So, let’s take fields in which it is pretty hard to find a woman, and then let’s intersect that with a field with even fewer women.
You know, as a lady programmer, I know what the score is. There just aren’t that many ladies out there in tech (and especially not within fields like engineering or programming). Once I was introduced to a potential client as one of my design firms’ programmers, to which the client replied, “You don’t look like a programmer.” But then again, you know, you don’t really hear a lot about ladies doing much of anything besides being in movies and being popstars (I mean, every now and then I guess a lady doctor saves somebody or Hilary Clinton does something/anything).
So let’s do a little thought experiment: let’s take this notion of being a trailblazer and inspect it within the domain of technology-driven innovation. Let’s distinguish qualities and characteristics of the trailblazer into categories like entrepreneurship, media presence, and technical skill (in this case, primarily software engineering). Is it really all that surprising that there are only 8 women on that list? (One of whom appears as part of a pair for a lifetime achievement. One of whom IS LADY GAGA.)
There have been some (e.g. here and here) mostly well-reasoned and articulated replies since the list, but I’d like to point out something that seems to be overlooked or glossed over. This is not a coincidence that we’re seeing a percentage even more embarrassingly low than usual re: the gender gap while looking at a list about technological (and really, computer/internet-based) innovation.
My roommate (a dude) is also a programmer, and we sometimes talk about my gripe with the extreme gender gap in programming. He’s a totally great dude, but he is also a straight white dude who has had the privilege of being in a field where that is not just the norm (because society does that), but that it is pretty much all you can see as far as the sky stretches across the digital valley. We often get stuck at the point where he asks me why it matters, why there ought to be an equal (or even slightly less unequal) distribution across the field. And to some extent he’s right, maybe. Maybe it doesn’t actually matter; I’m a woman, I have the success I want in the field I want to be in, maybe other women just aren’t interested.
But why? Why aren’t they that interested? Is there really a biological imperative that runs so deep and strong that women can get themselves interesting in fields where they can create new technologies, where physical strength — often cited as why women can’t excel in some professions — will never hold them back, and where the strength of your work, at the end of the day in programming, will stand on its own regardless of what anybody else says? I wonder — back in the days of the race gap in literacy (well, at least when it was significantly more severe), did conservative white people defend the gap as just a matter of differing interest levels? Maybe people of color simply don’t care to read, so why should we make them?
See when you get there, you’re stuck. The only place you can go if you wish to maintain that the gender gap is not something that is due to bullshit sexism, and rather, something that is how the cards played out with respect to a distribution of capabilities is to say that women just don’t have the stuff to be as good. That of course any system has outliers, but on the whole, you know: there’s a reason why there aren’t that many women carpenters out there. See what you are committed to, when you say that women just aren’t naturally inclined to be interested in programming, is that at the heart of it, there is a biological difference between men and women that extends to their analytical capabilities. You’re saying that there are other physical differences between men and women, and maybe the brain is, at the end of the day, a physical organ, controlled by hormones even, and so maybe, maybe this is why things happened as they do. It’s nobody’s fault, this inequality, and maybe it’s not a big deal either.
And maybe you’re not a woman, you’re not a programmer. So you don’t know that that is preposterous and complete bullshit. You don’t realize that there is nothing special about the way a programmer’s brain works, that in a way we all move through life constructing algorithmic shortcut behaviors, and that this is all that a programmer does with code.
And beyond not realizing those things, you don’t feel the sting of the gap. I am very lucky to work with some great dudes. Seriously, they have all been absolutely cool and fabulous about making me feel like my voice is heard, like I can ask dumb questions without being judged for anything but the stupidity of the question, and like I can crack as many poop and Star Wars jokes as I want to. Most dudes I talk to in the field are actually, once I start talking shop and they realize I’m part of the in-group, pretty awesome.
But it stings just knowing the gap is there. It stings knowing that I am very lucky, it stings feeling like maybe I really am just some freak of an outlier, but mostly, it stings not having very many ladies to swap stories with. The gap is so severe that I can count the number of women I know who I can talk some shop with on one hand. One finger.
This has to stop. And not just because I’m a little lonely and would like for my company to have a reason to call the dude in to fix the tampon machine in the bathroom at work, and not just because equality is important and we need to be empowered with the tools that are being used to make the future: ladies, writing code is fun. It’s so much super fun. Whatever that’s making you say “I would love to learn, but my brain just doesn’t work that way” is basically like your mom who wouldn’t let you stay up late to watch Twin Peaks on tv. You’re missing out! Don’t! The solution to the gender gap in technology is not to just start talking about it a lot and create awareness programs; WE ARE AWARE OF THE GAP BUT WE HAVE BECOME COMPLACENT TO IT. Don’t be complacent. If you’re reading this now, you have a machine and access to a network that will allow you to begin doing something about that gender gap right now. Let’s get to work, ladies.
I was meaning to write about switches vs. function pointers this week but got sidetracked by a friend challenging me to run an Arqua analysis on the Linux kernel. After cloning the kernel git and building an “unconfigured” kernel for arm with the appropriate Makefile patches I was struck by a storm of bugs in Arqua. After days of fixing my own unpleasantly unstructured code and adding new visualisation elements to Arqua, I managed to produce the following image of the recently tagged v3.5-rc4:
Moving in even further we get this:
Now, I am extremely naive when it comes to the Linux kernel, so this is strictly an objective analysis of the structure only. The result also assumes that the directory structure is a reflection of the architecture. This is of course not necessarily true, but the way Arqua operates (and in my personal opinion a sound way to structure source code).
“Yes, I’ve forgiven them, because I’ve been forgiven many times,” he said. “My country’s been good to me … This country is my house, it’s the only home I know, so I have to be able to forgive — for the future, for the younger generation coming behind me, so … they can understand it and if a situation like that happened again, they could deal with it a lot easier.” - Rodney King
I was five years old and living near downtown Los Angeles when Rodney King was beaten. I remember seeing that video play over and over and over again. I remember the riots. I remember the helicopter engines and searchlights keeping me awake at night. I remember not understanding why any of this was happening, why the people who were supposed to keep the peace were hurting this man, why they seemed to be at war with the people they should have been protecting. I was five years old and living in a very poor part of Los Angeles, so race was a thing that was happening all around me, and yet until Rodney King, race was not a thing I knew divided people in any significant way.
I don’t think Rodney King wanted to be anybody’s symbol for the fight for equality and justice, but he stepped up when the call to action came (because really, what choice does anybody have?). It has been twenty years since the beatings and the riots, and between Trayvon Martin and the Occupy raids, I think it’s obvious that we still have a lot of work ahead of us.
We would all do well to remember what Rodney has taught us, not just about the justice we all deserve, but also the forgiveness we are capable of for those who have yet to join our fight. Forgiveness is an act of compassion that allows us to remember what has happened while moving forward towards change. It’s remarkable to me that for all Rodney King has been through, he’s able to practice and recognize the importance of forgiveness and empathy, that he can articulate that antagonizing and holding anger in our hearts towards the human manifestations of oppression only holds us back from being capable of seeing the future we are working to achieve.
Watching that video of Rodney King when I was five years old was heartbreakingly confusing, and I can only imagine what it must be like for the children of today to see the perpetuation of that violence. But we must forgive so that we can understand, so our future generations can understand, and so we can, indeed, deal with it a lot easier.
Rest in peace, Rodney King, and thank you for everything that you have taught me.
consider the error that watson, the jeopardy-playing supercomputer, makes at the end of its first round of otherwise flawless competition.
notice that it’s not just that watson gets the final jeopardy question wrong, but how he gets it wrong that makes this particularly interesting—naming a canadian city when the answer was asking for an american one.
why that makes this example so interesting is that it reveals a fundamentally important part of how watson works—it’s not that watson understands the questions presented before him, but rather that he understands how to find the answers in a systematic and algorithmic way. watson, and many other artificial intelligence systems like him, are designed to take a natural language expression of a certain grammatical structure (in this case, one presumes that watson is trained to understand the particularly unique structure of a jeopardy prompt, which incidentally, is actually presented as the answer which the player has to guess the question to) and parse this structure for its constituitive parts (e.g. generally basic syntactical roles like part of speech—noun, verb, adjective, and other more exciting things like semantic association—like synonyms, antonyms, related terms). to retrieve the relevant information related to the structure, the system then takes the parsed structure and then queries a similarly tagged data set to retrieve the most likely relevant terms—but this itself is not language comprehension, or understanding.
as a former linguist working primarily in the field of semantics, where one is really concerned with understanding what understanding a word or an expression means, this sort of thing was what i would point to to prove that computational linguistics would never yield a truly intelligent artificial being with just algorithms and syntactic parsing.
and yet it now occurs to me that what is most interesting about watson is how much he does get right without ever even having to understand what “blue” or “wine” really means to you or i. because while we have been long in pursuit of designing machines which act like us, trying to teach them to understand like we do (and asking ourselves what it really means when we understand something), perhaps another question can be raised: could we act like machines? could we understand as machines do and still make quite a lot of progress in our individual pursuits? are we not already just blindly navigating through the infinite streams of data, making significant process while never truly understanding?
alan turing, father of modern artificial intelligence, proposed in his 1950 paper “computing machinery and intelligence” that rather than ask “can machines think?” that we ask “can machines do what we as thinking entities do?” it’s becoming increasingly clear that not only that they can, but that they can so much better than we “as thinking entities do” when it comes to many tasks.
perhaps, then, what watson’s error reveals is not a new fact but a new question: can we as thinking entities do what machines as computing entities do? or rather, taking a cue from turing himself, can humans compute understanding/understand computing?