 |
|
My LiveJournal blog name - valhenson - was the last major holdover from my old name, Val Henson. I got a new Social Security card, passport, and driver's license with my new name several months ago, but migrating my blog? That's hard! Or something. I finally got around to moving to a brand-spanking-new blog at WordPress:
Valerie Aurora's blog
Update your RSS reader with the above if you still want to read my blog - I won't be republishing my posts to my new blog on this LiveJournal blog.
If you're aware of any other current instances of "Val Henson" or "Valerie Henson," let me know! I obviously can't change my name on historical documents, like research papers or interviews, but if it's vaguely real-time-ish, I'd like to update it.
One web page I'm going to keep as Val Henson for historical reasons is my Val Henson is a Man joke. Several of the pages on my web site were created after the fact as vehicles for amusing pictures or graphics I had lying around. In this case, my friend Dana Sibera created a pretty damn cool picture of me with a full beard and I had to do something with it.
It's doubly wild now that I have such short hair.
14 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
ZFS now has data deduplication - with the right configuration options for safety and performance in a compare-by-hash based storage system. From Jeff Bonwick's ZFS deduplication blog entry:
Given the ability to detect hash collisions as described above, it is possible to use much weaker (but faster) hash functions in combination with the 'verify' option to provide faster dedup. ZFS offers this option for the fletcher4 checksum, which is quite fast:
zfs set dedup=fletcher4,verify tank
The tradeoff is that unlike SHA256, fletcher4 is not a pseudo-random hash function, and therefore cannot be trusted not to collide. It is therefore only suitable for dedup when combined with the 'verify' option, which detects and resolves hash collisions. On systems with a very high data ingest rate of largely duplicate data, this may provide better overall performance than a secure hash without collision verification.
What I like is (1) the user chooses the hash function based on their security and performance needs, (2) the system can optionally check for hash collisions, and (3) the ZFS storage pool design makes it easy to migrate data to a new hash function if necessary. ZFS is the first deduplicating storage system I know of with these features. (Do let me know if there are others out there!)
3 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
For my money, the Bay bridge can stay closed. I couldn't believe what a difference it made when the Bay bridge was closed over Labor Day weekend. My crappy, noisy, stressful SOMA neighborhood became quiet and pedestrian-friendly. Birds sang. Property values would skyrocket. Even just closing half the lanes would make a huge difference.
Anyway, to do my teensy-tiny part in making this a possibility, I just want to remind people that you can work around the Bay bridge closure even if your ultimate destination isn't on public transit. Just take BART across and get a Zipcar the rest of the way.
Post A Comment | Add to Memories | Tell a Friend | Link
Today I was bicycling back from the Golden Gate Bridge and got trapped in a surprise Columbus Day parade in North Beach. This happens to me a lot - I am out minding my own business somewhere in SF and suddenly I am surrounded by thousands of strangely dressed, probably drunk people who are in between me and my home. I have been surprised by Bay to Breakers, Pride, Folsom Street Fair, SantaCon, a zombie flash crowd, the Blue Angels, and now a Columbus Day parade. Did you know there was a Columbus Day parade in San Francisco? Neither did most of the people watching it.
Is there an RSS feed somewhere out there for "things happening in San Francisco involving several thousand people that might screw up your day OR even be fun to attend?" 'Cause I would subscribe to it.
5 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
I just posted the union mounts (now called writable overlays) design doc to LKML and fsdevel:
[RFC] Union mounts/writable overlays design
Along with it is a new release, featuring:
* Rebase against 2.6.31
* pivot_root() and rename() support
* jffs2 support (thanks to Felix Fietkau)
I booted this one on my laptop with a root file system as a writable overlay, and got all the way up to the (text) login prompt with only a couple of complaints about missing chmod() and link() support. Yeah!
The patches that need to be rewritten have useful comments in their commit messages, and the fixes are described in detail in the design doc. So if you want to jump in and do a little bit of development, you can do so even if you're not a VFS expert (which is most of us).
As usual, all the good stuff linked to from here:
Writable overlays (a.k.a. union mounts)
Have fun!
Post A Comment | Add to Memories | Tell a Friend | Link
My laptop has rivets in it:
My cat knocked it off the desk about a year ago and broke one of the hinges. It spent most of the following year wrapped up in gobs of electrical tape with an assortment of picture hanging parts as braces. Last weekend, my boyfriend basically splinted the broken hinge with a combination of sheet metal, double-sided adhesive tape, and rivets. I think the new hinge is actually smoother and less floppy than the original Dell XPS M1330 hinge.
4 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
I only work for Red Hat part-time; the rest of the time I write and consult on file systems. I'm experimenting with a new consulting service: on-site training classes on the file system topic of your choice. The goal is not a canned PowerPoint lecture, but a combination of in-depth personal instruction and an opportunity to pick my brain on whatever file systems problem is keeping you and your engineers awake at night. If you've seen me speak at a conference, it will look a lot like that plus an extra hour of question and answer time.
For the test run, I'm offering an in-person two-hour class, unlimited attendance, Bay area location only, at the discounted rate of $2500 for the first three customers to set a date. If the experiment is a success, I'll offer more classes at a substantially higher rate. :)
If you're interested, talk to someone who can approve training expenses at your company, and then email:
info@vaaconsulting.com
Post A Comment | Add to Memories | Tell a Friend | Link
|
|
I just read Charlie Stross's rant on reducing his household's carbon footprint. Summary: He and his wife can live a life of monastic discomfort, wearing moldy scratchy 10-year-old bamboo fiber jumpsuits and shivering in their flat - or, they can cut out one transatlantic flight per year and achieve the equivalent carbon footprint reduction.
I did a similar analysis back around 2007 or so and had the same result: I've got a relatively trim carbon footprint compared to your average first-worlder, except for the air travel that turns it into a bloated planet-eating monster too extreme to fall under the delicate term "footprint." Like Charlie, I am too practical, too technophilic, and too hopeful to accept that the only hope of saving the planet is to regress to third world living standards (fucking eco-ascetics!). I decided that I would only make changes that made my life better, not worse - e.g., living in a walkable urban center (downtown Portland, now SF). But the air travel was a stumper. I liked traveling, and flying around the world for conferences is a vital component of saving the world through open source. Isn't it? Isn't it?
Two things happened that made me re-evaluate my air travel philosophy. One, I started a file systems consulting business and didn't have a lot of spare cash to spend on fripperies. Two, I hurt my back and sitting became massively uncomfortable (still recovering from that one). So I cut down on the flying around the world to Linux conferences involuntarily.
You know what I discovered? I LOVE not flying around the world for Linux conferences. I love taking only a few flights a year. I love flying mostly in the same time zone (yay, West coast). I love having the energy to travel for fun because I'm not all dragged out by the conference circuit. I love hanging out with my friends who live in the same city instead of missing out on all the parties because I'm in fucking Venezuela instead.
Save the planet. Burn your frequent flyer card.
11 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
|
| 2009-09-01 06:29 |
| Spears |
| Public |
|
I was rummaging through my old pictures and found this one of the sign at the security checkpoint at Lihue airport in Kauai:
Damn, I forgot and left my spear in my purse!
2 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
I have started blogging over on the Geek Feminism blog. My first post:
The Thank-You Meme: The coreboot project
This one's for you, Matthew. :)
Some years ago, I quit women in open source activism because it wasn't clear to me that my work was having any positive effect on the world. (It was definitely having a negative effect on my world!) I still haven't answered the question of how to best use my time to help women in open source, but I'm dipping my toes back into the water for this one. Thanks to Kirrily Robert for getting me inspired again.
2 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
The gods confound the man who first found out
How to distinguish hours! Confound them too,
Who in this place set up a sundial,
To cut and hack my days so wretchedly
Into small portions!
-Maccius Plautus, c. 200 B.C. When I look at my calendar and see a pristinely empty weekday, a feeling of joy wells up inside me. "Yes," I think, "I can get so much code written!" A day without meetings is a joy forever. And whenever I have the misfortune to glance at a manager's calendar, filled with meetings from beginning to end, I feel both pity and wonder. Paul Graham wrote an excellent essay explaining the difference between programmer time and manager time in Maker's Schedule, Manager's Schedule. Basically, programmers need large undisturbed tracts of time, something a manager doesn't need, can't get, and can barely comprehend. Next time your manager wants a meeting with you, send them a calendar invitation with subject "Read this" and the URL to this essay. But remember, it takes two to protect your time. It's your job to refuse to go to unnecessary meetings. If someone wants me to go to a useless meeting, I make them a deal: Send out an agenda in advance and I'll consider it. I never, ever get an agenda. I am proud to say that Red Hat (my corner of it, at least) gets programmer time implicitly. I average perhaps one meeting a month, and it is usually one I have organized myself. The biggest problem is remembering how to get a conference call number.
1 Comment | Post A Comment | Add to Memories | Tell a Friend | Link
Gregg Pollack over at Envy Labs put together a video of OSCON speakers giving their talks in 60 seconds (or frequently less, post-editing). I'm in this segment about 0:55 in, talking about btrfs.
My main take-away: I need to wear lipstick more often. This is representative of the kind of deep thoughts that fill my every waking moment.
Post A Comment | Add to Memories | Tell a Friend | Link
|
|
I recently read Womenomics: Write your own rules for success, a practical guide to working fewer and more flexible hours aimed squarely at women with both children and jobs. It doesn't just tell you how to renegotiate your job, it also spends a good chunk of time pumping up your self-confidence enough that you are willing to actually try negotiating. At the end of that part of the book, the average woman's confidence might rise to, say, 50% of that of a man with comparable qualifications - truly a magnificent achievement.
The book was thought-provoking for me, despite being mostly review for me as a veteran of Women Don't Ask and Ask for It. The main focus of the book is women who want more time with their children, which made me more than a little queasy since I believe that the path to equality lies in men doing more of the work of raising children. Instead of finding ways for women to cope with the second shift, we should encourage men to work 50% of the second shift.
Yet, it was undeniable that I made the exact same changes in my career for reasons having nothing to do with kids. A few years ago, I realized that climbing the corporate ladder was not going to make me happy, quit my job, turned consultant, and then got a steady gig doing 30 hours a week and writing on the side. Any temptation to go back to my old workaholic programmer lifestyle was obliterated when a bizarre and mysterious constellation of minor health problems was finally diagnosed as a major health problem. I simply don't have time to be a workaholic any more.
When you spend the slightest amount of time thinking about it, you realize most people want fewer and more flexible hours. Here, as often is the case, women with children are just the canary in the coal mine. What makes them happier and more productive is what makes nearly everyone happier and more productive; they are just the first to reach the breaking point.
The second take-away was a clear view of modern work ethic as, basically, stupid - a holdover from the relatively brief factory era of labor. Hours of work became tied to a massive constantly moving production line; the institution of fixed unvarying hours of work grinds on unthinkingly into modern society even where violently counterproductive or completely unsuited to the task. Every time I trudge off to another mid-week doctor's appointment, I wonder guiltily what I would do if I had to work a "normal" job. I think I'd work myself into miserable government-funded disability by age 40, which makes no sense for anyone.
"Womenomics" pained me in many ways, but I consider it redeemed by one factor: the pep-talk buildup explaining why women are valuable employees and have the power to get a better deal. Many a book on negotiating skips straight to the how-to without covering the why-for. Why bother learning the strategy if you don't have the guts to apply it? If all a reader gets out of this book is greater self-confidence in her value to her employer, then it's well worth the investment.
Many thanks to Valerie Bubb Fenwick for giving me a copy of "Womenomics," continuing a long tradition of shameless book-swapping. You da best.
8 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
I wrote an article on the origins and design of btrfs, which, because I was too busy with OSCON to post about it last week, is at this very moment available free to all:
A short history of btrfs
I originally intended for it to be an overview of btrfs from the user/administrator point of view, but it morphed into this weird present tense gonzo journalism pseudo-documentary. Kinda weird, but people seem to like it, and I didn't want to claw my eyes out while writing it.
In the article, I briefly compare and contrast btrfs and ZFS architectures. After you read this, you should be able to explain the basic differences between ZFS and btrfs in 60 seconds or less while twirling your glass of champagne. The other web 2.0 nerds will be TOTALLY IMPRESSED. I swear.
5 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
|
|
Okay, I gave the Facebook thing a try. And I like being able to keep up with two or three of my friends who live far away. But that is the shiny little gold coin buried in a mountain of shit.
What initially started out as an irritation - people trying to use Facebook's internal messaging system to communicate with me in a real-time manner - has become symbolic of why I hate Facebook. Basically, Facebook is trying to create a second Internet, one in which all communication and information is linked to particular people and relationships - and, oh, by the way owned and controlled by Facebook and hey, we can use your photo in ads or whatever we want because all your data is under their terms of service, etc. Wired has a great article on Facebook's Alternate Internet Reality strategy:
The Great Wall of Facebook
Facebook's internal messaging system is 100% pure evil and completely representative of what I hate about Facebook. I can get an email notification that someone has sent me a message, but replying to them via email requires me to look up their email address on their profile by hand - so you don't, you just reply using internal Facebook messaging, which requires me to go back to their damn web site every time I want to communicate with this person. I can see this being very attractive if you have a sucky email provider with bad spam filtering, but you could also write this in such a way that it integrates smoothly with your existing email account. Facebook didn't, because they want Facebook Internet to partition from Real Internet, leaving them with far more control over your online data than Google could ever dream of acquiring.
Don't get me wrong, Google has issues - but they also always implement all the obvious compatibility and data export features you would want. IMAP access to Gmail? No problem. Export your contacts? We got your back. Last week, I met someone working on making it efficient to get all your Google Docs out of Google and on to your own hard drive - it isn't good enough that it's possible, he wants it to be easy.
So, screw you Facebook people, I like Real Internet. I'm going to keep my Facebook account as a bridge between Facebook Internet and Real Internet, but I'm going to use it as little as possible. My great act of protest? Disabling Twitter forwarding to my Facebook status. Ha! Take that, enormous corporation!
13 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
People often ask for slides from a talk. I personally believe that if my slides are useful as a standalone document, then my slides suck. But I'm happy to post the demo scripts I use. Here's the btrfs demo script from my OSCON btrfs talk:
btrfs demo script
Many, many thanks to Chris Mason and Yan Zheng for helping me iron out the wrinkles.
5 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
|
|
The CFP for my favorite Linux conference, linux.conf.au, closes on Friday. That means if you want all the fabulous perks of a speaker at LCA (no sarcasm here!), you need to send in your talk proposal by Friday. I mean, how many conferences take all their speakers on a sunset dinner cruise past the Sydney Opera house?
If you're not sure whether to submit a talk, go ahead and submit one anyway. If you are worried about expenses, LCA organizers put together a (limited, small) travel fund to help cover expenses for some speakers. They also go to a lot of effort to find inexpensive lodging options in addition to the usual corporate hotel rooms.
LCA is still worth attending for even the most jaded open-source globe-trotter. I have a box full of conference badges (literally); out of all these conferences, LCA ranks first or second in nearly every category: technical content, quality of talks, professional organization, networking opportunities, speaker perks, nice people, fun parties, beautiful location, low cost. If you are organizing or planning a conference, you should attend LCA at least once just to see how well it can be done.
5 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
Here's an update on my battle to get my security deposit back from Citiapartments. The check just cleared at my bank; I got the whole thing back, including the legally mandated interest. After leaving messages for two weeks, I went in to their office in person and talked to the person in accounting responsible directly. She said something like, "I think that's one of the buildings I can still cut checks for," and went back to her office to investigate. Two hours later I had the check. Even then I wasn't entirely certain that the check would clear, but I checked today and it's in my bank account, only 3 weeks late.
During this process, I have heard of three lawsuits against Citiapartments: a state lawsuit, a federal lawsuit, and an up-and-coming class action lawsuit. I got my money back right away, but I think most other people won't be as lucky.
2 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
My security deposit from Citiapartments is 3 weeks overdue. When I finally got someone to return my call, they told me that the money is "in lockbox by the bank" - whatever that means - and they'll see about getting me my deposit. Maybe. Which is of course complete bullshit and I'm going to keep working on getting my money back.
Turns out I'm not alone: Outlook Not Good For CitiApartments Tenants Complaining Of Illegally Withheld Deposits
So, if you rent from Citiapartments, you should consider taking your deposit out of your last month's rent. The worst they can do is start eviction proceedings, by which time you've already moved out.
8 Comments | Post A Comment | Add to Memories | Tell a Friend | Link
|
|
|
|
|
![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small}