Morning Ed: Society {2016.05.04.Th}


Will Truman

Will Truman is the Editor-in-Chief of Ordinary Times. He is also on Twitter.

Related Post Roulette

98 Responses

  1. Avatar LeeEsq says:

    That was one depressing comic.

    The first link in the CV of failures lead to a blank page. Anna Peak’s CV points to a big issue in home schooling, it works better with the wealthy than the non-wealthy. Colleges would not accept her unless she could demonstrate that she had the requisite academic skills. Her family couldn’t afford all those SATIIs though.Report

    • Avatar LTL FTC says:

      Reading her failure CV, I thought to myself, “I bet she’s a lot of fun at parties.”

      But seriously, if I had a nickel for everyone I know personally who has complained how unfair it is that nobody will pay them to write about feminism and Victorian literature, I’d have a quarter. Which strikes me as rather a lot.Report

      • Avatar LeeEsq says:

        I didn’t google Anna Peak so I had no idea what field she is in. Education has economic and non-economic functions and both of these are kind of required. Most people expect to be capable of getting some type of gainful employment after completing their schooling and the skills necessary for said employment. At the same time, its also supposed to produce people capable of being good citizens and curious about the world rather than good employees. We forget that a lot.Report

      • Avatar Oscar Gordon says:

        She does sound bitter. Instead of being inspired, I kinda want a drink…Report

        • Avatar LTL FTC says:

          It’s hard to write a piece about how every institution you’ve come in contact with has failed you in some way without appearing bitter, whether it’s true or not.

          But part of me thinks it may be counterproductive to bitter-shame her. People don’t need to be cheery whatever their situation just because the wallowing may make someone else feel bad.Report

  2. Avatar Kolohe says:

    May the fourth be you…always?Report

  3. Avatar notme says:

    CA nanny state strikes again. Under 21, no cigs for you!

  4. Avatar Jaybird says:

    AI-driven sex.

    In the same way that cheap and easy calories eventually led to an obesity crisis due to our bodies enjoying empty calories as much as (or even more than) the calories that are more difficult, the emotional experience of having a soft, pleasant voice tell us that we’re looking like we’ve had a rough day and that we’re appreciated and then to ask, hey, want a tumble? is something that will appeal and tickle us even if the emotional calories are empty.

    Personally, I blame The Pill.Report

  5. Avatar Saul Degraw says:

    Everyone has failures or set backs but I can see why Anna Peak wanted to push back against the first CV of failure. The original one was written by a professor at Princeton. He made it despite the failures. Anna Peak is an “assistant professor of instruction” at the Intellectual Heritage Program at Temple University. Can’t tell whether she is full-time or adjunct. She seems to teach one class.

    I wonder if Lee is right in his observation on unschooling and whether you need to be of high socio-economic status or connected to a very specific community for it not to backfire.Report

    • Avatar LeeEsq says:

      Its not so much the connections but most colleges and universities are going to want some indication that home schooled and unschooled students can hack it in a traditional academic environment. This is true even if they aren’t elite schools. It isn’t really unfair either. The indicators are standardized tests. If the kids can’t afford to take the standardized tests than they are screwed.Report

      • Avatar Oscar Gordon says:

        This. If you are going to homeschool your kids, you’d better be ready to help them navigate the testing gauntlet they’ll need to undergo if they want to get accepted to college.

        If her parents refused/did not offer to pay for her testing, etc. that’s kinda crappy of them.Report

        • Avatar LeeEsq says:

          Its borderline neglect considering that a college degree is kind of a necessity for gainful employment these days.Report

          • Avatar Oscar Gordon says:

            Any post secondary is, and any place offering training worthwhile is going to want to put a homeschooled kid through some testing.Report

      • Avatar dragonfrog says:

        It seems odd that there are charges for taking the standardized tests.

        There is public education at taxpayer expense, and for tax-funded students, the tests are also tax-funded right?

        So, on a callous dollars-first logic: if someone is saving the state money by not using the rather expensive public school system, what is the rationale to discourage that by denying them the quite inexpensive taxpayer funding of access to standardized tests?

        On a less callous, children-first logic: if someone is home schooling their child, isn’t it important to make it wasy for them to check their work, discover and mitigate the weak spots in the education they’ve been able to provide?Report

    • Avatar Kazzy says:

      It stands out that she referred to her lack of second-language fluency as being the result of it being something her “college training did not provide”.

      Does she mean to say that her college offered no foreign language courses? Or only that they didn’t require it? Or that they did but she did not walk away fluent?

      Saying that her “college training did not provide” her something implies she denies any agency in the matter, which I find really off putting and the sort of attitude that is likely to lead to more failure than success.Report

      • Avatar Saul Degraw says:


        Foreign-Language proficiency for a PhD program is not even really native level. What usually happens is that you are given a passage to translate, a dictionary, and time, lots of time. This is more like intermediate to advanced level proficiency but at the low end of advanced.

        Though she could also be captured by the unschooling ethos still.Report

        • Avatar Kazzy says:

          But my point is that, regardless of the specifics of the requirement, saying that she does not possess it because her “college training did not provide it” is to deny herself any agency or control over the situation.Report

      • Avatar aaron david says:

        One of the things I remember from schooling was of a certain group of students, who uniformly did poor, saying that the “teacher didn’t teach me.” One can only assume that they expected the teacher to sit down with them al all times and hold their hands for each little bit of instruction. When that didn’t happen, they of course blamed it on anyone but themselves.Report

        • Avatar Morat20 says:

          I said that once, about one single class, during my undergraduate years. The guy was clearly skilled, knew what he was doing. He couldn’t teach worth a flip, and the book for the class was atrocious.

          All the knowledge on the subject, no ability to teach it. I heard the phrase “That’s how we do it in the industry” when students asked about basic knowledge he’d just flat out skipped imparting (and the books lacked) used about a MILLION times until i was ready to scream “We’re not IN the industry! That’s why you’re teaching it!”.

          This was back in 97 or so, and I found another book on the subject in a half-price bookstore that was about a thousand times better. (Every contemporary book on the subject was written by the same moron that wrote the crappy course book).

          I did well because I finally found a freaking 5 year old book that covered fundamentals of the course that the current book nor the teacher covered, but was required to actually understand what they were doing.

          Seriously, it would be the equivalent of teaching C++ as an “intro to programming” and teaching high level OOD and syntax, but never actually covering loops, variables, variable types, pointers, decision statements…..because “Everyone in the industry already knows that stuff. Let’s talk about polymorphism”. (And for some reason the book lacked it).

          Crappiest class I ever took. I can handle a crappy professor OR a crappy book, but both? Prior to the internet, at least, that was really hard to fix.Report

          • Avatar aaron david says:

            Yes, well being the son of a professor, who actually liked to teach! makes a bit of a diference in how you see things. I was mostly refering to the fact that they used this logic for every class that they didn’t quite get intuitively, that they would not ask questions nor apply a trial and error method to start looking at problems they encountered.

            Worst prof I ever had was for logic, but the book was good so I got through in the end, but I am with you on the “if both are bad…”Report

            • Avatar Morat20 says:

              In college, it was pretty taken for granted (even for the classes they might bring in an adjunct for), that the professors knew the material. You could spot, really easily, the ones that either had grokked basic educational theory or had enough expertise to handle it — they not only knew the material, they could teach it.

              A lot of professors weren’t bad teachers, but they weren’t great. Honestly, a bit of guidance would have probably done wonders for their ability to impart knowledge, but by and large between them and a semi-decent book, any mildly motivated student could get by.

              Of course “19” and “motivated” don’t always go together. (That’s back to executive functioning issues that don’t resolve for years).

              I found college an entirely different experience at 30 (went back for a Master’s on the company dime) than at 19.

              I still struggled with database theory. Screw relational calculus (or algebra or whatever it’s called) right in the ear. I’m still 50% sure it’s just all made up so that DB experts can charge banks out the ear…..:) (No, it’s a real thing and really useful. It just was always so alien to how I thought. Give me a good evolutionary algorithm any day!)Report

              • Avatar veronica d says:

                Relational calculus includes higher-order operators, looks like predicate logic, is the basis for datalog and such, and can express recursive relationships. Relational algebra cannot. Basically, if your logic can operate on functions/predicates as objects, it gets called a “calculus”, as an analogy to traditional calculus, where you have “functionals” such as the derivative operator that take a function and return another. An “algebra,” on the other hand, tends to be a first-order theory with simple equational logic and limited recursion.

                (This has been a pedantic math post compliments of me.)Report

              • Veronica is like this in person too. It’s kind of scary. 🙂Report

              • Avatar Oscar Gordon says:

                I gotta go drinking with @mike-schilling & @veronica-d; the casual conversation alone might fill in some of the gaps in my math education…Report

              • Avatar veronica d says:

                Plus I like to dance!Report

              • Avatar Morat20 says:

                I feel the same way. 🙁 I’m awful at abstract math. I just don’t grok it, and the DB stuff was too abstract (or perhaps I just didn’t find the problems interesting enough) to really connect them.

                I finally understood Laplace transforms about a year after I learned them, when they were useful to solve circuit problems with inductors and/or capacitors. Suddenly they made sense. What they were. How to use them. How they worked.

                It was literally a moment of enlightenment.Report

              • Avatar Oscar Gordon says:

                I’m still a bit fuzzy on Infinite Series. I get the general concept, but for some reason I don’t grok how people use them to do things.

                And I had a similar experience with Laplace & Z transforms. Was a big WTF? until I took an automatic controls class. Once we started doing signal analysis, they both gelled in my head and I breezed through that part (come to think of it, quite a few things regarding imaginary numbers & polar plots made a hell of a lot more sense after automatic controls).Report

              • I don’t particularly recall Laplace transforms. (DiffEQs was a pretty distasteful class: Here are some techniques and hints about when to apply them, but we’re not going to prove that any of them work or make sense, or define precisely the circumstances over which they’re valid.)

                But if you have specific questions about infinite series, I could write a post to try to answer them. I did one about power series a few years ago.Report

              • Avatar Oscar Gordon says:

                If you could, that would be awesome! It’s one of those parts of Calculus that I never got in class, but really wanted to. I always kind of figured I’d run across them in my professional work and have another opportunity to grok them, but oddly enough, I haven’t (& you’d think, given my line of work, I would have).

                So I have no trouble understanding the general concept expressed by the sigma notation (i.e. as k increases from a given starting value to infinity, each computed value of the summation becomes smaller until we can essentially call it zero for whatever application you have, so while the value of the summation is always increasing, it’s asymptotic to some value, or expression).

                Where I stumble is how to go from something like the Sine function to an infinite series, or the reverse, given an infinite series, how do you jump from the series to some known function.

                Basically, when I was learning it, this part always reminded me of this cartoon.Report

              • Avatar Morat20 says:

                Going from 20 year old memory: In terms of circuits, there’s a nifty method used to determine voltage and resistance at nodes in a circuit. SUPER easy. That method doesn’t work with variant currents (capacitors or inductors).

                So you use a Laplace transform, which removes the time-dependent variable from the equations you’re working with. You can now use the simple and easy method to find the voltage or resistance you’re interested in, and then transform it back to put the time-dependent aspect back in.

                In short, if you have an annoying variable that gets in the way of solving the problem (like the fact that voltage will vary with time because of a capacitor), you can temporarily remove it, solve the problem, then stick that time-dependent variable back in. (You can’t solve for that. Your answer will always involve “what time is it”)

                In diffy q, we just learned “you can remove a variable, do stuff, put it back” which seemed like a lot of pointless work to move from one equation to another. Nobody really said “You do this because the variable you’re removing is integral to the solution — you’ll have it always, like “what time is it”, so you remove the variable you WON’T be using to simplify or solve the other stuff that can be solved.

                Derivatives and integrals made perfect sense (doing them was easy, but the specifics on what they were doing) with basic physics. Oh, the derivative of velocity is acceleration — and the integral is distance. THAT’s what’s going on.

                I’m just awful with abstracts like “slope of the curve”. I can follow what you’re doing and how it works, but I need a “why” if I’m going to remember it. An example where I fully understand the front and back end.Report

              • Avatar Michael Cain says:

                Certain classes of stochastic optimization problems. Intractable as you would normally write them, you can apply a Laplace or Laplace-like transform to get a solvable problem, then invert the transform to obtain a solution to the original problem. Lots of heavy lifting proving that a particular transform works that way for a particular class of problem.Report

              • Avatar veronica d says:

                @michael-cain — That sounds positively delightful. Do you have a reference?Report

              • Avatar Michael Cain says:

                @veronica-d — Not off the top of my head. It was a long time ago, and stochastic optimization always made my head hurt. I was a deterministic problems and algorithms guy.Report

              • Avatar veronica d says:

                @morat20 — I’d say you are describing more the difference between applied and pure math, rather than the difference between the concrete and the abstract. After all, if you are doing transforms on actual equations, that is rather concrete compared to the Bourbaki type stuff.

                For myself, I quite believe one should learn the concrete before encountering the abstraction, in the sense that a person does not need to know about matroids until long after they’ve encountered linear independence and probably quite a few graphs. The point is, one should have at least one, and better yet three, concrete examples of a thing before they struggle with the unifying abstraction.

                The idea that one would learn about cartesian closed categories before they learned about typed lambda calculi is literally horrifying.

                Don’t learn Category Theory until you were about to invent it yourself! (Or if you are a Haskell programmer, but never mind that.)

                That said, the pure-versus-applied distinction is quite a bit different. First, so much of computer science is completely-and-totally “pure” math, inasmuch as “applied” tends to refer to calculus and various method of crunching big matrices over the reals, neither of which are foundational to CompSci, compared with graph theory or logic.

                Anyway, math classes teach math, taught by those who focus on math-for-math’s-sake. If you’re learning (traditional) engineering or science, I think you should take math on its own terms, while understanding that plenty of applications will come along. Those are taught in the engineering classes.Report

              • Avatar Morat20 says:

                I pretty much did. I’ve actually been toying going back for another bachelor’s degree (I have a BS and a MS in Computer Science, but getting the PhD would drastically limit my employment options AND require time off work I don’t want to take).

                My company has rather generous education benefits, so I could probably push out a BS in a few years. (Two classes a semester). I’ve been toying with a pure engineering degree. I could claim several fields of engineering as relevant to my job (materials, mechanical, aersospace, and if I stretch chemical).

                I’m not sure I want to, but I’ve been working with engineers so much the last decade that I’ve been having the urge to do something a bit less…virtual. Plus I’ve been feeling a bit of brain rot. 🙂Report

              • Avatar Oscar Gordon says:

                I’ve toyed with getting a CS degree, but at this point in my career, focusing on project management has more value.

                Although some in depth study of higher order concepts would be interesting, I’m just not interested in chasing a sheepskin.Report

              • Avatar Morat20 says:

                I’m not gonna say it isn’t useful, but it’s more…situationally useful. Sufficient experience can easily compensate if you’re programming. The closer you are to innovation, the more the all that background helps — but if there’s a single field where you can access the theory (everything from “Dummies” to “Post-Doc” level) online, it’s CS. 🙂

                In practice, I’ve found that the degree gave me a rather robust framework that’s been quite useful. I’ve noticed my peers (that went into coding from engineering degrees) often have some odd blind spots or gaps (which makes sense) even the ones that are far better coders than I am.

                Offhand, the most useful courses I took in CS were on algorithms, data and file structures, operating systems and advanced operating systems, database design, and a few specialist classes — numerical methods, boolean algebra, and some high level design stuff.

                The absolute MOST useful I ever took was an advanced OS class that, over the system, had us basically implement a micro-kernal as an ongoing project. (I mean it sat on-top of Unix, but by the end it let multiple users log in, create, read, and write to files, etc. So you had to implement shared memory, file structures, basically design and handle threading and locking, semaphores and such….)

                It not only gave me a solid feel for how OS’s worked, at least the root bits, it was a very hands on practical class on a lot of advanced programming techniques.Report

              • Avatar Oscar Gordon says:

                I took a handful of CS courses along the way, not enough to qualify as a minor, perhaps just a focus. I also took a bunch of numerical methods classes, and having algorithms & data structures* under my belt for those let me cut loose on my program designs in ways many of my classmates just couldn’t.

                My wife is probably going to be taking some classes soon, so no school for me until she’s done, but I might see about some texts or “at my own pace” online classes for OS design. I’ll be curious if it will be helpful with my API rebuild.

                *2 of my most satisfying classes. Challenging, but when I was done, I actually felt like I could write actual software.Report

              • I only took four CS classes, from introduction to programming (in FORTAN!) up to data structures [1]; everything else was learn by doing.

                1. Few of which I’ve ever used, much less implemented from scratch [2]. But learning to think about data structures as at least as important as algorithms was extremely important.

                2. I did implement red-black trees in C# for compatibility with Java TreeMaps, but the only balanced binary trees the course mentioned were AVL trees.Report

              • Avatar El Muneco says:

                Heh, that crap can be important. I was in an interview a couple weeks ago, and basically scrambling to answer based on 25-yo memories and working things out from first principles. And I got dinged for only remembering the name of the second-best string comparison algorithm, not the best one. I know that question is important in certain domains, but it literally had never been an issue for decades.

                TL,DR – CS classes will come back to haunt you at the strangest times…Report

              • Avatar Morat20 says:

                I get irritated by those questions in interviews. “If I need to code a string comparison algorithm from scratch, I’ll google it”.

                But the odds of the std::string class not being sufficient in C++, for instance, are pretty freaking low.

                (Having said that, I actually DID once have to modify a string comparison algorithm because I was using strings to store a sort of compressed time-series of decisions and I wanted to know how much alike the output of multiple decision trees was. I googled the algorithms….)Report

              • Avatar El Muneco says:

                I have a BS and MS also, but my impression – and it’s a bit old, so it might not be current – was that the PhD track in Computer Science was so bound up in the theory track that you really can’t leverage it in the real world. That the people who go on past the MS are the ones who want to teach at some point.Report

              • Avatar Morat20 says:

                That’s my impression as well.Report

              • I used to work for a guy who refused to hire PhDs because they didn’t like to get their hands dirty doing the messy kind of work that gets products out the door. Though where I work now, there are PhD’s everywhere, and many of them have gotten over that.Report

              • Avatar Michael Cain says:

                For myself, I quite believe one should learn the concrete before encountering the abstraction…

                Although, as I always used to say, theory is important because it tells you where the ice is thin…Report

              • Avatar veronica d says:

                Truth is, I’m a bit of a theory snob — with the full acknowledgement that no one likes a snob. But all the same, I like theory. But more, I think it makes a big-giant difference in how your career goes.

                Okay well, maybe sometimes, it depends. Careers are tricky things, and your personality has as much to do with progress as anything —

                — for instance it’s probably an advantage not to come across as a snob!

                tee hee

                But still, much of what happens in software engineering is people using standard tools to solve typical problems in routine ways. Which fine. This is needed. Software eats the world.

                But still, how many software engineers write code that moves data back and forth between databases and screens, doing only the most rudimentary transformations on that data in flight?

                This is a solved problem. If your problem is “small-ish,” you can hack it together in a couple days using Ruby on Rails. If it’s bigger, you have to start looking at sharding and “eventual consistency” and that “noSQL” stuff. Fine. There are a billion article written at this stuff, which require little math or theory to grok.

                On the other hand, the folks who wrote the Deep Mind stuff to beat Go — they know more than average. And I promise, they know much theory.

                Anyway, my point is, I don’t want to be the “move data back and forth between the database and screen” person. It’s dull. I want to do new shit.

                Which actually, I largely don’t. But that’s the luck of the draw, and the limits of talent. Which, it turns out I’m not as smart as the Deep Mind folks.

                Likewise, Edward Farhi consults with my employer on QC stuff, and he’s done talks at my office. I’ve chatted with him. Actually, after one of his talks I got the crazy idea that I’d end-run the non-abelian hidden subgroup problem and find some QC algorithm for short vectors — which is a fun NP-Intermediate problem that I rather like. Of course, I didn’t get very far, cuz I’m not smart enough. I guess. Or something. But still, I learned a lot in failure.

                All the same, even if I’m not the software-superhero I wish I was —

                — and neither am I the sexy, gorgeous, and dangerous cyborg battle chassis that I wish I was —

                But anyway. I do have a cool job working on cool software. It’s actually a nightmare of a program, about a million lines total, a mix of C++ and Common Lisp. It does airline pricing optimization, and is hyper-optimized, hyper-complex, and really it takes a certain kind of brain —

                Like mine! (she says with a big dumb grin and a flip of her hair.)

                However, the program has flaws, and many. In fact, it would have benefited greatly from a certain kind of theory. Now certainly, it uses plenty of advanced algorithms. But this-or-that fancy algorithm is not the key. The key is language design, the expression of airline industry rules in a computational semantic structure, rather than a mishmash of Lisp code spread haphazardly across a billion files. If we had a semantics and a formal expression of the rules, we could do transformations, generate code, perform “global search” style optimizations and heuristics. Basically, you hire a few folks who’ve written a proof assistant or Prolog compiler or whatever. You set them loose on the problem. You send me also, for moral support! (Plus I’m pretty.)

                Except we can’t really get there from here. Once you have a million lines of code, it’s kinda not possible to rewrite, not for any formal reasons, but more for a “beyond the capacity of human project management” reasons.

                Or not. I dunno.

                Haskell cannot get null pointer exceptions, which if you are a Java programer, you are probably familiar with NPEs. For myself, I consider a NPE a kind of type error. You declared the variable was a String. A String can do certain things. However, the variable does not reference a String. It is null. Null cannot do what Strings can do.

                This is a type error, even if Java advocates can insist that “well exceptions and non-local termination are part of the language semantics and thus part of the type systems and thus…”

                (At which point I throw rocks at them.)

                It’s a fucking type error. Get over it.

                The reason that Haskell cannot get NPEs is because its creators recognized that an object that can be either a String or a Void is a kind of tagged union. Furthermore, they recognized that tagged unions are a kind of “sum type,” which is dual to a “product type,” which together form the algebraic data types. The short version is, you need two values for form a tuple (A*B), but when you consume the tuple, you can look at either the A or the B. You don’t need to handle both. For the sum type (A+B), however, you can build an instance of A+B with one type, either an A or a B, but when you consume the value, you need to allow for either value type. This must be explicit, else it is not type safe. In other words, you need a case statement.

                In Haskell, if you say something is a String, then it is a String. If you say it is a String+Void, then it can be either, but you have to handle both cases. It’s really nice.

                The point, by starting with this theory, in fact with an algebra, Haskell programmers avoid a big giant source of bugs. In practice it works very well. You seldom asked for String+Void unless you really need to. (Note: in actual Haskell, the X+Void thing is built into the library as “Maybe x.” So you can declare a variable “Maybe String.” It’s goofy and fun!)

                Database theory, the relational algebras and calculi, they are only marginally useful for your average engineer putting together fairly standard databases. Most of the various “normal forms” are pretty common sense, once you’ve seen it done a few times. But still! Data is forever. Your applications will change. Your data model — it lives on, its mistakes calcified into the scar tissue of your organization.

                Mathematical soundness is not the only thing worth caring about, but it matters.

                For example, the airlines have an entire system for automated ticket processing based on a complex set of rules, filed by individual airlines, according to a semantics designed by an industry consortium.

                The people who designed this, they were “systems analyst” types, and not (OMG NO!) computer scientists. And it shows. These fuckers couldn’t even get basic boolean algebra right, as the entire systems is built out of a weird three-state logic, which is inconsistent from one rules-category to another.

                It’s messy-messy-messy, a poorly specified mass of flim-flam, only partly compositional, with inconsistent semantics. We all pay for these mistakes. (On the other hand, I have a job dealing with how hard it is. On the third hand, this is the “broken windows fallacy.” If this was easy, we’d still want to do amazing things, and we could do more amazing things.)

                In short, if you are doing something new, please base it on sound theory.Report

              • if you are a Java programer, you are probably familiar with NPEs.

                Whose name, by the way, demonstrates that Java has pointers. Calling them “references” was marketing.Report

              • Avatar Oscar Gordon says:

                Pointers was a dirty word.

                Learning Java & grokking references helped pointers make sense.

                C++ implementation of pointers/references is still annoying.Report

              • All true. Though the C++11 smart pointers (especially unique_ptr) that clarify who owns what piece of memory are a big improvement.Report

              • Avatar Oscar Gordon says:

                I’m glad someone else recognizes that C++ pointer implementation is annoying. Whenever I say that to C++ devs, I get a round of, “You just don’t understand pointers”.

                If I find myself having to code in C++, I’ll see about using 11.Report

              • Avatar Morat20 says:

                There’s a reason I debugged someone’s C implenetation of a self-growing array by just loading it to the failure point in the debugger.

                I didn’t want to wade into the pointers, pointer arthimetic, and figuring out where it went off it’s defined rails.

                Instead I just used the debugger, isolated the crash, and located the index and saw it had abruptly become negative. Changed an “int” to “unsigned int”, added a “If you get X big, throw a “You’re loading too much data” pop-up so they know we only support a half-million or so inputs of that particular complex format” and abort the process so you don’t end up eating five gigs of memory to throw to a computation engine that has hard limits.

                It was a LOT faster.Report

              • Avatar Oscar Gordon says:

                My most successful math classes were the ones that covered concepts we were using in one of my engineering classes. It always felt like one of those magic eye pictures, were you just let your eyes lose focus, or cross for a bit & the POP!, it all makes sense.Report

              • Avatar Morat20 says:

                Yeah, that’s the feeling. 🙂 At 18, I struggled with it — I’d often be learning the theory the same week I was trying to apply it in another class (if I was lucky, I’d learn the math before!).Report

              • Avatar Morat20 says:

                Thank god people like you exist to do that crap, because I could only barely do it with a book in front of me and a simple problem.

                I didn’t get nothing out of the class, though. I now shudder when I see what people consider “databases”. (“That’s a spreadsheet, not a database. Why do you have a field for a numeric ID? For identification? Why do you need that. You already have three fields that can individually be keys, and one obvious pair that’d work exactly like what you want….).

                I think the last time I touched a database, I spent three weeks forcing them to use GMT, splitting out some obvious tables, and basically turning it from “This is how you tracked it in Excel” to something vaguely efficient.

                But I couldn’t prove that mathematically. 🙂Report

              • Avatar Road Scholar says:

                Back in my Navy days I was tasked with being the Department computer security petty officer. Basically I was supposed to make sure the AV software was up to date on a bunch of standalone PCs running DOS, no network. Standard software on those machines were WordPerfect 5.1, dBase IV, and Formtool.

                Nobody working in those offices had a clue what dBase was about much less how to use it. So the guys running the tool crib were keeping an inventory as a WP doc. Another guy was using Formtool for form letters (not its intended function). Frankly, I was pretty impressed just to find someone using the right software for a job, even if clumsily.

                BTW, that’s when I taught myself the basics of DB design.Report

              • Avatar Michael Cain says:

                An economics professor was removed from a flight and questioned because, well, working rather single-mindedly on a differential equation.Report

              • Avatar Oscar Gordon says:

                Note to self: don’t write math beyond simple arithmetic, or use Mathematica for some calculus on an airplane, because people are fecking stupid.Report

              • Avatar Road Scholar says:

                There’s a certain irony in an innocent being called out on account of stereotypes by… a stereotype, a blonde woman. Just sayin’…Report

              • Avatar Oscar Gordon says:

                One of my friends & I were talking about it today at my son’s birthday, and he was wondering if the woman thought the guy was writing Runes of Explosion & trying to cast the spell.

                There’s a fun magic system for ya.Report

              • Avatar Michael Cain says:

                The most trouble I ever got in on a commercial flight was caused by this bad boy. Sometime in 1992, on a long flight after dark from New Jersey to Denver, sitting in the aisle seat. Had Linux and Bellcore’s mgr windowing system loaded on it. Editing a batch of C code in one window, a “make” job scrolling busily in another, the clock ticking away up in the corner. Some guy headed back to his seat from the restroom stops at my shoulder and then yells up the length of the plane, “Hey! This guy’s got Unix running on a laptop!” Followed by a stampede of geeks, all headed for my seat. The cabin crew was not pleased.Report

              • Avatar Snarky McSnarksnark says:

                I hadn’t quite realized how many of us on this site are professional programmers (I’m a mostly-database guy myself).Report

          • Avatar Oscar Gordon says:

            That sounds exactly like the very first OOP class i ever took. It was billed as an introduction to OOP, but the reality was that it was an introduction to C++ for people who had experience with Fortran, Pascal, etc.

            I had to drop it, took me a few years before I was comfortable trying my hand at learning OOP again.Report

          • Avatar Brandon Berg says:

            The professor for my theory of computation class apparently did so badly that the class was rebooted, six weeks in, starting from the beginning with a new professor. Honestly, I couldn’t tell the difference, but I, as usual, mostly just spaced out in lectures and read the textbook at home.Report

        • Avatar Kazzy says:


          While @morat20 is correct that there do exist inept/incompetent teachers who can genuinely undermine a student’s ability to learn, the reality is that — by the college level and beyond — the onus is on students to learn the material.

          I’ve discussed this very issue quite a bit with my mom, who spent much of her career as a 5th/6th grade teacher before moving up to 7th/8th (and then on into admin). My take has always been that if you were to graph the responsibility of the student for his own educational outcomes, you’d have a pretty even diagonal line moving up from left to right. Conversely, the responsibility of the teacher for the student’s educational outcome would be the inverse, though perhaps up a notch or two on the graph as a whole. So basically, as students get older, they assume more and more responsibility for their learning and teachers assume less, though teachers never quite get to zero (and maybe never below 10-20%) while students might spend their first few years at or near zero.

          I can’t expect my students to adjust their learning style to fit my teaching style. Instead, I need to adjust for them. But students can and should adjust over time (and given certain instruction/feedback about their learning style and how to adjust for different contexts). By college, you have fully opted to be there, chosen your school, chosen your major, and choose most if not all of your courses and instructors. Get your shit together.Report

        • There are two many comments in this thread to read them all. Which ones will be on the midterm?Report

      • Avatar Richard Hershberger says:

        That jumped out at me, too. My guess is that she did what was required for her B.A. without looking into what graduate programs would expect. As an undergrad I heard an astonishing number of stories that amounted to someone never looking into what classes were required for the degree or when they were typically offered, and being surprised by the discovery that the last class they needed to graduate wouldn’t be offered for another year. I never had any sympathy for those guys. My freshman year I bought the full course catalog, studied the class requirements and the prerequisites for each, and compared them with the class schedule. I knew what courses I would be taking and when. This Ph.D. program second language requirement has the same odor to it.Report

        • Avatar Saul Degraw says:


          You had to buy the course catalog? Ours were distributed every year.

          I think it depends on how often these things are planned in advance and how specific a program is. During the height of the recession, there were stories of classes/sections being canceled at the last minute to cut costs and/or filling up way too quickly. Interestingly this is another example of a benefit of going to a selective/elite. I never had to deal with not getting a class I needed in order to graduate and getting into a class was relatively easy except for a few popular ones. The only class I needed to wait for until senior year to take was a popular history course called “Justice.”Report

          • Avatar Stillwater says:

            You had to buy the course catalog? Ours were distributed every year.

            So, you didn’t go to a state school, eh?Report

            • Avatar Will Truman says:

              I did and ours were distributed every semester.Report

            • Avatar Kazzy says:

              Mine was online.Report

              • Avatar Stillwater says:

                {You must be younger than me…}Report

              • Avatar Kazzy says:

                I didn’t go to college until I was 37.Report

              • Avatar Stillwater says:

                I was implying that when I went to school there wasn’t any internet.

                Add: well, undergrad anyway…Report

              • Avatar Kazzy says:

                Heh… sorry I ruined the joke. I knew what you meant. We had wired high speed internet in all the rooms. With multiple jacks!Report

              • Avatar dragonfrog says:

                We had a paper catalog, the size of a mid-sized town’s phone book, great stacks of which were placed in each department’s admin office for anyone to pick up.

                We’d get a letter giving us an assigned time when we could phone the amazingly awful voice menu enrollment system (UStar). So there you are at 3 AM on Tuesday, jittery with coffee, with the catalog and your list of preferred courses and sections laid out in front of you, improvised sudoku-style charts of section conflicts (if you can’t get into the Wednesday evening art history section then the 9 has to be in the rightmost column…).

                By 3:30 you might finally get through to the phone system, and you get to learn which of your first pick sections is full, then do the mad scramble through the calendar to find another section you can get into where the conflicts don’t completely crush your whole schedule, and start enrolling in that one before the phone menu system decides you must have fallen asleep and hangs up on you.

                I went to a lecture at one point, where the fellow giving the lecture stopped a few minutes in and said “If any of you are having inexplicable shudders right now – yes, I am the voice of UStar”Report

  6. Avatar Damon says:

    Dude bro to woman: This date’s going very well. Want to come back to my place and have a threesome with my sexbot and me?

    Woman: That’s disgusting!

    Dude bro: My sexbot doesn’t think so! She’s totally into it. Why can’t you be?Report

  7. Avatar Richard Hershberger says:

    Re second-tier schools and football, I don’t have a lot of sympathy, especially for the ones flailing around in Div. I. It seems to me that they want to pretend that they are Alabama or Notre Dame, with results that are mostly pathetic. This is all very well and good if it is just empty posturing, but it is very expensive empty posturing, with the students paying a big chunk for the privilege of not caring about a losing team. What is the point? I’m sure that if pressed, we would hear the usual talk about how the alumni is happy to pony up the big bucks so long as the school keeps its embarrassingly bad team, but would abandon it to its fate if it didn’t have that team. Even if true, this is simply pathetic.

    I considered the absence of a varsity football team a strong plus back when I was looking at colleges. I expect to reprise this in a few years, when my kids reach that age.Report

    • Avatar Will Truman says:

      It’s more directly positional than that. It’s not that Texas State thinks that with a good football program they’d be Texas. Rather, it’s that each one is looking directly ahead and behind them.

      So Texas Tech wants to be alongside Texas and ahead of Houston. Houston wants to be alongside Texas Tech and ahead of North Texas. Texas State wants to be alongside North Texas and ahead of Sam Houston State. I mean, they’d all love to be alongside Texas, but it’s only Texas Tech’s (and maybe Houston’s) standard of success.

      I think that the reason that Eastern Michigan is so reluctant to drop down has less to do with the University of Michigan and a lot more to do with Central Michigan and Western Michigan.

      Montana was looking at making the jump (from FCS to FBS) a few years ago, and I saw the various arguments for and against. The non-economic arguments in favor tended to directly relate to who they wanted to be peer institutions (Idaho, Utah State, Wyoming) and who the current state of affairs seemed to be putting them with (Idaho State, Weber State, Eastern Washington).

      Given that they have a stake in the success of their institution and I don’t, I tend to assume that their eye is on the ball. The five institutions that I’ve questioned the most are Idaho, UAB, EMU, UMass, and Louisiana-Monroe. Idaho is dropping down, UAB actually killed their program but the community response revived it, and EMU and UMass both have had faculty rebellions (faculty at UAB supported the revival). ULM seems to be the only one that hasn’t considered it. (Hawaii also considered killing their program and declined, they’re something of a special case.)Report

  8. Avatar Oscar Gordon says:

    So apparently Iran wants to export goods, but not so eager to import them? Wasn’t that the point of that huge nuclear deal, to lift sanctions?Report

    • Avatar notme says:

      The lifting of sanction was more of a benefit to them so they could sell oil. The Obama admin naively thought they would buy stuff from us.Report

      • Avatar greginak says:

        A bunch of Euro countries were working on deals with Iran. Those were going to happen sooner or later and much more likely sooner. I don’t remember anyone saying Iran would start buying all sorts of stuff from us. We actually still have some other sanctions on them but in the long term trade with them would be good.Report

    • Avatar notme says:

      Obama’s Foreign Policy Guru Boasts of How the Administration Lied to Sell the Iran Deal

  9. Avatar notme says:

    DOJ Decides It Won’t Call People ‘Felons’ Or ‘Convicts’ Because Hurts Their Feelings

    This might be funny if it weren’t so pathetic.Report

    • Avatar dragonfrog says:

      I chose to skip straight to the actual piece by the assistant AG, rather than the Daily Caller op-ed on it. And I’m all for the measure.

      I think the change makes perfect sense – if we want people to successfully reintegrate into wider society upon release from prison, we need to actually do things to support them in that attempt.

      And this thing in particular, (1) while it may have a relatively small effect, will almost certainly have some positive effect, and (2) costs exactly nothing, so the effectiveness / cost ratio is infinity.Report

      • Avatar Joe Sal says:

        Who gave the thumbs up for the DOJ to label anyone in the first place?Report

        • Avatar dragonfrog says:

          Well, the DOJ is going to have to write things about people who are and have been imprisoned using some kind of words – press releases, background checks, staff training materials, briefings to congress, etc. etc..

          Hopefully they’ll use consistent words across individuals and contexts. So they have to choose some kind of wording.Report

      • Avatar Jesse Ewiak says:

        But how will folks like @notme feel above former inmates if they don’t get to lord it over those people forever? I mean, aside from the lifetime of increased earnings, better place in society, et al.Report

    • Avatar Joe Sal says:

      DOJ wears their politics on their PC statist sleeves. How long ago was John P. Carlin rubbing elbows with SPLC, grouping anti-government views on par with “racism, bigotry” and other “despicable beliefs”.

      All these guys are is an agenda in the flesh, wearing a tie.Report

    • Avatar Art Deco says:

      “Felon’ and ‘convict’ amount to plain language which would be used by ordinary people. They also differentiate between the sort of banal virtue ordinary people have and which the felons and convicts have lacked. Ordinary virtue is ordinary, and a certain sort of bourgeois is motivated to establish his bona fides as someone who stands outside the ordinary. Anything will do, as long as its perverse and it would not have occurred to an ordinary person. A fondness for Frank Gehry’s architecture or a fondness for criminals – its all one. That aside, it’s a literary trope that the truly decent and honorable people are never who they seem to be. As for this change in terminology, it has to be taught and provides opportunities for petty scolds to exhibit socially sanctioned aggression against ordinary people.Report

      • Avatar dragonfrog says:

        “Person in prison” and “person who has been in prison” seem like even plainer language – if you sort all the word in English from most to least used, I think those expressions are made of words entirely higher in the list than “convict” or “felon”. And they are clearer as to their exact meaning too.

        Except I suppose “English” isn’t just “English” – there is regional variation. So just because I pretty much never hear ‘felon’ or ‘convict’ (as a noun) in use, only means they’re not common words in the part of Canada where I live.

        (You could make the English even plainer if you substituted “big house” for “prison” but that’s getting silly).Report

        • Avatar notme says:

          Yes, a convict means a person that was found guilty of a criminal offense or a person serving a sentence in prison. Or a felon meaning a person convicted of a felony. It’s so important not to hurt folks’ feelings these days the gov’t must get involved.Report

          • Avatar dragonfrog says:

            I don’t think it is mostly about “hurting feelings” as about actually shaping perception in a way that leads to productive outcomes.

            Another example – my city has recently switched its language in discussion of tranportation planning. ‘Cyclists’ became ‘people riding bikes’, ‘motorists’ became ‘people driving’, ‘pedestrians’ became ‘people walking’, and ‘transit riders’ became ‘people riding transit’.

            “How ridiculous,” one says, or “PC nonsense,” or “government busybodying.”

            But it has had a real positive effect – because when the topic was about how “cyclists” will have this infrastructure available to them, folks used to get all pissy about “those darn scofflaw cyclists” and “war on cars” and such nonsense.

            Because a ‘cyclist’, as subconsciously perceived by the citizens at the planning meeting, is defined by their cycling – they’re some mythic creature like a centaur, where their lower body is a bicycle and they have a lycra exoskeleton – they hate you because you usually commute to work by car, and it only makes sense to hate them back.

            But when the topic is how “people riding bikes” will have this thing available to them, things get much more sensible and calm – because a “person riding a bike” is a person first and foremost. It includes you, even though you’re not a “cyclist” personally, since you take easily twice as many trips by car as by bike.

            Words matter. They shape actions, even if perfectly rational vulcans we would see through to the Platonic meaning and be unaffected by the choice of words.

            Similarly, a “convict” is forever defined by their having once been convicted, and that’s the entire content of their soul – a propensity to commit crimes. A “person who has been in prison” is defined first by personhood.

            Scoff all you want – this is real and if the DOJ is serious about following through, it will have the effect of improving many people’s lives a little bit.Report