This is a plausible hypothesis. I’m curious whether OpenAI has considered this already and examined it I feel like an average senior eng could eval this in under two focused days, but maybe OpenAI has less unit-testing than I expect.
Don’t feel like you need to understand 100%. You can always give yourself an hour to read a paper and gloss over some notation. If you read 5 papers over the course of a month, you can go back to your favorite and dive into the notation.
Very cool. Some questions - How many games (?) in a row did you win and how much money was this? How many episodes did you appear on? How does this compare to other top players? Thanks.
We won four in a row, then lost the fifth one in the final round due to two mistakes on our side and a bit of bad luck.
> and how much money was this?
When you reach the finals, you take part in what amounts to a lottery for which the number of tickets you get is determined by how well you play. If you win, you get €2500, if you lose, you get the amount of money you earned in the first round, which is comparatively negligible. We consistently got a reasonably high score in the finals, and with that won the lottery once, which I'd guess is about or slightly lower than the expected success rate.
Do note that the amount was per-team and before taxes, so in the end it amounted to a little bit more than €1000 per person. A nice little bonus, but probably lower than my usual rate given the amount of work put in, so it really was mostly about the fun we had doing it. (And an additional unexpected bonus is that it gave my brother quite some exposure as a comedian.)
> How many episodes did you appear on?
Five: the four we won, and the one we lost.
> How does this compare to other top players?
There haven't been that many episodes yet, and to be fair, I don't really watch the show usually, but it's certainly better than average (since most people just lose the first game they're in), and probably slightly better than the mean winner - though of course, we're doing it with no natural talent and being loyal viewers of the show. There are certainly quite a few who do better than we did though.
There isn't actually a set number of rounds, I think it mostly depends on how long the rounds take. (I think there's about 15 minutes playing time for the first round.) I haven't rewatched an episode to keep track, but my guess would be it's about four five-letter words and four six-letter words. After that, there's always one seven-letter word (in which both teams take turns making guesses).
Curious, not trying to be a jerk. For something like ordering pizza (or even plane tickets), how much worse is calling to make the order as opposed to ordering online?
It depends. If I know I'll make an order only once, calling phone is probably easier. If that's a pizza, and say I have expectation that I might be ordering same pizza every week or every month, then ordering through website is easier, since you can save your credit card information there instead of repeating it on the phone every time.
But even for one-time orders it is good to have an accessible site, since what I mentioned above - I would waste my time if the web site is not accessible. I assume if they had put a notice on their web site saying that it is not accessible for screenreaders - please call this number instead - that would be already a better solution rather than just an inaccessible web site.
From another angle here, I occasionally have problems that make me use a screenreader.
When the nerves that make my eyes work stop behaving correctly, it's also normal for the nerves in my throat to flip out so that speaking feels like breathing razor blades.
Making a call isn't an option, but using a web browser still is, if they haven't gone out of their way to break all the ways a web browser is supposed to behave.
Same. I have severe convergence insufficiency (that is highly variable and cannot be corrected, from a practical standpoint) due to a rare immune-mediated neurological disease.
I either use Kurzweil 3000 (Windows|Mac|Browser) (paid)|JAWS (Windows) (paid)|Voice Dream Reader (iOS|Android)| VoiceOver (iOS) (free)| Talkback (free), depending on what I am doing and how my much my eyes are affecting me at that moment.
I taught myself braille and I have a couple of refreshable braille displays, which I use, depending on how well my eyes are working.
I can digitize printed material well, including STEM material, using a program called InftyReader (Windows).
It is no fun, but you have to do what you have to do.
Blind people are the same as everyone else, but with reduced vision. So when you ask "Why don't blind people just call?" you actually need to be asking "Why don't I just call?" Blind people will have the same reasons.
Most people I know prefer to use an online system to avoid talking to people especially if there's a complex process or lots of information to enter. Blind people feel the same way. A phone based option is not "offering the same service" to blind people.
If there was an online system that only worked with screen readers and the company said "sighted users can just call instead" I wouldn't use that company, and if they had the best price offering I'd be very annoyed. Making an inaccessible website is exactly the same.
I’m not blind. But I would be furious if I had to call to place an order rather than submit a web form, merely because the site couldn’t be bothered to adhere to basic standards and instead used meth-addled “code artisans” practicing resume-driven development.
Coming from somebody who is disabled: this is what life is like for a disabled person.
There are many, many indignities that we encounter, even nearly constantly due to the way society ignores our needs.
If you go and check out some disabled activists' Twitter accounts, you may be shocked at the anger and the lack of "decency". But, put yourself in their shoes, and realize that they have been forced to deal with systematic and near-constant indignities, and many of them have been forced to fight for their mere existence as human beings.
I agree with such activists. It's stupid to have to make a phone call when a website should be perfectly capable for this role.
While I don't claim to know what such disabilities are like, I feel I have experienced analogous frustrations. Back when Linux on the desktop and Firefox were catching on, I remember it being a crapshoot about whether critical sites (like banks and tax filing) would play nice with non-IE browsers, and I had to have a PC/Windows/IE setup as a fallback. Same issue: the use the meth-addled design that breaks any non-mainstream clients.
I also use Tridactyl (and before it, Vimium and pentadactyl), an extension that lets you click links from the keyboard, which is a huge UI improvement and (along with other keyboard input methods) speeds up web browsing significantly. It's generally good at detecting links, but the same sloppy design and over-clever features make clickable elements undetectable and frustrate this enhancement.
And for the kicker ... often times, these improvements "for the disabled" end up benefiting everyone else even more, but designers/buisdev people don't get it! See my previous comment about the Curb Cut Effect [1].
The web was designed with screen readers in mind. It is a serious regression to find major sites telling blind users to call in. This isn't the 60s.
Are those standards sufficiently complete as to be referenced by regulatory agencies?
If so, what's keeping us from adopting standards or at least recommendations for legal compliance?
I think that Domino's assertion is that the current framework is too vague, but I regularly see the argument that there are standards to follow. So... What are those standards, and why aren't they law?
Obviously you should just be allowed to use the same services as everyone else. The web is accessible by design, so all sites that are not accessible are broken. The people that made them broke them.
But, just for the sake of your curiosity, American Airlines charge $35 per ticket[1] when you call them to book.
Stupid franchisee. I kept a coupon active for years after the promo ended for one guy. He liked his $11.99 Veggie lovers. The coupon was hidden, but if you manually typed the code, it worked. I called him personally after he complained to corporate about the promo ending. I walked him through how to get the deal and he ordered weekly until I sold the store. The new owners deactivated all old promos. He had my cell phone number, so I got to find out, after I sold the store, thst he now orders Papa John's.
Source: former multi-store, multi-Rolex winning Domino's franchisee.
It was $5 cheaper than regular menu price. But he ordered a literal 100 times a year with the coupon, tipped the delivery driver $5 every time and the order still was more profitable than $15 of chicken wings.
Once the coupon left, he went to the competition. I used to spend about $1,000 a month in advertising just to get a few new customers to call. Keeping the ones you have is orders of magnitude cheaper, even if they complain about "cold pizza" once a month or so to get a free pie.
I didn’t quit fb but I unsubscribed from pretty much everybody. I think I’ve posted once in the last 6 years, though occasionally I come up tagged in friends’ pics. I think I’m better off. I occasionally miss engagements of my friends though so this is a downside. (I’m 29). I continue to use messenger to chat with friends, including one good friend who has a budget phone plan (meaning fb messages are free while texting this friend isn’t for her).
I guess I’m posting this here to say that there is a middle(?) ground that works for me. I rarely find myself envious or jealous on fb. I will confess that I do experience these emotions in my offline life (much less frequently than daily, more often than monthly). Come to think of it, while I spend under 20 minutes a week on LinkedIn, I feel some negative emotions while browsing this feed. I’ve just realized I should limit my LinkedIn better.
I’m not posting this to brag, just to share another perspective because it felt relevant.
