Forgot your password?
typodupeerror

Comment: Private Data Walls Work (Score 3, Informative) 110

by enderwig (#46313159) Attached to: All In All, Kids Just Another Brick In the Data Wall
FWIW, I am a public high school physics teacher who has taught physics to the bottom half and top half of the student population. The school I teach at is majority minority with a population that identifies as Caucasian at around 30% and African-American around 40%.

Nearly ALL students (and teachers for that matter) would like to see how they rank against others. Nearly all students also want their exact rank to be a secret. Highest grade, lowest grade, highest average or lowest average does not matter. One of the skills I had to learn was how to DISCRETELY pull struggling students aside to give them pep talks and advice on what they could do to improve their grades.

The struggling kids are shamed even if they publicly tell everyone they are ranked 99 out of 100. Adding another bad grade is just another poke at an open wound. ACTING stupid is okay if everyone thinks you are smarter than you look. No one wants to BE stupid. By being discrete, I've gotten quite a few that would do work for me.

I've also had to learn when and how to give kudos to the top achievers. For honor students, its a competition. Unless you are in the top 3, there is some shame associated to being "only" 5th. Knowing someone's rank is a little bit like knowing someone's true name in fantasy universes: there is some power in that knowledge.

Dealing with teenagers is like the super-position principle: it works until it doesn't.

+ - Why Standard Deviation Should Be Retired From Scientific Use-> 1

Submitted by Anonymous Coward
An anonymous reader writes "Statistician and author Nassim Taleb has a suggestion for scientific researchers: stop trying to use standard deviations in your work. He says it's misunderstood more often than not and also not the best tool for its purpose. 'It is all due to a historical accident: in 1893, the great Karl Pearson introduced the term "standard deviation" for what had been known as "root mean square error." The confusion started then: people thought it meant mean deviation. The idea stuck: every time a newspaper has attempted to clarify the concept of market "volatility", it defined it verbally as mean deviation yet produced the numerical measure of the (higher) standard deviation. But it is not just journalists who fall for the mistake: I recall seeing official documents from the department of commerce and the Federal Reserve partaking of the conflation, even regulators in statements on market volatility. What is worse, Goldstein and I found that a high number of data scientists (many with PhDs) also get confused in real life.'"
Link to Original Source

Comment: Re:Casio fx-115ES Plus (Score 1) 328

by enderwig (#45448095) Attached to: Ask Slashdot: Cheap Second Calculators For Tests?
I like this calculator (I have the plus variant), too. It's a great scientific calculator and can do integrals, too. It has a constants library as requested by the OP and a conversion library. It does stats.

The package says: AP, SAT I/II, PSAT, NMSQT, ACT, and is listed on the NCEES website as acceptable.

The only issue I've had is that you never know when the battery has died on the thing until the lights get too dim and it shuts down.

Comment: Re:The sky is falling... ? (Score 4, Informative) 158

by enderwig (#42268717) Attached to: Frame Latency Spikes Plague Radeon Graphics Cards
However, if you actually read the Tech Report's review of the GTX 670, you will find they say the exact opposite. The GTX 670 has ridiculously low latency compared to the Radeon 6990 and just a bit lower than the 7950 and 7990.

As clearly seen on page 3 of the 670's review.

Comment: Re:many answers, but so many questions (Score 1) 138

by enderwig (#37733412) Attached to: Scientists Discover Mechanism That Gives Shape to Life

so is there another clock that turns on this clock?

My guess is yes, there is something else. It may not be a protein but a small nuclear RNA.

but when new cells form in that area weeks later, how do they know their place?

Molecular landmarks sort of like what makes one intersection different from another even if both have a coffee shop, a fast food place, and a gas station. The landmarks could be on the cells, on the extracellular matrix, a diffusable protein gradient, or some other way to differentiate an environment.

Is the HOX system reused to control the layout of my arm down to five jointed fingers? If not, what takes its place at lower levels?

Actually, they are. Nature likes to re-purpose genes temporally and spacially to do more than 1 thing when in the correct environment. Scientists have a decent understanding of how fingers in a mouse and a rat are made. It's quite interesting how expression of one Hox gene creates fingers while the another one is required to create the space between our fingers.

Comment: Re:It's in Science 'cuz now we can SEE it? (Score 1) 138

by enderwig (#37731610) Attached to: Scientists Discover Mechanism That Gives Shape to Life

Precisely. From the abstract and press release, the authors imply that the opening of the super-coiled DNA is necessary and sufficient for the HOX genes to be temporally regulated. Now parsimony and K.I.S.S. usually are the correct ways of thinking about things, but based on what we already know from 10 years ago, simple unwinding can not be the temporal mechanism.

tl;dr summary: We still don't know what starts the cascade of temporal regulation. I don't think this work moves us very far upstream in the regulatory chain.

Comment: It's in Science 'cuz now we can SEE it? (Score 4, Interesting) 138

by enderwig (#37730346) Attached to: Scientists Discover Mechanism That Gives Shape to Life

A caveat as I write this critique, I have only read the linked article and the abstract of the original scientific article, not the full Science article.
Also, I'm a Ph.D. in Developmental Biology from 2000.

If unwinding the super-coiled DNA is considered the chronometer for embryonic segmentation, what makes the DNA unwind at such a specific time? I'm not sure how much new light is shed by this work. We've known for >20 years that transcription factors help "open" DNA for the transcription process. We've also known for >20 years that HOX genes in their clusters are the masters of structural differentiation. Put these two facts together and we can see it should be obvious that the HOX genes need to be "opened" sequentially.

  In the end, we are left with the still burning question of "What controls the HOX genes and their clusters?"

Comment: Fitting the Data to the Question (Score 2) 118

by enderwig (#35742614) Attached to: Dearth of New Nintendo Games Could Indicate Wii 2

A single game drought in early 2002 is somehow strong enough evidence for the author to verify his hypothesis. There is no pattern since the data analysis only began using information from 2001, and there is only a single gap. Not only that, the Wii was released a full 4 years later. If you believe his guess, then Nintendo started making Wii exclusive games approximately 3 months after releasing the Gamecube.

Due to the known lack of software support for the Gamecube, a more likely scenario is that lack of interest in the Gamecube prior to its release is a main component of the game release gap.

Comment: Re:Too Much Imagination Required? (Score 1) 429

by enderwig (#34688084) Attached to: <em>Tron: Legacy</em> &mdash; Too Much Imagination Required?

Completely agree with the parent. Tron was a movie that showed imagination about the inner workings of computers and networks. Tron Legacy grounded too much of the inner workings with the Real World. That is my main problem with the movie. The plot devices used to ramp up tension were mostly problems that exist in the Real World, but shouldn't in The Grid (e.g. the crash landing). The original movie got away with it, since Kevin Flynn had to learn how to be a User. In Tron Legacy, Kevin Flynn has been a User in The Grid for over 20 years. He should know how to make things in The Grid. Sam Flynn is part of the computer generation. He should have been experimenting with being a User from the get-go, but he does nothing User-like in the whole film. The is especially galling since he grew up listening to User stories from his dad.

Another weakness with the movie was the dialogue, but the original Tron's dialogue was just as bad. In addition, the big plot hole of "why" reintegration was bad. That plot device came out of nowhere with a single throw away line midway through the movie. Finally, I found Olivia Wilde was horribly miscast for the role of Quorra, and/or the writers couldn't figure out whether Quorra should be a sex symbol, an innocent, a little girl, a world-weary survivor, or all of the above. I guess Quorra was just a poorly conceived character.

Comment: Give them the basics first (Score 1) 434

by enderwig (#32305186) Attached to: Most Useful OS For High-School Science Education?

Since my experience is primarily in molecular and computational biology, my opinions are obviously biased towards those fields. I have worked in both academia and industry (i.e. >15 years of "Science" experience with 9 years at the PhD level). In my opinion, you should be concentrating on these science skills in upper grade level high school (11-12 grades, preferably just 12).

1) Really get to know MS Office or some other package of word processing (with references support, like Endnote), spreadsheet, and presentation software. You will need a good understanding of the word processor to write grants, reports, and manuscripts. A good understanding of the spreadsheet to organize and analyze your data, with special attention on doing correct statistical analysis. A good understanding of the presentation software for ... presentations. Macs or Windows since it doesn't matter.

2) Really know how to use websites such as http://www.ncbi.nlm.nih.gov/ and http://expasy.org/. The biological science world revolves around biomolecule and biopolymer databases. Then, make them find manuscripts in http://www.ncbi.nlm.nih.gov/pubmed and evaluate them. They also need to learn to evaluate the quality of the information that they find.

3) Make them do journal club. That will hit all levels of Bloom's taxonomy, is student-centered, and buzzword compliant!

Other skills that maybe useful:

3) Entering, searching, and retrieving information from a SQL-like database. Oracle, MySQL, or PostgreSQL servers are everywhere in both academia and industry. Maybe industry has enough resources to create a frontend for their scientists, but most likely they will have to wait for the comuputer analyst group to provide the data you want. Better to ask for read access and do it yourself. Any operating system can be used to access the data. MySQL and PostgreSQL are well supported in Linux.

4) Industry is moving to Lab Information Management Systems (LIMS) and large data generating academic labs are also using LIMS. I don't know if there are free/low-cost LIMS software available, but this would be extremely nice exposure considering most universities won't have such a system for undergraduates. Most LIMS are web-based so it really doesn't matter about the front-end. The back-end is probably Linux or Windows Server.

5) In academia, knowing Linux/Unix/BSD is very useful as most academic software packages are made to run on a Unix-like OS. MacOS X support is actually pretty decent for academic software due to its BSD underpinnings. CygWin is a must if you want to run on Windows. Academics program for the computers that they have, and they mostly have Macs and Unix-like systems.

5) Programming languages that are used extensively by computational biologists are C/C++, PERL, PYTHON, JAVA, and Fortran (more legacy now). From what I saw, PERL and PYTHON dominate on the bioinformatics side.

As for hardware/OS...
For computational biology or computers in biology, Windows is winning that market share. Macs are pretty much only found in academia and mainly for MS Office. They can be used as front-ends obviously, but the general trend of specialized software is to run on XP, for now. I don't know how many science software developers have moved their code to support Win7 natively, but probably not many as these companies are rather slow in adopting new tech. Still, obtaining these licenses is pretty much impossible for a high school. I doubt even the district could find the budget for them.

Setting a Linux cluster for computational number crunching is seen very often in academia and probably in industry, too. So, maybe you can salvage some of the older computers and turn them into a small computational cluster. However, setting up things like this may be impractical with your IT department...

Overall, I think it is very ambitious to provide "real world" science and engineering experience at the high school level. Leapfrogging 4-year university experience, when that is currently required even for an entry-level laboratory technician position, seems to be putting the cart before the horse. True, the university experience is not even close to the "real world" either, but I think interning or setting up summer research jobs in university labs would provide a much better experience than a high school course, unless that high school course is setup like a university lab. Your IT department may not allow any of this and if they did, they probably couldn't provide support.

Help the students build a solid foundation to carry them through life, whether or not that includes undergraduate or postgraduate education.

Comment: Re:It will never be big (Score 1) 116

by enderwig (#31440672) Attached to: Sony's PS3 Motion Controller Gets Demoed and Named

I agree. There have only been a few peripherals that I can think of that have succeeded.

Dual Shock - became standard equipment for the later PS1's
Dual Shock 3 - became standard equipment for later PS3's
Guitar Hero guitar
maybe the Rock Band stuff.

I like my GunCon and Street Fighter TE Fightstick, but I'd be hesitant to call them truly successful. Both are pretty much niche controllers and probably are/were profitable.

The Move controller has a high barrier to entry since it requires an Eye and the ball controller that comes in a $100 start pack. The "nunchuk" is sold separately, or you can use your existing DS3/6-Axis as a stand-in. I don't think this will be more than a niche and may not be profitable.

1 + 1 = 3, for large values of 1.

Working...