and gals. Why would the title do that? Well, being a member of the ATLAS collaboration means agreeing to present our results collectively; that means that anything I learn about our collision data will only be publicly available after everyone’s had a chance to look at it and make sure it’s work they can stand behind. That’s only fair; after all, everyone contributed to the results; the plots I make are only the final few steps in a chain of design, construction and commissioning more than half as old as I am! (I’ve written about what we can and can’t blog about before.)
So why am I saying I can show you a plot? Well, take a look:
No jets here!
See the problem? It’s empty. What it claims to show is that I looked at 500 events and saw exactly zero track jets in every single one. But actually I had something set wrong when I made it, so it doesn’t really tell you anything about ATLAS data at all.
I’ve fixed the problem now. Did I see any track jets after that? Well, I can’t tell you — not for a few months at least!
Hi,
I’m not convinced by the writing style of this post.
Whilst I agree that science should be accessible to all I find that this post comes across as being rather childlike. Is LHC supposed to be writing it?
Any science experiment takes a while to get set up properly and that didn’t come across. It is good to know however that the equipment is functioning!
Ian
Well, I am not used at all to confidential science financed by the public. Even it’s just for a period of time. But the fact that it is not at all an urgent issue and the pubilc could easily wait for another twenty years for the result makes up for that. So take your time.
However, I’m puzzled by the fact that everybody in your business seems to use root, which I have come across as one of the meanest tools, and one of the hardest to use. Every simple operation, it seems, one must either know by heart or else it takes more than an hour for a plot, or a fit, or something.
What is your average workflow? What do you spend the most time with using root, and how long did it take you to grow accustomed to it? As an undergrad, root seems to be the greatest hurdle approaching experimental particle physics. I’d really like to read a blog post on that.
There is more of that, according to the RESONAANCES blog:
CDMS submitted a paper to Nature, and they were recently accepted. The paper is embargoed until December 18.
CDMS may have found indirect proof for WIMPs. Is there any other interpretation of that announcement ?!?
Great post! I find it interesting to see the little problems that arise in these experiments. I work in the realm of molecular biology & bioinformatics, so “minor hiccups” are common. Thanks for sharing a part of your work.
Hi Mike,
The point is not that the data is “confidential,” the point is that we have to understand what we’re looking at before we announce results. As far as I know, experiments in other fields also analyze their data and sources of experimental error before releasing their results!
The key thing about root, and the source of some of the difficulties in using it, is that it’s very good at storing and accessing extremely large amounts of data, both quickly and compactly.
Seth
And Atlas shows collisions a 2.3 TeV is ok?
http://atlas.web.cern.ch/Atlas/public/EVTDISPLAY/events.html
Events OK. Analysis and conclusions no?
Hi Harbles. Approving events is a quicker process than approving analyses (let alone doing them!). All the events displays are designed to show is that everything is working and looks pretty. Actual measurements of physical quantities require more care and take more time.