Here's a fun site in KAIST that pokes-fun at it https://www.facebook.com/englishmistakeskaist/

Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.

I recommend this website. It not only has the various color palettes for different purposes but also allows you the method described above of uploading photographs / extracting colors.

The curious exception with unique z-order from last reading: last child added (highest index) is drawn first

Because some events may not be handled immediately, and we need the reference time to correctly interpret the events.

It's also possible to reach the same effect without pressing multiple keys simultaneously, with Sticky Keys. It is an accessibility feature that allows users with disabilities to reproduce the same effect by pressing one key at a time. https://www.wikiwand.com/en/Sticky_keys

left

From "0.0 (transparent) to 1.0 (opaque)", I thought 100% alpha was opaque?

What is the correct way to read this? There is a big debate in the internet about this

So we are experts? :D

Existence of similar models is starting to confuse me, and I'm guessing that this variant is also called MVP (Model View Presenter). Reference

What does fair selection of subjects mean in this context of "Justice"? I'm slightly confused as the Belmont Report seems to explain justice as a concept of fair distribution of burdens/benefits.

Is it suggested by the IRB guideline to compensate users who quit mid-experiment?

I think part of this can be attributed to number of observers in the room. (I wonder what is an average # of observers for the user to feel overwhelmed / uncomfortable)

Despite these guidelines, it seems there are loopholes that allow conduction of unethical research even with an IRB approval. I wonder if the recent scandal with Cambridge Analytica, posed as a research, did have an IRB approval. The Problem isn't Cambridge Analytica: It's Facebook.

This rule is relevant not only for preserving users' privacy but also for preventing biased performance due to evaluation apprehension. It doesn't only apply to bosses, but also coworkers (people whose judgment can impact user's work). I feel that the user from above story would still have felt the burden on her first day with her coworkers laughing, even without her boss's presence.

Maybe the human computer can give signals accordingly (ex : Similar to when hovered over a hotspot, the mouse cursor turns into hand, the human computer may make ambient beeping noise with higher pitch.)

I also think that low-fidelity designs are easier to critique than high-fidelity designs, because the visible lack of finish and effort gives you less burden/(guilt?) in critiquing.

So have some individual thoughts to start off before the group brainstorming to share / build upon others' ideas, aka never start brain storming from a blank state.

One question -- if we want to comment on linked pages, can we use hypothesis in that linked page or should we just comment on this linked text to aggregate all comments in one page?

Couldn't this be somewhat necessary in our DP, depending on our target users? I don't get how "having a potential user" in our team is discouraged, whereas participatory design is another method for needfinding.

It's funny to see that the simple addition of "I'm Sorry" makes the system look very passive aggressive. Another good reason to be extra selective in choice of words for systems.

The major difference between mistakes / lapses is whether users already have executed it.

Would a good analogy understand these slips as the following?

A procedure : Given current state, read available options, execute accordingly.

Where

capture error : same state, correct read, erroneous execution;<br> description error : same state, erroneous read, correct execution<br> mode error : different state, correct read, correct execution?

I think this part suggests that the error was partly due to incorrect recall from memory, which makes it more similar to lapses.

I think some of "Homing" is included in this part; I feel that calculating the next step while moving between keyboard / mouse (two different input types) is more costly than calculating between different keystrokes within a keyboard (same input type).

That's why we love vim!

In Mac, it's interesting how in grid view you can select individual items in different locations for aggregation, whereas in list view you have to select all items between the first and last selection.

On a distantly similar note, I think it's interesting how some interfaces make you retype the entire password when you get a password wrong, whereas some other interfaces allow you to start off where you last left off. What are the reasons behind these different practices?

Merits of well-meant metaphors : recognition! (as opposed to recall)

This may be an interesting context for the ESP game, where two users play until they come up with a same label for a given picture. http://ael.gatech.edu/cs6452f13/files/2013/08/labeling-images.pdf