bartonfink

I bought my wife's ring from Blue Nile - www.bluenile.com - and would absolutely recommend them. They made it very easy to get what I wanted, and did a good job of educating me on what options were available to me. It's entirely an online operation, so it's very easy to browse their inventory over a long period of time and research any ?'s you might have somewhere else (if you don't trust their shopping guides). It's the antithesis of what I experienced at a couple of brick and mortar stores, where salesmen were pushy, unwilling to answer ?'s about their product, and at one point actually told me "if you have to ask, you can't afford it" when I asked about a ring of lesser quality than what I wound up buying. I don't plan on making it a habit to buy jewelry that expensive again, but if I did, I'd use Blue Nile in a heartbeat.

Also, they did probably the coolest thing I've seen with shipping. I had the ring shipped to my office address, and spent much of its arrival date refreshing the tracking page. Almost immediately after I saw that our shipping department had "signed" for it, my phone rang. Someone from Blue Nile called to say that someone other than me (the name of someone in our mail department) had signed for the package, and they wanted to make sure it was in the right place. I didn't pay anything extra for this service, but it was a really nice touch and I've told that story to everyone I've recommended the company to.

MapQuest offers a geocoding service which doesn't tie you to MapQuest's maps the way Google does. It's free up to some usage amount (I think it's 5k requests a day). You can see more here:

http://www.mapquestapi.com/geocoding/

Where are you that a PhD is a three year programme of study? It's absolutely not the case in the United States that a PhD is expected to be completed in three years, and a hard deadline of four years would be exceptionally tight.

The Open Boat, by Stephen Crane. Ecclesiastes. Siddhartha, by Herman Hesse. All would be considered "classics", all have much to recommend them, and none is so long that you'd lose interest.

In my opinion, no. In this day and age, there are far better ways of obtaining preliminary knowledge on a subject besides asking another person to fill you in. There's obviously Google and Wikipedia as starting points, and there are any number of paths those might lead to which would help you form a specific ? - e.g. "what would you have to do to make true a la carte channel options from a cable provider profitable?" That ? shows me that the person asking has done at least preliminary research and, if I have any thoughts, I don't have to start from square one.

Asking the community to spend its time doing preliminary research for you is, to put it bluntly, rude.

Interesting. You're ahead of US Netflix on that - just checked and I do NOT have access to Breaking Bad Season 5.

We looked very heavily at Disqus for comments on a product I worked on that was released recently, and decided against it in favor of rolling our own commenting system. Ultimately, what it came down to was that we could build an equivalent system that met our needs for less than they wanted in a contract. I'm sure that there are things Disqus does that would show improvements over our homegrown commenting system, but ultimately it just wasn't worth it. A basic comment system just isn't that hard to write, and it's hard to imagine a scenario when added functionality in your product's comment system would justify paying for an external system but that functionality isn't important enough to justify building it yourself.

Having a suite of unit tests I can rely on. It helps me avoid the paralysis that occurs when I wonder "what else might have been touched by this code I changed?" I can give that ? a couple minutes thought, write whatever changes I need, and then run the suite to see whether I missed anything. That factor more than almost anything else helps me write code quickly.

Mapquest.com

Primarily in Denver, CO and Lancaster, PA, but we're a relatively distributed company so go ahead and ask if you live somewhere else. As an AOL subsidiary, we've got employees who work out of AOL offices around the world so it wouldn't be unheard of to work from e.g. Seattle or New York if you decide Denver's just got too much sunshine. I've also been told we will sponsor an H1B.

Over the past year, MapQuest has brought on an almost entirely new management team to shake things up and change the direction of the company from answering the ? "How do I get there?" to answering the more interesting ? "Where do I want to go?" We've released two new products in the last year (Discover and Local, formerly known as Vibe) and are aggressively working on creating an engineering-driven culture that actually builds things people might want to use instead of polishing products that were cool in 2004. We've had a banner year, and are looking to hire engineers as well as product managers to continue this upward trend.

Among the technical roles we're looking to fill are several Ruby developers at varying levels of expertise to help develop new products as well as one Java developer to work with our data team and revamp various backend services (i.e. search, geocoding and a data ingestion service that populates our location databases). Ever since Valve leaked their company handbook, the phrase "T shaped" has been been thrown around a lot by upper management, so you'll need to be able to do full-stack development. We do a lot of pair programming right now, so if you've got experience doing that, it'd be a big plus. We're particularly interested in someone who's got very deep Rails experience to serve as a sort of tech lead, but we have many positions open for Rails people so if you'd be interested at all drop a line.

Product is looking to bring on people with some combination of travel industry knowledge and mobile applications to help get the company past maps and directions. Most of our product people have development experience, so you'll be expected to be able to "talk shop" with the engineers you work with, but your day- to-day responsibilities aren't going to be producing code as much as producing ideas, pitching them to the rest of the company and helping the engineers and designers see them through to completion.

I can definitively say that MapQuest is among the best places I've ever worked. The office space is fantastic (Downtown Denver right off the 16th Street Mall) and the new management team is very serious about doing things right as we try to pivot the company away from directions and into product development. The brand is practically a household name, the product team is happy to listen to engineers for feedback or ideas, and we've got the interesting position right now of being able to work like a startup except with a large pre-build audience (I've been told that we had 40M UV's on a brand new product this past Thursday, for example). You can read the company postings here (http://company.mapquest.com/careers/) but I'd really enjoy being able to answer ?'s directly. My e-mail is in my profile and please feel free to contact me personally if you have further ?'s, want to chat, or anything else. Happy holidays, and hope to hear from some of you soon!

I don't know the scope of Colin's cron problems, but filtering out mail in this case seems like a less preferable solution than simplifying the way cron does email to begin with. Filtering e-mail in this case is like putting your fingers in your ears when you walk by a loud stereo you've left on instead of just turning it off. As a developer, cron doesn't give me a lot of control over where e-mail is sent unless I jump through hoops to give myself that control. Ignoring the complexity of the reports, and the fact that filters like that don't easily scale across different recipients, the fact is that cron's e-mail capability is extremely coarse-grained. For every run of every job, the entire output is e-mailed out to the address specified in the crontab. This gets very unwieldy.

I've had to write cron table entries with a blank mail recipient in the crontab itself, and handle the specific mailing cases in the job script. This sucks. I'm not sure I think the solution is something that reads cron-mail and automatically generates filters, but I would very much like it if cron supported e-mail lists based on return codes or something else like that.

I've had similar results by informing recruiters that I'm planning on moving abroad in 2-3 years, and that I really don't want to switch jobs unless it directly furthers that goal. If I get even one recruiter who has contacts in New Zealand or Australia, that's a big win - otherwise, it politely tells them that I'm not on the U.S. job market.

Richard Feynman.

Just to be clear, token - does the supervisor have anything up his sleeve with "flawless?" The story that it took years, for example, to produce a "flawless" binary search isn't as stupefying when you consider that many of the flaws dealt with things like integer overflow and floating point precision. If this guy wants this to work on a 4-bit microcontroller, for example, getting to 100 is going to be significantly trickier than what many would consider a straightforward problem.

I certainly wouldn't pay $150 for that, and was asking myself the same ?. It's not going to appreciably increase the value of my home, so an argument that it's home improvement doesn't hold water for me. It doesn't seem to have that much hardware, which would help me understand the price point even if I still wouldn't pay for it.

It simply doesn't solve a problem I have, and at $150, I'm not likely to give it a shot to see if locking my doors manually is really that much of an inconvenience (I doubt it is). I might pay $50, but even then that's a stretch because, as I said, it doesn't solve a problem I've got.

Larry Page does have a bachelor's degree. He dropped out of a Ph. D program, not college.

For what it's worth, I now describe myself as a full-stack developer even though I cannot make something pretty. I see that as a designer's job, and while a front-end specialist should be able to whip something up in photoshop, I think that it's perfectly reasonable to say that your part of the work ends with something being functional.

It's almost essential to be familiar with jQuery for in-browser Javascript development these days (I'm not sure of its market penetration but I'd be shocked if fewer than 33% of websites use it). There's lots of cool new libraries on the front end with various pros and cons, but jQuery is a really good start.

I've read something similar by Michael Abrash.

Based on a press-release for the capabilities of a new graphics card that would have blown his project out of the water, an engineer spent a long time thinking about what they must be doing to get that sort of performance. He figured they were using a FIFO somewhere in the circuitry, and had enough space to implement a 1-deep FIFO for a certain operation, figuring that it wasn't enough. Turns out, his improved chip beat the benchmarks that were in the original press release because they had implemented a FIFO for a different operation that had less impact on performance.

That's actually not a bad idea - Google sponsored hikes.

You should put your e-mail address in your profile on HN. The e-mail field isn't publicly visible.

I would absolutely hire an agent. I have a day job, but freelance on the side. I'd estimate I have about 20 hours a week of available time that I can use for freelance coding.

I would LOVE to hire someone to help me keep those hours full. I have a day job that pays me well, so I can afford to be generous if it means I never have to wonder "Barton, are you just being strung along by this lead?"

There's a guy on here - mindcrime - who's got a startup that's doing something like that. I think the company name is Fogbeam Labs or something to that effect. I'd look him up.

I'm very happy to be mistaken. I was a casual fan in college, but the worm turned for me when I started hanging out with my brother in law who plays at Ball State. Hearing someone accurately predict what's going to happen before the ball is snapped, and then explain why (e.g. "they ran it up the gut because the linebacker had his weight back on his heels") really opened my eyes to the depth. If there's any HN football fans out here in Denver who'd like to get together one weekend to watch a game, I'd love to meet up!

Good point. All-22 tape, which is defined in the article, is footage of a football game that shows all 22 players on the field simultaneously. It is almost certainly not the footage you'll see if you watch a game on television, which follows the ball or, more rarely, individual players.

This is very exciting, as there's a lot of strategy to American football that isn't plainly visible when the camera just follows the ball as it does on television. It's not going to showcase technique very well, but it'll go a long way towards explaining e.g. why some wide receivers are always open, why some players always seem to be in the right position to interrupt a pass, etc. I'd actually find it very interesting to see something like a Coursera course using film from this to explain different plays, packages, etc.

However, I'm not sure it's of wide interest to HN. As far as I can tell, I'm the only one here who likes NFL football.

I had a fantastic internship many moons ago concurrently with my studies, which could be a synonym for an apprenticeship. If not for that, I'd probably be teaching Latin somewhere. I still keep in touch with a couple of the engineers from there.

Mapquest does not currently have anything like that, and I'm not sure we're set up well for that sort of thing at the moment (I'm not sure what they'd do). We have interns in our marketing department and I think the new, very young looking dude in finance is an intern, but nothing in development.

Edmond - I'll take a couple. My e-mail address is in my profile - please ping me if I can help.

I don't see it that way at all, and would never infer either of the statements you gave if someone told me they were self employed.

It takes a lot of courage to admit that you struggle with depression, even with an anonymous account, so first off I want to say good on you for writing this.

I've had severe problems with depression before, and one of the biggest realizations that helped me out of it was that I have the power to change nearly every aspect of my life. There were many things in my life I wasn't happy with, and it took me a long time to realize that it was in my power to change them. It took me a much longer time to actually implement a plan and change them, to the point that I feel pretty comfortable in my own skin right now, but that realization was the start. I spent some time on medication (Lexapro) and in therapy, but I'm not a psychiatrist so I can't make any specific recommendations on that front. I can say that I viewed the therapist less as a doctor and more as an outside observer into my life, to whom I could talk about almost anything and get some trusted advice. I think it's paramount that you talk to somebody about this before you hurt yourself or, less dramatically, before you spend more time feeling depressed and questioning why you're even alive. Talk to a therapist, talk to a hotline, or even talk to a bartender somewhere if it's this bad. You've taken a first step by posting here, and that's a really brave thing to have done.

Please e-mail me if it's this bad and you genuinely feel you've got nowhere to turn. My e-mail address is in my profile, and while I'm not a psychologist/psychiatrist, I will gladly give you what assistance I can. What've you got to lose?

US only? I sure would like to get on the radar of companies hiring in Australia/New Zealand. Any chance you'd be able to expand down that way?

Your original point was that we might not need server-side languages or frameworks at all. "The grid" is still a server - it's just not one you control directly. I don't think we'll ever get away from servers because there are definite advantages available by having computation centralized - you can have more powerful hardware available, you don't need to worry about outdated versions of code running, etc.

More broadly, though - there's nothing specific to a language that allows or disallows communication with a database. That's really the realm of a library or a driver to interact with the database (an external service) through some established protocol. People have written databases with JS hooks before, but it's just not common, likely because of the security problem.