ShabbyDoo

"more cells, elliptical shape" Yeah, a relatively rigid wing which reacts quickly when deformed by pulling a toggle.

"Are high-performance canopies restricted to certain levels of licensed proficiency?"

Circa 2004 (when my knowledge ends), rules were informal and largely enforced on a case-by-case basis by dropzone owners who didn't want the bad publicity of an inept person splattering himself into the ground. "No Bob, you're not going to jump with that canopy here!"

I do recall the wingsuit manufacturers at least having minimum standards and then asking questions when someone called to purchase one. It also helped that most wingsuits were sold at skydiving conventions where the manufacturer was on-site and could directly assess the buyer's competence. This sort of regulation is quite prevalent in the BASE jumping community where it's even frowned upon for individuals to sell used gear to someone without first assessing competence (or at least the existence of a mentor).

At least as of a few years ago, the rate of death per skydive had remained largely constant for years. What changed drastically was that people died from injuries sustained after a successful canopy deployment. Concurrent with Cypres's adoption was the emergence of high performance canopies. With large, "boring" canopies, it would be hard to kill oneself if already under a good canopy above cornfields. However, small, high-performance canopies provided thrilling swoops toward the ground along with long plane-outs. Half the fun of the skydive now could be had during the landing. With this fun came the risk of a poorly executed turn slamming a person into the ground hard enough to kill him. So, having achieved greater safety in the air, perhaps skydivers adopted add-on risks? The other factor which probably reduced deaths due to low deployments was the widespread adoption of larger aircraft which could quickly take large groups of jumpers to 15K ft. When one's only option for skydiving was to ride for 20+ minutes in a Cesna 182 to 10K, he wanted to maximize his airtime and might deploy somewhat low. However, the extra time in the air isn't so significant if one can exit 5K higher after a mere 10 minute ride. Deploying higher means extra time to cut away a malfunctioning main canopy and deploy a reserve parachute.

tl;dr; Skydivers traded increased safety levels for newly-found risks.

In the US, perhaps the real market might be for motorcyclists where the surrounding culture attaches negative traits to helmet use -- dislike of freedom, lack of bravery, etc.

It is now so common for skydivers to use automatic activation devices (deploys one's reserve canopy if he passes below a certain altitude traveling at least a certain velocity) that many dropzones refuse to allow the use of "rigs" without them installed. I recall being told that, when the Cypres (www.cypres-usa.com) was first introduced, a major marketing feature was that it is not visible to those casually observing the skydiver or his rig. So, a user could avoid the stigma associated with choosing to reduce the risk of death from skydiving. Over time, the culture reversed so that those not using a Cypres were deemed foolhardy.

I should note that I stopped jumping a few years ago, so my information is somewhat out-of-date.

I live in a relatively wealthy school district in Ohio where the schools generally are well-regarded. Last year, I went to the schools "parents math night". I had expected to be exasperated by, you know, those idiots who will be teaching my children (my older son is entering first grade this fall). However, I ended up impressed with the teachers and disappointed by the parents in attendance. The school had chosen the "Connected Math" curricula (connectedmath.msu.edu) for its middle school math program, and some parents were outraged that Little Johnny wasn't learning long division (or whatever). Their arguments could be summed up as, "Well, that's not the way we learned it." I angered the other parents by playing a bit dumb and asking for a short description of the curriculum. Then, I stated loudly that it "sounded delightful." The teachers barely contain their smiles. I ordered a couple of the books in used form from Amazon to see for myself, and I think I would have found the exercises fun as a middle schooler. I really wanted to ask the parents doing the complaining, "Don't you love your children? You pay more for a house just because it's in this school district, and then you lobby to dumb-down the math curriculum?"

We have a standard IDE set-up for our team and require more junior devs to use it -- a least when starting out. It makes pairing with them easier.

I see your argument. Unless an employment contract stipulates specific work deliverables (even showing up, etc.) , it really is just a non-compete which might require the employee to pay the employer in the case of early termination by the employee.

This is especially true for software companies where it is difficult to prove what the market value for a company ought to be within an order of magnitude (if not more!). Did Facebook overpay for Instagram?

While I am reluctant to suggest that the courts require specific performance (continuing to work) for employment contracts, awarding damages to the other party does not seem terrible. Does California law allow for damages to be awarded in such cases?

Most web email providers stamp outbound messages with the source IP address used when sending the message as a way of discouraging illegal activity.

Can't Burner collect money from users for temporary numbers and then delete all records of number assignments when the numbers are "burned" by users? Twilio's recordkeeping might be more persistent though. Are there laws which effectively require Burner to maintain records? If not, Burner would keep a persistent record that User123 paid $2 for a number but could conveniently delete other identifying information.

I'd like to see a graph of average spreads between 2007 and now.

>Google warns that “there’s still a long road ahead.” The cars >still need to learn how to handle snow-covered roads, ...

What if the car was smart enough to identify situations where it was unable to auto-navigate safely? It could pull-over and ask the human to take-over for awhile. I would rather see a mostly capable self-driving car come to market sooner than wait around for near perfection.

On this note, it seems that the trucking industry could adopt a self-driving truck even if it was only good enough to travel along long stretches of highway. The first/last miles would be manually driven just like today. Could the trucking companies use remote navigation (much like military drones) to handle first/last mile and perhaps problem spots along the way?

What's the minimum viable product?

If the Google Car had gotten into 100 accidents over 300 miles, we easily could claim it much worse than most drivers. However, as with most rare events...

Agree. Did humans step-in when they noted conditions which they believed the car unable to handle and perhaps prevent the majority of likely accidents?

I see two interesting aspects of the issue: (1) the astounding value in affecting perceptions of one's popularity and (2) the implementation schemes for #1. Whether Romney/Obama/whomever took any particular action is only a concrete instance upon which to consider these two ideas.

Consider getting a low-level job in a shop which is doing Java/Android work (presuming this is your interest). You might at first do 2nd-level customer support, testing, or other non-programming things. However, especially if you position yourself to employers as someone taking on such a role so you can learn on-the-job, you'll be able to finagle your way into more challenging work. Maybe there's some manual testing which can be automated. Now, you've shown you can write code. Before long, you'll find yourself in a development role, and you'll have been paid while learning. By letting potential employers know of your intentions, you can induce selection bias in your favor -- the employers who want testers who will shut-up and do as told will throw your resume away, but those who want motivated people who can take on greater responsibilities will love your goals.

The astroturfing seems so obvious (rapid ramp-up) that I had to wonder if Romney's campaign was actually behind it. Why not pay a little more to do a good job -- at a minimum, ramp-up over time. Two possibilities (among many, I suppose) are that an Obama supporter paid for the fake followers so that a story could be planted about Romney's deceptive tactics. Another possibility is that a fake user farm set-up its fakes to follow a few random, high-profile users to make the fakes seem more legit. If this was the case, why choose Romney but not Obama? Based solely on the data, we can't know what actually happened, but it doesn't seem like the article's conclusion was drawn after exploring alternate possibilities and finding reasons to dismiss them.

100K in a day seems like sloppy astroturfing. Why not add on smoothly over the course of weeks? At only $800/thousand followers, perhaps $80K was spent by an Obama supporter to create a news story and discredit Romney? Perhaps Romney was the inadvertent target of these astroturfing firms' efforts to make their fake users seem more legit? I checked out Obama's follower stats, and he experienced no such bump. Why pick sides if your goal is simply to sell fake followers random folks?

[I'm not necessarily a fan of either major party candidate, so please don't consider this a political discussion.]

I rarely tweet and have few followers. Every month or so, I get a new, usually female follower (whom I always flag) which I always presume to be the first step of some Russian-visitor-ish sort of scam. Perhaps I simply am being used to make these fake accounts look more legit?

Yes. If I don't trust the charity to decide how best to spend $50, why should I be willing to give them the time of a person who is paid that much? It's not like the charity itself can't use Exec to purchase the same services.

I did some work for an ecommerce consultancy a couple of years ago where we were literally looking a a customer's request for "ability for customers to upload photos of themselves wearing our clothing" and saying, "Why isn't there a white box image service?" Now there is.

To those who have not dealt with user-generated media content, it is easy to underestimate the amount of effort involved in building a secure, monitorable system. Building an image grinding pipeline is non-trivial, even though all the open source tools are of high quality. Once users start uploading photos, marketing will realize that they need to be cropped, resized, and enhanced. As soon as your users upload the first pieces of questionable content, you will have to build an image moderation app. Then, you have to build a feature where the user sees his own photos on the site privately when logged in because users hate the lack of immediacy. Then, your marketing people will ask for video. Totally separate pipeline. Then, someone will upload custom-generated 40Kx40k single-color jpgs to see if your image crunching pipeline allows denial of service attacks. Once you get your first upload of child pornography, your legal team will become much more interested and ask how you are ensuring the destruction of these images from all your servers. Now, you'll learn that a reasonable portion of your bandwidth is being used to serve up deep-linked content for sketchy sites. Oh, and other stuff will happen too. Then, you'll see the value of Chute.

Did you contest the charge with your credit card provider? I can't imagine that Amazon would contest such a claim.

Agree. Whenever I hear the words better/worse used without an obvious axis of measurement, I am suspicious of the speaker.

This is an apples-to-oranges comparison. The author admits this, and I'm not finding fault with the analysis. I do worry that those not familiar with various ser/deser schemes for Java will draw the wrong conclusions. My take-away is that you should not blindly use Java serialization in cases where performance matters and you don't need the broad spectrum of use cases it covers.

The main reason that Java serialization is so slow is that it is designed to persist self-describing, arbitrary object graphs.

Self-describing = Full package+class names and versions. Any object graph containing Serializable classes available via the classloaders of both the sender and receiver.

Arbitrary object graphs = The identity of objects is preserved. Let's say that 17 SalesOrder objects point to the same Customer object. salesOrderA.getCustomer() == salesOrderB.getCustomer() (identity in Java for those who haven't had the pleasure) is maintained.

Oh, and JDK dynamic proxies are handled as a special case. And some other stuff, too. I had the pleasure a few months ago of copying the JDK's serialization code into my own package structure and adding logging so we could debug deser issues in an OSGi environment. From what I can tell, the impedence mismatch between OSGi classloading hierarchies and Java Serialization makes this combination a non-starter.

The very-performant sample case provided in the blog posting works only for a list of flat objects. As soon as you try to preserve identity, you must implement a two-pass system where you deser placeholders and then resolve them later. I'm not suggesting that it isn't useful, but it probably constrains your serializable objects to such a degree that you'll be unhappy in other areas of your system.

Kyro (as another commenter noted) seems like the best combination of small size/speed/general-ness that I've encountered thus far. There seem to be a reasonable number of pre-built Kyro serializers for common libraries such as Joda time.

For a project I worked on recently where size was more important than speed, I took advantage of the similarity of the objects I was serializing and wrote a custom serializer which created dictionary look-ups on the fly. Repeated strings were turned into small integers, etc. I wish I didn't have to write it.

In a nutshell, if you constrain the scope of the problem you are attempting to solve, the maximum possible performance of the system across covered use-cases goes up. Java Serialization's performance is relatively poor compared to less generalized schemes -- as expected.

The phrase 'better for consumers' doesn't seem to have any testable meaning. What hypothetical experiment, if performed, would determine whether a proposed rule change would be "better for consumers"?

Let's pretend that we can clone the universe and run an A/B test. Universe A receives the new rule treatment and Universe B does not. In both universes, we inject happiness meters into every living person's head. These meters also inject clones of themselves into fetuses for the next 100 years. So, we can compute the average happiness levels of all people on earth over the next century. Is a net positive change in happiness levels the definition of "better for consumers"? Must we also consider changes in the distribution of happiness among individuals? Is happiness even the right metric to consider? When the word 'consumers' is used, does it imply that everyone's happiness should be considered?

So, back to some reality. It's certainly possible that an Apple-like walled-garden approach might, for some particular technology/situation/problem/whatever set-off a chain of economic events which has a positive net effect on overall human happiness. However, I have no idea if I'm making a meaningful argument to those using the "better for consumers" phrase as no one seems too interested in attaching a strong definition to the term.

W.r.t. anti-trust litigation, the phrase is oft thrown about. What has bothered me is that the "obvious" answer is provided by a static, short-term analysis of current economic conditions. In the short-term, it's better for consumers to regulate the price of electricity. After all, it costs "too much", you know. However, the disincentive to build new power generation facilities likely leads to an outcome very negative for consumers -- shortages, rationing, etc.

tl;dr; Stop using the phrase 'better for consumers' unless you bother to attach to it a testable definition.

I wish Plex on the Roku just turned my TV/Receiver/etc into a dumb yet high-quality media player. I don't want to use the awful Roku remote with a 10 foot interface -- I'm happy to use my laptop or phone as the control module and have the TV be little more than a digital "projection screen".

I get the argument from a purist's perspective, but it's a minor problem when compared to the set of impediments "normal" users face if they decide to install a Linux distro.

No -- my stereotype is that sons often are treated with more reverence than daughters. This is in opposition to Western stereotypes of "Daddy's little princess" and other celebrations of daughters.

A reputation economy for tenants/landlords would be worth a lot to both parties. I just bought a rental property in Chicago, and I'm terrified of a bad tenant because the state/local laws are so pro-tenant. It could take me months to get rid of someone, and I risk penalties if I don't do "essential" things like provide a timely statement of interest earned on a tenant's security deposit. I'd gladly accept $100 less/month for someone who has proven himself reliable in the past. Conversely, I'd like to develop a verifiable reputation as someone who hasn't screwed over any tenants.

A friend who is an attorney lives in Texas (a very pro-landlord state) and has had to threaten legal action against his past two landlords to get back security deposits. Even as someone who is more capable than most at fighting back, he would love to avoid the trouble by picking from landlords who haven't screwed over others.

"I was a pampered little brat who was waited on hand and foot."

Are you a male? Were your sisters/cousins/etc. treated similarly? Just confirming my (presuming) Indian stereotypes :)