terriko: (Default)
2017-02-22 09:50 am
Entry tags:

Home!

Made it home from India without incident, by which I mean my green card was accepted at the border and no one asked me to unlock my phone. It's weird how I just went on a trip to a country where I couldn't drink the water and the front page of the newspaper had multiple rape cases and an acid attack against women and yet, crossing the US border was *still* the most scary part with the constantly changing rules.

The trip was great. I saw so many things I never expected to see, ate so much delicious food, and met so many people that I'm not sure I'm ever going to get everyone's names straight. The PyCon Pune conference was *amazing*. I keynoted to a room of over 500 people, and I've never had such an engaged audience! I did code sprints with people who were awesome, too -- we discovered that Mailman had something like 9 different dev setup guides, many of which were out of date, and yet somehow everyone got things up and running *and* folk helped patch up the docs to be consistent. If you ever get a chance, seriously, go.
terriko: (Default)
2015-04-14 11:16 pm
Entry tags:

"but it is not going to be installed"

Playing with docker and the mailman bundler today in an attempt to get some images running, and I hit a point where I got this:


root@fd97a058ea83:~/mailman-bundler# apt-get install python3-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
python3-dev : Depends: libpython3-dev (= 3.4.0-0ubuntu2) but it is not going to be installed
Depends: python3.4-dev (>= 3.4.0-0~) but it is not going to be installed
E: Unable to correct problems, you have held broken packages.


It was very non-obvious to me what was going on and google wasn't super helpful, so I'm posting in case someone else searches for the same "but it is not going to be installed" installed error message.

Turns out, what happened is that I needed to run apt-get update, and now it's happy.

Well, at least relative to that... buildout in mailman-bundler is complaining about numpy so I've still got some work to do!
terriko: (Default)
2014-04-26 11:33 am
Entry tags:

Mailman 3.0 Suite Beta!

I'm happy to say that...


Mailman logo

Mailman 3.0 suite is now in beta!

As many of you know, Mailman's been my open source project of choice for a good many years. It's the most popular open source mailing list manager with millions of users worldwide, and it's been quietly undergoing a complete re-write and re-working for version 3.0 over the past few years. I'm super excited to have it at the point where more people can really start trying it out. We've divided it into several pieces: the core, which sends the mails, the web interface that handles web-based subscriptions and settings, and the new web archiver, plus there's a set of scripts to bundle them all together. (Announcement post with all the links.)

While I've done more work on the web interface and a little on the core, I'm most excited for the world to see the archiver, which is a really huge and beautiful change from the older pipermail. The new archiver is called Hyperkitty, and it's a huge change for Mailman.

You can take a look at hyperkitty live on the fedora mailing list archives if you're curious! I'll bet it'll make you want your other open source lists to convert to Mailman 3 sooner rather than later. Plus, on top of being already cool, it's much easier to work with and extend than the old pipermail, so if you've always wanted to view your lists in some new and cool way, you can dust off your django skills and join the team!

Hyperkitty logo

Do remember that the suite is in beta, so there's still some bugs to fix and probably a few features to add, but we do know that people are running Mailman 3 live on some lists, so it's reasonably safe to use if you want to try it out on some smaller lists. In theory, it can co-exist with Mailman 2, but I admit I haven't tried that out yet. I will be trying it, though: I'm hoping to switch some of my own lists over soon, but probably not for a couple of weeks due to other life commitments.

So yeah, that's what I did at the PyCon sprints this year. Pretty cool, eh?
terriko: (Default)
2014-02-16 10:48 pm
Entry tags:

Photogenic Mailbox in Snow

The snow is gone at this point (now it's all "risk of flooding" and "high winds" out here), but I thought I'd share a photo from last weekend:

Photogenic Mailbox in snow

This mailbox has a little spotlight above it (presumably so people don't drive into it or so that the mail carrier can find it), which always amuses me. I personally refer to it as "photogenic mailbox" because of the spotlight. Photogenic mailbox is apparently also photogenic in the snow, not just the dark.

I imagine I'll use photogenic mailbox in a presentation about GNU Mailman, someday!

Note: you can actually see the spotlight, or evidence of it, if you look at the photo carefully. I should have photoshopped that out, but it turns out photoshop was installed on the hard drive that died, so I haven't sorted that out and gotten it re-installed yet.
terriko: (Default)
2013-07-14 12:42 pm
Entry tags:

Mailman Virtual Hackathon

We're having a mailman virtual hackathon right now on #mailman on freenode. The plan is to run 'till around 2300 UTC today, so another 4h or so. Link for figuring out what that means in your time zone.

We're doing a variety of things: bug triage and fixing, discussion of architecture, new feature development, helping each other with any blocking problems, spouting off crazy new ideas, code review and merging, etc. We're especially hoping to make sure we clear any issues we can relating to GSoC projects, but there's plenty of work to go around. New folk are welcome too.

If you don't read this 'till after the fact, don't despair! There will likely be another such hackathon next Sunday, July 21. Keep an eye on the mailman-developers list for more details.
terriko: (Default)
2013-03-25 11:31 pm
Entry tags:

Back from Pycon!

I should write up a proper trip report with pictures and stuff, but as it's nearly midnight and I don't want my sleeping patterns to stay on California time, you get some short highlights:

1. The conference itself was awesome. Recall: I attended the sprints last year but not the main conference, so while I had high hopes I didn't know that the content would be so good. I attended a lot of great talks and no doubt missed quite a few as well. I'll be making heavy use of the conference recordings over the next little while, I expect.

2. I am really excited about my free raspberry pi. While I know lots of folk who frequently get given cool toys and told to go hack them, this is the first time someone has gifted me with such an item/mission, and it feels great. I haven't figured out what I'm going to do yet, but there was this great talk about hooking one up to a $300 CNC machine, and another great one about home automation that could be useful...

3. The sprints were super-productive! You can see our todo/completed/waiting list here if you want the nitty gritty. I'd been joking earlier to anyone who asked that we were totally going to release by Friday, and while we didn't do that, we *are* very close and you should all expect a beta release of postorius + Mailman 3 very soon. I can't wait to show it off!

4. Perhaps later I'll do up the stats on exactly what I was doing to our repository, but I should tell you that not only did I make plenty of my own code commits, but I also got to merge code from new contributors. This was totally my favourite part, seeing new folk get their code accepted and in the main tree. And it wasn't just the people who were physically at the sprints with us: I also merged code from people contributing remotely, most of whom are prospective GSoC students. Way to impress me, students!

5. I got to talk to a bunch of people about GSoC. I do this all the time by email, but it was especially fun to talk to folk in person about what's involved, why it's awesome, how to be good at it, and why they should sign up.

6. And post-con, I got a few days to catch up with friends in the area and visit the Japanese Tea Gardens in Golden Gate Park, which I've wanted to do ever since I read Seanan Mcguire's October Daye books. As I processed a few photos for this week's assignment, you get one here:

1/400s of meditation in a tea garden

And with that, midnight has rung and it's bedtime. I have a long week of catch-up ahead of me at work, but expect some more pycon / mailman / gsoc posts out of me over the next little while as I internalize all the things I've been thinking about this past week.
terriko: (Default)
2012-08-10 09:36 pm
Entry tags:

Mailman: Note to Self

Not sure this will be ever useful to anyone else, but just in case it hits me again later or when I'm helping someone else set up mailman on mac osx 10.7...

When I tried to run python bootstrap.py I got an error that looked like this:

AttributeError: 'module' object has no attribute '__getstate__'

The solution, as implied here is to fix my setuptools which is somehow wrong. On my mac, that meant clearing:

/Library/Python/2.7/site-packages/

Simple, easy, except that I have about a billion copies of python installed so finding the right one took some work. To figure out what to remove, I did the following:

(a) Ran the version of python I was actually trying to use to get a command line shell
(b) loaded setuptools (e.g.: import setuptools)
(c) Checked where it actually was (e.g.: print setuptools.__file__)
(d) Removed stuff from that directory (I could have just removed setuptools, but since I was trying to set up a fresh environment anyhow, I actually emptied the entire directory and let mailman reload)
terriko: (Pi)
2012-03-16 11:28 pm
Entry tags:

Home from Pycon!

The Pycon sprints were amazing, and the GNU Mailman team got a lot of work done on Mailman 3, the web UI, and the archives. I've never done a 4-day hackfest format, and usually for the short hackfests I'm one of the volunteers helping people set up their environments so I barely do any coding myself. But this? This may have been GNU Mailman's largest gathering of Mailman core devs ever, plus other experienced hackers to boot. It was 4 days of glorious code, architecture discussions, bugs and features. It was energizing, productive, excellent and exhausting.

I'm eager to talk about the UI work I got done and what is coming next, but I think it should wait 'till after I've slept!
terriko: (Default)
2012-01-16 09:01 pm
Entry tags:

"Accomplishments" of a technical nature this weekend

The scare quotes are because neither of these things should have been accomplishments at all, since they should have just worked. Since they didn't, though, I'm blogging for posterity with links to the things that helped me solve the problems.

Short version: I now have 8gb of ram that works, a backup drive that doesn't, a Mailman dev environment that half works, and I kinda hate Apple. )

I'm tired and cranky, but I'm determined to win this... tomorrow.

Also, I made myself cookies, so that's something.
terriko: (Pi)
2011-11-15 12:24 am

Trying to use my post-GHC energy wisely

Honestly, I think I make more resolutions after GHC than I do at new year's. I'm always so inspired!

Thing 1: Pushing the development of the GNU Mailman UI



Two things came together for me at the conference:

1. One thing I heard frequently while working the free and open source software booth is that there are plenty of folk interested in getting involved with open source, but they're not sure where to start.

2. I came home with a suitcase full of paper prototypes and pictures from the Mailman 3.0 part of the codeathon for humanity on Saturday. I was looking at spending my evenings digitizing them and turning them into functional prototypes.

So... I asked for help! Transcribing paper prototypes isn't the most glamorous of work, but it's a great place for a beginner to start, and given that we're hoping to have a Mailman 3.0 release as soon as possible, new contributors would have a chance to ramp up to doing real code commits very quickly. Plus they'd be able to see their code go out and be used in the real world sooner rather than later!

I posted to the Systers list knowing I wasn't the only one feeling the post GHC rush, and I posted to the Mailman list knowing we had a would-be contributor who wanted to help.

What I wasn't expecting was that I'd have talked to NINE volunteers in less than 24 hours. How awesome is that? And most of them are women as well!

Now I have the problem of making sure I have enough for everyone to do, but with a variety of skill levels I'm sure we won't have any trouble finding stuff for everyone. I'm so excited, and I hope they are too!

Associated goals:
- Allocating more of my time to serious Mailman development.
- Getting more women involved in open source.
- Improving the usability of Mailman 3.0
- Speeding up development of the Mailman 3.0 UI.
- Doing some teaching/mentoring since I love it but won't be doing it at work this year.

Thing 2: e-textiles



The first thing I did after I got home from GHC11 was sleep. But when I woke up in the middle of the night, the second thing I did was order stuff from SparkFun. :)

I've ordered a couple of simple e-textiles kits and the goal will be to play with them. I made an awesome monster at the GHC e-textiles workshop and I was eager to do more. The end goal is to build a set of lights into my new coat that respond to my movement in some way (See the tentative wishlist), but for now I'm going to make a lit cuff/armband for walking at night and experiment with the neat little aniomagic chip 'cause it looks like so much fun!

Associated goals:
- meeting more people in the local community
- actually becoming a member of a hacklab to support my projects
- making it safer for me to walk home in my beautiful-but-not-visible new black coat
- experimenting with e-textiles
- doing some more hardware-oriented projects
- making sure I had a project that would take me away from the computer

Not-quite-a-Thing 3: Not biting off more than I can chew



A common theme at GHC is reminding people that we have to really be careful about time management so that we don't get overloaded, so I'm choosing those two things that cover lots of my personal goals, and I'll aim to do them well and save the other things I want to try for later. Wish me luck!

I'd love to hear how other people are using what they learned at GHC11!
terriko: (Default)
2010-07-08 04:47 pm

HotSec & LinuxCon or How I wound up speaking in 2 cities in 3 days (totally different topics too!)

My paper was accepted to HotSec! This is the web visual security policy research I've been working on for a while in various forms, but this is my first proper paper on the subject (although some of the related issues were touched upon in my W2SP paper). Getting in to HotSec is rather a big deal, as it's among the top publishing venues available to me. I was one of 11 papers chosen (out of 57). Go me! So I'll be heading down to DC on August 10th to present it. If you're curious, we should have the final camera-ready copy done in a few days.

My HotSec acceptance causes a bit of a logistical problem, though, since I've also been accepted to speak at LinuxCon on August 12th. It's a bit of a long story as to how I ended up applying at all, but the short and relevant part is just that I wasn't originally planning on submitting to HotSec and didn't realise I'd have such a conflict. (There's a longer story involving speaker diversity issues and good folk willing to go out of their way to work on solving them.)

Anyhow, I really *should* send my regrets to LinuxCon as, academically speaking, it makes a lot more sense for me to go to USENIX Security immediately following HotSec. Especially this year, as I'm hoping to graduate soonish (more ish than soon; don't get too excited) and should be networking as much as possible. But I chatted with my supervisor, and he agreed that it's a bit of a toss-up as to which is more valuable to me: it's nearly as likely that the person I need to meet will be at LinuxCon and that I'll wind up finding a job through open source connections. Raising my open source speaking profile may be just as useful.

What's clear is that Mailman benefits more if I go to LinuxCon, since I'm going to be talking about upcoming awesomeness in version 3.0. The other day, I had someone comment that they didn't even realise Mailman was in active development... ouch. I think getting people interested now, while we're in alpha, is probably absolutely perfect timing. Plus I'm hoping to have some nice stuff to show off from my excellent GSoC students, if they're willing to let me talk about what they've been doing with the archives, and maybe some of the other projects as well.

If you're interested in coming out to LinuxCon, they helpfully gave me a 20% discount code to share. Drop me a note and I'll pass it along (they asked we not just post the code publicly, but I can pop it in a private post later). If you can offer me a job then I'll be able to tell my supervisor I made the right choice. Heh. No, seriously, it's just nice to see people.

Anyhow, I'll make my final decision when I see if the travel arrangements are ridiculous, but it *should* be relatively easy to go from DC to Boston after HotSec, so let's hope this all works out!
terriko: (Default)
2010-06-01 01:11 am
Entry tags:

How do you use mailman mailing list archives?

I'm helping to mentor some totally awesome students working on improved archives for Mailman as part of Google Summer of Code this year, and before they start writing code we'd like to gather some data on how people really use the Mailman archives. If you have a minute, it'd help us a whole lot if you could fill out this survey to tell us how you use the existing archives and how you'd like to use them:

http://spreadsheets.google.com/viewform?formkey=dF9XcGRsYUpsOUtxYjBWRUdnVXN4X1E6MQ

I know I'm always inundated with survey requests, but I promise this one will help us create Archives Of The Future. ;) Seriously, this should lead to more usable and useful Mailman archives, and who doesn't want that? Feel free to pass it around to anyone else who might want to help, and if you have any questions or concerns please contact me.
terriko: (Default)
2010-04-20 02:31 am

Women in computing groups considered harmful?

I saw Hilary Mason's post, "Stop talking, start coding" and realized she had put in 4 words what I'd been debating taking as a personal philosophy.

Theory: The more time we spend on women in computing initiatives, the less time we have to actually get stuff done.

I've been turning down a lot of opportunities lately, and most of them have been in relation to women in $foo initiatives. Where $foo can be all manner of male-dominated geekdom. I've turned down chances at serving on a board of directors, recruiting, mentoring, speaking, giving campus tours, or running new women in $foo groups.

Why? Because I sat down and looked at my time a few years ago, and decided that I wanted to be the sort of person who gets stuff done, much like Sarah Mei articulates the answer in her post, "Why I don't work at Google." I like groups of smart people, but smart people like the GNU Mailman team who were working on version 3 held a lot more appeal that the Linuxchix folk who were just talking.

It'd be easy to blame women's groups as the problem, but then you'd miss the thing that I love most about women's groups:

The best women's groups aren't about separation and segregation: they're about providing an incubator for people who need a leg up to be part of the wider community.

That pretty much sounds like a recipe for making change and getting stuff done, and means the wider communities I care about are getting more awesome people. I love teaching. It's such a rewarding part of my job that I never feel that my time in the classroom working with my students is a waste. So why had I begun to feel guilty about my involvement with incubator organizations?

I recently went to a talk by Jane Goodall. She didn't talk about being a woman at all: she talked about the positive changes she's seen in the world, and how talking about these positive changes helps to inspire people more than shaking her finger seemed to. She believes this so strongly that she spends 300 days a year travelling and talking. But she says she's very careful to choose the right initiatives: Sometimes people are so desperate to Do Something that they sometimes lose sight of the bigger picture. This isn't a problem exclusive to women in computing groups.

So I'm working on a checklist for choosing the right things for me:
  1. Do I want to do this?
  2. Am I the best person for this? (Or can I refer them to someone else?)
  3. Can I do it without negatively impacting my other commitments? (Will it take up too much of my time? Does it happen at a time when I'm busy?)
  4. Am I reasonably sure this will result in getting stuff done, so I'll be able to look back and be proud of what I accomplished?
I still answer my email and occasionally post a blurb from an organization that doesn't otherwise know how to reach women. It takes minimal energy to be polite and provide basic help, and I know I appreciate it when people do the same for me.  But the initiatives that get the bulk of my energy are going to be the ones where I feel like I'm really making change.
terriko: (Default)
2010-03-19 01:11 am
Entry tags:

Mailing lists of The Future! [cue sci-fi sound effects]

I was really thrilled to see both Máirín Duffy and Luis Villa talking about mailing lists this week.

Before I talk about their posts, though, I want to tell you two awesome things related to Mailman:

1. The Mailman logo contest is now closed and the steering committee (including me!) is deciding how to narrow down the field before we put it to a vote. Check out all the awesome entries here!

2. Systers has been accepted as a mentoring organization for Google Summer of Code. And I'm going to be part of their totally awesome team of mentors! They're hoping to get a whole lot of improvements for their mailing list, and a lot of these are the same sort of improvements we generally want for Mailman, so I'm really hoping we'll be able to use some of that code in Mailman 3. Check out our proposed list of projects: I'm most excited about the archives one, but there's lots of awesome in that list. (Student looking for a summer project? Apply!)

Okay, now that I've talked about my announcements, let's talk about those posts I mentioned, and how they're connected to my upcoming work with Mailman.

In which I talk about the awesomeness of dynamic sub-lists and improvements to the archives )

I'm really excited to see other people thinking about the problems of mailing lists... and I'm even more excited to say that thanks to Systers, Google and a team of SoC students, we may be seeing the mailing lists of The Future sooner rather than later!

Expect me to be talking about this stuff a lot over the coming months, and if you've got suggestions regarding Mailman, now's an even better time than usual to tell me about it. Don't be shy!
terriko: Yup, I took this one. The eyes are paper, not photoshop (chair)
2010-01-22 02:14 pm
Entry tags:

Mailman is looking for a new logo

The current GNU Mailman logo
The current GNU Mailman logo
My favourite open source project, Mailman, is looking for a new logo.

I volunteer my time to Mailman because I like working with software that helps people communicate, because I like the Mailman community and because I like the other communities it's helped to build. Programming languages may form the foundation for open source software, but Mailman mailing lists often form the foundation for open source communities. And it's not just the open source folk affected by Mailman: Even my local friends have been using mailing lists to communicate for over a decade now. It's just one of those things that has quietly become infrastructure in my life, and in the lives of many other people. Although we don't have particularly precise numbers, it's a reasonable guess that millions of people use Mailman daily.

I'm working on scheduling some of my time so I can work on Mailman 3. Yes, despite the thesis deadlines breathing down my neck. It's worth it.

So if you've got some graphic design ability, I hope you'll be willing to volunteer some of your time for that new logo. :)

Submissions will be open until February 28, 2010.