My homepage and default search engine is now Duck Duck Go. Unlike Google, it makes it a selling point that it doesn't track your searches, or "bubble" you - and if you aren't concerned with those things then you should click on the "track" and "bubble" links on the main page for a very clear explanation (it's even worse than the concerns I express in my old entry on web privacy)
https://duckduckgo.com/
I've been using it for about a week now, and for pure search the results are at least 80% as good as Google's - it almost always gives me what I'm looking for in the first few results. It doesn't have as slickly dovetailed features as Google, especially the integration of Google image, news, etc, but there's a search syntax that makes it easy to conduct the same search on Google. And it has a couple of nifty touches of its own.
I feel more at ease knowing that my searches during this period are not contributing to the giant database of intimate knowledge Google has assembled about me. Of course I have no independent confirmation of the claims that Duck Duck Go makes for its privacy measures, but I find the slight homemade quality of the site and the earnestness of their privacy essays encouraging on that front. And at least we now have a way as consumers to say to Google, "being creepy will cost you hits."
Thursday, November 03, 2011
Freedom from Internet Self-Distraction
If you have a mac, and struggle with internet self-distraction like I do
("I just don't think I can write another word until I find out what
Phoebe Cates has been up to lately") there's a beautifully simple
application that might help you called Freedom. It blocks your
computer's internet connectivity until a predetermined number of minutes
has passed, or you restart - which is nice as a barrier high enough to
stop me cheating, but low enough enough that I'm not terribly
inconvenienced in an emergency.
http://macfreedom.com/
For some reason the idea of becoming too dependent on it makes me uneasy
(shouldn't I learn some self-control eventually?) but used like
caffeine, as some extra mojo to power through something once in a while,
it's awesome.
("I just don't think I can write another word until I find out what
Phoebe Cates has been up to lately") there's a beautifully simple
application that might help you called Freedom. It blocks your
computer's internet connectivity until a predetermined number of minutes
has passed, or you restart - which is nice as a barrier high enough to
stop me cheating, but low enough enough that I'm not terribly
inconvenienced in an emergency.
http://macfreedom.com/
For some reason the idea of becoming too dependent on it makes me uneasy
(shouldn't I learn some self-control eventually?) but used like
caffeine, as some extra mojo to power through something once in a while,
it's awesome.
Monday, March 07, 2011
Steps for Canadians to enter the U.S. under a J-1 Visa
I've just lost some time and money figuring this out, and so I'm putting this up in hopes that googlers in my situation will find it and benefit. This applies to a J-1 visa specifically, but should also generalize to many other types. This information is current as of the date on this post.
My main source of information is this webpage from the U.S. consulate in Canada:
http://www.consular.canada.usembassy.gov/usa_visa.asp
The most important point is that if you are a canadian citizen, you don't need to schedule a visa interview in person. "Most Canadian citizens can travel to the United States with a valid passport and without a visa", and that includes working and studying (though there are some exceptions). However you do need to have documentation with you at the point of entry (e.g. the airport or the border). Here's what I need for entry under a J-1 (selected from the list on the same webpage above). Later I'll talk about what you need to do to get these things.
You'll need your passport to follow all of the subsequent steps.
HOW TO PAY YOUR SEVIS FEE (and thus get a receipt)
Go to https://www.fmjfee.com/i901fee/ and fill out the form. You need your SEVIS # (starts with an N), your Exchange Visitor Program Number, and your Exchange Visitor Category, all of which are on your DS-2019. You must have the physical receipt that is mailed to you (3 weeks) or couriered to you (1 week) if you pay an extra $35. Besides that I had to pay $180. If time is tight and you're waiting for your DS-2019 to arrive in the mail, all you need from it is the information I listed above, so you might be able to get that directly from your hosting institution.
HOW TO FILL OUT A DS-160
This is your application to enter the country, a Nonimmigrant Visa Application, and it is found here: https://ceac.state.gov/GENNIV/
It is a long (and nosy) form, plus a buggy web application, so I suggest clicking the save button frequently and also using the option to save the form data in a file on your own computer, so you can save your progress if you find there's something you're missing. Besides everything described for the previous steps, here are a few of the things you might need to gather before you start:
My main source of information is this webpage from the U.S. consulate in Canada:
http://www.consular.canada.usembassy.gov/usa_visa.asp
The most important point is that if you are a canadian citizen, you don't need to schedule a visa interview in person. "Most Canadian citizens can travel to the United States with a valid passport and without a visa", and that includes working and studying (though there are some exceptions). However you do need to have documentation with you at the point of entry (e.g. the airport or the border). Here's what I need for entry under a J-1 (selected from the list on the same webpage above). Later I'll talk about what you need to do to get these things.
- Valid Passport (for at least the next 6 months after the arrival time).
- Older passports containing previous visas.
- DS-2019
- Proof of sufficient funds to cover all expenses while in the U.S.
- SEVIS fee receipt
- DS-160.
You'll need your passport to follow all of the subsequent steps.
HOW TO PAY YOUR SEVIS FEE (and thus get a receipt)
Go to https://www.fmjfee.com/i901fee/ and fill out the form. You need your SEVIS # (starts with an N), your Exchange Visitor Program Number, and your Exchange Visitor Category, all of which are on your DS-2019. You must have the physical receipt that is mailed to you (3 weeks) or couriered to you (1 week) if you pay an extra $35. Besides that I had to pay $180. If time is tight and you're waiting for your DS-2019 to arrive in the mail, all you need from it is the information I listed above, so you might be able to get that directly from your hosting institution.
HOW TO FILL OUT A DS-160
This is your application to enter the country, a Nonimmigrant Visa Application, and it is found here: https://ceac.state.gov/GENNIV/
It is a long (and nosy) form, plus a buggy web application, so I suggest clicking the save button frequently and also using the option to save the form data in a file on your own computer, so you can save your progress if you find there's something you're missing. Besides everything described for the previous steps, here are a few of the things you might need to gather before you start:
- A digital photo of yourself following very rigid passport photo requirements. I uploaded a scan of my actual passport photo, and that seemed to work.
- The date and duration of your last 5 US trips
- The address and phone number of your US contact
- The address and phone number of two people in canada who can vouch for you
- The " Passport Book Number", which according to several online sources is the number that appears on a canadian passport above the barcode on page 3. I have not been able to verify this from an authoritative source, but I decided to go ahead and use it. Hh
Saturday, February 19, 2011
Essential Software for a Scientist
I was thinking about the tools I use every day and which ones I would insist my students learn when they arrived in the lab. Here's the list I came up with:
- Microsoft Excel (or OpenOffice) and PivotTables (called something else in openoffice). So, so helpful for analyzing data in tabular format. I would even call them magical. With this and a few other Excel tricks and formulas you can avoid SPSS for pretty much all undergraduate-level statistics - and go from raw experiment output to final inferential statistics in under 30 seconds.
- Inkscape. Terrific free vector graphics editor. So useful for preparing figures and diagrams, especially with programs that let you export to its format from Matlab (search for svg and matlab)
- Version control software (if there is going to be code involved anyway)
- Software-based calendar program, not paper based. That way it can be backed up, and you can do things like schedule repeating blocks of time for classes etc. Then you can feel confident that it really represents your committed time and you won't double book.
- Citation-management software, that integrates reasonably with a word processor. I've talked about EndNote here, but if you're just starting out I'd suggest the free, browser integrated Zotero instead.
- Dropbox
- Automated backups to an external hard drive (I might handle this as part of the lab infrastructure, but if *any* data is to be stored on lab members home computers then they need to have this). For the mac, Carbon Copy Cloner has worked well for me.
- Microsoft Excel (or OpenOffice) and PivotTables (called something else in openoffice). So, so helpful for analyzing data in tabular format. I would even call them magical. With this and a few other Excel tricks and formulas you can avoid SPSS for pretty much all undergraduate-level statistics - and go from raw experiment output to final inferential statistics in under 30 seconds.
- Inkscape. Terrific free vector graphics editor. So useful for preparing figures and diagrams, especially with programs that let you export to its format from Matlab (search for svg and matlab)
- Version control software (if there is going to be code involved anyway)
- Software-based calendar program, not paper based. That way it can be backed up, and you can do things like schedule repeating blocks of time for classes etc. Then you can feel confident that it really represents your committed time and you won't double book.
- Citation-management software, that integrates reasonably with a word processor. I've talked about EndNote here, but if you're just starting out I'd suggest the free, browser integrated Zotero instead.
- Dropbox
- Automated backups to an external hard drive (I might handle this as part of the lab infrastructure, but if *any* data is to be stored on lab members home computers then they need to have this). For the mac, Carbon Copy Cloner has worked well for me.
30 Minute Writing
I'm very excited about this approach to writing right now:
http://jimdavies.blogspot.com/2010/12/write-in-how-can-i-get-more-writing.html
Essentially, you write every single day, and for 30 minutes of concentration without switching away from your word processor. I've been doing it for about 3 weeks now, and it's helping me just crush through difficult bits of writing, with almost no angst. (I do more than one 30 minute block per day, but widely spaced apart) Awesome stuff. I hope to keep it up for the rest of my professional career, no matter what stage my research is at.
I think the 30 minute unitasking approach could help in other areas of life too. This week I've been doing 1/2 an hour of housework a day, and that has had amazing results too. Hard focus is the magic sauce that makes tasks easy and even pleasant. More and more I realize how precious it is.
http://jimdavies.blogspot.com/2010/12/write-in-how-can-i-get-more-writing.html
Essentially, you write every single day, and for 30 minutes of concentration without switching away from your word processor. I've been doing it for about 3 weeks now, and it's helping me just crush through difficult bits of writing, with almost no angst. (I do more than one 30 minute block per day, but widely spaced apart) Awesome stuff. I hope to keep it up for the rest of my professional career, no matter what stage my research is at.
I think the 30 minute unitasking approach could help in other areas of life too. This week I've been doing 1/2 an hour of housework a day, and that has had amazing results too. Hard focus is the magic sauce that makes tasks easy and even pleasant. More and more I realize how precious it is.
Science-based Advice on the Strategies that Work to Achieve Your Goals
http://www.spring.org.uk/2011/02/reaching-life-goals-which-strategies-work.php
I like this first because it's based on putting these ideas to the test, and because it generally coheres with other things I've learned (and occasionally posted about) I'm not sure about the advice about avoiding fantasizing. I'm quite interested in the idea of positive imaging, that you should visualize what you want in as much detail as possible, and my intuition is that it could help. Of course that intuition could easily be wrong.
- Make a step-by-step plan: break your goal down into concrete, measurable and time-based sub-goals.
- Tell other people about your goal: making a public declaration increases motivation.
- Think about the good things that will happen if you achieve your goal (but avoid fantasizing - see this article).
- Reward yourself for making progress in your goal: small rewards help push us on to major successes.
- Record your progress: keep a journal, graph or drawing that plots your progress.
I like this first because it's based on putting these ideas to the test, and because it generally coheres with other things I've learned (and occasionally posted about) I'm not sure about the advice about avoiding fantasizing. I'm quite interested in the idea of positive imaging, that you should visualize what you want in as much detail as possible, and my intuition is that it could help. Of course that intuition could easily be wrong.
Thursday, January 13, 2011
Much Better Error Bars for Within-Subjects Studies
For any scientists reading this blog, and of those, the ones who use within-subjects designs, this will be a revelation. Everyone else should skip. There's a problem that came up in our last set of reviewer comments, that if you have a within-subjects factorial design, standard error bars or 95% confidence intervals on your bars representing means do not portray the results of the repeated-measures ANOVA. Basically they're way too big, because they don't incorporate the benefit of comparing people to themselves; they include the between-subjects variance. So the basic trick for comparing two means by eye to determine significance, as described here
http://scienceblogs.com/cognitivedaily/2008/07/most_researchers_dont_understa_1.php
(in a nutshell, if they represent standard error the error intervals have to be separated by about 1/2 interval before the difference between the means is signicant at alpha = .05) doesn't work. You lose the very desirable property of being able to tell the story of your results purely in your figures.
To the rescue comes Cousineau (2005)'s within-subject confidence intervals.
http://www.tqmp.org/Content/vol01-1/p042/p042.pdf
The idea is so straightforward and easy to implement: if your data is organized with participants as rows and conditions as columns, simply take the mean of each row and subtracted it from the items in that row, making a new table. Then add the overall mean of the original table to all the entries of the new table. Each column will have the same mean as in the original data, but the row means will all be identical to each other and to the overall mean. Now construct your standard error bars or 95% confidence interval bars in the usual manner. Then the error bars will represent only the difference due to condition differences, and visually comparing any two error bars in the manner described above is the equivalent of doing a paired-samples t-test between the means (I haven't doublechecked that) When we did this to our latest paper, the difference was like night and day: all of a sudden nearly all of our significance findings were clear and easy to read off the bar graph.
There's a risk here, that your readers may not know what the heck you're doing, or even be suspicious that you are trying to make your results look better than they are. But the visual pairwise comparisons will be very close (not necessarily exactly the same) as the pattern of results from the corresponding ANOVA (and at least one reviewer out there is certainly applying that kind of visual test even when inappropriate, that is, for a within-subjects design), and there is a paper to cite for the idea. It has now been cited 67 times so it appears the idea is catching on.
Read the Cousineau paper first, but as a late breaking correction to it there's this paper, Morey 2008:
http://www.tqmp.org/Content/vol04-2/p061/p061.pdf
It appears that the error bars are slightly too small when done the Cousineau way, but can be fixed by an easy numerical correction.
http://scienceblogs.com/cognitivedaily/2008/07/most_researchers_dont_understa_1.php
(in a nutshell, if they represent standard error the error intervals have to be separated by about 1/2 interval before the difference between the means is signicant at alpha = .05) doesn't work. You lose the very desirable property of being able to tell the story of your results purely in your figures.
To the rescue comes Cousineau (2005)'s within-subject confidence intervals.
http://www.tqmp.org/Content/vol01-1/p042/p042.pdf
The idea is so straightforward and easy to implement: if your data is organized with participants as rows and conditions as columns, simply take the mean of each row and subtracted it from the items in that row, making a new table. Then add the overall mean of the original table to all the entries of the new table. Each column will have the same mean as in the original data, but the row means will all be identical to each other and to the overall mean. Now construct your standard error bars or 95% confidence interval bars in the usual manner. Then the error bars will represent only the difference due to condition differences, and visually comparing any two error bars in the manner described above is the equivalent of doing a paired-samples t-test between the means (I haven't doublechecked that) When we did this to our latest paper, the difference was like night and day: all of a sudden nearly all of our significance findings were clear and easy to read off the bar graph.
There's a risk here, that your readers may not know what the heck you're doing, or even be suspicious that you are trying to make your results look better than they are. But the visual pairwise comparisons will be very close (not necessarily exactly the same) as the pattern of results from the corresponding ANOVA (and at least one reviewer out there is certainly applying that kind of visual test even when inappropriate, that is, for a within-subjects design), and there is a paper to cite for the idea. It has now been cited 67 times so it appears the idea is catching on.
Read the Cousineau paper first, but as a late breaking correction to it there's this paper, Morey 2008:
http://www.tqmp.org/Content/vol04-2/p061/p061.pdf
It appears that the error bars are slightly too small when done the Cousineau way, but can be fixed by an easy numerical correction.
Subscribe to:
Posts (Atom)