lambdaelite

Practically speaking, none I can think of. But I completely disagree that the networking aspect has no benefit.

At least in my discipline, authors either retain copyright to their manuscript and can disseminate that freely, or they are able to personally disseminate the final published article (sometimes including on their own website). No copyright transgression occurs in this case.

One guidance for the ethical side are the guidelines agreed upon by professional associations. I can't think of any that condone copyright infringement.

So why not email the corresponding author? I have yet to not get (or give) a manuscript that way. From my own perspective, each time I respond I'm possibly getting another citation. It's also a great form of networking.

If they want to seriously engage regulated industries like finance, medical, energy, etc., they need to. As it stands now, an association with them is suspect.

Silicon Valley and YC don't exactly have a stellar reputation for ethical behavior. Having a "pirate website" at the top of the news page doesn't exactly change that perception.

I totally get that journals are evil, and charging money for research generated with public funds is questionable. It's very frustrating as a small entity needing to view articles, and being asked to cough up $25-50. That said, there are legitimate alternatives (like emailing the corresponding author, or professional society memberships, or alumni library access, or DeepDyve). The linked website is flagrantly violating copyright and that should be cause for concern; not breaking the law is part of every engineering (and professional) ethical code.

I find this is generally a poor place for hardware-related startups.

Places to check (and network with)—I'd try checking in with your university's tech transfer office: they might be able and willing to clue you into recently-formed startups licensing university IP. Another would be your university's professional practice office (may go by another name), i.e., the office that sets up student internships. Another is with your department's industry relations rep (if you have one). A fourth place would be in any hacker spaces or tech villages in town. A fifth would be with the local angel investor groups.

A really great possibility is to take your senior design project seriously. It's not unheard of for those to turn into something commercially viable.

I'm not a big FTIR person, but at least for Raman, laser power is far less important than the linewidth (and you definitely don't need a pulsed laser) and the rest of the optical engine. I'm actually pretty bearish across the board on these consumer spectroscopy products mostly due to the importance of sample prep. Your use case statement is correct, distinguishing paracetamol from candy should be easy, but is really that a common need or a novelty? It seems to be the latter to me.

If you consider it functional, Common Lisp is still used by some very large and established companies: the Lisp HUG mailing list (LispWorks) will frequently have email signatures from some of them. There is some truth to the notion that CL is a competitive advantage and so its use tends not to be discussed in the open.

The right way to do that would be with FTIR or Raman spectroscopy, and this sensor is capable of neither.

At best, this is an ersatz replacement for the tunable filter on a hyperspectral imaging system. There are some advantages (off hand, acquisition rate is a big one) to a more diverse mosaic filter, but other than in using QDs as the actual filter medium, this idea is not at all new.

Aren't recreational drones limited by a 400ft ceiling?

Why are concurrent GCs rare?

I agree, the choice of programming language is one of the less important parts of the SDLC. In the case of Rust for SC work, as the linked article alludes to, what doesn't make sense to me is that there is no industrial-grade tooling or support software. It seems like an ill-informed statement.

As to proofs, I thought some level of formal proof was required at SIL4?

I go by what I thought was the accepted definition, which is a failure or error presenting a risk for temporary or permanent harm to people. Financial risk like data loss would fall under mission critical.

The Cartoon Guide to Physics might be what you're looking for.

Rust’s strengths are Clojure’s weaknesses, and vice-versa. Rust isn’t as expressive or interoperable, and its concurrency story isn’t as complete. That said, it’s much better for performance or safety critical needs, and it can be embedded inside other programs or on very limited hardware.

I find this hard to believe. Is anyone actually using Rust for a safety-critical application?

Funny you mentioning L4... based on the HN title, my first thought/hope was someone got seL4 running with the NetBSD userland.

If you're going to use the backlight it's a worthwhile upgrade. Otherwise, not sure.

Yeah, I wasn't clear on the versions.

At night, I have to use level 7 on the backlight for comfortable reading without any other lights on. It's noticeably blue to my eyes at that setting. I get that Amazon's trying to make the display look whiter in normal light, but I (and I think others may) place more value on a warmer backlight for nighttime reading. Outside of vacations and doctor appointments, bedtime reading is my most frequent use case.

Yawn.

I upgraded from a Paperwhite 1 to a Voyage because I thought the screen and haptic buttons would be improvements. I can barely see a difference in the screen quality when I'm deliberately looking (and don't notice any while reading), and the haptic buttons are poorly placed so I rarely use them. The Voyage is nice but was questionably worth the extra money over the Paperwhite 1. Now with the Paperwhite 3? Fuggetaboutit.

You know what feature would be a great improvement? A backlight without blue light, for reading at night without screwing up sleep cycles.

edit: clarified

Moleskine paper is notorious for feathering and bleeding when used with fountain pens, even fairly dry writers. I've actually had better luck with a no-name bound book from Office Depot than I have had with Moleskines.

Happy to help! I myself just learned about that nuance the other day.

I don't understand why the Moleskine brand has grown so popular when the paper quality is so poor relative to other manufacturers, e.g. Rhodia/Clairefontaine.

Were you using encrypted Google, or via !g on DDG? The trick only works through the non-encrypted one.

That sound you hear is the laughter of Toughbook owners.

Anecdotes? Sure. My wife loves to trade options. I generally avoid them, except for buy-writes. Across investments, she enjoys taking higher risk positions than I.

I can't say that gender doesn't play a role, and it would make sense if gender were a factor, but I don't see how it could be anything close to a dominant factor.

I think caution is advised at applying these studies broadly, as you are doing. Not only are they done on WEIRD populations, but a limited subset of that grouping.

Software engineering blogs.

Not interested in opening up that debate, but I thought based on the title this would be something else.

A dual income couple can tolerate more risk, but in my experience they are unwilling. After marriage, a couple's risk tolerance skews a lot closer to a woman's risk tolerance than to a man's.

How is someone's gender in any way relevant to their financial risk tolerance?

Inexpensive ones.

Kidding aside, as a startup, our needs and capabilities are quite different from a more established company. We can't afford to pay for consultants to fly business class, nor expensive per diems. We can't afford to implement heavyweight development processes, nor should we since being nimble is one of the few advantages we have as a startup. Software licensing costs is a concern at this stage, so we'd rather do a subscription for a (for example) CAD package even if it's not the industry standard over buying a license for the industry standard package. Also frustrating in general is that as a medical device, our volumes are never "large" in the consumer sense and so many common manufacturing processes (e.g., injection molding) are not practical, nor are companies excited to supply parts in our expected volumes—our consultants need to consider what are sometimes less common processes or smaller, "hungrier" suppliers. For many things, the budget is driving the bus, and it's within reason acceptable to need to spend more in the future if it means we can spend less now.

It can be hard to find the a consultant at an experienced group (and domain expertise is what we need) that is willing to work with us. Our future business is not guaranteed (statistically we are likely to fail), the business we will be bringing is small, and we're pinching pennies on what we bring. What seems to have worked is to find a "hungry" associate-level consultant (keeps costs lower) at a larger group who can tap into the experience of senior/principal-level consultants on demand. We point out that, if successful, we'll be able to bring more business to the associate-level consultants, so we both will grow together.

I'll give a specific example, stream of consciousness style partly out of laziness. We have a software component, and a portion will be safety-critical... crap, things just got more expensive. We know we need to develop the device under a risk-management plan (ISO 14971), so we don't need consultants to tell us that, but we may need them do to a design review because we don't have enough in-house experts (how many do we need?). We know that the software component must be developed under a SDLC (ISO 62304). Should we do it agile (as per AAMI TIR-45, not cowboy agile nonsense), or a more traditional V-model? Does CMMI or PSP apply to us? Should I listen to the nice people at IBM and use Harmony, or are they preying on a poor (emphasis on poor) innocent startup? It'd be nice to have a consultant advise us, based on their experience, that one or the other works better for a startup. Have we properly identified the safety-critical portions of our design? What should the development environment look like: is MISRA-C everywhere the right answer? Probably not. So MISRA-C for the safety-critical portion, and regular C everywhere else? Or should we do Java/RTSJ with a fancy JVM? Ada/SPARK? We could justify any of these, and probably others. What are the costs of tooling for each choice, licensing costs, how hard will it be to find employees/contractors (should we use contractors? how many will I need to hire?), and what will productivity look like? What about maintainability? Is QNX better than VxWorks? Is model-based systems engineering worthwhile, or is it an expensive buzzword? Those nice people at IBM said Rational is great, and it'll tie into DOORS! And they said DOORS is great! Maybe I should use Excel instead of DOORS, I heard from a guy on the internet that works just fine. And so on...

So for much of this, we want an experienced consultant that can review case histories and show us that in the past, comparable companies were successful by doing X, and so it makes sense for us to go that route too. Many times we know we need to do something, and frequently how to do something at a larger company, but not the optimal way to do it at a startup. I can do a decision matrix, but without the case histories, there's too many gaps in the matrix to be useful. We also need to have external people for design reviews, and figure out when we need to pull in those external people, when we don't, and when we should. I need to be able to tell the regulators that our processes were based on industry best practices. I need to be able to tell board members the same, and be able to tell them that our processes and plans have a successful track record at comparable companies to justify the time/cost/etc. I want to be able to show shareholders that we're being good stewards with their investments, and we're doing that by making careful, informed choices.