HNHacker News
TopNewBestAskShowJobs

carbonica

544 karma · joined November 23, 2009

Carbonica - http://carboni.ca/ - OSS projects, usually involving Ruby.
submissionscomments
carbonica··on How Twitter Could Win Local
It also bugs me that @refs count against the limit. When I come across 3 developers having a conversation and I'd like to chime in, I'm already down to <120 chars just by including their names. And since Twitter's model means they can't reasonably find out about my tweet if I don't put their name in it, the more people you want to talk with, the less you can say.
carbonica··on Lion's Mission Control: UX fail, especially for multiple displays
I personally only use OS X on my laptop, and here's what I had before Lion:

1. 3x3 Spaces grid 2. Each major application assigned to a space. Primary editors (vim, Textmate, etc.) in space 5 (center), others surrounding it. 3. Each application typically maximized to fill its space. 4. When I need to switch applications, command-tab between them.

This gave me each application with almost a full screen of space, isolated to itself, with little cognitive overhead. I don't need more than command-tab at this point to manage my workflow.

I installed Lion today. Now, I have a linear set of spaces, but some of the applications I was giving full-screen to can actually use the whole screen. Mail is one of them. Chrome's support is a bit buggy so I'm avoiding it for now.

I still command-tab between applications, and that works fine, giving me all the space I had before. It seems to me that Lion's replacement for spaces was designed for me: someone who assigns one space per major application and likes to have those applications full-screened. I can 4-finger swipe left/right between the spaces (I had to set 3-fingers for page swiping so I could keep using the trackpad for Chrome forward/back). That means I can't use 4-finger-swipe to pull up the command-tab switcher, but 3-finger-swipe-up brings up MC, which is close enough, now.

Just food for thought. If I had to deal with the multiple-screen issues OP mentioned, I'd probably go nuts, but I've only used a customized Ubuntu for such situations long-term. As a laptop user, MC seems designed for me almost to the T.

carbonica··on Twitter Closing $800M Funding Round at $8 Billion Valuation
The only new scalability issue that Facebook has tackled is how to scale one enormous MySQL database. They've primarily done this by sharding it thousands upon thousands of times and putting over 10,000 memcached instances in front of it.

Most everything else they've confronted was done by Amazon and Google years before Facebook had to.

carbonica··on Google In-App Payments for the web
I'm a bit confused - this is about in-app purchases for the web. Are you saying that on Android, a web in-app purchase shows a lightbox for adding a credit card and does not temporarily redirect to Google?
carbonica··on Google In-App Payments for the web
Ah, I hadn't noticed that if you were already logged-in and had already entered CC information, that it uses an iframe lightbox, which you're right, naturally does not show a URL.

But then, they won't be entering any CC information without being at Google's domain. You only enter CC information at checkout.google.com, and it initiates a popup to go there if you are either not logged-in or don't have a CC entered.

So, as an attacker, all they're doing is getting you (the naïve user) to click a button that looks like Google's button, and since they've already gotten you to click on a button to begin with (to initiate the transaction) they've already gotten any clickjacking exploit you need out of the user.

carbonica··on Remove any Site From Google (even if you don't control it)
Those flaws were all over their entire codebase in very basic parts of the site's functionality. They had literally zero authorization checks.
carbonica··on Show HN: A Bitcoin casino with 23 games.
PayPal does more than just send money - it can hold a balance for you securely, which given all of the hacking incidents (MtGox, BetCoin), is a pretty important thing for the BitCoin community to get right. Sending someone money without having to worry about the security is a pretty important service.

Who knows how you make money off it, though.

carbonica··on Google In-App Payments for the web
The same way you already tell if the Google checkout dialog is legitimate? HTTPS?
carbonica··on Aaronsw indicted for hacking MIT network to download millions of JSTOR docs
I think the parent was just disappointed at using "hacking" to refer to something as mundane as getting guest accounts to the network.
carbonica··on Aaronsw indicted for hacking MIT network to download millions of JSTOR docs
I wonder what they'll push for. He sounds pretty screwed if this evidence pans out. Looks like he could even end up with a few years' time if the prosecutors want.

1. Wire fraud maxes out at 20 years outside of a presidentially-declared emergency. No fine cap, it seems. http://uscode.house.gov/download/pls/18C63.txt

2. Computer fraud under 1030(a)(4) caps out at 5 years with no prior offense, no fine cap. http://uscode.house.gov/download/pls/18C47.txt

3. 1030(a)(2), (c)(2)(B)(iii) looks to be another cap of 5 years. Ibid.

4. 1030(a)(5)(B), (c)(4)(A)(i)(I),(VI) looks like another cap of 5 years. Ibid.

IANAL, just trying my best to read the code itself.

carbonica··on If Dropbox Used GitHub’s Pricing Plan
I laughed at the article, but you have a really good point. If you have 30 projects and only want a place to put your code, you're far better off just using a private server, which you undoubtedly already have. If you have a development culture that is well-modeled by how GitHub approaches collaboration, that's where you'll see a lot of added value over a private server.

I think the tough part is that GitHub's innovations in collaboration are primarily a huge win for open source. GitHub makes discovery of these projects so much easier, connects disparate people across communities (and countries!), and provides a unified technical stack and process "stack" for those people to contribute (same bug trackers, same "send me a pull request" approach, same wiki). That's something that's a pretty big deal for OSS.

Most of those issues aren't as big a deal for a business in my experience - most businesses either don't have such problems (discovery) or have their own solutions (process). But hey - you guys have got customers, you're hiring like crazy, and I love your stack personally; I'm certainly not judging! Just offering my perspective on why some folks might not buy in to the collaboration stuff from a business point of view.

carbonica··on Why I will never pursue cheating again
> Sometimes it is, sometimes it isn't.

This isn't code, these are written assignments we're discussing. English is pretty damn flexible, and nobody gets suspended for "Abraham Lincoln died in 1865." There's no point invoking degenerate cases to argue against a policy. That's just silly.

When a nontrivial sentence appears both in an essay and in one of the listed sources, the Wikipedia page, or a top Google hit for the subject, you can draw inferences. If you can argue that you just happened to write:

> However, just six days after the surrender of Confederate commanding general Robert E. Lee, Lincoln was shot and killed by Confederate sympathizer John Wilkes Booth at Ford's Theatre in Washington, D.C.

just as it appears on Wikipedia at present, and convince your peers it's more likely than not that you did, then you won't be punished. Not unsurprisingly, such pleas typically (but not always) fail.

carbonica··on Why I will never pursue cheating again
28 students last year (out of about 4,000) were reported. 17 were suspended, 7 got probation, 3 were found not responsible, and 1 had the allegations formally rescinded.

This is roughly consistent year-to-year.

carbonica··on Why I will never pursue cheating again
Where I attended college, you get a year's suspension for getting caught copying a sentence without citation. Anything as egregious as this presented would've meant a class full of suspended students.
carbonica··on Introducing BrowserID: A better way to sign in
Dartmouth does this as well, and I agree the interface sucks in browsers. Especially since most browsers will gladly show expired certificates for you to choose so you have to manually prune your list.
carbonica··on Google+ Is Built Using Tools You Can Use Too
The ex-googler who compares Protocol Buffers to MessagePack is either lying about his understanding of Protocol Buffers, or hasn't looked at MessagePack beyond a simple glance.

First of all, MessagePack throws up a benchmark on their front page bragging about how it can deserialize 3 integers and a string faster than Protocol Buffers. Lovely. Of course, besides being a silly microbenchmark, that's in an array - not a keyed map, and since your protocol is going to evolve, you'll need a keyed map. Since MessagePack serializes map keys with the message, you're going to take a big hit there. Let alone when you have a protocol that grows over several years and has to support a service with over a hundred keys. Eventually, your string keys will become too big of an overhead, and you'll switch to integer keys. Then you have to roll your own translation layer between your space-saving integer keys and useful names for your keys. It's roughly at this point you wish you had picked the "creaky" solution which has done the right thing™ for almost a decade (though of course not open source for that long).

carbonica··on A change thousands want made to Google+ circles
> Google+ doesn't support the broadcast/subscribe model that well, and although it helps posters keep stuff private when they want, it ultimately leaves the decision of how much information is too much or too little about some person up to the sharer, when it should be dictated by the recipient.

Social Networks are not supposed to be the same as a broadcast medium. What is being discussed in this thread is how to bridge the gap, but to suggest that information flow "should be dictated by the recipient" instead of permitting fine-grained sharing misses the entire point of Google+'s approach to sharing.

Allowing the recipient to dictate information flow is excellent for self-promotion, marketing, and news broadcast. It doesn't make sense for how people interact with friends, family, drinking buddies, and colleagues.

carbonica··on Google on today’s massive Google+ spam influx: “We ran out of disk space”
I guess dropping tidbits of inside information for HN karma will never go out of style. NDAs apply after you leave, be careful...
carbonica··on Ex-Wave, Ex-Plus engineer on Google Plus
Personally, I think you're right, but you sort of take a risk when you go for hyperbolic sarcasm. I'm not sure how much of the first paragraphs he believes; much of it, if he did, somewhat discredits him in my eyes as an engineer (let alone as an unbiased source). The takeaway certainly doesn't depend on whether the preceding paragraphs are gospel truth, but one's judgement of the author's credibility, in my opinion, does.
carbonica··on A second is a long time
> This effect is made even worse by most monitoring tools because most take samples every 5 minutes.

Who exactly is using monitoring tools that sample every 5 minutes? Goodness. Maybe I've been spoiled, but I couldn't imagine using such a blunt tool.

carbonica··on Google on today’s massive Google+ spam influx: “We ran out of disk space”
Eh, I was definitely lowballing a lot with "a couple grand." You also need to factor in power and replication.
carbonica··on Google on today’s massive Google+ spam influx: “We ran out of disk space”
Yeah, I was a tad hyperbolic in just referring to the disks. I would expect the costs to be around $20/GB/year when you also factor in power - bigger drives are making a difference, but the other factors always cost more than the disks themselves.

It doesn't change the fact that 60TB is tiny for a company whose every product involves storing enormous quantities of data and serving them at monstrous scale.

carbonica··on Google on today’s massive Google+ spam influx: “We ran out of disk space”
How do you know "the minimum size of a Colossus cluster"? Are you a Google employee leaking information?
carbonica··on Google on today’s massive Google+ spam influx: “We ran out of disk space”
60TB? That's about a couple grand worth of space. Why would that be a lot of space to Google?
carbonica··on Google Singleton Detector
It operates on JVM bytecode, not on java source code.
carbonica··on Light - If you view the Earth from far enough away can you observe its past?
> The Moon is almost exactly 1000x closer to the earth than Mars[3], so I wonder if I used it for my mirror array instead if I could just put a station with 500 mirrors on it to accomplish the same thing.

Plus, since the moon has synchronous rotation, the mirors would at least always be facing us. Assuming all the other massive difficulties got worked out, it'd be a good place for those mirrors.

carbonica··on Why the #AskObama tweet was garbled on screen
> Or maybe I've just been fortunate enough to be in an environment where an occasional goof of this caliber doesn't have any serious consequences.

Primarily, this means you don't have to support internationalization - which is hardly a bad thing, especially if you work on a startup, where worldwide distribution should be the last thing on your mind. When your product is rendered in over 80 scripts, including right-to-left languages, you can't afford to figure encoding will sort itself later.

carbonica··on Chrome: Playing the browser game for keeps
I'm referring more to the fact that he suggests increasing a near-hidden version number every 6 weeks to be potentially criminally anticompetitive.
carbonica··on Chrome: Playing the browser game for keeps
"This smells to me like an anti-trust move"

People really need to learn what anti-trust is and what it isn't. My goodness, the label gets applied to everything.

carbonica··on US claims all .com and .net websites are in its jurisdiction
Again, if you are incapable of seeing the distinction between a general-purpose search engine which indexes the entire web and a website dedicated overwhelmingly to links to copyrighted content (including categorization features, rating systems, subsections devoted entirely to individual television programs), then we must just view the world differently.
← PreviousPage 3 of 4Next →