[syndicated profile] sumana_feed
Sometime around 1999 or 2001, I first heard "King of Spain" by Moxy Früvous. The UC Berkeley a cappella group DeCadence performed it during one of their lunchtime concerts near Sather Gate. (Four out of five weekdays one of the a cappella groups would do a noon concert -- DeCadence, Artists in Resonance, the Men's Octet, the California Golden Overtones -- and I caught as many of them as I could.) And then Steve Shipman introduced me to more of their songs and albums -- it was Bargainville, which ends with that haunting a cappella "Gulf War Song", that I was listening to on September 10, 2001.

In 2014 it came to light that band member Jian Ghomeshi had a fairly sordid history, and for a while I couldn't listen. Now I seem to have the ability to listen again; that change I don't have as much insight into as I'd like.

Just now Leonard and I were singing bits of "King of Spain" to each other; he sang:

Royalty
Lord, it looked good on me

I said "What?!" Because back around 2000 and through all the years to the present, I heard those lyrics as:

Royalty
Lord of the good ennui

So for the entire time I've been with Leonard, he and I have interpreted that song slightly differently. He heard the narrator figuratively wearing royalty like clothing, like a fashion statement, which connects to the silk he mentions in the next line, and which logically connects to the garment swap later in the song. Through my mondegreen, I heard an emphasis on the narrator's malaise and boredom (a reason for the prince-and-pauper swap) and a connection to the literal meaning of an additional French loanword, laissez-faire, that he uses later.

A quick web search tells me that Leonard's version is the consensus, that to join intersubjective reality I would let go of "Lord of the good ennui". I shall bury it here, with due ceremony. Goodbye, old mondegreen friend! You were a lot of fun.

[syndicated profile] sumana_feed
The principle at stake in California v. Johnson: due process requires that we be able to examine the evidence used to convict someone. Kern County got a $200,000+ grant and started using closed-source software to perform a new kind of DNA testing for criminal forensics. You are not allowed to audit the software to check for bugs, but the company founder will fly in and testify in court to say he attests to the validity of the results it finds. Uh, no, we need to check, and the ACLU and EFF have just filed amici curiae* briefs before California's Court of Appeal for the Fifth District, saying so.

Man at lectern in front of screen displaying 'Winning Raffle Numbers: 12345 12345 12345 12345', photo (used by permission) by Mike Pirnat at the PyCon PyLadies auction in 2017As I've written and even testified, we need more auditability, transparency, and security in software governments use in laboratories and field tests. Heck, we need it in software governments use to make decisions more generally -- lotteries for visas, school assignments, parole and prison sentencing, and so on.

So I was delighted to learn of bill Int 1696-2017, currently before New York City's City Council. Summary:

This bill would require agencies that use algorithms or other automated processing methods that target services, impose penalties, or police persons to publish the source code used for such processing. It would also require agencies to accept user-submitted data sets that can be processed by the agencies' algorithms and provide the outputs to the user.

I applaud James Vacca, chair of the council's Committee on Technology, for introducing and sponsoring this bill, and for citing/shouting out to danah boyd, Kate Crawford, and Cathy O'Neil as people whose work has shaped this legislation. The New York Times says: "As a committee chairman, he plans to convene hearings before he leaves office in December." I'm looking forward to attending those hearings.

If you live in New York City, you can contact your councilmember and suggest they cosponsor this bill. If you live elsewhere, consider telling your local elected officials that they oughta introduce legislation like this. When writing or calling, if you're a programmer or other technology expert, say so -- our voice matters.

I have more links in the algorithmictransparency tag on Pinboard.


* Many years ago, Seth Schoen made me an illustration that we still have somewhere. Reconstructed from memory:

[one smiling stick figure, male, near a courthouse] Sum amicus curiae.
[one smiling stick figure, female, near a courthouse] Sum amica curiae.
[many smiling stick figures of various genders, near a courthouse] Sumus amici curiae.
[one stick figure, male, holding a finger to his mouth as though shushing you, near a courthouse] Tacit! Sum inimicus curiae!

Edited Tuesday Sept. 19th to add: The Committee on Technology is holding a public hearing to discuss Int 1696-2017 on Monday, October 16th.

[syndicated profile] sumana_feed
The principle at stake in California v. Johnson: due process requires that we be able to examine the evidence used to convict someone. Kern County got a $200,000+ grant and started using closed-source software to perform a new kind of DNA testing for criminal forensics. You are not allowed to audit the software to check for bugs, but the company founder will fly in and testify in court to say he attests to the validity of the results it finds. Uh, no, we need to check, and the ACLU and EFF have just filed amici curiae* briefs before California's Court of Appeal for the Fifth District, saying so.

Man at lectern in front of screen displaying 'Winning Raffle Numbers: 12345 12345 12345 12345', photo (used by permission) by Mike Pirnat at the PyCon PyLadies auction in 2017As I've written and even testified, we need more auditability, transparency, and security in software governments use in laboratories and field tests. Heck, we need it in software governments use to make decisions more generally -- lotteries for visas, school assignments, parole and prison sentencing, and so on.

So I was delighted to learn of bill Int 1696-2017, currently before New York City's City Council. Summary:

This bill would require agencies that use algorithms or other automated processing methods that target services, impose penalties, or police persons to publish the source code used for such processing. It would also require agencies to accept user-submitted data sets that can be processed by the agencies' algorithms and provide the outputs to the user.

I applaud James Vacca, chair of the council's Committee on Technology, for introducing and sponsoring this bill, and for citing/shouting out to danah boyd, Kate Crawford, and Cathy O'Neil as people whose work has shaped this legislation. The New York Times says: "As a committee chairman, he plans to convene hearings before he leaves office in December." I'm looking forward to attending those hearings.

If you live in New York City, you can contact your councilmember and suggest they cosponsor this bill. If you live elsewhere, consider telling your local elected officials that they oughta introduce legislation like this. When writing or calling, if you're a programmer or other technology expert, say so -- our voice matters.

I have more links in the algorithmictransparency tag on Pinboard.


* Many years ago, Seth Schoen made me an illustration that we still have somewhere. Reconstructed from memory:

[one smiling stick figure, male, near a courthouse] Sum amicus curiae.
[one smiling stick figure, female, near a courthouse] Sum amica curiae.
[many smiling stick figures of various genders, near a courthouse] Sumus amici curiae.
[one stick figure, male, holding a finger to his mouth as though shushing you, near a courthouse] Tacit! Sum inimicus curiae!

[syndicated profile] sumana_feed
The principle at stake in California v. Johnson: due process requires that we be able to examine the evidence used to convict someone. Kern County got a $200,000+ grant and started using closed-source software to perform a new kind of DNA testing for criminal forensics. You are not allowed to audit the software to check for bugs, but the company founder will fly in and testify in court to say he attests to the validity of the results it finds. Uh, no, we need to check, and the ACLU and EFF have just filed amici curiae* briefs before California's Court of Appeal for the Fifth District, saying so.

Man at lectern in front of screen displaying 'Winning Raffle Numbers: 12345 12345 12345 12345', photo (used by permission) by Mike Pirnat at the PyCon PyLadies auction in 2017As I've written and even testified, we need more auditability, transparency, and security in software governments use in laboratories and field tests. Heck, we need it in software governments use to make decisions more generally -- lotteries for visas, school assignments, parole and prison sentencing, and so on.

So I was delighted to learn of bill Int 1696-2017, currently before New York City's City Council. Summary:

This bill would require agencies that use algorithms or other automated processing methods that target services, impose penalties, or police persons to publish the source code used for such processing. It would also require agencies to accept user-submitted data sets that can be processed by the agencies' algorithms and provide the outputs to the user.

I applaud James Vacca, chair of the council's Committee on Technology, for introducing and sponsoring this bill, and for citing/shouting out to danah boyd, Kate Crawford, and Cathy O'Neil as people whose work has shaped this legislation. The New York Times says: "As a committee chairman, he plans to convene hearings before he leaves office in December." I'm looking forward to attending those hearings.

If you live in New York City, you can contact your councilmember and suggest they cosponsor this bill. If you live elsewhere, consider telling your local elected officials that they oughta introduce legislation like this. When writing or calling, if you're a programmer or other technology expert, say so -- our voice matters.

I have more links in the algorithmictransparency tag on Pinboard.


* Many years ago, Seth Schoen made me an illustration that we still have somewhere. Reconstructed from memory:

[one smiling stick figure, male, near a courthouse] Sum amicus curiae.
[one smiling stick figure, female, near a courthouse] Sum amica curiae.
[many smiling stick figures of various genders, near a courthouse] Sumus amici curiae.
[one stick figure, male, holding a finger to his mouth as though shushing you, near a courthouse] Tacit! Sumus inimicus curiae!

[syndicated profile] sumana_feed
The principle at stake in California v. Johnson: due process requires that we be able to examine the evidence used to convict someone. Kern County got a $200,000+ grant and started using closed-source software to perform a new kind of DNA testing for criminal forensics. You are not allowed to audit the software to check for bugs, but the company founder will fly in and testify in court to say he attests to the validity of the results it finds. Uh, no, we need to check, and the ACLU and EFF have just filed amici curiae* briefs saying so.

Man at lectern in front of screen displaying 'Winning Raffle Numbers: 12345 12345 12345 12345', photo (used by permission) by Mike Pirnat at the PyCon PyLadies auction in 2017As I've written and even testified, we need more auditability, transparency, and security in software governments use in laboratories and field tests. Heck, we need it in software governments use to make decisions more generally -- lotteries for visas, school assignments, parole and prison sentencing, and so on.

So I was delighted to learn of bill Int 1696-2017, currently before New York City's City Council. Summary:

This bill would require agencies that use algorithms or other automated processing methods that target services, impose penalties, or police persons to publish the source code used for such processing. It would also require agencies to accept user-submitted data sets that can be processed by the agencies' algorithms and provide the outputs to the user.

I applaud James Vacca, chair of the council's Committee on Technology, for introducing and sponsoring this bill, and for citing/shouting out to danah boyd, Kate Crawford, and Cathy O'Neil as people whose work has shaped this legislation. The New York Times says: "As a committee chairman, he plans to convene hearings before he leaves office in December." I'm looking forward to attending those hearings.

If you live in New York City, you can contact your councilmember and suggest they cosponsor this bill. If you live elsewhere, consider telling your local elected officials that they oughta introduce legislation like this. When writing or calling, if you're a programmer or other technology expert, say so -- our voice matters.

I have more links in the algorithmictransparency tag on Pinboard.


* Many years ago, Seth Schoen made me an illustration that we still have somewhere. Reconstructed from memory:

[one smiling stick figure, male, near a courthouse] Sum amicus curiae.
[one smiling stick figure, female, near a courthouse] Sum amica curiae.
[many smiling stick figures of various genders, near a courthouse] Sumus amici curiae.
[one stick figure, male, holding a finger to his mouth as though shushing you, near a courthouse] Tacit! Sumus inimicus curiae!

[syndicated profile] sumana_feed
I was talking with a fellow consultant about what to do if you have a gig getting you down. Especially when you realize that the client isn't being helpful, and there's a bunch of learning curves that are exhausting you, and you still have several weeks to go.

In my master's in tech management coursework, I learned the lens that thriving is a function of a person times their environment. I think those of us who are used to trying harder, overcoming obstacles, etc. can be -- kind of out of self-protective instinct -- bad at noticing "this environment is so crappy it makes it systematically hard for me to achieve and thrive". Especially with short-term projects. At first, things like "I feel tired" or "ugh, new thing, I don't want to learn this and be bad at it (at first)" and "I'm worried that person doesn't like me" or "they missed the email/meeting/call and now it's harder to execute the plan" are identical to problems that we are reasonably sure we can overcome. Maybe we notice patterns about what's not working but think: I can take initiative to solve this, myself, or with my few allies.

Several papercraft pieces I made out of gold-colored wrapping paper, some alike and some differentThe data points accumulate and we chat with other people and, in the process, learn more data points and shape our data points into narratives and thus discover: this is a bad environment, structurally. But by the time we really figure out the effect a short-term project is having on us, it's supposedly the home stretch.

I'm looking back at gigs that I found draining, where, eventually, I had this realization, although I have never quite framed it this way before now. On some level I realized that I could not succeed by my own standard in these projects/workplaces, because there was so much arrayed against me (e.g., turf war, a generally low level of skill in modern engineering practices, lack of mission coherence, low urgency among stakeholders) that I could not do the things that it is kind of a basic expression of my professional self/competence to do.*

So I had to change what it was I aimed to achieve. For example, I've had a gig where I was running my head into the wall constantly trying to bring better practices to a project. I finally talked with an old hand at the organization and learned the institutional reasons this was practically impossible, why I would not be able to overcome the tectonic forces at play and get the deeper conflicts to resolve any faster. So we changed what I was trying to do. Running a daily standup meeting, by itself, is a thing I can do to bring value. I changed my expectations, and made mental notes about the pain points and the patterns, because I could not fix them right away, but I can use those experiences to give better advice to other people later.

An editor recently told me that, in growing as an editor, he'd needed to cultivate his sense of boredom. He needed to listen to that voice inside him that said this is boring me -- and isn't that funny? Parents and teachers tell us not to complain about being bored -- "only boring people are bored", or -- attributed to Sydney Wood -- An educated man is one who can entertain a new idea, another person, or himself. But pain is a signal, boredom is a signal, aversion and exhaustion are signals. Thriving is a function of a person times their environment.

Also, the other day I read "Living Fiction, Storybook Lives" (which has spoilers for Nicola Griffith's excellent novel Slow River).

How come I spent many years living a rather squalid existence... yet managed to find my way out, to the quite staid and respectable life I have now, when others in the same situation never escaped? In the course of writing the book, I found that the answer to my question was that the question itself was not valid: people are never in the same situation.

It takes substantial introspection and comparison to figure out: what kind of situation am I in, both externally and internally? Is it one where I will be able to move the needle? It gets easier over time, I think, and easier if I take vacations so I can have a fresh perspective when I come back, share my stories with others and listen to their stories, and practice mindfulness meditation so I am better at noticing things (including my own reactions). Maybe "wisdom" is what feels like the ability to X-ray a messy blobby thing and see the structures inside, see the joints that can bend and the bones that can't. In some ways, my own motivation and mindfulness are like that for me -- I need to recognize the full truth of the situation I'm in, internally and externally, to see what needs changing, to see how I might act.

The thing that gets me down most, on exhausting projects, is the meta-fear that nothing will improve, that I am stuck. When I realize that, I try to attend to that feeling of stuckness. Sometimes the answer is in the problem.


* As Alexandra Erin discusses, regarding her political commentary via Twitter threads: "I do the work I do on here because I feel called to it. For the non-religious, I mean: I have a knack for it and I find meaning in it."

In Black

Sep. 7th, 2017 01:43 pm
[syndicated profile] sumana_feed
Grabbing a book to read, thought I was grabbing Johnny Cash's autobiography, was actually grabbing Jonathan Zittrain's The Future Of The Internet And How To Stop It.
I'd like to code a rainbow every day
And tell y'all it's 200 OK
But cyberspace is crying out: the commons we shall lack
Until....
Till we steal it back



Page generated Sep. 21st, 2017 01:51 pm
Powered by Dreamwidth Studios