Author Archive for Nik

Gender and the semantic web

Published
by
Nik
on August 26, 2008
in Uncategorized
. 0 Comments

Caveat: I’m not entirely sure I understand the semantic web, but I do think it’s a fascinating idea and that there is some great work going on in this area laying the foundations for what promises to be a fantastically exciting development for the web. Despite my relative lack of knowledge, I’m going to express an opinion anyway. Please do correct me if you can.

So the semantic web is a mix of technologies which essentially aim to make the web more readable by computers. In order to do this, coders are trying to come up with technologies that go beyond HTML’s description of content to a description and classification of things themselves. For example, there will be defined, in this semantic web, an entity named “Barack Obama”. It will have attributes, such as gender and party affiliation, and various elements of his life history too. Crucially, it will also have relationships. So the web will know, for example that <Barack Obama><is the husband of><Michelle Obama>, and <is the 2008 presidential candidate of><the Democratic Party> and so on.

Much of the power of the semantic web comes from defining what kinds of these triples (<object> <relationship> <object>) exist, and from filling out the definitions of things so that computers can search them. The object is to come up with a system in which, for example, I could ask my computer to find out who I am meeting in three weeks time, when their birthday is and what the company they are working for is currently up to that I should know about. While much of this information is available to people searching the web, computers are much worse at interpreting the text form in which it currently resides and so we have to do the legwork of looking for it ourselves. The semantic web aims to make more of the web machine-readable.

After that long introduction, here’s the problem. A researcher from Austria believes that the semantic web is in danger of becoming gendered because of the gender of people who are building it. As I have noted before, men are way more likely to get their fingers dirty with the business of tinkering with hardware, writing code etc. And so it is that the semantic web community who are setting out the relationships, entities and protocols that will define this new web are mostly male.

Are the male and female points of view so different that the basic ontologies we give to computers to make sense of the world might be inadequate if only one gender contributes? I am struggling to think of ways in which this might actually make a difference, but then I am somewhat handicapped by my (60%) male brain. Perhaps someone can enlighten me?

It does, however, moderate my earlier optimism that the new web would be female-friendly and easy to use. For all that coding is no longer as important for the average user, it still builds all the technologies we love. Perhaps women will find it just as easy to use as men, but the fact that they are routinely not involved at design time may mean that gendered ways of thinking are hardwired into new technologies.

(This obviously applies to other groups who are excluded from the design process. There is a lot of interesting work on how basic philosophical concepts, such as ‘knowledge’, differ between ethnic groups. The same arguments as applied to men above may well apply to north Americans, or white people (not that most programmers are white, but you get the idea, hopefully). This blog is about gender, though, so I’ll stick to the point for the time being.)

The end of male geekery

Published
by
Nik
on August 11, 2008
in Uncategorized
. 2 Comments

A conversation I was having with my girlfriend’s father sparked a thought about how computing structures can lead to gendered outcomes. He was an engineer in Germany, and for his undergraduate dissertation, he programmed a computer to play a game that sounds like a cross between connect-4 and Go. He did it on punch cards. We talked for ages about the virtues of Fortran versus machine code, C versus Pascal and other geeky things I pretended to understand.

I would describe myself as a novice programmer at best, but like many guys I know I spent significant portions of my youth in the 1980s tinkering with IBM compatible computers that mainly ran MS-DOS. The first thing the computer would show was not a friendly desktop with windows and icons, but a prompt:

c:\>

There is something about this blankness that means you have to begin to get under the hood of your computer, and have a dim idea of how the bits work and communicate with each other. And in fact it encourages you to tinker and tweak. I may have messed up my father’s computer so badly it needed an engineer to come over and spend half a day fixing it, but I learnt a lot. Now, this sort of thing is clearly gendered. The male brain loves getting stuck into machines and playing with them, whether these are cars, computers or bikes.

I am going to resort to purely anecdotal evidence, so if there are any ladygeeks out there who love to program and know their way around autoexec.bat, then apologies, but hopefully what I’m saying will still ring true to some extent. I didn’t know any girls in the 80s, but even since then, I can only think of one girl I have met who could (as far as I know) be remotely interested in the conversation I was having with my girlfriend’s father. For a long time computers and the concept of geekiness were organised around the idea that geeks could program, that they could code their way out of trouble and would take the time to run through system files tinkering and tweaking to accomplish what they wanted to do. And during that time, computer geekdom was a resolutely male domain, as it was largely men who actually enjoyed doing this sort of thing.

That began to change over a decade ago, when Windows 95 banished the C prompt to dire emergencies only. But emergencies still happen, and the traditional male geekiness is still called upon. However, the move towards cloud computing is going to strike another blow at this predominantly male domain. Processing power, programs and the problems they cause will no longer be stored locally, and will no longer be sorted out locally. Your laptop, like Belinda’s Eee PC, will not be a fully-featured powerhouse, but mainly a way of accessing the internet, where most of your computing needs will be met by a variety of services provided by Google, Amazon, Microsoft et al. With this model, there is so much less that can go wrong, and correspondingly less need for the male geek types.

And this unleashes productivity, too. Instead of spending countless hours fiddling with registry files and secretly enjoying it, the focus of computer wizardry shifts to the wonderful things that you can actually achieve with them, and this field is not gendered. Women are just as interested in technology as men, now that the technical fiddling is no longer required (most devices actively discourage you from opening them up and tinkering), and in every way, they are an equally important part of the technology marketplace. Mobile phones, content creation and social media are all areas where women are just as likely as men if not more so to participate and produce. These are also areas where the internal workings of the system are best hidden from its users.

The new geeks will be male and female, they will have no need to know how computers function internally, but they will be masters of manipulating symbols on screen and in the cloud, and the things they accomplish will be awesome.

How female are your browsing habits?

Published
by
Nik
on August 11, 2008
in Interesting
. 0 Comments

This is pretty cool. Click the link and you are taken to a site which analyses your browsing history to assess how likely you are to be male or female. The idea is that men and women have different likelihoods of visiting certain websites, and so using that information backwards, you can tell a user how male or female their browsing is.

A reader of Andrew Sullivan’s points out:

An interesting irony of it is that the highest ratios you can get are from gay websites (adam4adam is 4.13 as a commenter points out) which ironically means that, in that world, a lot of gay men get a 100% while most heterosexual male get at least something like a 10% female side.

I got 69% male at home, but on my work computer I got 57%, which means I’m fully 12% more in touch with my feminine side at work. I do live with my girlfriend, but also two gay guys, while my office has mostly women and straight guys. Maybe the more feminised context of the office affects what I am interested in online, and probably much more besides?

The mamas, not the papas

Published
by
Nik
on July 10, 2008
in Uncategorized
. 2 Comments

I was playing with some old survey data the other day and found some interesting effects of life events on how men and women use technology. We asked about the number of people who had done various activities online in the last 6 months.

On the whole, the numbers for men and women for most activities were broadly equal. But for certain activities, the presence of children changed the picture completely. Take blogging. Among those in the pre-family lifestage, the proportions of men and women who were bloggers were about equal. If we now compare these figures to those who have children, the proportions who blog are similar for women, but men with children barely blog at all.

It’s a similar story with social networking sites. Mothers are almost as likely as not-yet-mothers to be active members of these sites. Fathers, on the other hand, just don’t seem to be as interested. Is this because mothers tend to have more time at home to spend on these sites? Given how much social networking usage seems to go on in offices, I doubt it. I think part of it is that mothers are more closely involved with their children’s social lives than fathers are, and so might have been persuaded to get a Bebo profile somewhere along the way. Might this also point to women becoming less set in their ways as they get older, perhaps again because they are around younger minds for longer?

Or perhaps it is simply a demographic effect of mothers generally being younger than fathers, and the difference in web 2.0 usage will disappear as younger cohorts come into parenthood and the fathers are equally tech-savvy as the mothers. This data is about a year old, and when I get a moment, I intend to check these results against more recent waves of the same survey to see if they show signs of shifting this way. [Disclosure: this data comes from (my employer) the Future Foundation's proprietary research.]

Finally, most people who have bet or gambled online are men without children. Most of them give up when they have kids, however. Young women are very unlikely to bet online, but when they have children, something seems to happen, and they become three times as likely to have a flutter.

All of the above are just observations from the data. I’d love to know your hypotheses for why older women seem to be more interested in technology than older men.