St-Clock

Although this is a good representation of a Ph.D. contribution, I don't think this is the point of the current debate. I would be perfectly happy to have made a dent in the knowledge circle and I think it would be worth the sacrifice, from a knowledge perspective. The problem is that you sacrifice a lot to make that dent and you often don't see the benefits (e.g., an interesting job). It's another story for the supervisor though...

One thing I try to do is to never stay logged in on any Google service like gmail or analytics (my browser remembers the password). I'm sure they could still track most of my internet traffic, but at least, their data becomes less reliable.

I also don't use Google DNS by principle.

1. This is a very limited subset of conferences/fields. 2. Google Research is younger than Microsoft Research. 3. Google Research is really mixed with their product development whereas Microsoft (and IBM) Research are more isolated so they can focus on fundamental research and publications (there is some transfer...). A Google representative once said in a software engineering conference: "Come work at Google Research, you'll work on real and interesting research problems like GMail". Gmail==Research? Really?

I'm not sure who voted you up, but the second part of this comment looks trollish (apologies if it wasn't).

For a starter, this is a very small subset of computer science conferences (e.g., the main software engineering and programming language conferences are missing). I'm not sure in which field Yahoo! Research works (looks like data mining), but in SE and PL, they are inexistent.

Best paper awards are just one metric among many. Citations, venue impact factor, number of publications are other (imperfect) metrics. It's probably fair to say that MIT and other big universities and corporate labs (IBM and Microsoft) have high scores in all of these metrics, whereas Yahoo! Research is still too young and too small to compete with even smaller but dynamic universities w.r.t. these metrics.

Is this still under development? The last blog post was in March 2010 and the website still says "beta".

Yup, "The service is not available. Please try again later."

"The main reason is that at this point, BitBucket is simple attempting to feature copy from Github except using Mercurial in the background."

I am so tired to read things like that.

Issue-tracking at github is far behind bitbucket issue tracking (try to make complex search queries... you cannot link a commit to an issue except when closing the issue... tags are nice, but i dislike having to create a tag for each milestone/release).

Wikis at Github have only been recently backed by a git repository: as far as I know, wikis at bitbucket have always been backed by a mercurial repository.

It's true that bitbucket is a lot like github and that github has a bigger and more dynamic community (from my experience, github's infrastructure is also faster), but can we leave bitbucket alone and stop the bashing?

Sorry, i'm probably weird, but I could see my vim experience described quite accurately in this post. It was funny and disturbing at the same time.

I particularly enjoyed:

"hjkl movement becomes natural. you mysteriously lose the ability to ice skate"

"oh balls. my .vimrc isn’t on the mainframez so all my cool shit is gone when I ssh to edit some configs."

"publishes obligatory blog post about how special and unique setup is"

"Which makes me wonder what is in the leaks that we haven't seen yet". Maybe there is nothing in those cables, but politicians must be very afraid of what could be published next... Although the irak cables were potentially far more damageable, they were not released in cooperation with major newspapers.

This is not politically correct to say, but given the gap in revenues between white and black, I wonder how the US would do if they had better redistribution of wealth.

"But now all sorts of well-established, multiply confirmed findings have started to look increasingly uncertain."

Good! That's what science is all about. Making theories that can later be "disproved" or "confirmed". It's when you cannot disprove something that it is no longer science.

I find LaTeX-Suite bloated. Since I already use snipMate and SuperTab which provides minimal but useful TeX support, I only had to install TeX-PDF (with Rubber) to ease the latex/bibtex compile cycles.

I agree, it was my first time on the website and I genuinely liked the idea of "design for hackers", but when I read things like see "How 6 of my last 9 blog posts made it to the front page of Hacker News", I got irritated. Was this book and newsletter supposed to be about design? Maybe a refocus is in order here.

+1 for duplicity. With the command line tool, you can also:

1- Easily go back in history (if you erased a file and want it again, it's here, in a previous incremental backup).

2- Ask to automatically make a full backup every X months.

3- Ask to automatically remove incremental and full backups older than X months

4- Select the granularity of the backup buckets (very useful to avoid having hundreds of files or if your connection is unreliable).

5- Upload a backup bucket while preparing the next one in parallel.

6- Use S3 reduced redundancy storage if you want (cheaper...)

Duplicity is written in Python and the API is easy to use if you want to rely on a Python script instead of a shell script.

The fact that you can use S3 was a deciding factor: it is a really cheap solution with a big ecosystem of tools and it should still be there in 10 years...

Just to clarify: i know Arch does some customization, but certainly not as much as Ubuntu. Arch also heavily relies on configuration files while I rarely edited a config file in Ubuntu. Presumably, it takes a longer time to adapt the GUI config tools in Ubuntu when new options are introduced.

Yes, and it seems that they don't customize the packages as Ubuntu does (Arch does not for sure and that's one of the reasons I'm using it).

So I really wonder how Ubuntu will pull that off (rolling release of customized packages?!)

Just want to mention that long-running applications written in 1. (e.g., C) can be slower than languages in 2. (e.g., Java) because the VM can optimize hot operations.

Other than that, I agree with your post!

There are many dimensions to consider when choosing a martial art and it may be wise to try a few if you never had experience beside fencing. I practiced Karate (shorin-ryu) for three years when I was young and I was never really happy. I've been doing flying white crane kung fu (fujian, not tibetan style) for many years now and I love it. It just depends on each person. Some dimensions to look for:

Soft (use the opponent strength) vs Hard (punch and kick hard!)

Traditional (deep historical background, traditional forms that have been taught for hundreds of years, at least some sparring without any equipment, spiritual dimension) vs Modern (sparring is almost exclusively done with equipments, more acrobatic moves)

Martial Arts vs. Combat Sport. I don't want to get too much flak on this one, but IMHO, the more rules you have, the more it looks like a combat sport. Which is perfectly fine if this is what you like!

Each school will favor some of these dimensions over the others, even among the same style (e.g., karate, kung fu, and even boxing vary widely across schools).

But the most important aspect is probably the teacher. If you are afraid of your teacher, if you are stressed every time you are on your way to the school, it won't do you any good, because clearly, you are not going there to become the next Bruce Lee or to learn how to save your family from burglars.

"some random people send you email and have such "eclipse plug-in" problem" -> They are not random people, they are Ph.D. students working on their thesis and you can be assured that they will find a way to publish their comparison one way or the other. Some Ph.D. students will go at great length to figure out other's code before asking a question, but some will suck your energy big time and still do a poor job at representing your work. Now compare this to those who do not release their code: they can only be evaluated on their published results and they can spend their time on their next study/research work.

My point is that there is not a lot of incentive to release code: it requires a lot of your time, it is not always clear that granting agencies/employers consider this as being more positive than publications, and it exposes your work to all sorts of unfair replication/comparison.

Now, if research work without released code were never cited, that would be a terrific incentive to release your code :-)

P.S. I speak from experience and I release the code of my most important research work.

Oh, you must do "customer support"! Otherwise, they will use your code and report 1) that your code does not work or that 2) their code is better because they did not use your code well.

And for code that is not easy to install (e.g., requires many libraries, frameworks, dev tools, etc.), you need to spend time on documentation and even then, you often need to answer questions like "how can i create an eclipse plug-in to use your code or lxml failed to install"...

Every time I see a reference to that website, it makes me cry.

As someone said: "The biggest non-technical difference between git and mercurial is the rabid culture surrounding git. mercurial users fairly happily and quietly use their tool, while I've had to send two separate door-to-door git missionaries away today alone." (http://www.rockstarprogrammer.org/post/2008/apr/06/differenc...)

Wrong, Ph.D. candidates get a lot of pressure to be associated with an idea, or a set of related projects, and they can get a lot of flak if they deviate a little because when you apply for an academic or industrial job, "you want to be known for an idea, not for your degree" (disclaimer: i'm a ph.d. candidate)

I 100% Agree. There is a difference between "it all started with a revenge stunt" (the lawyer say something like this at the end of the movie) and "I worked all these years motivated by a revenge stunt".

Font rendering is horrible (at least on linux), you have to try various l&f before finding one that is "acceptable", etc.

It's a great IDE, general usability is certainly better than Eclipse (it was meant to be an IDE as opposed to Eclipse which is a platform with an IDE built on top), but personally, font rendering was a deal break.

I think he was being ironic (look at how it was quoted: @bar.com and not @http://bar.com)

On the whole, I agree with you, but replication is often expensive and in some fields (e.g., software engineering research), it's not enough to warrant a publication in a high profile journal or conference (especially if you confirm the original findings). Heck, you may not even get funding to do that.

Regarding "if you seem to have done everything correctly", I don't think that any honest scientist can claim that his/her study had no limitation or flaw. I regularly review papers for big conferences and there is no such thing as a perfect paper/research project/study. It's more like a threshold: despite the issues, were the findings novel, relevant, and found through a rigorous process? Would the community learn anything valuable by reading this?

Articles like the one cited by OP are useful if they make scientists and normal "folks" realize the limitations of alpha values, they are useful if they make scientists reconsider some of their methods (and way of presenting findings), but they can be harmful if the readers conclude that most scientific findings are "false" and thus, that science is bogus because it cannot find the "truth". Science is rarely, if ever, about true and false, religion is.

P.S. I realize this answer was more about the article, and less about your reply, it's just that your reply prompted me to write something :-) Again, I agree with you!

I disagree. Give me a hard problem and a short deadline, leave me in front of my computer with my environment set up (so I can do some quick trial-error), don't look over my shoulder to evaluate me (but you can help me if we're doing pp), and I'll succeed. Give me an easy problem over the phone (I'm a visual) or on a whiteboard (my handwriting is not good) while timing me, and I will fail.

But in the end, I like this kind of interview, because I don't want to work for anyone who would evaluate me like that.