decklin

I'm not looking at this very deeply (on phone), but I think this does not make a good landing page. My impression of the copy is "sync what queues?" I have a Netflix queue; I don't want to have another one that needs to be synced with it. It sounds like it's creating a problem I don't have. There's nothing that explains the benefits of GoodFilms to people who aren't already users (like the front page does). Netflix already shows me "new" and "popular" films, which is not very useful anyway.

I suggest that if a user is not logged in, show something that focuses on what makes GoodFilms useful and that you can use your NetFlix queue on it instead of having to make another one. You can just say that you can browse on-Netflix films specifically; taking up a bunch or screen real estate with what amounts to an example of that does not explain it any better.

This is good advice, but I think using the active voice as well as the present tense helps. For example, instead of "...a Preview is shown", say "Show Preview when...". Simple sentence structure is easier to skim.

The suit is back!

I'd like to be able to drag the two thumbs on each "Lightness" slider past each other, to invert the usual relationship. (Just splitting them into two sliders each would be fine.) Otherwise, this is really great.

Indeed. The twitter announcement had the vague feeling of something produced after many heated words between product and engineering. This sounds like what it is, someone trying to sell their own personality.

On a similar note, I like to periodically clean out my shell history by dropping commands less than a certain Levenshtein edit distance from earlier commands[1]. Looking at the removed lines would probably produce further ideas for optimizing commands; I'll have to try it next time.

[1] rudimentary script: https://github.com/decklin/bin/blob/master/hprune (haven't bothered to add real option parsing or make it efficient; for shell histories of thousands of lines you probably want to up the search depth to something in the hundreds)

Please, please do not scrape the HTML pages. The XKCD admin has put a JSON API (http://xkcd.com/json.html) in place for people who want to do things like this. Thanks.

That article is mainly about the strawman argument that if Dvorak is better, then free markets are bad, because they lead to the kind of network effects that gave us QWERTY. (As a happy Dvorak user, I find this argument silly.) The author is very motivated to prove the premises false, for political reasons. I don't think it's very insightful.

For the same reason that they should make sure every tech has a functioning screwdriver? Everything that adds 5 minutes to an appointment adds up.

The success of e.g. Apple is because they wake a product that works, not because of some nerdy argument about settings. If your product is getting a tech to show up within an N-hour timeframe, you need to find ways to engineer that.

Of course, as has been pointed out up thread, most cable companies don't have to care if their appointments are reliable, so they don't, and whichever department made that web form probably wasn't accountable either.

And in case anyone is not familiar with just how long "long before" was: http://itre.cis.upenn.edu/~myl/languagelog/archives/001604.h...

Everyone seems to be talking about the force-yourself-to-code part of this, but I recently read something about "write every morning" and as a tool for that this sounds appealing -- but only because my job is coding, not writing. Sort of like stretching before athletic activity, maybe?

The only thing that worries me is I have to inspect and trust the code to not do anything with potentially sensitive text. something local would be nice.

The first thing I thought of when I realized the author was seriously using SLoC as a performance metric was this story: http://folklore.org/StoryView.py?story=Negative_2000_Lines_O...

Interesting! I wasn't sure if anything like this had been done. I guess it shows that there's nothing good that can come from change here.

I wonder why browsers with a modern automatic-update process don't set their user-agent to something that discards all this madness ("Chrome/23.4.5678 (Windows)", or similar) for the cutting-edge/nightly builds only (or even betas, if they wanted to discourage casual users from switching to them, but I don't think that's the case at this point). Surely their users have signed up for a little breakage in exchange for the latest features? And if they actually get website operators to stop or at least fix their sniffing, the whole prisoners-dilemma situation would disappear.

(I guess this assumes that the huge user-agent that my Chrome is currently sending is necessarily bad, and in the real world maybe no one really cares...)

Actually, the order in which function arguments are evaluated is undefined. http://c-faq.com/expr/comma.html

You can use a comma in other expressions to introduce a sequence point: http://c-faq.com/~scs/cgi-bin/faqcat.cgi?sec=expr#seqpoints

You can communicate with the Skype client over DBus, which allows you to replace a lot (although not all) of the UI with something else. Two examples:

http://code.google.com/p/skype4pidgin/ https://github.com/technomancy/dotfiles/blob/master/bin/skyy...

The protocol is incomplete (the "native" Qt UI isn't built on it), but it is at least documented on their dev site.

If you're interested in learning music theory (as I am currently) and how the the Beatles' music evolved, this series of articles by Alan W. Pollack is fantastic:

http://www.icce.rug.nl/~soundscapes/DATABASES/AWP/awp-notes_...

They were originally Usenet postings(!) starting in 1989 or thereabouts, but have been revised a bit. Warning: you could lose hours reading through them all.

One thing that comes up many times there that I'd be interested in seeing added to a quantitative analysis like this one is how many times something can "seem" like it might be in one key but ends up really being in another, but I don't know how you would count that.

OS X turns off the delay if you map that key to something else. The only workaround I've found in Linux is to force the LED on with setleds so that the hardware thinks it doesn't need to enable the delay.

Sonian (Newton, MA) - Full time, local or remote

We're a funded (recently closed Series C) email archiving startup with many large corporate customers. We are looking for an additional engineer for our operations (or "devops" if you prefer) team, to take us from merely using configuration management and automated deploys to delivering our internal services as dependable, multi-IaaS-portable tools that empower everyone from support engineers to Clojure backend developers to solve customer issues without paging us and use our monitoring API to automatically spin up instances to process work.

We have offices outside of Boston (Newton), but our dev and ops teams are mostly remote, and we meet over VOIP and pair-program in tmux.

Some highlights of what we do:

  +   We manage hundreds of compute instances across multiple clouds, including over a petabyte of data.
  +   We write the code to help facilitate deploying our application to new public clouds.
  +   We automate everything we possibly can
  +   We build up awesome applications, and then opensource them (Perhaps you've heard of our monitoring framework Sensu - https://github.com/sensu )
  +   We build applications to manage every aspect of our compute and storage infrastructure, from deploying code, to provisioning systems on multiple clouds, to securing access control.  For every problem - there is a code solution.
  +   We have an awesome time working together on technologies that very few (if any) other companies are doing today.
  +   We have days where all we do is hack on a specific project (http://blog.petecheslock.com/2012/02/13/why-your-company-should-have-internal-hackdays/) - leave the day-to-day tasks for the next day.
  +   We meet up 2-3 times a year in Boston - with the rest of the company.

What are we looking for in a new member of our team?

  +   Excitement, energy and a willingness to learn something new.
  +   The ability to take a task from idea to code to testing to deploy (all within a short amount of time).
  +   Bring new ideas around project or processes to the team - if we like it - we'll do it.
  +   You can work from home, from our Newton, MA headquarters, or your local coffee shop.  Anywhere with Internet access.
  +   You'll be part of an on-call rotation with the rest of our team.
  +   Flexible person who can work by themselves or pair with their team to teach or learn.

What are some things you should know or have worked with before? (Some, many or all)

  +   Ruby (must know or want to learn)
  +   Not required, but pluses: Python (used by some parts of our infrastructure), Clojure (used by our backend team)
  +   Chef (or other configuration management systems), Fog
  +   Rails, Sinatra (both also used by our front-end team)
  +   Git or other DVCS
  +   System administration with Ubuntu (includes security, logging, monitoring, web/SQL admin, etc, etc)
  +   Knowledge of Openstack (especially Swift) is a plus
  +   A love for making complex configuration and systems simple via automation
  +   Continuous Deployment, Agile Development/Kanban
  +   Contributing to open source

Interested? - Contact pete.cheslock@sonian.com or https://twitter.com/petecheslock, and let him know you saw this post. Please include a short summary of why you think you would be a good fit, resume/blog/etc, and a link to a Github account or other code samples.

You might want to take a look at this:

http://www.whatwg.org/specs/web-apps/current-work/multipage/...

I need some sort of AI filter for HN that will not only hide blog posts from a particular domain, but hide other people's blog posts "responding" to those blog posts and then submitted as separate articles.

Most people who buy and/or torrent lots of music have a very large amount of music available to listen to. Listening to a full album by a band you may or may not like is an investment of time that you could have spent listening to something else. A free sampler implies that someone has tried to pick out only the best tracks for you, and removed any "do I really want to buy this/do I really not-care about the ethics of torrenting this" friction involved in adding it to your playlist.

If you do listen to the sampler, that someone's objective is accomplished - now the album is in your head. Later on, when you think, "hmm, is anything good coming out this week?", you might think about it and buy it. (Of course, if you never buy and always steal, that means you'll steal it, but then you wouldn't have paid for it anyway.)

One thing in this article I do agree with is that you should try to write "as little code as possible". But it doesn't follow that if you don't know how to code, you'll write zero code, and do well by that metric! You'll be unable to identify people who can code well, and hire poor coders to solve your problem, ending up with lots of low-quality code, in situations where an experienced programmer could have come up with a better solution ("hmm, I've seen this before...").

Even the slightest familiarity with programming here will help someone understand that when an offshore development house says "oh, sure, we'll just throw 20 more people at the project", that will not necessarily make it done better or faster. This is a Good Thing. And this has nothing to do with whether or not Mr. Bloomberg is implementing his own Ruby or .NET scripts.

I'm still getting this error. OP, if it helps, I have click-to-play on for Flash.

I don't know, my "cool Apple owning friends" told me for years that unhinted font rendering was better, and I always thought they could pry my Bitstream Vera from my cold dead hands. Then I used OS X full-time for a while. Upon switching back to Linux, I immediately turned hinting off. It just... looked "wrong".

Previous discussion: http://news.ycombinator.com/item?id=3667797

There's something discomforting about seeing an apology "for any offense" from a party who is not the offender. A common derailing technique is the non-apology-apology: rather than apologizing for what I did, I apologize for your taking offense, which makes it look like you're the one who has a problem. Here, Cloudmine does this on behalf of the creeps at Boston API Jam!

Why? It seems clear to me, based on their actions, that Cloudmine's intentions are good and that they don't mean to blame the victims. Comparing this post to other blogs and Twitter messages, I think I see why, and it's the same reason the post doesn't quote or directly link to API Jam: it's written in a different register -- "professional" communication.

It's absolutely necessary to do that if your company takes a public action, like pulling sponsorship, and needs to publicly address it. What interests me here is that the non-apology-apology is creeping into the English language by sounding "more formal", because our ruling classes use it so much (and no one seems to care when they're called out on it). So, I hypothesize that whoever wrote this up was trying to stay calm, detached, and respectful, and unconsciously hit on this rhetorical device because of it.

Note, I do not think that CloudMine is under any obligation to apologize for other people's actions, or that the writer is stupid. I'm just rubbernecking a collective lingustic maladaptation.

You may enjoy: http://languagelog.ldc.upenn.edu/nll/index.php?s=crash+bloss...

Wow. I thought this might be something I could send to my parents to explain this stuff, but it's just some guy being snarky and sarcastic and playing to "us" for laughs. I guess if your goal is feeling superior, it's a worthwhile diversion.

No mention of M-x grep?