jofish22: (Default)
I just noticed there's an article about technology in long distance relationships over at Forbes.com. It talks about the VIO for a while, and a few other prototype devices, some of which you might have seen in the same alt.chi session last year.

Nice wood.

Feb. 11th, 2008 12:43 pm
jofish22: (Default)
http://groovewoodworking.carbonmade.com/

My friend Leah has put up a site about her custom woodworking. Some truly glorious work here.
jofish22: (Default)
i'm giving up on travel today and staying in sf with jess and colyn. Same flights tomorrow. Now to use some of that amex travel delay coverage..
jofish22: (Default)
Those of you planning on attending CHI might want to check out the prices this year. The registration process strangely obscures the final price until the end of the process, but if you step through, you'll note that

One day workshops are $225 (up +66% from last year's $150)
Two day workshops are $300
Student early-bird registration is $400 (down 6.25% from last year's $425)

I understand that the CHI committee did try to find a way that we wouldn't have to pay VAT on the conference fee, but they were unsuccessful, so it turns out that these prices are subject to +20% VAT. (Which, I assume, was a surprise.) That means that an early-bird student registration for a one-day workshop and the conference -- a pretty reasonable undertaking -- comes to

($400 + $225)+20% = $750, which is up 42% from 2007's $375 + $150 = $525 [EDIT! I can't add.]

I suppose I should be over complaining about the price of CHI by now (here, here), but it's just feeling... obscene.
jofish22: (Default)
I'm co-chairing alt.chi at CHI this year; alt.chi is where to publish things that you can't get published at chi. As we say in the call:


With alt.chi 2008, we want to open up for unusual, challenging and thought-provoking work that might not otherwise be seen at the conference. alt.chi is a place to experiment with how CHI submissions are presented, submitted, reviewed and selected. alt.chi is CHI's breathing hole, the space for change, where new ideas can be tried out and experienced. alt.chi 2008 is your chance to present that paper you always wanted to write but you knew would never get through the conventional review process.
...
Alt.chi invites controversial ideas, novel prototypes, failed but valuable user studies, bold experiments, and anything else that can give a fresh perspective on CHI. We invite submissions that explore technical or practical limitations in technologies or methodologies; that introduce promising, although currently non-viable techniques; that critique the current state of the field; and that explore topics outside of current discussion. We particularly welcome topics on CHI 2008's theme, art.science.balance. We invite work that would otherwise not have been presented at CHI 2008, because it is too controversial or outside of the norm.


In addition to providing an opportunity for novel and interesting work, we're also doing the process in an interesting way: it's open, non-anonymous submission, open, non-anonymous review: anyone can submit, anyone can review. The big change this year is that we've completely integrated reviews with discussion: there's the opportunity for discussion between authors and reviewers in a constructive way, rather than the hands-off review-rebuttal series of normal CHI reviewing.

Submit your papers to: http://www.chi2008.org/altchisystem/
jofish22: (Default)
Uh, what's this plant called? I know it's a common house plant, but I'm not sure what it's called. EDIT: A spider plant! Solved in less time than it took me to edit the post to put in gratuitous amounts of kitten info.



Thanks.

I'm trying to eliminate the houseplants as possible poisioning sources: the current list is

- this one above
- rosemary (recent; no signs of eating)
- ficus (which she does chew/eat, but I think it's harmless)
- bamboo (barely eaten)
- philodendron (but it's in Jenn's room, which she *never* goes in. really.)
- seed onions (which she does play with, maybe eat)
- dumb cane (which could be indicated, but I've never seen her eat it and it looks unchewed)

other sources
- rodentcides (unlikely; they'd be left over from previous owners at least four years ago, in the basement, which has had major work on it)

I'm growing grass for her to nibble on anyway. But there you go.

It's a bit of a stressful time for the Annie.

Well, not really. She's perfectly happy and has no signs of ill health barring a tendency to pee on the futon. But Jenn and I have our knickers in a bit of a twist about it.

Ok, now I think the focus of this post has changed a bit... Has anyone got a vet in the family? Or a good friend who's a vet? I do like my vet, having someone else who knows this stuff look quickly at this to see if rings any bells would be great. (I'm thinking like House, MD, but for cats. Like, House, AVMA.)

Pusscat Annie is having some kidney problems; we've not found anything yet for sure, but still looking. Here's some data:

She's six months old, inside cat, spayed.

She has hypercalcemia;
- blood calcium (14 <11.5 normal)
- ionized calcium levels are high (1.69, <1.47 normal).

- high creatinine levels (4, <2.1 normal)
- blood urea nitrogen (BUN) levels (116, <35 normal),
(which I think that means she's azometic, right?)

Urine protein:creatinine ratio is normal (0.3, <0.4 with azotemia)
Na:K ratio 26, a bit low, I think.
Alkaline phosphatase is high (138, <96)
Potassium a weeee bit high (5.7, <5.6)

Nothing from a urine culture; ultrasound showed small mineraliziations in kidneys, small acoustic artifacts in bladder, no thymic tumour. She drinks a lot more water than I've ever seen a cat do before.

The current plan is to do a PTHrP/PTH/Vitamin D test; looking at the flowchart from Small Animal Clinical Diagnosis by Laboratory Methods which the vets kindly provided for me (p 138 if you're following along at home), that's supposed to be related to distinguishing between hyperparathyroidism (less serious, I suppose, than super-hyper-open-apple-para-thyroidism) and occult Neoplasia (gulp. Although it doesn't, in fact, mean anything about skulls and pentacles.) However, given she seems to be azotemic, then both increased and decreased PTH point to occult Neoplasia, so I'm not sure why the vet seems to think hyperparathyroidism is a possibility. Then again, I'm reading some poor quality photocopy from a book, and she's all, like, qualified and stuff.

Any thoughts? Or know of a good place to ask questions like this?
jofish22: (Default)
We're accepting applications for a tenure-track position at Cornell in Information Science:

"...
Candidates in information related fields are invited to apply. The ideal applicant will have a strong mix of both technical and social science research skills and accomplishments. We are especially interested in the areas of human-computer interaction and user experience research, but will also consider candidates in other areas of Information Science such as digital libraries and game design.
..."

http://www.infosci.cornell.edu/apply/

Figure this might be of interest to some of you.
jofish22: (Default)
I got interested in trying to understand to what degree the conferences in my field were self-similar; how often had people who published there, published there before? It turns out that on average, the answer is about 24% of the time; here's some more detail.




To read it, start on the left hand side. That first column in red is all the authors who published in CHI'83. Anytime one of them publishes again, they show up in red at the bottom of the column. The next column is CHI'85, in orange. Anyone who first published there will show up in orange, just above the authors who first published at CHI'83.

The colors just repeat every seven years because I just did ROY.G.BIV for colors rather than coming up with anything fancy, so the red column in 1983 doesn't have anything to do with the red column 1991.

Have a look at http://posters.jofish.com for links to the large images (nifty for zooming in and looking at the details) and to links to buy posters of it if you don't have a handy-dandy large format printer.
jofish22: (Default)
This might be useful for others if you need to hack the names out of the ACM Conference Proceedings from the ACM DL, since it produces code that's so far away from XML it's not even funny. (How about "class" tags, guys, in the meantime?)

I'll post why this matters soon. Still cleaning it all up.

Python code below the cut )
jofish22: (hard at work)
I have the entire conference proceedings for several years of several conferences on my hard drive. This is for a number of reasons:

- I generally find that if I went to a conference, that massively increases the chance of me refering to a paper from that conference in the future.
- It means that I always have papers I'm looking for around, even if I'm not currently connected to the internet
- It means that keywords from those papers show up in Google Desktop searches, increasing serendipitous finding.
- Deep down inside, I think I don't trust network connectivity.

However, the ACM Digital Library is poorly organized to make this possible. It's impossible to download even metadata in any kind of reasonable format, like XML. (I'm happy to share my code for extracting names from ACM DL proceedings, for example: it involves searching for strings like <BR> <STRONG> to find the start of names.) However, it is possible to download an entire conference proceedings with reasonable ease.

The best way I've found (and I've tried several) is to install http://flashgot.net/ into Firefox, then go to a proceedings page and Flashgot All. Then click Mark Highlight to remove all selections, right click on the menu that comes up, sort by URL, and select all the ones with the title PDF. Hit Mark Highlight (you'll forget that bit again and again) and Ok, put in a reasonable directory name to save to, and boom.

As far as I can tell, this is entirely within the bounds of the ACM DL's terms of use -- and it's defintely ethical on top of that. Hope you find this useful. I do.
jofish22: (Default)
This is the final version, for now, at least.

Evaluation is the process by which we determine the 'value' of a technology; in other words, the quality of the object of study. In HCI, we evaluate at the end of building a technology, although this may frequently serve to inform the design of the next iteration of the technology in question. This evaluation functions much in the way that a drug trial works in pharmaceutical research: it studies whether the technology does the job it was intended to do.

For much of the history of HCI, the tendency has been to measure the value or quality of a technology against task-oriented metrics that reflect HCI's origins in commercial and military enterprises; metrics like 'efficiency', 'time on task', and 'speed'. However, computers and related technologies have become part of life outside of the workplace, in which they have become less of a tool for task completion and more of a medium for socializing and expressing oneself to the world. As such, it has become necessary to measure the value of a technology against a different set of metrics. For example, a blogging technology may need to be evaluated against the amount of control it gives the user over their presentation of self and sense of identity, while the evaluation of a cellphone may need to take into account the physical aesthetics in addition to more task oriented metrics such as a battery life.

In this thesis, I present a conceptual framework, methodology and methods to understand and determine appropriate metrics for the evaluation of non-task-focused technologies, and demonstrate techniques that suitably characterize a technology in terms of those metrics. I draw on the notion of experience, defined by Wright & McCarthy as "felt life", as a useful starting point for these metrics. I integrate this with ethnomethodology, which, with its focus on the behaviors by which people make sense of situations in everyday life, provides an approach that allows for the characterization of experience in a meaningful way. Both of these emphasize the individual and subjective nature of experiences in a manner that would may appear irrelevant when the focus of evaluation is on task completion, but allows for the rich characterization necessary for the evaluation of experience-focused HCI. I use this foundation to demonstrate that approaches that encourage the defamiliarization of everyday experiences by the user can produce rich stories that characterize experiences in a powerful way to inspire better technologies.
jofish22: (Default)
this actually seemed funny when i was in the shower.

mysql_connect(localhost,'wen','good2bedak1ng'); //good king wenceslas looked out
$date=feast('stephen');                         //on the feast of stephen
exec('add weather;weather bohemia -o');         //while the snow lay round about, deep and crisp and even
exec('pom');                                    //brightly shone the moon that night
exec('add weather;weather bohemia -s');         // tho the frost was cruel
$poorman = $_GET["fuel.winter"];                //when a poor man came in sight, gathering winter fuel

mysql_select_db("page");                        //“Hither, page, and stand by me, 
$query = "SELECT dwelling.location, dwelling.type, peasants.name ".
         "FROM dwelling, peasants ". 
         "WHERE peasants.currentloc = 'yonder'";//Yonder peasant, who is he? Where and what his dwelling?”

$results = mysql_query($query))
if(mysql_num_rows($results) > 0)): {              //if thou knowst it, telling,
       {               
       $location = $results["dwelling.location"]; //Sire, he lives a good league hence, 
                                                  //underneath the mountain,
                                                  //Right against the forest fence
                                                  //by Saint Agnes’ fountain.”
       } 



ok. i really should get some work done.
jofish22: (Default)
Thanks for your input, particularly [livejournal.com profile] perlick and [livejournal.com profile] orbitalmechanic. Comments welcome.

The Evaluation of Experience-focused HCI


In this thesis, I will present ways to understand and determine appropriate metrics for the evaluation of non-task-focused technologies, and demonstrate methods that suitably characterize a technology in terms of those metrics.

Evaluation is the process by which we determine the 'value' of a technology or situation; it's the process by which we determine the quality of the object of study. In HCI, we evaluate at the end of building a technology, although this may frequently serve to inform the design of the next iteration of the technology in question. This evaluation functions much in the way that a drug trial works in pharmaceutical research: it studies whether the technology does the job it was intended to do.

For much of the history of HCI, the tendency has been to measure the value or quality of a technology against task-oriented metrics that reflect HCI's origins in commercial and military enterprises; metrics like 'efficiency', 'time on task', and 'speed'. However, as computers and related technologies have become part of life outside of the workplace, and have become a medium and a tool for socializing and expressing oneself to the world, it has become necessary to measure the value of a technology against a different set of metrics. Furthermore, studying the uses of technology in everyday life has emphasized the importance of these non-task-focused metrics within commercial enterprise.

I hypothesize that the notion of experience, as characterized by Wright & McCarthy as "felt life", is a useful starting point for these metrics, and that ethnomethodology, with its focus on the way people make sense of situations in everyday life, provides an approach that allows for the characterization of experience in a meaningful way.
jofish22: (Default)
I'll be giving a talk at Stanford on April 27th as part of the Human-Computer Interaction Seminar.

I'll be talking about evaluation in HCI, how it's changed over time, and how that has to change as we do new and nifty kinds o' HCI.

It'll be broadcast by the Stanford Center for Professional Development and so y'all can get professionally developed. I'm told that "people from local industry will attend to hear a speaker they are particularly interested in", so presumably, local industry people, you could attend if you're particularly interested.

I'll be going out for coffee with the Berkeley/Stanford HCI/STS/New Media group that [livejournal.com profile] chimerically's been organizing afterwards.

This is going to be fun.

BostonCHI

Feb. 25th, 2007 07:35 pm
jofish22: (Default)
I'll be speaking at the March meeting of BostonCHI, about Experience-focused HCI.

March 13, 6:30pm, MIT, Stata Center, Room 32-141.

http://www.bostonchi.org/

Here's an abstract:

We can trace the history of HCI from early experiments in interactive computing to current work on blogging, instant messaging, affective computing and more. Over this period, the ways in which we create and validate knowledge have changed to reflect the problems facing the field at the time. In this talk, I'll discuss the development of HCI's notion of evaluation, starting from its pre-history in electrical engineering and computer science, incorporating methods and metrics from experimental psychology and cognitive science, and debate around expertise that accompanied the rise of in prominence of usability as an approach to HCI. I conclude by discussing what appears to be a growing emphasis on experience-focused, not task-focused, HCI, and how we can evaluate technologies to support such uses.

Should be fun!
jofish22: (hard at work)
Ever wanted a quick-and-dirty gender breakdown of a conference, class, program committee, list of awards or guest list?

Use my new, handy-dandy Genderyzer to get a quick count of men, women, people using only their initials, and people it's just not sure about.

http://cemcom.infosci.cornell.edu/cgi-bin/genderyze/genderyze.py

Genderyzer takes as input a text file of names, like this:

Jimmy Cricket, Ph.D
John "O'Groats" Highlander
F. Scott Fitzgerald
Jane M. M. Murphy
M. Peterson
e. e. cummings
Winnie the Pooh
...

and will tell you that there's three men, two women ('Winnie' is generally female) and two people using just their initials. And there's more! The Genderyzer's information is drawn from a wide variety of sources on the web, and even learns from its users. Try using this list of authors from NordiCHI'06 and contribute to the program's expertise about Scandinavian names! (Well, until it knows all of them, of course.)

Use Genderyzer today!

jofish
jofish22: (Default)
So I'm teaching a sung improv comedy class tonight. With a bit of help from [livejournal.com profile] lemurtanis and [livejournal.com profile] thedan, I threw together this quick cheat sheet for improv songs. Posting because it may, just, be useful to some other people. It's probably not useful unless you already know the pieces, but it's useful for teaching things like this. Let me know if you spot any mistakes. The Great Improvised Song Cheat Sheet 1.1 ) Edit (after teaching): changed order to something more reasonable in terms of sequence -- it was only the other way around so that the page break would be a nice place. We got all the way through to Madrigal, impressively. Blues Jam is so very easy. I wonder why we never did it in Roadkill.
jofish22: (Default)
I went ahead and bought carbon credits to offset my recent slew of flights. Somewhat held back by the stunningly poor and contradictory array of carbon calculators out there on the web: most seem to be incredibly poorly written, contradict even themselves, and lousily designed. But after running the same flight through a handful of places (LHR-LAX), and looking at a bunch of places to buy credits from, I decided to buy up 12 tonnes of carbon credits from carbonplanet.com. They claim to actually plant trees; they claim to keep them for 100 years (Kyoto specifies 30 years); they're in Australia (so I believe they'll actually plant the things; I just *don't* believe it for the ones that claim to plant in the UK, particularly "near you" in the UK. There's just not enough space). They're certified up down and center, their website doesn't suck (although their page o'faqs is a bit overwhelming), and the price AU$23(= US$18)/tonne is low enough to be affordable but high enough to actually be reasonable.

Maybe I'll try and specify that I want all my flights to be carbon-neutral into my next contract. That could be fun...

nice

Oct. 10th, 2006 08:26 pm
jofish22: (Default)
my homie josh g. of epistimographer.com fame has just released a nice firefox extension for reference managing. it rocks. as soon as they get portal.acm.org supported, it's true gold.

think go to the webpage of the book or article, then click on the titlebar and the book or article gets added to your reference page in full detail. boom. just like that.

http://www.zotero.org/
jofish22: (Default)
I was playing around with the laser cutter this afternoon.



Inspired by http://www.oncotton.co.uk/peter/index/A4PAPERCUT_000.htm
Page generated Sep. 25th, 2017 12:41 am
Powered by Dreamwidth Studios