Last post I mentioned the LHC update that was given at a recent meeting at CERN. One cool thing Steve Myers’ showed during his talk was a discussion of the quality of the splices and how it might affect the LHC’s ability to run.

For a sample of the trade-off, check out this plot, stolen from page 46 in the talk.

Along the x axis is the measured resistance between two magnets (across the splice). The units there are nano-ohms – something only the most expensive multimeters can measure. If you remember your Physics 101 course, you remember P=I^2R (power is current squared times resistance). The units of P are Watts (!) – just like your light bulb. These are superconducting magnets, of course. The magnets are very powerful and so have 1000’s of amps of current flowing through them. So even small R’s mean decent heat sources. Heat warms up the magnets, and makes them no longer superconducting – and that can be a disaster (a few of these is not a problem – it happens every now and then – but a chain reaction is what caused the last September accident). So – the splices, which aren’t superconducting, need to be excellent and have almost no resistance. Like 10-15 nano-Ohms.

The Y axis is how much current you are pumping through the magnet. Current is proportional to the magnetic field, which is proportional to the energy we can run the LHC at. As you can see, if you can run at about 6700 amps you can run at 4 TeV. If you run at 8300 amps then you can run at 5 TeV.

The red and green lines are the keys to reading this plot – they are two different conditions for the state of the copper joints. The LHC machine folks always talk about the worst case scenario (the red line) – but I’m not 100% what the difference between the two is. Lets say you want to run at 5 TeV. Follow the 5 TeV line over from the left of the plot until it hits the red line. You see that it drops down to 58 nOhms. That means all splices have to be less than 58 nOhms in order to run at this energy. The machine is full of these splices. So this is a bunch of work checking these guys!! [listen to the video on the agenda page at about 30 minutes in]. So, one of the things the LHC engineers are doing is measuring all the splice resistances and then putting them up on that plot to see where they are.

BTW, nominal is 10-12 nOhms, and they need to be less than 25 to run at a full 7 TeV (two beams at 7 TeV gives you 14 TeV, the design of the LHC).

Sorry if this is old news…

CERN management recently had a council meeting. These meetings take place between the council and the CERN directory general. Big funding changes, new projects, major schedule changes, a new country wants to join CERN, etc., all have to be approved by this council. As you might imagine the recent council meetings have been dominated by the “schedule changes” (I don’t actually know as a function of time if that is true, but I would imagine).

What is nice about the current CERN DG is that he usually immediately sends a message out to the public and the the CERN folks. Much better than reading about an updated CERN LHC schedule in the Geneva newspaper. Even better, a presentation to all of CERN (and an open webcast) is schedule. The last one just happened (there is a video link and slides link at the top of the agenda, just under the main agenda title).

Everyone is eager for data. I’ve discussed what I think are some of the pressures on the accelerator division previously. This meeting is a continuing part of that conversation.

It is clear they are doing a huge amount of work. They have a lot done. A huge amount. From a physicist’s point of view, the most frustrating thing about Steve Myers’ talk was there is no date and no energy. It wasn’t clear to me what the plan was until someone asked a question at the very end of the talk. Basically – they will have measured the splices (electrical connections) in all of the LHC in early August. Those splices are what caused the disaster last September – so it is important that all of them be carefully measured. And once they have measured everything – then they can start a discussion with the experiments on start up schedule and energy.

Next time something on the trade offs…

Well… I used to think it Washington was the worst. It still is, but California is doing is best to take the crown. I’m talking about the budget of course.

Washington state legislature cut support for University of Washington 26%. That would have been the end of UW as I know it – and I guess the state legislature knew it too. To prevent that they allowed us to raise tuition on undergraduates by 30% over the course of two years (30%!!!). Which we will do. This is old news now – the reality came down at the end of April. Final budgets are being drawn up and they are every bit as bad as we were worried at the time. But the institution will survive.

Most states escaped such dramatic changes. I think only Nevada was going to be hurt as badly as we were. California was bad when I wrote that post – about 10%. What I didn’t realize was the 10% depended on a bunch of initiatives being passed that would raise various taxes. Apparently the complete political establishment (and a lot of people I know, including me) fooled themselves into thinking this was going to pass. Ops! I think the only budget related initiative that passed was one making sure the legislature didn’t give themselves a raise!

The upshot is going to be nasty. Berkeley’s chancellor just laid it on the line:

…the campus now facing a budget shortfall of around $145 million — "a shocking number," he said, more than twice the size of the deficit expected just six weeks ago…

…Birgeneau said all campus units will be asked to cut their budgets by an average of 20 percent over two years, instead of the 8 percent cuts expected as recently as mid-May. Staff who survive these contractions — and, the chancellor emphasized, "there will be eliminations of staff positions" — will see their paychecks shrink.

"We can all collectively expect wage reductions in the neighborhood of 8 percent," reported Birgeneau, adding that the formula could include some combination of furloughs and actual pay cuts.

Wow – 8% pay cut? That sucks. I’ve got a lot of friends in the Ca system. If only UW had some extra cash I’d say we should be off hunting there!! Too bad we will also have a multi-year hiring freeze.

To be fair, the Ca budget isn’t final yet. And it could be some of this is a warning shot at the legislature. But if it is anything like the Washington legislature, all the Ca university system will get back is a big ***ger.

Term is done. It was done a week ago. Next week I move back to Marseille for three months. Those of you who have followed this blog for a while know that I spent 2007-2008 there. Well… I’m going back. Ironically, in 2007-2008 I was really hoping that I’d see first collision at CERN when I was there (it is less than 4 hours door-to-door from Marseille to CERN). This time there is no hope (there was a power outage there today – I wonder if that affected the LHC’s state? Nope, not really!). I was lucky enough that IN2P3 had some money available to help fund my three month trip (thanks IN2P3!).

To celebrate my return there I put together a synth of some 1000-odd pictures I took of the train station when I was there last time. There is lots of cool stuff up there (check out this one  of the Hubble repair mission).

Synth’s are a very cool way to arrange a large collection of photo’s of a single subject. At any rate, enjoy it. I’m going a little crazy trying to get everything ready for the trip!

BTW – I wasn’t able to make this 1000 image synth until I got my new computer with huge amounts of memory – during the synth process it always ran out of memory! The irony here is that synth code is 32 bit. I guess Windows 64 bit gives just a bit more memory space to 32 bit programs than the 32 bit version of windows!

I read this in an article on ars technica. First the setting:

When San Jose State University student Kyle Brady published the source code of his completed homework assignments after finishing a computer science class, his professor vigorously objected. The professor insisted that publication of the source code constituted a violation of the school’s academic integrity policy because it would enable future students to cheat. Brady stood his ground as the confrontation escalated to the school’s judicial affairs office, which sided with the student and affirmed that professors at the university cannot prohibit students from posting source code.

And second, the thing that made me decide to write this post:

Cory Doctorow shared his thoughts about the issue on Thursday in a blog post on BoingBoing. Doctorow suggests that assignments are ultimately more valuable to the students when the work that they produce can have broader purpose than merely fulfilling academic requirements. He also rightly points out that peer review of source code and studying existing implementations are both common practices in the real world of professional software development.

These are both compelling points and they illustrate how traditional academic sensibilities can be detrimental to the intellectual development of students.

Give me a break. This has nothing to do with any high end ideals. It has to do with work. In lower division courses there are only so many types of homework problems you can write without making something really complex, and in upper division courses creating a good problem that is hard, solvable, and interesting takes an immense amount of time. The professor of the course just wants to be able to re-use the homework problems – and cut/pasting the answer from the web is something he/she wants to make as hard as possible.

I wrote a bunch of problems for my graduate course this last year – they took a lot of work – I spent hours on them. I’d very much like to be able to reuse them – or reuse the core of the problem. If the solutions were widely available then that means I have that much more work I need to do next year.

I think, on its own, the answer to the question about a student posting their source code is clear: they should be allowed to do it. But the issue isn’t black and white when you get right down to it – that solution is a product of both the students and the professor’s sweat. Finally, actually saying that you can’t post code is totally unenforceable in this day and age (e.g. RIAA). There is also the basic fact that a student that decides to cheat is only cheating themselves… boy, that sounds kind-a lame, doesn’t it?

I don’t have an opinion in this particular instance. But I think the overly simplistic view that is taken is a bit sensationalist. Cory Doctorow wrote a much more nuanced bit:

But the convenience of profs must be secondary to the pedagogical value of the university experience — especially now, with universities ratcheting up their tuition fees and trying to justify an education that can put students into debt for the majority of their working lives. Students work harder when the work is meaningful, when it has value other than as a yardstick for measuring their comprehension.

I disagree. It isn’t just about convenience of the profs – it is about having a good course for the next student. It is about the professor learning what worked to teach the students this time and refining it next time – these things have to be factored in. Both the students and the professors, it seems to me, have a shared responsibility here.

Though a bit later on he seems to go off the rails (no pun intended):

I’ve always thought it was miserable that we take the supposed best and brightest in society, charge them up to $60,000 a year in fees, then put them to work for four years on producing busywork that no one — not them, not their profs, not other scholars — actually wants to read.

Well, gee. When you start learning something you have to start with the basics. You can’t start with Quantum physics – you need to understand a bit about mechanics, E&M, and other things – after all, quantum had better devolve to those in the macroscopic world! You can start out students on cutting edge research – we all do it when we take on undergraduate researchers – but they have to learn the basics too. It is, sadly, a fact of life. If you don’t know how to start a program, how can you learn how to write cool code!?!?

BTW, if you read down, you can find a response from the student involved in this case.

Emails which contain links to images have long bothered me. I get a lot of them in my day job. They usually contains some descriptive text, then a link to a jpeg or an eps file, and then more text, and then more links.

Often they contain text that says “compare x and y” or similar. This is actually pretty painful: I have to click on all the links. Each image opens up in a separate web browser tab.

To compare them I have to click on each tab, back and forth. Never can see the images at the same time.

On my last plane ride back from Chicago I wrote a tiny little utility, TextToHTML, to take care of this:

You simply paste in the text from the email, click Render, and it will pop open one web page with all the text and images interleaved:

It also knows how to deal with eps files (it will download them locally and render them).

This thing is dirt simple. It can’t do password protected files yet, for example. Suggestions here in the comments or on the project web site are welcome, of course. Feel free to download and play!

In one of my last posts I wrote about Bill Hill and his efforts to improve onscreen reading. For me this has always been about high resolution displays. It would seem that Bill, who has thought about this a lot more, agrees. When I buy a portable there are two criteria. First, it has to be a convertible tablet. Second it has to be high resolution. The high resolution is for one reason only: gorgeous rendering of PDF documents. The more pixels per letter the better the letter looks. It definitely works!

Bill is constantly raving about his brand new Apple (he is running Windows on it):

But I have to say, their displays were not in the same class as this 133ppi display from Apple. It’s stunningly bright and crisp.

The key here is crisp. I have a large 1280×1024 monitor at home. It has one of those glass screens – the ones that reflect and are really really bright. I’ve never seen my photographs look so good as on that screen. But pop-up a PDF and compare it to my 1080p 24” screen? No comparison. Looks a lot better.

There is one problem. The OS and web pages. They all render things much smaller than they are designed to.

In Bill’s blog postings, however, I discovered there was a way to configure Windows to set the DPI correctly. Basically, if you asked your computer to draw a line an inch long on your screen, it would come out an inch only if your screen was built at 96 DPI. My X61T is 147 DPI! That makes the text too small. So my 9 pt text appears as something much smaller, which can be a problem:

It’s especially complicated because humans need (not want, need) to read type which is between 9 and 13 points high. This dimension is dictated by the size of the foveal area in the retina of the human eye, which is only 0.2mm in diameter, with about 1.5 degrees of visual arc.

Ok. So at least Windows has something in it that will allow it to set the Pixel DPI. I set it to 147. Boom. The inch was really an inch (I’d never seen that before). But it was also clear that not much else would work! Web pages were a disaster. Many programs I depend on acted very oddly (who would have thought RealVNC would have so much trouble with a setting like this!?). Such a pity, but MS Office, and the OS itself looked pretty good. So I had to set everything back (to about 120 DPI). My impression from reading around is because most programming is done in terms of pixels, not in terms of actual sizes or lengths. I’d never given this any real thought when writing my own code.

BTW, I have a Lenovo X61T for my portable. The high resolution screen is great (hey, better than the Apple), but it isn’t as bright or crisp as I’d like it. Apparently that is because it is a tablet, so the digitizer takes the edge off it a bit. So it is a compromise machine in that sense.

I’m spending a week at an ATLAS meeting here in Geneva. The day I flew was the day of the horrible Air France 447 crash. The plane just disappeared over the Atlantic on its way from Rio de Janeiro, Brazil, to Paris, France. Just disappeared. It flew through heavy turbulence and a thunder storm. And just disappeared.

As I’m sure you’ve read by now modern airplanes (which this A330 is most certainly one of) are built to deal with these sorts of conditions. Being struck by lighting, flying through strong and gusty winds, etc. So what happened?

Of course, none of us know. But many of the group of people over here at CERN that I hang out with are from the USA. Which means we take these long distance flights all the time. In the past month I’ve taken the equivalent of two of these flights. The uncertainty is really getting to us. Most of us will be on one of these long distance flights (and some on a A330) in less than a week. So you can imagine…

There is another plane crash I remember – Swiss Air 111. This flight was going from JFK to Geneva in 1998. It happened during the CHEP 1998 conference. There were a bunch of people on that flight from the CHEP conference.

I guess one comforting thing for those of us that have to fly is that this is the first A330 crash ever (besides one initial test flight). Still, it is the main topic of conversation over beer and lunch right now. Which is probably just making it worse.

Regardless, our hearts go out to the folks that survived the passengers of 447. We are all connected to the people on that flight one way or the other.

UPDATE: Now that they have found the wreckage they are sending a submersible down there to search for the black boxes. Apparently, the submersible they are using is the same one that is used to repair the Antares experiment (a neutrino experiment on the floor of the ocean).

UPDATE II: I went to dinner last night with someone that was actively trying to get on that Swiss Air 111 flight. It was sold out, fortunately. He knew some people from BNL that were on that flight, on their way to CERN.

We like to say that science is deterministic. You do x,y, and z, and then a will happen. I 100% agree with this. Except when it comes to noise.

What is noise? This is part of the problem – it comes in so many forms. Noise is like a weed. We call a plant a weed when it doesn’t belong. Got grass in your garden? It is a weed. Got grass in your lawn? Definitely now a weed. Is your cell phone making buzz-buzz-buzz sounds on your computer speakers? Noise! Are you using the cell phone signal to talk? Not noise!

But electronic noise is more insidious than the weed analogy implies. If you have an iPhone you can do the following experiment. Sit down at a computer that has separate speakers – the key is that an exposed wire connects the speakers to the computer. Place the iPhone near the speakers*. Wait for a few minutes. Now you’ll hear the buzz-buzz-buzz. So you know the source – you put the iPhone near the computer and it makes that buzzing sound. Lets say you didn’t know that. You had the iPhone in your pocket, and you walked up to your computer – it would start buzzing. You’d notice that every time you were near you’d get the buzzing. You might conclude you were the source of noise (talk about a loud heartbeat!!). You’d be both right and wrong.

The beauty of science is that if you do a, b, and c, then x will happen. Electronic noise is very much about science – if you do x, y, and z then you’ll get noise. The problem is it is very very difficult to determine what x, y, and z are. Above x might be “walk near your computer”, and y might be “have iPhone in your pocket”. You might get x right away, and then give up on fixing the speaker. A computer is no good if you can’t be near it! If you figured out y you might be able to do something different (like leaving your iPhone across the room, or better shield your speakers).

This is why finding and diagnosing electronic noise is an art.

I was on shift for two nights last week. On Friday night we saw a strange noise pattern in D0’s calorimeter. The pattern of noise is called high noon. And it was in four crates of electronics – typical high noon noise is never seen in more than one crate at a time. Experts were here until 2am trying to figure out what the source was. No clue. Eventually they went home – have to investigate the next morning when everyone was awake. Impact on data seemed to be minimal, and the automatic noise detection algorithms seemed perfectly capable of detecting this type of noise.

There were other problems with the detector that night too. A muon phototube kept looking synchronization with the readout. Each time this happens an automatic system would rap its knuckles and bring it back into line. Finally, after it had lost sync about 20 times, we decided to hit it with a hammer: reset it. And lo… the noise in the calorimeter, which had been mocking us for the last 6 hours disappeared.

Keep in mind – the muon system is meters away from the calorimeter system. They share no common electronics – not even racks of electronics are shared by the two. But the muon system was acting like the cell phone, transmitting electronic noise and the calorimeter was acting like an antenna – the speaker wire – and picking up the noise. D0 has been running since 2001. As far as I know, no one had seen this particular failure mode before.

If anyone tells you science isn’t an art – that there is no room for creativity – they are full of it. There are a lot of unknowns in unsolved problems – and it takes some creativity to guess what those unknowns are. And that is also the point behind reproducibility – to make sure you found x, y, and z that produce effect a.

Long live art!

* BTW, the iPhone isn’t the only one that will do this, it is just the worst offender of any of the smart phones I’ve seen. And easy way to fix it is place the iPhone on some tin foil. I’m not sure if that is because the tin foil becomes an antenna and so the iPhone needs less power to communicate with the tower or because it provides some shielding. But it works. See? Art!

Last time I used iPhone app’s as a lead in to writing about why PDF is no longer the best format to get papers around and read. The problem is that your-screen-is-about-the-size-of-a-page no longer holds. Here is what a single page from a typical paper in ArXiv looks like:

It looks like that no matter what your screen size is. So if you zoom in to a single column you’ll have to scroll down, then across and up and then down, etc. Yuck.

Here is an example of one program that does what I want – the New York Times reader*. Here it is reading an article on a “small” screen:

Now, for a “larger screen” (I’m simulating this by resizing the window, but you get the idea):

Note how the text columns changed width and resized – and now it added a picture as well (in the first screen size it was on page 2). This is what I want for the ArXiv papers. Automatic reflow depending on the screen size being looked at. On the iPhone you’d imagine that it would do only a single column. Perhaps you could render it in-line in a web page as a single column if you wanted – or just render it in columns. Your browser could become the display application and the document could reflow depending. How sweet would that be?

So how close are we to that? I don’t really know. HTML will work for most text. I suspect you could do figures pretty well. To first order if you didn’t need them to float it probably would be possible to do them in HTML. But for us in physics the killer are equations. There seems to be one possibility out there: MathML. For an example of what this looks like – see UT Austin physicist’s Jacques Distler’s blog Musings. As Distler points out, WebTex along with modifications to include MathML makes for a pretty decent solution. And the source for the ArXiv files is available too. Perhaps a first solution only takes a server in the clouds doing the conversion and acting as a front-end for CPU-weak devices like smart phones?

Bill Hill, who was one of the inventors of Clear Type (which makes text look so much better on a computer screen) writes an interesting blog where he talks about some of these issues (though not the Math aspect). As you can see from spending about 30 minutes there he claims everything is now in place, but someone has to put it all together…

I think we are at the brink of an explosion in different sized device displays. It is time our display rendering technology caught up with that!

* These images are from version 1 of the reader, which is Windows only. They have a version 2 out now, which is cross platform (on Windows, Apple, and Linux), but I like the Windows only interface better so I’ve not converted yet (but it does the reflow as I’m describing here). But if you own Linux or Apple and you want the offline experience of reading the paper in a much more comfortable environment than the web, I’d definitely recommend checking it out.

A friend of mine, Ann Heinson (thanks!), sent me a link to a UK Telegraph blog posting: ArXiv on the iPhone: a pocketful of Science. ArXiv is, of course, the main paper repository for physics papers of all sorts. The iPhone applications allows you to browse the latest papers, or perhaps search for a particular paper. It was particularly gratifying to note that the best written one was done by David Bacon, a fellow faculty member at University of Washington and also a great blogger. Included in his app is the ability to store papers for later offline reading.

Which brings me to my own personal soap box: offline reading. Here is an expert from the original text on the UK Telegraph:

As the full texts of the papers are not available and the PDFs are often divided into columns, reading on the iPhone’s small screen is not always a very pleasurable experience. The only way to change the font size for easier reading is to zoom in on the text, but as the columns do not change you will probably find yourself scrolling horizontally as well as vertically, which can be wearing.

Before I rant on PDF, let me say that when it arrived, and mostly to date, there isn’t much better than PDF. But it’s time for it to go. Its biggest sin, in my mind: static formatting. I’ve been known to read ArXiv papers on my portable 1440×1050 screen, my 1080p desktop screen, my 48 inch TV screen, a old 1280×1024 screen, and a tiny 800×600 screen. And that doesn’t include a Kindle or an iPhone. PDFs problem: it displays exactly the same way on all those screens! More next time.

Disclaimer: I do not own an iPhone myself, so I’ve not actually tested any of these apps.

Saw this in What’s New:

CANADA: RESEARCHERS GO WHERE THEY CAN DO RESEARCH.

Even as researchers in the U.S. are looking at record basic research increases, Prime Minister Stephen Harper’s government next door plans cut basic research to help pay for his "stimulus package." According to Monday’s Globe and Mail, one of the worlds leading immunologists, Dr. Rafick-Pierre Sekaly, is leaving the University of Montreal for Florida and taking 25 scientists on his team with him.

This is so true. A number of years back, when science budgets weren’t going anywhere, Canada’s looked pretty nice – they were seeing constant increases and seemed to be quite serious about attracting Canadian Americans back up north. Especially the good ones. Then Harper showed up. Things haven’t been as nice since, and they have gotten much worse as this economic slump has bit in. The US’ approach has been to try to spend its way out, and Canada is attempting to keep costs down while redirecting spending. You can see one affect here – someone who will create new ideas and research is shifting countries. Of course, no one knows if he is going to generate some new cure that will be worth millions, but if he does he will be in the USA now and that will go towards our GDP and our taxes and…

Part of the reason I never totally understand people who don’t want to attract and let in more highly trained foreigners.

While the plane was bouncing all over the map, it occurred to me that ROOT is a lot like Microsoft Office (indeed, any product like Office).

Looking at the list of libraries/packages in ROOT you might be tempted to call it bloatware. You’d be right, of course, but, just like MS Word and other similar programs, the libraries you think are important are different from the ones that someone else thinks are important. And almost all the libraries are available for separate use as well! But you’ll notice that few people are using it that way.

The reason is obvious – simplicity.

The brilliance of large software packages like MS Office and ROOT is not that they innovate (though both certianly do), it is that they have taken tried and true ways of doing things and made them work together by packaging them up and building bridges between them. Take TMVA, for example, the package that allows you to easily implement various multivariate analysis techniques (like a boosted decision tree). You can get that software separately – there are lots of packages out there. But TMVA is specifically designed to work with a TTree and other things in ROOT.

Once you get yourself into ROOT you have access to all these tools – and the potential barrier you have to get over to use them is minimal. You want to switch to something else? Of course you can do it – but it will be a lot of work! Just like MS Office.

I think people complain about ROOT and how hard it is to use, but use it anyway for the same reasons they complain about MS Office and its daughter programs. Fortunately, for MS Office, there is real $$ involved so other companies are finally starting to compete – which I hope will make MS Office better. I don’t see anything like that coming along to challenge ROOT until after I retire (or there is a complete revolution in how we do data analysis in HEP).

A Boeing 777 is a big plane. Apparently, however, it is not big enough. I’d swear going over Canada we were actually a paper airplane. Or the pilots were entertaining themselves.

There is a great Farside cartoon (which I can’t find) – it shows two pilots in a jet and one saying to the other “Let’s do that again! Announce more rough air ahead”. Brilliant – it was about how I felt during this flight!

Ever since 9/11 turbulence has been a fascination of mine. Before 9/11 I never cared about it. I was in a plane on 9/11, on the way back from China. It was a smooth ride. But something about that whole experience…

How many people listen to Channel 9 when it gets rough? Available on United, it enables you to listen in on radio communications. When flying through heavy turbulence the first thing you learn is that… no, it isn’t heavy. “Light Chop”, as defined the by pilots, is the stuff that makes the plane go bumpity-bumpity-bump. You, the non-flight attendant, can still walk, but you might hold onto something as you do it. At the upper end of the light scale the pilot turns on the seat belt sign. Then comes moderate.

I think that is worst I’ve ever bin through. I’d hate to know what the other two feel like (“heavy” and “extreme”). At the low end of moderate you could walk if you first violated the seatbelt sign (who doesn’t!?) and really held onto things tightly. Flight attendants even start holding onto things here – and at the upper end they have to sit as well. Now the plane is making decidedly un-plane like noises (thump! Groan! Creeeeek!). Listening in on Channel 9 I heard another pilot describe +- 20% pitch, gusts across the plane that caused “significant sheer”. In the understated language of jet pilots, significant is well, probably significant (the flight controller responded with “Wow!”). Another guy said something like “we are being tossed all over the place – like salad!” No one, of course, had any fear in their voices. We were decidedly below a plane’s breaking point

Uncharacteristically, however, our plane wasn’t changing altitudes to find smoother air. This is what caused me to listen into Channel 9 in the first place. Every other plane was hunting around, trying all sorts of different altitudes – looking for a bit of smooth air. That is what I’m used to. It sounds like way down at 29,000 feet it was smooth. We were up at 37,000 and it was decidedly unsmooth!

Every time the flight controller would ask how the air was up here, our pilot would always say light chop with intermittent moderate chop. This is exactly what everyone else was saying, and what every other plane was going up and down trying to figure out how to avoid. But not our guy. He just powered through it. The result: I didn’t sleep a wink until we were well over the Atlantic and things smoothed out.

That was another thing. As long as we were over North America all conversations were between the flight controller and the pilot. As soon as we left Canadian airspace, there was a “switch to frequency xxxx, end radar coverage”. After getting a backup frequency, the pilot switched over, announced himself (as they always do when they change). And then the pilots talk to each other “Hey – anyone know who won the Preakness?” The other thing is it sounded like the planes, once they start the “crossing” can’t change altitudes. They are allocated a slot before they start, which is cleared in some central clearing house, and then they have to stick with it for the whole flight.

In my last post I mentioned that long lived particles break some basic assumptions that we make in the way we design our software and hardware in HEP. One fascinating example of this that was brought into clear relief for me at this workshop is the interaction between Monte Carlo generation and detector simulation. Look again at the picture I had up last time:

While what I’ve shown above is real data, lets imagine it was a simulation for sake of discussion. Simulation is crucial – it allows us to compare what we think we know against nature. You might imagine that the code that generates everything that happens at the very center of that picture on the left is different from the code that propagates the particles out through the detector (the green, yellow, and blue lines). In fact, this is exactly how we structure our code in HEP – as a two step process.

The first step is to generate the actual physics interaction. Say a top quark production, or Higgs production and decay, or  Hidden Valley decay. As output the generator produces a list of particles and what directions they are heading. Most of them will then stream through the detector leaving tracks and data similar to the right side of that above picture. At this point we’ve got the starting point for all those “lines” or particle trajectories on the left.

Then the detector simulator program takes over. Its job is to simulate the detector. It takes each one of the particles and steps it, a millimeter at a time, through the detector. As it moves through the detector it decides if it should loose some energy interacting with the material, or leave a signal in a detector, etc. Once the simulation is done what we have is something that looks like the experiment was actually run – we can feed it through the same software that we use for real data to find electrons, tracks, etc.

But some of these long lived particle models have particles that interact as they move through the detector. The Quirk model is the poster-boy for this (odd, a model without a web page! At least that I could find). As pairs of these move through the detector they interact with each other and with the material they are traveling through. In short – the detector simulation has to act a bit like the generator – we are mixing these two things.

The main detector simulation program (GEANT4) – written in C++ and carefully planned out – does not look anything like an event generator – written in FORTRAN (common blocks!? ‘nuff said – wait, that was flame bait, wasn’t it?). My guess is it will take a year or so to get GEANT4 updated to accommodate models like Quirks. While it isn’t a complete rewrite of the package – it was quite generally designed – the GEANT4 folks probably didn’t think of a modification to allow interactions like this as a possibility.

Which makes me wonder if in the future generators will really just be subroutines (methods, sub-classed objects, etc.) in detector simulations? We all know that detectors are the most important things out there, after all!

I spent a very enjoyable week attending a workshop here at UW – Workshop on Signatures of Long-Lived Exotic Particles at the LHC. These workshops are funded by the DOE – and allow us to fly in a small list of experts to discuss a particular topic for a week. As you might imagine, things can get pretty intense (in a good way!).

The point about long lived particles is they are long lived! And not much else in the standard model is long lived the way these guys can be. Sure, a bottom quark might travel a few millimeters – and most of us tend to call that long-lived. But the things considered at this workshop can go much furthers – meters even. All sorts of models can generate these particles – like SUSY or Hidden Valley.

Nothing in a particle physics experiment is really designed for these things – not the hardware and not the software, certainly. Not clear our brains are thinking about them too well either! This is part of what makes them so fascinating!

Take the hardware, for example. Just about everything in the Standard Model decays very quickly after it is created in a collider. Millimeters:

That is an exploded schematic view of what happens in our detector (this is a CDF event, I’ve stolen, from Fermilab). The inner circle on the left is about 2 inches in diameter. You see the exploded view on the right? The distance between the vertex and the secondary vertex is about a millimeter or so. That is normal long lived particle for particle physics. All of our code and the design of our detectors are built to discover exactly those kinds of long lived particles.

That picture is from the small conference dinner we set up at Anthony’s, a local nice fish place here in Seattle. I’ve got more pictures from the dinner posted on my flickr account.

Ugh. I hate computer viruses. I hate them because they make me do extra work. I don’t like the work much because I don’t think of it as interesting – why spend my time programming security fixes when I could be trying to search for the Higgs or something similar to that. However, it must be done.

Up until very recently I’d always thought that putting a set of machines behind a firewall, and not letting anything through that firewall, was a great way to secure them. Since an attacker from the outside couldn’t penetrate, you’d be safe, right? A random port-scan of a computer behind the firewall will just turn up a “no one home” answer because the firewall will block it. This makes security a lot simpler.

Until the Phalanx2 virus came along. Actually, virus is the wrong word here. It is really a Trojan horse. Or perhaps a combination of a Trojan horse and a virus. It is insidious. Most viruses break into a computer and then start scanning for any other computer near by that might vulnerable, break into that computer, and then repeat. As a result they can spread very very fast. However, the firewall I mentioned above totally defeats them. They try to scan a machine behind a firewall and… well, the firewall blocks them. So those machines are safe.

But us humans have a way to get to those machines behind the firewall. We have a special set of encrypted commands that we execute that will poke a small hole through the firewall and allow us to access these machines. I use this technique all the time to access computers at Fermilab and CERN. It allows me to help maintain the data acquisition system at DZERO, for example.

And this is where Phalanx2 is clever. Once it gets onto a machine it watches for me to make these very types of connections, memorizes the keystrokes I used to make them, and then later, perhaps while I sleep, it will repeat the same actions and gain access to that machine behind the firewall. It then uses a bug in the Linux kernel to worm its way into the system and set up shop. Ops! Infected machine behind the firewall!!

The upshot is it is suddenly as important to keep those machines behind the firewall as up-to-date and patched as the machines that are outside on the internet. This is more work. A lot more work. And, frankly, it isn’t very interesting work. Bastard virus writers.

After getting back from CERN on Friday I spent a few hours on Saturday night fixing a few bugs (you know… relaxing!). The result is a new version of the deeptalk desktop application. This version has numerous fixes since the version that was released to the web and CHEP. Frankly, it is the version that should have been released at CHEP! Among other things, it will do the layout for a very large conference correctly, and also knows what to do with pptx files (office 2007 PowerPoint).

Eventually I want to be able to do password protected web sites this way (i.e. ATLAS talks, for example). At least, that is the reason why I’ve also got a non-web version of this.

The website remains unchanged for now. I want to make a few very minor updates to it and then roll in these big changes above – and reimage all the conferences. There should be a noticeable improvement when I get to that. As always, this is a hobby, know telling how long it will take!

I’ve been here in Geneva for almost a week now – and it has been all rain. For the first time today it was really sun, however. I almost missed my first meeting of the day to snap a few pictures of the amazing fields of yellow (click for much larger versions!).

And this one for dramatic effect:

I mentioned a few weeks ago ROOT Builder, an app that allows you to build ROOT on your windows machine fairly easily. It turns out that version worked very well on x64 bit machines, but not so well on good-old 32 bit machines. I’ve fixed that issue now – v1.32 (a one line change of code, duh!).

BTW, it isn’t often that you would want to use this tool: it is pretty much meant only for experts. If you need a special build, or special command line switches, or you need source level debugging of what is going on inside root, etc…  Then this might be useful to you!