Older blog entries for clarkbw (starting at number 57)

ATMs

I've got a few suggestions to those interaction designers who work on ATMs. I know Seth has already discussed his issues with ATM interfaces so I thought I'd throw out my recent experience as well. This is a quick story with some helpful tips along the way.

I drove up to the ATM to make a deposit of a check I had gotten from taking a standby flight on Delta. I entered my PIN and navigated my way to the screen where it asked for the amount of the deposit. I wasn't sure how much the amount was so I grabbed the check to take a look. I see the amount and start typing in the amount on the touch pad. As I enter in the numbers, but the ATM takes me to a different screen before I hit OK where it asks me if I need more time.

The screen that I was taken to asks if I need more time or would like to cancel. As luck may have it the Cancel button was lined up exactly where the OK button used to be on the previous screen. So as I go to press OK for the amount I entered the screen changed and I hit Cancel which then spit out my card and required me to start the whole transaction all over again.

In Interaction Design we have this great thing that I like to incorporate as much as possible and it's called "Feedback". Users like feedback and here's why. Feedback essentially means not surprising users by giving them as much relevant contextual information as you can about the situation. There are several places where the ATM could use lots more feedback.

The ATM has some kind of timer that it uses for security, so people don't leave the machine and the next person can walk up to it and start where the last left off. A dialog screen kicks in because I hadn't switched screens fast enough and it assumed I had left or just needed more time. Now the ATM should be smart enough that the activity of pressing keys pauses the timers count down instead of just timing between screen switches; however it is what it is. One small improvement here in terms of feedback would be a nice progress bar that tells me exactly how much time I have before this timeout system kicks in.

Another little addition to the design of the system would be to switch locations of the buttons for the dialog screen that pops in. Instead of the button location that used to be an OK for my last screen now turning into the CANCEL button for the new screen. Take the time to determine a new location for these NEED MORE TIME or CANCEL buttons that take over without warning.

COSI

This weekend is Open House for Clarkson so I'll be hanging around the lab as usual helping out. Read into that as "I'm here for the free food". However afterwards it looks like I might go hiking, so that's exciting. Hope for nice weather.

Thesis

I've got great review comments from Jeanna, my advisor and I'm putting together the final copy to send back to her before I go out tonight to Flip-Night.

End of line.

QOTD's

Right, so I never answered Matt's original QOTD and now it looks like we're on to another one. Honestly I don't know if I can keep up with a question everyday, but I can try. Two or more questions a week might be more my style, so TOMQAW sounds pretty cool to me. :-)

First Distro I ever used?

It was red hat, I don't remember what version it was because I didn't care back then. (I guess I don't really care now either as long as its the latest). So I talked about this in my IBM Linux Scholar Challenge article in developerWorks. Basically my experience went like this: I tried to install Linux, it took forever because I didn't know how to navigate through all the mirrors to get to the files I needed. Oh but then I didn't know the files I needed, all the instructions seemed cryptic with using special DOS tools to format a floppy in a special way. Then boot off of that and know what kind of network card I had so I could start a net install. Well I got that far and then I borked my MBR and lost my Windows partition thanks to not knowing the ins and outs of that Druid partioning software. I installed windows again and didn't touch Linux for another two years.

Now to Donnie's question

What was the first computer you ever used?

Ha! I'm too old to be blogging about this with you guys. My first computer was a Commodore 64. It didn't have a hard drive, but did come with a tape drive. I played a game called 4th and Inches on it a lot with my Dad. I also used to get the Commodore 64 magazine which had programming code that you could type in (which I did) and then run; the code was usually new games. Remember that there wasn't a hard drive, so everything I typed (pages and pages of code) was lost as soon as I turned it off. I used to program a game and then play it for a week at a time leaving the C64 on the whole week. Aparently there is still a large community of people who love the C64, I'm not sure why. After that computer we got a Tandy Sensation! It was the first computer to come with a CDROM drive builtin, and boy did that thing run a mean version of windows 3.1

Now to Kyle's question

What is the first programming language you learned, and what did you use it for?

The first programming language I learned was Basic in a High School programming class. I used Basic to create a great game of Pong. Well actually it wasn't a great game due to an extra feature or too that I put in the code; the ball would just leave the game every now and then. I think it was a good feature to add as kids today play video games way too much, however my teacher didn't think so and took off points. What a jerk! I had such foresight into this video game problem that so many teens are having and he couldn't recognize that in all my if statements.

Thesis

I made mad progress on my thesis yesterday. I wrote and rewrote over 8 pages as well as coded a whole lot more. My demo app is pretty much ready to show off the thesis ideas. Today I'm going to rework the rest of it and then I can send it out to my committee after a little review from Jeanna.

Not here for long!

I'm leaving on Wed. to go hang out in Saratoga Springs for St. Patty's day. I have a bunch of friends going down there that I'd like to see. After that I'll be leaving to go to Boston the next day for an interview on Friday. I'm hanging out the weekend in Boston so I won't be checking emails or anything during that time.

I have to say that this is the funniest email I've ever gotten from Tim:

[Timothy Fanelli no longer checks his Clarkson e-mail]

I will be out of the office starting Mon 01/19/2004 and will not return until Fri 05/01/2015.

Due to the numerous problems with the Clarkson mail servers as of late, I have opted to no longer check my Clarkson email account. If you emailed me from a clarkson.edu address, then your message has been forwarded on to my correct email address and I will respond to you shortly; otherwise, your message has been marked for deletion and will not reach me. Sorry for any inconvenience this may have caused you.

Austin, Texas

I'm still in Texas for the next couple days. I'll be home on Friday, just before vacation starts. I haven't been very active with GNOME lately, I've set a goal to finish a first draft of my thesis by Wed. the 17th. I've been working on code for my thesis but as of late I've stopped coding and I'm starting to write more. I've accepted that my work only has to be a proof of concept and therefore just needs to be up to "demo" specs.

Soon enough time will clear up and I can back into the GNOME of things. I'm excited about this new release coming up. Of course I'm not excited about all the crap on the mailing lists lately, people need to relax.

I'm meeting with an IBM DE on Thursday for an interview like discussion and to meet others in their group. I sent off an application the other day to another place where I'd like to work, interviews are these coming weeks so we'll see what's up.

Chili

I need to remember to pick up those Texas spices for the Conry chili.

Kyle

Even funnier because the grammar is slightly off

And what is this about?

2pid

And the results for CMU are in! It looks like I'm on the job hunt after all.

Thesis

Now the the grading is finished for this week the race is back on to which one of us can finish our thesis first!

Grading

Again I don't understand why there isn't a better system for these classes. I'll outline my major improvement to the way I would run classes when/if I'm a professor.

Students grade their own papers. That's right, students will do the work of the assignment and then on the day it's due I'll post the grading criterial and the students will get two days to hand in their graded assignments.

My job as the professor would then be to approve or disapprove the grade. If I disapprove of the grade the student gets a 50% of the grade they submitted. If I approve of the grade they gave themself then the student gets that grade.

Not only do I think this will provide students with a valuable lesson on grading assignments, but they'll really realize the value of the work they put in. Also I think the 50% part forces you to be a little modest on the grade you give yourself, so as not to really lose points. Of course if you low-ball yourself chances are I'd give points where points are due, but I wouldn't count on that as a good strategy.

This came about because honestly I can probably tell the grade range a student will receive just by glancing what they hand it. The hard part is assigning an actual number value to that work. Most often I spend lots of time developing formulas to make sure that students grades will all be equal, even, and fair. If the student gives themself the grade it has to be fair to them, they can complain that Jenny or Johnny got 5 more points on the same assignment than they did, but they gave themselves the grade so it's their fault for not giving themselves more points.

Schools

Sorry about the CMU letter Brendan, they are really missing out on a great researcher and a hard worker; it's their loss. (Oh, and Lime Cat is not pleased about this either) I suppose this means I'll be hearing something from them soon as well. Right now I'm a little uneasy no matter what the answer is, a lot more things are depending on my decision with this than I had orginally known.

At least I know that Mike McCabe will keep looking like Janet Jackson long after I make my decision.

Notes on the new FOSA

I talked with Seth for a while on the new File Selector for GTK and I think we had some good conclusions.

The path finder needs to shrink some, we're thinking of removing the items that are to the left of $HOME. The items are statistically used less so they will be only taking up extra space as well as constantly being in vision but having no real meaning to the user. Outside of the $HOME directory it is reasonable to have the full path appear as normal.

Seth and I didn't get to this one, but I believe it was discussed on the fury of messages on the usability list about the FOSA the path finder should keep the full tree path as long as the user doesn't go down a different branch. So if you go into first/second/third and then use the path finder to go back to first/ the second/third/ directories remain until you diverge on a different path.

Grading

I'm grading right now, which is something I really despise doing. If these undergrads were so good, their homeworks would come pre-graded and I could just approve or disapprove the grades. :-)

Thesis

I'll be done the grading for this locking homework tonight, which gives me tomorrow, the weekend and next week to work on my thesis! I should fix up the next assignment for OpSys somewhere in there, but I really want to crank on my thesis for a while.

Tuesday's Gone With The Wind...

Bored

Well it's not my homepage, but I was messing with Google for a second today and noticed that I'm back on top. I think it's because of all the blog links from Planet COSI and the COSI bloggers. I'm finishing up the next-generation of Planet COSI while I have a little down time in Texas. I'm using the jdub Planet code now, which is much easier to control and easier to modify as they have done a lot of work in cleaning things up. The new code shouldn't drop as many feeds as the old stuff was, I was spending too much time hunting down UTF encoding errors that I want something that really just works.

Grading

I'm working on the OpSys grading now, I really don't want to look at another homemade shell for as long as I live.

OpSys Hacking

I'm looking into the last assignment for OpSys while I have some time. I think it's going to be a cool break the linux kernel assignment that actually reads in the kernel memory and tries to mess with the internals.

IBM

I emailed Tony Temple while I was down here to see if we could meet. He's a pretty busy guy but I'm setting up a time to call him to talk about what kind of work he and his team are doing. Tony is the VP of Ease of Use for IBM and works in Austin, Texas.

I also gave a talk at IBM the other day about GNOME and the future collaborative roadmap that will be coming out soon. The talk was focused on showing off the desktop and discussing ways in which GNOME has integrated it's applications (like evolution) to provide better interaction and more secure desktop environment.

I'm working in the Soft. Eng. lab right now, I've been here all day. I really like it in here, there's rarely another soul in the lab with me, so it's always quiet and the computers are smoking fast. I was logged into 5 of them and switching from station to station as I waiting for things to load on the web. The only problem I see so far is that AFS is crapping out on the Linux installs here. I don't know if it's the AFS servers at Clarkson or the installation of the AFS clients on these machines, but every 20 min or so the machine locks up and I lose all my files. Then they all return after about 2 mins of hang time.

Damn Thesis!

If I could only get this damn thesis done I can't even imagine the free time I would have. I've done nothing but work on this for weeks I feel like I'm getting barely anywhere. I'm about finished a conference paper on my thesis topic, I'll be submitting to Interfaces for All this Sunday, but it appears as though it may be a photo finish like usual.

Oh yeah, I saw Mike McCabe the other day and he is looking good. :-)

Hacked away on my thesis today

Tomorrow is going to be about writing more of the draft conference paper that I need to have done for Sunday. I have about 2 pages written right now and I need to write about 2 more pages before I can start organizing and reviewing it. Plus once I get some more screenshots put together I think it will all fall in place pretty swiftly.

COSI

Tonights meeting seemed to go really well. COSInauts were hanging around getting things done, I think everyone learned a little bit about how to use the tar and ssh commands today, along with some shell scripting. I hope the monthly project idea is going over well, it's really not meant to take over any other projects people are actively producing. The goal is to get things done in COSI that are really junk work for an entire 'team', but everyone can learn something from doing it and if we can automate the entire build/backup system that will be one headache gone. Perhaps Justin can handle organizing the next COSI meeting since he's in charge of web services this meeting could be about subversion/cvs or maybe about setting up an MTA with mail aliases.

Evolution

I have to say that I really love the new evolution, it is coming together so nicely that I've been using it for my daily email client for about a month. I'm on the CVS HEAD version, which is the only one i'd recommend right now. I really enjoy the multiple calendars, it helps a lot to be able to sort out where the different calendar entries are coming from. The system is similar to categories for your calendars, which has made me wonder if having these different calendar sources makes calendar categories obsolete in some way? I don't use the categories at all, I just use the check boxes for the calendar sources, which is about as detailed as I think anyone would get.

ITL

I used the Windows machines in the ITL lab the other day and I'm sorry to say, but those things are riddled with viruses and other junk that I don't want anywhere near my stuff. I'm worried about just typing up my paper in there, I don't want some kind of macro virus embedded in my doc file. Plus I need to upload the file somehow, which means I need to enter in my password while all those keystroke loggers are running. I'm not parranoid about this stuff, I think it was the popup ads that appeared on my screen when I wasn't even web surfing that clued me in.