Hacker Newsnew | past | comments | ask | show | jobs | submit | cpobrien's commentslogin

I used to work in AWS, and had a less-than-positive experience with the amount of hours I worked in Amazon. I hope that reduced-hour work becomes successful inside Amazon and out, but I worry that those in the position will sleepwalk into working far too many hours, as is what happened with me.


I disagree that it's unfair to expect apps to work without reading. Some apps, sure, but I think a taxi application ought to be usable without being able to read. Consider

1. The user is drunk, and can't understand meaning of the text.

2. The user is preoccupied with other tasks, and skips over the words.

3. The user is illiterate. Functional illiteracy is prevalent in the United States, and being unable to recognize the characters of one's own language continues in many countries, too

My company invests in ensuring many of their products are usable for the illiterate. I doubt it's the only company that does so.


Designing for the illiterate (or inebriated) is a nice goal, but designing for people to change their app into a language they cannot read, and then expecting it to all be easily usable, is not really a particularly great design goal. In that instance, a button could have a single word that 99.9% of people can read, but if you've changed it to Spanish it may render the button incomprehensible. I personally can't think of a single app I use that is completely usable without basic reading comprehension. I'd be curious about examples though (assuming reasonable complexity) because it'd be some potentially useful design that I'd be interested in seeing (and potentially using).

Adding to this, I'm curious how someone who cannot recognise the characters of their languages would use input fields, too? I don't think I know of many input fields which are particularly accessible if one doesn't recognise language characters without using speech recognition (which kind of sidesteps the issue).


As a (hypothetical) example, I'd imagine a pizza app to be pretty usable if I didn't speak the language (illiterate may be a bit much as I'd need to know my address):

Put in my postcode & choose my address; choose a pizza size & toppings from icons/photos.

Add credit card info into a standard looking form & that's it (or even, touch the fingerprint sensor when the fingerprint icon comes up)

I think the trouble isn't that you wouldn't know the Spanish for "OK" (or "pepperoni") but if the app lacks proper information hierarchy so you don't know what to do next.


If you're a foreigner you would probably be tripped by the post code/address. I certainly was when the petrol station asked for my post code in the US! (My card postcode does not fit the US format.)


> is a nice goal

It is the goal, if you’re designing any kind of graphical app.


I know this sounds awful, but perhaps Uber drivers don't want drunk people who might vomit in their cars?


Services like Uber protects the road from drunk drivers. We want it to be easy to use while drunk.


Source: used to work for a direct Uber competitor (not in the US)

Drivers and platforms definitely don't want too intoxicated passengers. Reasonably drunk? Sure. Completely high/passed out/etc? Nope. It's just too much of a mess to handle. There is indeed a vomiting fee, but it's more used as a deterrent for the users: it does not cover the actual cleaning + lost rides on a busy Saturday night...

Bit of anecdata: sexual intercourse in the car is also not ok, and gets you banned. Yes, it happens. Drivers don't like it.


Exactly. The response to my comment misses the point that what is best for the rest of us is not necessarily what is best for the poor driver...


Ironically, part of the reason traditional taxi networks enjoyed their local monopolies in many cities was to compensate them for being considered part of public transport infrastructure, and forced to do the things that's not in their best interest (but is in the best interest of citizens).


I have to +1 this. Not only is it easier to navigate, AWS always manages to run pretty smoothly, while Azure's website makes my 2014 Macbook Pro slow to a crawl. It often takes up to a second for a CSS :hover effect to appear!


There is a mention of the 66 parent merge from Linus himself:

http://marc.info/?l=linux-kernel&m=139033182525831


Whoa, you're right! He referenced it neither by 7-character short hash nor by full hash, and I didn't think to check intermediate hash lengths. I've updated the post to reference that email.


The kernel devs recommend using 12-character commit hash abbreviations: https://www.kernel.org/doc/html/latest/process/submitting-pa... (bottom of the section has a git config snippet)


> "Christ, that's not an octopus, that's a Cthulhu merge"



Never change, Torvalds, never change.


>Anyway, I'd suggest you try to limit octopus merges to ~15 parents or less to make the visualization tools not go crazy. Maybe aim for just 10 or so in most cases.

I'll file this under "problems I'm glad I don't have".


When I opened that page and emptied my cache, I got a lot more than 8MB...

http://imgur.com/a/qt0Zd (59.9MB)

I don't think JavaScript is really to blame for this though, the problem here is they're dumping a whole bunch of full sized images when they could have used thumbnails.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: