Category: Tech community

Posted on

US-based websites and COPPA

Alex Sutherland, who is not yet 13 years of age, told Google+ his date of birth and promptly lost access to his Gmail account.

I’m not posting this to join any obnoxious blamestorm aimed at Alex or his parents: it sucks he lost his email archives and I hope that his parents are able to get it back for him. It sucks he had his trust breached and there’s no getting that back for him.

But I’m mostly posting because people are seeing the provisons of the US Children’s Online Privacy Protection Act (COPPA) for the first time and saying “pfft, not that hard to comply, why ban under 13s at all?” There was an illuminating comment on Making Light that is helpful there:

COPPA has a lot of “common sense” provisions which no doubt sounded great from the point of view of legislators and parents, but which are pretty appalling from the point of view of the operator of a Web 2.0 service. They’re burdensome enough, that, to my knowledge, only sites intended specifically and exclusively for children trouble to implement them. That is, no Web2.0 websites operating in the US permit users under the age of 13, except for specialty children’s sites. Not Google, not Facebook, not MySpace, not Livejournal, not Twitter, not Flickr or Picassa or Photobucket, not any web service here in the US.

Why? Well, you know how when you have a problem with your Gmail, you can pick up your phone and call Google’s tech support line? Ah ha ha ha. Right: no such thing. Well, one of the provisions of COPPA is that there has to be a phone number through which parents can call the service, as well as an email address at which they can email the service. Google doesn’t particularly want to have to pay operators to be standing by. No Web2.0 startup wants to be staffing a phone number open to the general public.

Google also doesn’t particularly want to figure out how to fulfill the provision of writing a statement as to what “information it collects” from (minor) users, since it allows users to type absolutely anything they want into those email bodies. Among sites for children, the open-ended TEXTAREA form field, like the one I’m typing this comment into, are seen as threats; highly structured or brief forms of input — pulldown menus and short text fields — are seen as safer. That prohibits most interesting Web2.0 applications.

Now Google is pretty big, it could afford to solve this if it wanted to, but has decided not to. But I think this is an issue worth knowing about in general: this means that children under 13 can’t participate in the Web as we know it today, essentially, because COPPA means that it’s prohibitively expensive to allow them to use websites that allow free-form content. Opinions might vary on whether this is a good thing (I certainly don’t think so, although I’m also not planning to turn my son loose on Google on his sixth birthday either), but it’s a thing.

Posted on

A petty rant

See the title? Consider yourself warned.

OK, geek culture. I am ambivalent with claiming particular things as being geeky or not in the first place, because half the time I fall outside it. (I’m not a night person, I don’t especially identify with or even like cats, to give some trivial examples.) And if it excludes me, it must be wrong. Duh.

But perhaps we should claim being petty and pedantic. Here’s my line in the sand: you do not count discrete things starting at a zeroth thing. Well, if you do, I say it’s not geeky.

There’s a sort of a general understanding that “geeks count from zero”. Here’s where it comes from: in many programming languages, arrays begin at zero, so array[0] retrieves the first item in the array and array[1] the second and so on. This is actually somewhat confusing, and results in plenty of off-by-one errors (for example, if an array has length l, then one is tempted to ask for the last item as array[l] when it’s actually array[l-1], and forgetting that is not at all uncommon).

It has meaning: it’s fairly obvious why this is done in C, it’s because elements in a C array are stored in contiguous memory and the name “array” is already a pointer to the first memory address. Say the array starts at memory address 7, then to access the items of the array you would do:

  • array[0], which is an alias for *(array + 0) or *array, ie, find out what is at memory address 7 (the dereference operator * means “look up what is at this memory address”)
  • array[1] or *(array + 1), ie, find out what is at memory address 7 + 1 = 8
  • array[2] or *(array + 2), ie find out what is at memory address 7 + 2 = 9

And so on.

Dennis Ritchie’s The Development of the C Language shows that this notation is inherited from C’s precursor B.

And it’s not a silly way to count some things. It’s the same way we count age in the Western world: at the beginning of your first year, you are age 0 and at the beginning of your second year, age 1, and so on. The first year of someone’s life begins at birth[0] as it were.

But it’s a silly way to count objects. It is not more geeky to count, say, two apples as a zero/zeroth apple and a first apple. You could perhaps refer to the zeroth apple offset, ie, the point just before the beginning of the first apple, if you had reason to refer to that point (I never have).

Try as I might, this has always bugged me about the history of linux.conf.au: CALU was not the “zeroth because it’s geeky!” linux.conf.au. It’s like someone made that up specifically to annoy me personally. I will find you, whoever you are, and I will take your zeroth apple away from you and unlike you, I will have my first apple. And I will enjoy it.

Posted on

Impostor syndrome and hiring power

This article originally appeared on Geek Feminism.

This is an Ask a Geek Feminist question:

What are some ways in which I can avoid rejecting people who suffer from impostor syndrome when they apply for a job?

I’ve recently been promoted to a position where I’m somewhat responsible for hiring people. I would like to increase the diversity of new hires, and so I’m more likely to put applications from women through to the interview stage.

Following that though I don’t want to lose out on quality applicants as they are modest about their achievements and abilities, due to impostor syndrome or otherwise.

Giving an automatic “+10 kickass” to every female applicant as they may suffer from impostor syndrome seems to be a strategy without much merit. Getting everyone to exhibit their full potential is clearly the better solution.

Can you suggest interview strategies that would create the environment in which women (and indeed anyone) will be better able to convince me of their suitability for the role?

I don’t have so many specific interview strategies, but I’ve got plenty of ideas for hiring strategies in general, I hope you can adopt some of them and perhaps our commenters can talk about the interview.

First, a should be obvious: a +10 kickass bonus may be illegal discrimination in your geographic area. If it is, definitely don’t do that.

With that out of the way, let’s talk about soliciting applications. Now, there’s a couple of things that stop some women at this point. First, there’s a tendency to regard themselves as underqualified for perfectly suitable jobs. Next, there’s concern that they needn’t bother, as a woman’s name will cause you to discount their resume. Some suggestions:

  1. get your signalling right. You want to say “women friendly employer” in your advertisements without discriminatory pro-women statements. This at least gets you past the “I’m not a man” part of impostor syndrome. Here’s some things you should be doing:
    • advertising all relevant open positions on a women’s job list such as, say, LinuxChix’s jobposts for open source jobs. This at least shows that you aren’t actively avoiding women applicants.
    • including on your full ads the “equal opportunity” boilerplate you might be able to find on other local job ads
    • including information on the “Careers” section of your website about your carer leave, your retirement contributions, your shared sick leave pool, your friendliness to part-time employees if any of these hold

    Not only are these things attractive to many women (and yes, some men as well) in and of themselves, they also signal in various ways that when you picture your new hire, the picture isn’t young, white, able-bodied, male, etc etc.

  2. if your employer has recently had a similar (especially perhaps slightly more junior) position available, get the resumes of the people who were considered the better applicants from the hiring manager, HR person or recruiter, and re-consider them for the new position (probably there would need to be some kind of process of tracking and perhaps re-application here, but I’ll handwave that problem to you).
  3. consider internal employees in more junior positions as potential applicants. Depending on the size of the company, other managers might be able to recommend people to you who are overqualified for their position (or possibly not, if they are getting good work from them)
  4. consider whether you really need experience that skews very very male. For example, does someone have to have open source development experience? Are there alternative ways that someone could have learned the skills you need?

And now for considering applications prior to interview:

  1. you may not be able to say you’re doing this, but in order to avoid bias on the basis of gender or other demographic characteristics, for as long as possible in the process keep names off resumes. Have names and addresses scraped from resumes by someone before you see them, and do as much ranking as you can prior to finding out the names and details of the applicants.
  2. avoid judgements about cultural fit at this stage.
  3. there are reasons companies rely on the recommendations of existing employees, but for each open position, try and select some applicants for interview who didn’t come in via the company networks in order to avoid duplicating your company’s present demographic by hiring all their friends

In the interview itself here is a strategy for getting people to talk about their successes when they are susceptible to impostor syndrome (note that any candidate might be part of an oppressed group, so don’t limit these to women candidates): ask about something the candidate did that benefited someone else. How did they save their company money or helped a team member learn what they needed to know? Present them with cooperative scenarios where they need to help you or your employer do something as well as or instead of competitive scenarios where they need to prove they are the single right person for the position. If anyone can flesh this out to specific example questions in the comments, that would be useful.

I strongly recommend reading Women Don’t Ask by Linda Babcock and Sara Laschever for good solid information both about women’s negotiation and self-promotion strategies and why they use those strategies, namely, that competitive and aggressive interpersonal strategies are simply not effective for most women because of negative responses to perceived aggression in women.

Posted on

Idea for the taking: Freedom Fest

Note: this isn’t commentary on linux.conf.au 2011 in particular, I’ve been thinking about this vaguely for a couple of years and it’s time to release the ideas into the wild where someone might actually do something about it. Also, it should in no way be read as a commitment to me actually ever doing this. Steal this idea.

Consider the linux.conf.au miniconf system, in which there are single-day community organised streams occupying the first two days of the conference. Now… consider that as its own conference. That is, I envisage an Australian open source conference that has the organisers take care of the boring chores centrally: insurance, registration, venues. Then the space is provided to representatives of various communities to run their own stream. Because I am a control freak, I would probably also do the following centrally:

  1. provide a common timetable for all rooms, to allow attendees to move around between talks
  2. provide conference volunteers to act as session chairs, in order to make sure the talks actually end on time
  3. check people’s program, and take away slots if they are filled with things like “TBA” and “Lightning Talks TBA”. Shorter streams than a full day should be possible.

(And yes, needless to say, I would want some kind of central management of conduct/harassment policies too. Which would be hard if the policy is to apply to talks that aren’t centrally selected. But then, LCA has this problem with miniconfs already.)

It would also be important to be more flexible on registration than linux.conf.au is (almost always, there were small exceptions in 2008), that is, to allow people to attend for a single day without paying for the whole event. Generous provision of hack or unconference space would be necessary!

This would mainly advantage communities that don’t overlap really well with LCA. Typically if they try and hold a miniconf they struggle both to get core members to attend (because they have no interest in Linux or in the main program) and to get LCA attendees along. One day registrations and the brand distinction would help a lot. It would also perhaps bring smaller communities together for the first time. The main disadvantage would be adding another major conference to the calendar, potentially competing directly with LCA if events like Haecksen moved to it. (People who use annual leave to go to conferences will likely only go to one long one.) If it actually replaced the first couple of days of LCA, perhaps not so much.

Posted on

"Just leave if you don't like it"

A note on the arguments following Mark Pesce’s keynote. There’s one in particular that bugs me: “just leave if you don’t like it.”

The thing is, it isn’t normal at linux.conf.au (unlike at a Bar Camp) to just exit a talk from, say, the front section in the middle of a row. Unless you are at the very edge of the room, it’s considered rude to just leave, to the point where some speakers or session chairs might actually yell at you. (I had university lecturers do that.) And I suspect LCA, for organisational reasons as well as for speaker comfort, would rather not encourage an atmosphere of people just traipsing in and out of talks through the centre of rooms. So… the environment is (somewhat) coercive: if you don’t like the talk, you have to be actively rude to the speaker and the rest of the audience in protecting yourself from the talk.

If an environment could be created where someone could leave a talk from any place in the audience with a minimum of fuss and without risk of social retribution, and if people really did do so for all kinds of reasons, and thus an exit during Pesce’s talk would not have been immediately visible to everyone as “I have a strangulation phobia, if you would like to bother me in future, please mime strangling me”[1], I’d at least take this argument seriously. But in the LCA context it currently equates to: “don’t like the talk? embarrass yourself and be rude to the speaker!”

(Note to LCA people: I have a comment policy, and if your comment annoys me I won’t publish it.)

[1] I do have a strong reaction to strangulation, although probably not technically phobic, and if anyone uses this information to harass me even as a joke[2], they will not be my friend thereafter.

[2] People who have physical triggers, like having sharp objects pointed at their eyes, or disliking their neck being touched without warning, and who admit them, do suddenly find that half their acquaintance immediately does that to find out what happens. Consider yourself warned about what will happen.

Posted on

linux.conf.au 2011: dinner activities

linux.conf.au has a charity auction over dinner. There are various failure modes:

  1. it’s a year of big corporate budgets, so bidding reaches about $5000, no one else can compete, and then it stops
  2. it’s not a year of big corporate budgets, so bidding reaches about $500 from a private individual and then it stops
  3. bids aren’t high enough, so there is some pressure for someone to donate something precious. This was how Bdale Garbee ended up being shaved by Linus Torvalds at linux.conf.au 2009. This can be fun, but it also at least tweaks and sometimes outright triggers people’s fear of coercion (having a lot of drunk people screaming for your beard is definitely coercive).

There’s always been a tradition of large consortia of private individuals forming to try and solve problem , in recent years these have even tended to win. The trouble then is what happens to the money that was pledged by losers: at lca2011 (and I think lca2010 too, but I wasn’t there) bids aren’t revocable. The donated money stays donated, the only question is whether you get a prize associated with it.

So far so good for money. And now for entertainment, as Rusty posts. The trouble with lca2011 was that the auction consisted of people walking up to laptops and having their donation amount entered and associated with their team. Running totals were displayed on a graph, but spectacle was lacking.

The ritual humiliation of Linux celebrities does have something in it. But, no more screaming for people’s beards. I think it would be much more appropriate, and probably fun, to organise something in advance to occur at the dinner, with celebrities volunteering. The closest model would be lca2004’s dunking of Linus Torvalds (which was organised in advance, the pressure placed on Torvalds to participate I can’t speak to but he gives the appearance of generally enjoying some mild organised humiliation for the benefit of charity).

Say, as an example, that five developers compete to throw three-pointers (actually, this is probably too hard, in addition to being difficult to stage at a dinner, but never mind). Then there’s a very short pre-planned set of auctions for things like being able to take steps forward to start with, extra shots, probably culminating in the right to substitute, together with a simple “highest amount, yay!” kind of contest. At least one or two bids to allow your celebrity to increase the challenge facing an opponent. Probably five rounds of shots total with bidding in between. You could probably solve some obvious problems (like everyone backing Torvalds or betting against him or whatever) with simple transparent manipulation: Linux Australia increasing their matching donations when tables back their assigned celebrity, or something.

Finally, since this is a developer conference, there should be some kind of application allowing people to pledge using their phones from their tables.

Posted on

linux.conf.au 2011: lightning talk take homes

As usual some rather important things went on in the lightning talks.

Rusty Russell got irritated at Geoff Huston’s “IPocalypse” keynote (which argued that the last minute no-options-left switch to IPv6 runs the risk of IPv6 being outcompeted by a closed solution) and he got coding. The result is a CCAN module (so, C code) to support simultaneous IPv4 and IPv6 connections, thus not penalising either. He’ll fix the dependency’s licence shortly. It might not work perfectly yet.

Donna Benjamin is trying to raise $7500 to get The National Library of Australia to digitise The Dawn, Louisa Lawson’s journal for women from the nineteenth century.

In intellectual property news (specifically, anti-stronger IP news) Kim Weatherall wants us to worry about the Anti-Counterfeiting Trade Agreement, which Australia will likely ratify, the Trans-Pacific Trade Agreement, which it would be really great to oppose, the impending result of the Federal Court appeal in the iiNet case, which iiNet may lose, and even if they don’t there will probably be legislative “three strikes” discussion about copyright violation.

Posted on

linux.conf.au 2011: Day 1

Slow first day for me. I had a stressful Sunday getting a toddler to the airport on my own and Andrew has just flown in from the US.

We weren’t very impressed with our hotel, iStay River City. For starters, it has extremely limited keys. Many, but not all, rooms have two keys, which would be hard enough with four adults per room, but one of the keys for our room is missing, which means one key (and suggests that somewhere out there a former guest still has a working key to our room). The hotel reception wasn’t even sympathetic. People steal our keys all the time! What else are we to do?!

There’s no way to leave a key with reception and get yourself back into the room unless you have a second key to the room. There are buzzers for the rooms, but the reception smilingly conceded that it does only get guests into the lobby. You have to go down the lift yourself to get them up to the room. (Interestingly, this has meant with a lot of confusion from other LCA attendees. “How hard is it to make a new keycard?” Bad assumption. They are using keys, as in, those chunks of metal with notches in them.)

There’s also several things broken in our apartment: a couple of lights, the phone, the bathroom fan.

Anyway, after a restless night, LCA! I mostly spent time at the Haecksen miniconf, although partly working on my laptop in an introversion bubble. I wasn’t really ready, after the travel and the settling in, to sit down and listen to talks well. Some talks I did catch in whole or in part:

  • Pia Waugh Applying martial arts to the workplace: your guide to kicking arse
  • Brianna Laugher An Approach to Automatic Text Generation
  • Andrew Gerrand Practical Go Programming
  • Noirin Shirley Open Source: Saving the World
  • Donna Benjamin We are here. We have always been here
  • Valerie Aurora and Donna Benjamin Training Allies (workshop)

I didn’t really fully follow any of them, except for Training Allies, which is of professional interest to me now. (More on that later, I guess.)

Posted on

LCA2011: unofficial planet

In lieu of an official Planet site for LCA2011, I’ve set up an unofficial one. http://lcaplanet.puzzling.org/

Q. I want to be on it!

If you would like to add your own blog to the site, please see http://conf.linux.org.au/wiki/Planet for info and contact me in comments if it doesn’t work out.

Q. What happens when there is an official planet?

I’ll add 301 redirects as appropriate as soon as an official planet is announced.

Q. Wait, can this BE the official planet?

LCA organisers, I’m happy to be the official Planet if it makes things easier for you. Get in touch with me.

Posted on

Linux Australia Council statement

I’ve accepted a nomination to stand for Ordinary Council Member of Linux Australia in 2011, and here’s my statement. If you have any questions and so on, please let me know! Ideally ask them on the linux-aus list but here is OK too.

Voting is at https://www.linux.org.au/membership/ until Jan 26th. You can sign up for free membership there at any time. I actually don’t want people to sign up for the sole purpose of electing me, but if you’re an Open Source developer or community member in Australia the organisation is designed for you.

I am keen to help Linux Australia with community and events in 2011 as an OCM. I am especially interested in events, especially liaison and helping Open Source events in Australia, which are increasingly affiliated with LA, share knowledge and resources. I’m also interested in being part of discussions about Linux Australia, Open Source in Australia and young people: entry level community involvement and entry level career outreach for the next generation of Open Source hackers.

Selected previous Open Source community involvement:

linux.conf.au:

  • chair or co-chair of paper selection committee 2008–2010
  • member of paper selection committee 2007–

LinuxChix/Haecksen miniconf:

  • (co-)lead organiser, 2007–2008
  • founder, 2007

AussieChix (Australian chapter of LinuxChix):

  • microconf co-organiser, 2008
  • systems administration, event organisation, website maintainence, 2007–
  • founder, 2007

Wikitravel:

  • administrator, 2005–
  • active contributor, 2004–2006

Sydney Linux Users Group (SLUG):

  • Secretary, 2003
  • Ordinary Committee Member, 2002