Looking at my inbox, I can only say - today will go down as one of those special days when my life changed. My most sincere thanks to everyone, I will get back to you soon (it's 1:30 in the morning here).
410 karma · joined June 23, 2010
Book on Node: http://book.mixu.net/
nwm, the Node.js window manager: http://mixu.net/nwm/
github profile: https://github.com/mixu
I would speculate that the tone might be in part due to cultural background as well - as Wikipedia says, "The Finnish language has no simple equivalent to the English 'please'." http://en.wikipedia.org/wiki/Finnish_grammar
As for the background story, I did start this project for fun (it was an old hackathon idea) but kicked it into higher gear due to this opportunity coming up.
Thanks!
So the number of threads you have (e.g. setting StartServers, MaxSpareServers/MaxSpareThreads) is way less important than the keepalive timeout: you can/should start enough threads to use all the available resources, but it will only make a difference if you aren't idling all those threads with a high KeepAliveTimeout. Apparently, that's the Apache default setting.
1) Layout and widgets only.
I start with a reset, then layout (block positioning), then widgets. Everything other than layout is a widget. There are no global styles. If there is one of it, it's an ID. If more, it's a class.
2) No (global) element styles! Ever!
No, not even a p { margin: xyz } or a ul { list-style-type }. Every element that cannot be referred to by class / id must look exactly like it would after the reset.
This avoids complex dependency chains where coincidence influences look. Widgets can be moved from one site to another and you actually have to think new widgets through rather than relying on the default values.
3) Avoid classes and id's as much as you can.
Have single base element with a simple but descriptive class name, and then specify the sub-elements.
For example: Instead of “div.content-title” use “#content h1″ (e.g. div#content with a h1 tag inside it).
Basically, design widgets which consist of one base element, and refer to sub-elements via longer expressions. Use indentation to separate sub-elements.
My strong preference is to maximize for human readability (=short dependency chains, descriptive widgets rather than single elements) rather than short CSS. Cascading within widget-scope is fine, but cascading with global styles should be avoided. I find that when I follow these rules, I like my CSS a lot more.
https://spreadsheets.google.com/ccc?key=0Ag-R_ZlGO21NdE9HSWR...
Also related: http://www.builditwith.me/
The problem with idea sheets seems to be that most ideas are terrible...
Didn't John Keynes say:
"The long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is past the ocean is flat again."
Though apparently, he was talking about inflation in this case... but the point is, the nature of risk is such that while it works out on average, you still end up with an individual outcome at a fixed point in time which may vary a lot from the expected payoff. Not that I do completely agree with tastybites, but I see the logic.
It's a simple, embeddable widget for getting feedback. The beta is free, final pricing will be inexpensive and the full "Show HN" thread is on it's way...
http://blog.peepcode.com/tutorials/2010/what-pythonistas-thi... (Blog post about the video referred to by the author)
http://news.ycombinator.com/item?id=1097723 (HN discussion about that blog post)
It's like finding out that IE9 only performs well on the subset of JS needed when you are drawing a fishtank..
For instance, I've never had to put in my 3D Secure code on Amazon, BackBlaze, Syncplicity or ZumoDrive. The problem is that at least here in Finland, the only company (representing all the local banks) offering credit card processing practically requires 3D Secure unless you implement everything yourself (e.g. can't use their CC vault) - and no, unfortunately the US subscription API services don't work here, unless you somehow manage to get a merchant account in a UK bank.
Experts can write the focal elements (e.g. pseudocode) first and then expand them into a full program. A good example might be the fizzbuzz test, or generating the permutations of a string, you'll have an idea of what the problem involves because you've done it before. If you can't visualize the problem in terms of something familiar, you'll have a hard time taking a top-down (e.g. memory-recall-based) approach.
To my understanding the issue is not whether "state" has side effects (e.g. your global vs. local example), but rather whether there is any concept of "state" (e.g. shuffling stuff from one memory slot to another).
My experience is from Prolog, where the statement "X is X + 1" results in an error, because once unified, the variable X cannot be modified - X is not a memory slot. Once you try to let go of the idea of a machine with memory slots, alternative-paradigm programming gets a little bit easier. Of course in practice state is often needed, so it often gets added through some mechanism.
Second, and just as important as the first issue, is that the Python core developers liked git the least out of the three DVCS options by a wide margin."
I'm using Mercurial for the same reason - Windows support without complex dependencies.
On a more serious note, I do agree with you. Some decent movies with a office environment/work theme: Boiler Room (2000), Duplicity (2009), Glengarry Glen Ross, Other People's Money (1991), Swimming with Sharks and Wall Street (1987).
I've also enjoyed the dramatized real-life stories in Rogue Trader and Barbarians at The Gate.
Only assuming that all people are equally likely to interact with one another. Which I doubt is strictly true, given that those cases in which I've actually interacted with everyone from an age group (e.g. military service) are strikingly different to the daily interactions I have. I would imagine people in places where opinions actually matter (such as the workplace) mostly interact with people of somewhat similar background and smarts, though obviously with different fields of specialization.
But even within groups of more-or-less equally smart people, the choice of topics to be assertive about and the style of argument are still factors that make people seem arrogant, so I do agree with you. But I would attribute it more to choice of topic and style of argument rather than "smartness". As you said, the definition of "smart" is problematic and potentially circular.
Example tables:
http://www.jbryanscott.com/2009/02/07/nasdaq-100-revenue-per...
http://37signals.com/svn/posts/2283-ranking-tech-companies-b...
A quick rundown:
The student unions (e.g. http://en.wikipedia.org/wiki/Student_unionism_in_Finland ) are part of the university administration, have 100's of millions in assets and provide basic services (healthcare, apartments, places to party) in exchange for 40+ euros per year.
The subject- and interest-based associations are separate entities that get some funding from the student union and organize events. They don't charge, and mostly get their cash from organizing events. However, since everybody lives around town, the interactions are more focused around specific events. The nations (http://en.wikipedia.org/wiki/Nation_(university) ) are a lot like the student associations, except they are specific to people from a particular area in the country (not strictly enforced), have a bit more money (e.g. may have their own apartments) and organize activities centered around traditions and whatever is popular.
Thanks for a good question, interesting to see the parallels in other countries...
Here is the comparison that convinced me (the Sennheisers are at the end): http://www.youtube.com/watch?v=cQdbcDTiIas
http://en.wikipedia.org/wiki/Student_benefit
(Though realistically the total sum is only enough for paying rent, so most people still end up having to work somewhat during their studies.)
Not that it isn't an important new idea in practice, since the psychological difference between 0.01 cents and 0.00 cents is huge.
If you end up doing the thesis with on freemium, you'll probably either end up doing a quantitative analysis (check out the quant research on shareware), or qualitative analysis around price discrimination, network effects and business models (check out this MA thesis https://jyx.jyu.fi/dspace/handle/123456789/23051).
Coming from one of the Nordic countries, I would say that I have accepted a social contract that says I pay more in taxes than in other countries, and for those taxes I get "free" healthcare and "free" university education. The contract is that people get something at the beginning and the end of their lives and aren't screwed when they are faced with injury or illness, and pay more than they get during their working years. As someone said on on /., taxes buy me (this particular type of) civilization.
I'd keep that separate from the economic (e.g. private ownership and allocation of resources via somewhat free markets) and legal (e.g. governmental decision making based on majority and strong individual rights) aspects.
Unfortunately, most people confound these different aspects for historical reasons, so they have come to mean "good choices in those factors" and "bad choices in those factors" depending on your POV.
This just seems a bit too much like a Rorschach test (http://en.wikipedia.org/wiki/Rorschach_test#Controversy ), and the interpretation that you can directly come to some conclusion about a job applicant based on what they say about a hypothetical person is even warned against in the actual article:
"First, the studies clearly demonstrate that perceiver effects represent more than simply the projection of an individual’s self-image onto other people."
Projective tests have their uses in clinical psychology (e.g. with the crazies) - but whether you would want to apply them to normal situations is questionable.
"In each study, we found regular relationships between personality traits and perceptions of others. Individuals who perceived others positively reported higher agreeableness (particularly aspects associated with friendliness and low hostility); higher life satisfaction; and lower endorsement of measures of personality disorders, depression, and antisocial attitudes."
In other words, the test is likely to tell you two things: first, how positive you are (which is unsurprising, since positivity is almost by definition related to seeing things and others in a positive light) and second whether you are not totally nuts (e.g. positive people are likely to be associated with lower measures of personality disorders, depression and antisocial attitudes).
But in non-clinical settings - such as in a job interview? "The magnitude of individual differences in the positivity of perceptions of others is fairly large, and individual differences in perceiver positivity impacted judgments of single targets to a considerable degree." That is, the positivity measure is stable, but there is a lot of variance. This suggests that you don't want to use this for finding out much more than how positive a person is - and if that is your sole criteria in a job interview, good luck!
Edit: to clarify that last quote - it basically says that among a normal group of college students, positivity measures have high individual differences and these differences impact judgments to a high degree. So big differences in the results of the measure may be simply explained by individual differences in positivity measures.