Jaunty notifications

I missed the best known feature of Ubuntu 9.04 in my roundup, the new notifications. There’s lots of information around about this, see Mark Shuttleworth’s announcement, LWN coverage of development debate and Glyph Lefkowitz’s criticisms for plenty of information (Matthew Paul Thomas comments at Glyph’s place extensively).

I just want to pass on one tip, which is how to turn off notification of new messages in the Pidgin instant messaging client, because good grief, is that annoying. My typical use of IM is actually somewhat asynchronous, as in, I will flip over to it and check for messages periodically, chat a bit, do some work, and back. The notification workflow assumes I want to know about every message as soon as it arrives, which I most emphatically do not.

Other people of my opinion can configure Pidgin’s notifications by going to ‘Plugins’ and then looking for ‘Libnotify Popups’, which you can either configure to be less talkative, or disable entirely. In an ideal world that would be slightly more discoverable: I expected the configuration to be associated with the notifications, not with Pidgin itself, and certainly not with a plugin in particular.

Ubuntu 9.04 (Jaunty Jackalope)

I didn’t do one of these for 8.10 Intrepid Ibex, I was a little worried to find not one but two links to my Hardy (8.04 LTS) beta writeup as reasons to be exceptionally cautious about upgrading. I discovered I did not want to be an influential commentator on Ubuntu, even for trivial values of influential.

However, I should note in the interests of fairness that both Intrepid and Jaunty have been vastly better experiences. Specifically, things that seem to remain reliably working across upgrades for me without requiring reconfiguration now include suspend and resume, sound, wireless and external monitors. I have not found Jaunty less stable than Intrepid and both are fine for my terminal+mutt+Firefox+LaTeX+Python+Pidgin use.

Fixed in Jaunty:

Still broken in Jaunty:

  • gnome-typing-monitor still doesn’t regard being suspended or hibernated as a typing break. This one is one of those bugs that gets triaged as ‘Low’, never gets fixed, and drives me completely insane, There is nothing like starting up to get some work done and immediately being locked out for a typing break. It’s bug 122038 against Ubuntu, bug 492662 against Debian and bug 430797 against GNOME. It’s nearing its second birthday. I might throw it a party and then never use gnome-typing-monitor again.
  • In both Intrepid and Jaunty I have occasional kernel panics resuming from suspend (I don’t know about hibernate, I use it extremely rarely because my install of Ubuntu 7.10 gave me a swap partition that is too small to hibernate into unless I close all interesting programs, and I haven’t bothered fixing it). I should attempt to file this bug, and I will get around to it any, uh, release now. The main thing stopping me is that it often happens when I resume on a train, which means I don’t have ‘net access, which means keeping copious notes for a later report, which means no report. Besides, I don’t think I’d ever seen a kernel panic before Intrepid. The novelty is wearing off, but only slowly.

Newly broken in Jaunty:

  • OpenOffice(!!) on my laptop dies consistently on launch with the warning unable to get gail version number. I have tried and failed to find an existing bug for this: there are a few isolated reports, either marked as fixed or useless conversations that end with oh never mind, I made a stupid mistake and now I’ve fixed it. (Dear Internet: always specify how.) There’s bug 350193 but it seems to require pasting some specific data into OO Calc, whereas I don’t even get to the stage of having a window open. OpenOffice works on my university desktop though. I’ve filed bug 365763.
  • Empathy, generally, although it falls into this category by default since I’ve never tried to use it before. I can’t get it to connect to any instant messaging service at all. I haven’t been motivated to investigate enough to file a bug report, because it doesn’t appear to support what for me is a key feature of Pidgin, namely, ability to have profiles where I connect to some IM accounts but not always all of them. (I also consider having to have several IM accounts in the first place something of a bug. With the universe.)

I don’t know the status of my very annoying bug in Hardy whereby whenever I adjusted the screen brightness downwards via the keyboard, my Ctrl key got virtually ‘stuck’ down. (So, for example, if I pressed ‘w’, a window would close, and ‘d’ would terminate programs.) Daniel Stone even kindly offered to look into this one for me, but shortly afterwards I poured hot chocolate all over my laptop and now the down arrow doesn’t work. I can’t try and reproduce this unless I finally get off my butt and get a new keyboard. And thus do promising bug reports vanish into nothing.

Plus, a bonus tip for people who use rdiff-backup. rdiff-backup is rather fiddly about version numbers, which is not a great attribute in a backup program, since you should be able to backup between two systems that aren’t running exactly the same version of rdiff-backup, Python, Ubuntu or anything else. (Quoth Martin Pool: I use duplicity as a workaround.) However, if you’re like me and your precious rdiff-backup backup regime took days to set up and you’re not doing it again for a while, Andreas Olsson points out that he maintains a PPA of the latest version of rdiff-backup for all supported versions of Ubuntu except Dapper, which is sufficient for my purposes at the moment.

Ada Lovelace Day wrap 3: new discoveries, among other things

Lest you worry that I am counting my posts before they’re hatched, a final Ada Lovelace Day 2009 wrap-up post. As should be evident from the many posts I’ve made, I hugely enjoyed this initiative. I especially enjoyed stories of women who barely or never knew their inspiration nevertheless gaining a lot from just knowing that someone else was out there.

I originally intended this final Ada Lovelace Day wrap post to be about five new women I had discovered through browsing the blog entries. But eventually it became five posts that spoke to me most, from outside my usual Internet haunts. You get two new discoveries, one rediscovery, one blast from the past, and one crack in the techno-utopia.

Number one comes in under the I should have heard of her category: Roberta Williams, Sierra On-line co-founder, and the brains behind the King’s Quest games among others, including Phantasmagoria. All games I never got to play, because I spent the latter half of high school reading about computer games in year-old gamer magazines rather than playing them. (We had a 486 my parents bought in 1993, when it was already past its prime, so from 1995 to about 2000 I was distinctly un-game-enabled. This explains the weekend long binges of Diablo II in 2000 though.) Williams was profiled in many places, see Happy Ada Lovelace Day at The Chaliceblog and Women in Tech: R. Williams for two profiles.

Number two is Shirley Jackson, theoretical and materials physicist, advised during her undergraduate years that colored girls should learn a trade, now president of Rensselaer Institute of Technology after a career spanning MIT and Bell Labs among other places. As with Karen Spärck Jones, one of the best profiles is by someone who has interviewed her, see a lab of their own.

Jackson was understandably reluctant to dwell too much on the pain and injustices of the past in our interview, and I hated having to even ask such questions — basically amounting to, “So, um, what’s it like to be black, and a woman, at MIT in the 1960s?” (Answer: not a hell of a lot of fun!) In a fair and just world, this would simply not be relevant; but we do not live in a fair and just world. So I asked those questions — awkward, embarrassed, stymied by white liberal guilt, and resenting the questions on her behalf — and Jackson graciously answered, choosing her words carefully, gently setting the neophyte science writer straight on a few things about life as the ultimate “Other” in physics.

Honestly, I was impressed by her lack of bitterness. It had to have been lonely, not to mention discouraging. One of her professors actually advised her that “colored girls should learn a trade.” Jackson didn’t take that advice, but she did acquire some measure of perspective by volunteering (in her copious spare time — not!) at Boston City Hospital’s pediatrics unit, to better “understand what real trouble is.” And she kept at her studies, because “If I give up, what have I done but allow the other guy to win?”

There’s another profile at Ada Lovelace Day: Dr. Shirley Jackson.

Number three is a woman whose name I knew. My eleventh year high school mathematics classroom was about 75% girls, of about twenty of us perhaps in the 3 Unit class. (A twelfth year extension to that, 4 Unit, was offered to a subset of the class: six of us, four girls and two boys, which meant we took extra classes in the mornings as well as the 3 Unit classes. I was more than a full year too young for that class, but I was put in it partly just because I could and partly because my own year wasn’t going to have the numbers for me to take a 4 Unit offering.) My teacher was used to having majority girl classrooms at the high ability end of high school mathematics (he never did really understand why), and tried to encourage us by having a poster on the back wall of well-known female mathematicians (of whom I recall three: Hypatia of Alexandria, Sophie Germain and Emmy Noether).

I was making up for being the baby of the class and the top of it by also trying for the trifecta: also being the most obnoxious person in it. I was sadly unobnoxious, for a fifteen-year-old, but I claimed it was hardly inspiring to know that in all of mathematical history there had been six women mathematicians. But I survived and live to tell, or at least link to, the tale of Emmy Noether, by Laura: Of honorary men. I was struck by this post touching briefly on several interesting issues: Noether’s being proved to be important because famous men thought well of her, the emphasis on her ‘eccentricities’ which probably aren’t terribly striking when you evaluate her by mathematical social standards and not early twentieth century female ones, and Laura’s friend failing to understand that ‘practically a man’ is not a unmixed compliment.

For the fourth post, a woman I know of intimately, in fact, but she gets a mention here because I had come close to forgetting her. John Gunders wrote about Isaac Asimov’s fictional creation, Susan Calvin: Ada Lovelace Day: Susan Calvin. In my rather extensive but admittedly not completionist teenage survey of Asimov’s fiction I read a lot of the Susan Calvin/US Robotics stories. The authorial treatment of her gender is irritating even when it’s not as explicit as in the Liar! story Gunders talks about, irritating enough that I could have cogently critiqued it at fourteen. But in some ways, it was useful. It’s more than a little annoying that Calvin is not really particularly happy, and that the author treated this as the price of admission for her career, but nevertheless, the fact that given the (not entirely false, but artificially stark) choice, she had gone ahead and chosen her work every time, was educational.

The final post I wanted to highlight is not about an individual woman in technology at all, it’s about the use of technology by abusers. Ada Lovelace day – empower the women!:

In a recent survey of 479 [domestic violence] victims aged 15-74, 25% of those had their browser history monitored. 24% had been repeatedly threatened, insulted or harassed by email. 18% had their email monitored. Did you see those numbers? At least 1 in 4 victims had technology used against them as a tool to continue abusive patterns.

I asked around and searched around after this and indeed anecdotally domestic violence survivors I’m acquainted with have often had their technology use monitored and used against them (or been suspected/abused for wanting to use it at all). I am not surprised, as such, but it was something that I was privileged to have not regularly thought about. (My previous contact with it has mostly been around the issue of the strong norm of using one’s legal name when interacting with the Free Software community. The number of people — I think all of them are women — who have major hesitations about this norm due to wishing not to be traced by a former stalker or abuser, is around about ten, and of course that’s just people who have become comfortable enough to reveal this in forums I read.)

There are several guides around the web to using computing resources safely when wanting to discuss or planning to escape an abusive relationship: National Network to End Domestic Violence: Internet and Computer Safety [USA], Washington State Coalition Against Domestic Violence: Internet Safety [USA] and Domestic Violence Resource Centre Victoria: Tip Sheet: Technology Safety Planning [Australia].

Ada Lovelace Day wrap 2: Karen Spärck Jones elsewhere

Yes, this does mean that a third of these things is coming, but I wanted to point to some other profiles of Karen Spärck Jones, aside from my brief one. At least at the present time, she’s on the first page of most profiled Ada Lovelace Day subjects. I was really pleased to learn more about this inspiring scientist.

Martin Belam has a long profile quoting extensively from Spärck Jones’s interviews and speeches and focussing on both her own career progression: she worked with Margaret Masterman at the Cambridge Language Research Unit. “You have no conception of how narrow the career options were [for women],” is one of Belam’s quotes. Another one of her stories reminds me of more recent stories Pia Waugh has told me about the resistance of parents playing a role in girls not choosing computing careers (these days it’s apparently the perceived low earnings and limited career prospects of programmers from the point of view of ambitious parents, so at least something has changed):

We were trying to get at girls in schools [to take up computing] and we knew we had to get to the teachers first. We found that the spread of computing in the administrative and secretarial world has completely devalued it. When one of the teachers suggested to the parents of one girl that perhaps she should go into computing the parents said: ‘Oh we don’t want Samantha just to be a secretary’. That’s nothing to do with nerdiness, but the fact that it’s such a routine thing.

Bill Thompson was a student of Spärck Jones’s, and writes about her influence on him as a fellow philosopher turned computer scientist. He also wrote her obituary for The Times (and, in 2003, that of her husband, fellow computer scientist Roger Needham).

IT journalist Brian Runciman remembers Spärck Jones as the most interesting woman he’s ever interviewed in Computing’s too important to be left to men. (I think it’s very important to get more women into computing. My slogan is: Computing is too important to be left to men. seems to be Spärck Jones’s best known quote.) In the interview with him, she talked about how her ideas permeate modern search engine implementations.

She scored smaller mentions from:

Ada Lovelace Day wrap: the usual suspects

I use the usual suspects here in the sense of folks in my feed reader. Over the next couple of days I am going to look around more of the posts and pick out a set of favourites, focusing on women I’ve never heard of. In the meantime, here’s the closer to home wrap.

I was profiled in a couple of places: by Jacinta Richardson in Ada Lovelace Day (with several others as described below), by Julie Gibson, the founder of Sydney LinuxChix, in Mary’s Random Curiosity and in one private post.

Here’s profiles pulled from my feeds, hopefully you find something and/or someone new.

Seen on Free Software planets (undoubtedly incomplete):

Ada Lovelace Day profile: Karen Spärck Jones

Let’s create new role models and make sure that whenever the question “Who are the leading women in tech?” is asked, that we all have a list of candidates on the tips of our tongues… To take part All you need to do is… pick your tech heroine and then publish your blog post any time on Tuesday 24th March 2009. It doesn’t matter how new or old your blog is, what gender you are, what language you blog in, or what you normally blog about – everyone is invited.

This is a profile of a woman in technology for Ada Lovelace Day.

Creative Commons License
Karen Spärck Jones by Markus Kuhn (modifications by Mary Gardiner) is licensed under a Creative Commons Attribution 2.5 Australia License.
Based on a work at commons.wikimedia.org.

I first heard about Karen Spärck Jones, who was a senior scientist in my field of computational linguistics, in 2007 as part of my paying job, which is as the editorial assistant for Computational Linguistics. Just before she died, Spärck Jones wrote Computational Linguistics: What About the Linguistics? which we published posthumously as the Last Words column for Vol. 33, No. 3. (Spärck Jones was aware both that she was dying and that her column was going to appear under the heading ‘Last Words’.) I was never able to correspond with her directly: she died before we even had the camera ready copies done.

Spärck Jones’s academic career began in 1957, and was funded entirely by grant money until 1994: most academics will recognise this as a hard way, requiring researchers to fund their own positions with grant money awarded in cycles.

Spärck Jones was the originator of the Inverse Document Frequency measure in information retrieval (1972, A statistical interpretation of term specificity and its application in retrieval., Journal of Documentation, 28:11–21) which is nearly ubiquitously used as part of the measure of the importance of various words contained in documents when searching for information. (The word ‘the’, for example, is very unimportant, as it occurs in essentially all documents, thus having high document frequency and low inverse document frequency.) She had a long history in experimental investigations of human language (most computational linguists are now in this business). She was also at one time president of the Association for Computation Linguistics.

Awards Spärck Jones won in her lifetime include Fellowships of the American and European Artificial Intelligence societies, Fellowship of the British Academy, the ACL Lifetime Achievement Award and the Lovelace Medal of the British Computer Society.

Elsewhere: Spärck Jones’s obituary in Computational Linguistics and Wikipedia.

Ada Lovelace Day profile: Allison Randal

Let’s create new role models and make sure that whenever the question “Who are the leading women in tech?” is asked, that we all have a list of candidates on the tips of our tongues… To take part All you need to do is… pick your tech heroine and then publish your blog post any time on Tuesday 24th March 2009. It doesn’t matter how new or old your blog is, what gender you are, what language you blog in, or what you normally blog about – everyone is invited.

This is a profile of a woman in technology for Ada Lovelace Day.

Creative Commons License
Allison Randal (Three photos) by Miles Sabin, Piers Cawley, Paul Fenwick, Mary Gardiner is licensed under a Creative Commons Attribution-Share Alike 2.5 Australia License.

Allison Randal is the chief architect of the Parrot virtual machine, which, I have just now discovered, had their 1.0.0 release a week ago today. I’ve known of Parrot for a long time, because of its posited relationship with the Python programming language (see the original April Fool’s joke), but I didn’t know much about the project beyond it being a VM until Randal’s linux.conf.au 2008 talk (see slides, Ogg Theora video, Ogg Speex audio).

I am not a Perl programmer and Randal is mostly known within the Perl (and OSCON, see below) communities, but Randal’s talk at linux.conf.au 2008 was the most memorable for me: she talked about bringing modern compilation ideas to the Free Software programming languages community, and then about the architecture of Parrot and the various intermediate languages it is possible to target.

The most striking thing about Randal’s work for me is that she combined high profile technical coding with deep community involvement (and technical writing). She is a past president and current board member of the Perl Foundation and chairs the talk selection for OSCON. In an ideal world I’d like to be able to straddle technical and technical community work in my own life, and Randal is one of the highest profile examples of this I know of.

Elsewhere: Randal’s homepage, Randal’s O’Reilly Radar blog, Randal’s use.perl blog and Wikipedia.

Why microblog when you can IRC?

I’ve been meaning to answer Glyph for a while:

Maybe the reason I don’t “get” this Twitter thing is I’ve been using IRC for much the same purpose for a decade, and the UI is better.

Finally I have an excuse, that is, I wrote it up in someone else’s blog comments and can just lift it and edit it here. I have a feeling that I am cheating by using more than 140 characters though.

Advantages of microblogging over IRC:

  • it’s easier to find out where the cool kids are playing; (Do you think Stephen Fry or even Sarah Haskins would tell me where their IRC channels of choice are? Me neither.)
  • I do not sign up for a whole ‘channel’ in its entirety, with an entire social group, ongoing conversations and complicated social conventions in order to microblog; and
  • it’s not real-time. I am basically the person email was designed for. I do not real-time. Even people I talk to when forced can vouch for this.

Bonus advantage over Facebook status updates: following my microblog does not require that we are friends, or even ‘friends’. You can just read them. I do not thereby have to share my high school graduation date and pictures of my hypothetical cat with you. (That said, I am now going to create a Facebook album entitled ‘my hypothetical cat’.)

Disadvantages of microblogging over IRC:

  • yeah, the clients suck;
  • when I do get in a conversation with someone(s) I am limited to 140 character messages (plus boring any onlookers), or figuring out how to switch media; and
  • the overhead of a social group and ongoing conversations do have some benefits: I’m more likely to admire someone’s wit and insight from IRC (or their blog) than from Twitter or identi.ca.

Cousins of spam

Spam is of course unsolicited commercial email, or at least any unwanted email obvious enough that we expect a spam filter can catch it. (They seem to regard filtering unwanted mail from senders we also get wanted mail from as outside their scope, which is probably fair enough considering that it’s false positives that really drive their best-paying users up the wall.)

There are terms for the niggles that get through, especially by virtue of coming from a trusted sender: I introduced the term tofu to several people on Identica and Twitter yesterday: it’s activist spam, those (usually manually sent but massively Bcc-ed) emails asking you to attend a meeting or a rally or sign a petition. (More alert activists send these with sorry for the tofu but… at the top.) I was in turn introduced to the term bacn, for email which is solicited in the ethical email marketing sense (you’ve subscribed to it) but which then becomes something to delete or skip.

Which brings me to a couple of types of unwanted email that need their own snackfood-related term. I suppose the bacn precedent means they’re supposed to have four letters, too:

reminders to participate in something I agree is very important but that everyone knows about

One of the things I love best about compulsory voting in Australia (FAQ: how can you bear to live in a country that would gaol you for not voting? I have no idea, in my country it attracts a fine of about $20, think of it as like skipping jury duty except a lot cheaper) is not having everyone nag me by email, phone and finally door-to-door approaches to get out the vote for weeks, at best, before an election. Fund-raising can become equivalently aggressive. (By extension from compulsory voting I suppose I should argue for such high taxes that no one has to, or can, donate money to charity.)

calls for papers

Many academics, especially in computer science, receive endless emails to their individual address reminding them that the deadline for the International Conference for Synergistic E-Knowledge in the Wireless Classroom, etc, has a deadline coming up soon. They’re somewhat targeted: I get a lot more to my .edu.au addresses than to my personal ones, and computer organisations like SLUG get them too. These should fall into the category of spam, since they’re unsolicited and you have to pay for the conference, but they don’t fall under the new definition of spam, stuff that spam filters actually recognise. (Mind you, at university I think the only thing Barracuda does reliably classify as spam is Mailman held-mail summaries, but it’s especially bad at these conference mails and also at anything to do with watches.) I get several a week and by the do I want these people up against the wall when the revolution comes? metric of spaminess, it is time to have a word for this blight.

Blood donation

There are calls for blood donations after the bushfires have killed and injured so many people in Victoria. I am a long-time donor and armchair expert, so I thought I’d share some tips.

First, info for those who just need a prod:

  • you usually need an appointment to donate blood in most places in Australia. Information at the Australian Red Cross Blood Service or by calling 13 14 95 inside Australia. If they can’t get you in this week, donate anyway, severe burns victims need long-term treatment, and there will, of course, be other ill people later too.
  • the Blood Service has some information about the process itself
  • do check the health guidelines as to whether you can give blood now, or at all. If you are sick right now, wait until you have been well for a week or more.

Pressuring people indiscriminately to donate is one of my (many) pet peeves, since so many people are excluded for health reasons or due to risk factors and all it accomplishes is a pointless guilt-trip about who is or is not a better person by virtue of being able to donate. So I want to clarify: I know many cannot donate for any one of hundreds of reasons (including, by the way, phobias and so on: the Blood Service would actually prefer that you not donate if you are going to faint or need a lot of care!). I’ve been one of these people at times. If you cannot or would strongly prefer not to for any reason, go about your business with equanimity. This post is for the curious and the undecided.

If you have donated blood before and it was fairly easy, please consider a plasma donation. Plasma is a crucial product in treating burns patients and there’s generally a strong need for it at all times. It’s taken in whole blood (standard) donations too, but dedicated plasma donations supply more of it. If you have blood type AB (Rhesus negative or positive, it doesn’t matter), you are a universal plasma donor, so you are a particularly good candidate for plasma donations. It needs good veins, so if you’ve had difficulty donating whole blood before stick to that. This type of donation takes a bit longer, and is done via apheresis. You can continue being a whole blood donor periodically and also be a regular plasmapheresis donor too. There are a couple of extra health questions for a plasma donation (because you will receive some anti-coagulants), so give the Blood Service a call if you are considering donating plasma for the first time.

Now for the ones who have trouble with whole blood. I am blood group O, Rhesus negative, a universal donor for red blood cells (excluding a few rare blood types). O negative people, while generous to others, can only be transfused with O negative red blood cells ourselves, so we are desirable as donors both because our blood is useful in emergencies if there isn’t time for cross-matching, and because O negative patients specifically need it. I am also, last time I asked anyway, one of the pretty small percentage of Australian adults who has no immune response to CMV, which means I’ve never had it and I can’t give it to anyone else. (Like chicken pox, it’s a kind of herpes and the infection is permanent.) So my blood is like liquid gold, taken straight from me and rushed to O negative immunocompromised patients via a quick stop off to exclude infections, as best I can tell.

And this is the only reason I am still a blood donor. The difficulty I have giving blood would have caused me to be excluded by now otherwise, I have veins that are both buried deep and very slim. So I have some tips for others who have hard to find veins:

  • if you don’t yet drink a lot of water before donating, start doing so. A litre or so seems about right, and then another half litre or so every half hour or so if you need to maintain. You want to be over-hydrated, not just the normal good hydration that you want going about your day. (No need to go nuts, I was once advised to drink two litres down and maintain, but my kidneys started hurting in the middle of the donation! And drinking crazy amounts of water can make you seriously ill.)
  • if you have ever had them actually not get a vein at all, or not get a full donation (this is known as ‘Did Not Bleed’ in your charts, I have it 4 times), go to the donor centres (not the mobile blood units) look your interviewer in the eye and calmly tell them you want to have the senior phlebotomist do your donation, please. I have this written into my donor record now, and I often get someone who has worked with palliative care or chemotherapy patients, there is nothing a blood donor’s veins can do to them that will thwart them.