ShabbyDoo

A few years ago, I bought a LaserJet 4200 used via EBay for < $200. Huge, probably sucks down a bit too much power in standby mode, and sort-of noisy. However, IT JUST WORKS. Because this series still is in offices all over (and a source of toner revenue for HP?), driver updates still seem frequent.

For those considering the printer mentioned in the article but need wireless connectivity, consider a cheapie dd-wrt router configured as a client. I use this set-up for my kids' computer in our kitchen, and it's been no problem.

I understand the cultural allure of a single event -- pizza, beer afterward, etc. If a company wants to allow all employees to participate, why not have a demo day every couple of months where teams could present stuff they worked on during that period. People could self-organize partially by their own time availability. Enforcing the time-boxing rules would be hard, but I don't think this is an integral aspect of what makes hackathons a cultural success.

At one company where I worked, the advertised benefit to employees was, "You get to choose what you build for the company! You can INNOVATE!" Not these words, but this was the message. If you want to take initiative to improve things around here, you have to do so on your own time.

Interestingly, I ended-up spending much of the hackathon day an evening fixing a production problem which was of obvious urgency. Some mild shaming from management ensued when I had nothing to present -- a clear message.

In total contrast, I once worked for a staid, rural insurance company where the majority of employees were middle-aged with children. "Flexible" hours meant that one could arrive at work anytime between 6AM and 9AM. Clearly, the flexibility was extended to those who wanted to arrive early so that they could be home by the time their children were getting off the school bus. Twenty-somethings who wanted to venture into the city to see a show (easily an hour drive) still ostensibly had to arrive at work just a few hours after getting home. These family-friendly policies sent a distinct message to the young/childless -- the company regards your happiness as secondary to that of those with children.

Edit: I regard such policy biases as companies' choices to make, but I doubt that many consider sufficiently the messages these events/policies/etc send to particular employee segments.

I'm glad this article implies that employees need not stay up all night, as Facebook Hackathons seem to promote. Such events are heavily biased toward the participation of the young and childless who have few mandatory scheduled responsibilities outside the workplace such as childcare. Of course, participation ostensibly is voluntary, but it's clear that there are career benefits such as making oneself known to upper management by presenting one's hack. Companies who schedule these sorts of events likely send a message unknowingly -- If you can't pull and all-nighter, you ought to consider working somewhere else.

An interesting lack of selection bias which I had never considered. I presume that those choosing to be career officers opt-in or at least are chosen based on desire/talent/etc.?

Gladwell notes how military culture differs from civilian culture in Korea. Surely, differences of similar magnitudes exist in the US.

I wonder how the US military has changed culturally over the past few decades. As an outsider, I have presumed that the purpose of degrading training programs, continual demonstrations of the power of officers (parade reviews, etc.), and required social proof of the officers' power and authority (stand and salute upon the officer entering a room) all were designed to avoid argument, hesitation, or doubt when an officer tells his soldiers in battle to charge forth toward an enemy firing upon them. To win an infantry battle, such obedience probably is necessary. Imagine if only a small fraction of the soldiers were willing to advance -- they would likely be much worse off than if all the soldiers advanced at once.

However, this nearly blind obedience surely comes at a price. Can a junior guy tell an officer that he is unsure of a drone target's validity? It strikes me that a modern military operation is much more like a large scale construction project or perhaps software development than traditional trench warfare. In these civilian endeavors, the price of hesitation to consider alternative opinions likely is low compared to the cost of being wrong. Has the military changed?

Imagine if Snowden had deployed a bunch of redundant crawlers of various news sites likely to cover him and quote him directly when he speaks. Their activity would be nearly undetectable in the traffic of the NYTimes, CNN, etc. He could come up with a bunch of seemingly innocent control phrases which he would use in soundbite quotes during press conferences, etc. He would say a phrase, the media would quote him, and the crawlers would identify this "control transmission" from Snowden and take action. Some phrases would be dead man's switches in that one of them would have to be observed every couple of weeks or documents would be released via mechanisms difficult to trace back to the server (Tor?). Other phrases would trigger incremental leaks to allow proof that he is still powerful and in control. "Tomorrow, I'm going to release a _mightily spectacular revelation_!"

Maybe there should be two levels of dead man's switching -- incremental leaks if a phrase isn't reported in, say, two weeks and a major release if no phrase reported in three months. This way, he figuratively would have multiple units of currency with which to bargain. Let's say he was thrown in jail. If he only could threaten to release a single, big bundle of secrets via a dead man's switch, all his bargaining power would disappear should a government call his bluff and keep in jail until after the switch fired. However, if he could threaten incremental releases and show that, when given full freedom, the releases stop, he would have power for quite some time.

How would one acquire the use of, say, 50 servers in various datacenters owned by various providers without leaving traces or implicit fingerprints (multi-year prepayment being the big one I am thinking of)?

Let's say he has N friends who have agreed to help. He doesn't want to allow any single individual to have the power to act alone and release information. And, he doesn't want to require all N of them to act together. What if one dies, is arrested, etc.? Could he encrypt the documents with, say, all combinations of three friends' own public keys such that a quorum of three friends would have to cooperate in the release? He would end up publishing N choose 3 bundles.

A couple weekends ago, I drove with my two young children to a restaurant 20 minutes from our house only to learn that it, a mom-n-pop operation, was closed for a couple of weeks. I suppose I could have called ahead to confirm my presumption that they were open, but I didn't learn about this "planned outage" until I saw the 8.5 x 11 sign on the door. Apparently, they closed for a family vacation.

It would be cool to have some sort of proactive notification when the many assumptions I make about the world around me become incorrect. Take a person's work commute as an example. Presumptions include road availability, public transit timeliness, lack of major events causing traffic delays, sufficient fuel in one's vehicle, the line at Starbucks being reasonable, no horrible weather forecast which would require bringing an umbrella along, etc. Imagine an app which invoked a bunch of figurative assert statements on the presumptions we make daily in life.

I'm posting this here simply because it entered my stream of consciousness when I saw the headline. At first, I thought that "real world" status is un-related to the problem StatusPage solves. However, sending notifications to customers upon an outage is a step closer to my handwaving above. Take as an example the availability of employee VPN access. How many times have you gone home early presuming that you would be able to work later in the evening and later realized that you can't kick off that build, etc. While learning that the VPN server is down doesn't help you do your work, you probably are more likely to find that your co-worker is still at the office at 7:00 PM than three hours later. Maybe you would have been lucky enough to learn about the outage before you left the office.

One problem I see is that being alerted about every presumption violation would be annoying. I'd like to infer intent somehow. "I'm going to work today" implies a lot about what's important for me to know. Being informed about "outages" certainly would be an incentive to put a zillion things on my calendar. What if you were planning to see a band at a bar next Thursday, and adding the event to your calendar somehow allowed proactive notification in case of cancellation?

We take it as a given that life will surprise us with disappointments, but this condition might only be the result of historically high costs of sending and receiving notifications. Does anyone remember phone trees? Watching the TV to learn of school closings? Making detailed logistical plans about how to meet-up with friends (along with various contingencies) in the days before mobile phones? Technology replaced these annoyances, and there's no reason to believe all progress has been made. I want the absence of notifications to imply that life will abide by my presumptions.

Yes. JVMs are slow for short-lived program executions compared to native code. By default, Hotspot does not compile a method until it is executed 10K times. Most libraries in the Java ecosystem are written to optimize performance over a long execution lifespan at the possible expense of startup time. The Apache HTTP libraries allow pooling of threads, HTTP keep-alive as an implementation detail, buffer re-use, etc. So, evaluating these libraries by starting up a new JVM for each run inherently tests them under a circumstance for which they were not designed/optimized. Gradle (a build tool) sort of solves the problem of short-run invocations by forwarding command line requests to a long-running Java background daemon.

I would like to know by what mechanism a powered parachute flight over a feed lot might result in humans consuming unsafe food. Presuming mad cow disease inducing "pellets" even could be produced, the pilot could have dumped some over the feed lot. However, this would be an absurdly complex mechanism of distribution and one with a high likelihood of future arrest. Does the feed lot allow inbound grain to be shipped in trucks without sealed covers? I could keep going, of course.

What's bothersome to me is that those using their coziness with law enforcement to dominate their "enemies" don't even feel the need to contrive plausible justifications for their bullying. This is how far America has come.

Let's say that someone sat for the JEE and ranked quite poorly. Who would know this person's actual ranking? Presumably, given the existence of the secret code for checking one's score, not all rankings are made public. Does the test taker himself even know below a certain level of mediocrity?

I ask these questions because they seem relevant in understanding the motivations of many taking the exam. Does it reflect poorly on an Indian family if a child (especially a son, I presume) does not sit for the JEE? If so, it might be socially rational for an academically mediocre child to go through the motions of test preparation. This incentive is heightened if no one would have to find out just how badly the child did on the test. Is it a source of pride or honor for a family that a child is preparing for the JEE?

If these social motivations are present, perhaps a talented student's actual chances at getting into IIT are much higher than the overall numbers suggest. I argue that admissions percentages at top-tier US schools are similarly affected. Little Sally, even after spending hours with a private SAT tutor, got only a mediocre score. Yet, her high-achieving parents demanded that she apply to Harvard, Princeton, etc. This way, they could signal to others that they deemed their child of high calibre and (the real incentive here) themselves parents so good that they want only the best for their child. Because the acceptance rates are so incredibly low, it will not be a mark against either Sally or her parents when she is rejected from all her "first choices" and ends up at a mediocre university somewhere -- her "safety school". Sally's parents might also be rational to avoid mentioning that she applied to and was rejected from a couple schools only slightly more selective than BeerCanU as these rejections would be seen by others as a legitimate measurement of their daughter's mediocrity.

A problem I see with many of these sorts of questions is that they often require the candidate to have some supposedly common knowledge which is not required for the job itself. Cryptic word games surely are much more difficult for a non-native speaker of the language in use. Questions related to facts about cities probably require local geographic knowledge. Surely the evacuation plan for SF must consider the capacity of various bridges? Someone who has lived in northern CA for most of his life would have a much easier time thinking through the logistics of moving people off a peninsula. And, of course, there's the Monopoly question (which I had to Google).

I like estimation questions in general for many of the reasons other commenters have cited. However, I wish those using them would consider the knowledge implicitly required of a candidate.

Interesting. I had presumed that the pilot would be violating some FAA rules, as I might be myself. If the pilot had a commercial rating, I presume the more costly maintenance requirements required of commercial carriers would apply here?

I'm interested in finding planes leaving from Cleveland to Chicago on Sunday/Monday and planes leaving from Chicago to Cleveland on Thursday or Friday. I'm flexible on airports. Furthermore, I don't even mind a seat on the floor (with a harness/seatbelt) -- I used to skydive a lot, so this seems like a reasonable and safe option. I know this isn't the purpose of your app, but I'm hoping there are options like Blackjet out there except they don't charge 5x the price of a Southwest flight. Ideally, I'd be happy to pay a private pilot in literal cash at the last minute for a seat on his plane. I'd win by getting to/from Chicago w/o fighting my way thru a major airport. They would win with up to $200 in their pockets for a flight they would be making anyway. Flying right seat would make it all the better :)

Given that flight crews have government-mandated work hour limits and most crew members are paid hourly, it would take a lot of WiFi and drink sales to make-up even the labor costs of an artificially lengthened flight.

Via eBay, I bought some 50's thru 70's era Chicago SunTimes vintage 8x10 file prints. Apparently, someone acquired their physical archives? The skill of photo journalists in those days was astounding. There were beautifully composed photos of the 1968 riots -- while people were rioting! I'm not suggesting that today's PJ's aren't as skilled, but it's certainly easier with a DSLR.

My name might be "Mud" a a previous client because I recommended MongoDB and implemented a first-rev persistence interface based on my presumption that this issue would be fixed in the near-term:

https://jira.mongodb.org/browse/SERVER-3253

To be fair, I educated the client on the alternatives should 10Gen not improve the aggregation framework so that it could do arbitrary transforms without failure should the output size exceed 16MB. [Use other MongoDB mechanisms, custom compilation with higher per-document size cap, etc.] What annoys me is that 10Gen did not mention this incredibly important limitation when they were touting the planned features of 2.2. My client would not have minded had it simply been the case that queries with large result set sizes were somewhat slow. What this client could not tolerate were failures to deliver any results at all. In retrospect, I wish that I had pushed harder for a solution based on the Hadoop stack. While it seems to have its own demons, at least there is an ecosystem dedicated to fixing the most blatant of limitations.

Whenever I read an article drawing conclusions about the world based on interviews with those attending a school reunion, I consider the inherent selection bias. If I'm an unemployed 60-something, I might consider a class reunion to be a great networking opportunity. And, of course, my time has a low opportunity cost. Conversely, am I willing to fly across the country for the weekend if I'm working a lot and need to be in the office on Monday morning? Also, those whose lives at age 60 aren't all that fulfilling professionally or otherwise might be more likely to attend a reunion where the could be reminded of better times. [I say this as someone excited to attend my 20 year high school reunion this summer.] A survey of Stanford engineering graduates done by random selection would be much more interesting.

Rather than using a hex editor, one might consider a tool like Soot from McGill's Sable group. It includes a DSL for dealing with bytecode:

https://github.com/Sable/soot

I had looked at Soot some time ago because of Dava, the Java decompiler included with the project.

Also, for programatic manipulation of bytecode, ASM is a great tool:

http://asm.ow2.org/

It includes a Java source generator which, given a .class file as input, produces a Java source class which will regenerate the input via the ASM builder APIs. Obviously, the benefit is that you get a starting point for building various re-writers, analyzes, etc.

Southwest apparently uses Row44 for in-flight WiFi. When it even works, it's absurdly slow. So slow that I'd gladly take a dial-up modem experience instead. On top of this, the company insists on slapping up a banner on the top of every webpage. If I click the "go away" button, I have to wait 30 seconds for the page to re-load. [Yes, I'm sure GreaseMonkey, etc. could rescue me] I fly on Southwest so often that I receive free WiFi, but I rarely use it. Even logging in is a multi-minute ordeal. I can't see how Row44 can generate enough revenue to justify its existence given the abysmal user experience.

"but you can much more easily upgrade the set-top box with new features every one or two years whereas there's very little reason to get a new TV less than once every 5-10 years."

This is the same problem I have with car audio (source,at least), built-in navigation systems, etc. The rate of improvement in cars and screens compared to the cost of upgrading is quite small compared to the value proposition provided by newer navigation/audio/etc. offerings every couple years. I want to buy a just a screen with a single digital video input -- no speakers even. And, I want my car to have a standard hole in the dash for an audio source box coupled with some power/speaker/GPS antenna/etc. inputs. Of course, my desires directly oppose the pipe dreams of industry executives who want to turn their products into Trojan horses of continuing revenue streams for years to come. I recall laughing as the Honda salesman tried halfheartedly to sell us a DVD/Nav system for our minivan. Would I rather pay $2K for a car DVD system or buy two, $500 iPads? Hmmm....

Edit: Another thing that bothers me about these artificial couplings is that the hardware-centric manufacturers rarely provide updates for more than a few months after a product's launch. How long until the browser in a "smart" TV can no longer render most sites? If Roku doesn't support its $50 box after a year or two, I can just buy another brand or a newer model. However, I'm pretty mad if I have to upgrade my $1500 TV.

I was told by a taxi driver in Chicago that bureaucrats love complaints from taxi patrons because they can extract bribes from the drivers to resolve them.

Were they discriminating against the individual or the likelihood that they would not be able to find a paying passenger in the immediate vicinity of the endpoint? Let's say they believed they would have to drive back to Manhattan for their next customer -- their effective earnings would drop by more than half (expensive backhaul) compared to a scenario where they could pick-up a paying passenger just a few blocks away.

I don't have a huge problem with your suggestion although I wonder if several competing Uber-esque services would solve these sorts of problems much more efficiently. The idea of "fair" prices is a difficult one because the definition of the idea varies so widely. I happen to like Uber's variable pricing strategy, but many people don't regard it as fair.

I work in Chicago and have a condo in the Near West area (basically Ashland and Madison). Cabbies who are already in the area love seeing me hail them for a downtown trip because they earn a few dollars driving where they would have gone anyway without a paying customer. If the industry was un-regulated, I might consider competing with Uber by offering people with less money than time a way to get an inexpensive cab ride from one of these "backhaul" areas. Imagine an Uber-esque service where a senior citizen who wanted to go to the grocery store sometime in the next hour but only wanted to pay $3 for her trip could place a bid for a pick-up. Cabbies who dropped off a premium customer nearby could at least pay for their fuel to reposition themselves for the next premium customer. Everyone wins. It's this sort of innovation which is stifled by laws deemed to be in the best interest of consumers.

You probably are correct. Initial regulations likely were earnestly enacted for the public good. I believe NYC implemented the medallion system (the first in the US?) because it was concerned that taxis would become so numerous that they would clog-up the streets. Over time, politicians figured out how to exploit their increasing power for personal gain. There's nothing unique to taxi service in this pattern.

"This is why we have regulations that attempt to provide some of the symmetry required for fair exchange."

In the case of Ebay, Uber, Airbnb, etc., technology can replace regulation as a means of providing power to the previously disadvantaged party. And, technological solutions almost always are more efficient than the regulation they replace.

Your driver sounds quite clever and profit optimizing. I had thought a couple of weeks ago about whether Uber should somehow allow drivers to specify destinations for more efficient routing. As a customer, what do I care of the guy 1/4 mile away has to drop someone off before picking me up, especially if the next alternative is a mile away? This was not an original thought on my part as I once worked on scheduling software for a large truck fleet, and we had implemented a similar feature.

One thing of many in Uber's favor is that its customers are of relatively high socio-economic status, and the company treats them well. When they hear that the company is being sued by the taxi industry, they take note. It's personal. Taxi service has been bad for years, and everyone's been a victim of the absurd regulations which have lead to this sorry state. Elected officials have been accustomed to taking advantage of regulated industries through corrupt actions (anything related to building or zoning permits as an example) without significant blow-back via news coverage or loss of reputation. This time, it's different. Look at how Washington DC's attempt to draft anti-Uber legislation lead to a national response via phone calls and emails. The proposed legislation was dropped. I suspect Uber will eventually prevail in most of its markets, but the politicians will take awhile to realize that the public is taking interest in seemingly esoteric taxi regulations -- the sorts of laws which have been used for years to transfer money from the pockets of the electorate into the hands of those close to power. I feel like I'm supporting democracy and freedom when I hit the "request taxi" button in the morning -- it's an action more powerful than my vote.