TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

8^8 – Find another you

306 点作者 justhw大约 10 年前

52 条评论

gattis大约 10 年前
&quot;Take the 8^8 test and the likelihood of someone else answering in the exact same pattern as yourself is 1 in 16,777,216&quot;<p>That requires A) the distribution of answers is even across all 8 options, and B) there are zero correlations between any two answers in the quiz.<p>Not to mention that the 8 questions have to pretty accurately split the space of human personalities down into eigen-answers that explain the most variance in personality.<p>And there&#x27;s just no way to do pick these questions without having the data to analyze in the first place.
评论 #9553541 未加载
评论 #9553349 未加载
评论 #9553506 未加载
评论 #9554376 未加载
jader201大约 10 年前
It&#x27;s fascinating (not being snarky) that this post has received about 50+ votes since I first saw it, and I&#x27;ve not once been able to pull the site up completely (never made it past the bot detection, and most of the time, won&#x27;t even come up).<p>So I&#x27;m trying to understand the phenomenon:<p>- HN folks are bookmarking it for later based on comments (even though it may&#x2F;may not be worthy of a +1)<p>- HN folks are +1&#x27;ing it based on the idea derived from the comments<p>- Those +1&#x27;ing it are actually able to make it through (though it doesn&#x27;t seem so based on most of the comments)<p>The fascinating part to me is that, assuming it&#x27;s one of the first two, how much +1&#x27;s an (arguably) broken link can get based on comments.<p>The idea sounds interesting, and the site may very well be +1 worthy.
评论 #9554804 未加载
评论 #9554428 未加载
评论 #9554498 未加载
braaap大约 10 年前
Ummm, I like other people precisely because they aren&#x27;t me. My soulmate is not a duplicate of me. They are a compliment.<p>At best, 8^8 might expose someone who ends up making a lot of the same decisions as me. Boring!
评论 #9553552 未加载
评论 #9553747 未加载
评论 #9553247 未加载
tjradcliffe大约 10 年前
The basic claim is false on many levels:<p><pre><code> Very few of us are lucky enough to come across our soulmate within our lifetimes. We tend to choose our friends and spouses from the limited pool of people available in our immediate vicinity: students at our schools, people who live in our neighborhoods, colleagues at our workplaces. Statistically, it is extremely unlikely that we will come across our soulmate. </code></pre> First off, the notion of &quot;soulmate&quot; is bogus romantic nonsense, and the notion that the person most compatible with us is <i>just like us</i> is demonstrably false. My father and I were very similar to each other and fought and argued all the time, because we were both cantankerous, ornery and contrary.<p>My life-partner and I--who are as close to &quot;soulmates&quot; as anyone can be to that basically ridiculous idea--are very similar in some respects, very different in others. The areas where we complement each other are as important as the ones where we reinforce each other.<p>More importantly, the claim that there is some great unexplored mass of humanity where our &quot;soulmate&quot; lurks is bogus. There are only 5000 people in the world. Maybe fewer. If there were more we wouldn&#x27;t keep running into each other all the time.<p>That is, the number of people in our tribe is surprisingly small, and anyone who is sufficiently similar to us to answer the questions the same way is already almost certainly a member of it, so dipping into the pool of random strangers across the world is unlikely to improve the odds much in most cases, and citing an anecdote or three--which some people will be tempted to do--does not change this fact. The human social graph is full of islands.<p>Finally, to work as advertised the test requires that answers to the questions are uncorrelated, which is almost certainly not the case. So most people will find themselves with hundreds of &quot;soulmates&quot;, a very few will have none. Unique matches will be extremely rare.<p>It&#x27;s a superficially fun idea that turns out to be more of a monument to the failure to understand probability than anything else.
评论 #9553885 未加载
评论 #9554776 未加载
btilly大约 10 年前
Interesting idea. Bad execution.<p>There is <i>NO</i> way that performance should be so bad given current load.<p>Some of the questions are frustrating. For instance question #4 is &quot;YOU SAVE AN OLD LADY&#x27;S LIFE. IN GRATITUDE SHE GIVES YOU $100,000. WHAT DO YOU DO WITH MOST OF IT?&quot; The obvious answer for me is, &quot;Save it.&quot; But all possible answers are ways to spend it or give it away. I chose &quot;Move to a nicer neighborhood&quot;, but that really is NOT who I am...
评论 #9555317 未加载
评论 #9554678 未加载
评论 #9557797 未加载
TheMakeA大约 10 年前
There seems to be an awful lot of gratuitous negativity going on here.<p>The site was on Reddit yesterday, now it&#x27;s on the front page of HN. Maybe the author can&#x27;t throw a lot of resources at it, or they used the project as an excuse to learn a new stack. Who knows.<p>Maybe it was slow for you. Maybe it&#x27;s not a mathematically sound idea. Maybe the questions aren&#x27;t ideal. The CAPTCHA isn&#x27;t ideal.<p>Who cares? There&#x27;s no need to be mean about it.
评论 #9555370 未加载
评论 #9554388 未加载
评论 #9553774 未加载
评论 #9554896 未加载
评论 #9553852 未加载
insickness大约 10 年前
This whole thing has been done before and in a way that allowed the questions to be crowdsourced, not just one person coming up with what he&#x2F;she thinks are meaningful.<p>On OKCupid, There are thousands of questions, many of them penned by users. You choose which questions to answer, then mark how important this question is to you. You also choose the answer you want your partner to have. For example:<p>Are you looking for a partner to have children with? [] yes [] no<p>Answer(s) you’ll accept [] yes [] no<p>Importance [] a little [] somewhat [] very<p>If this question were not at all important to you, you would just not answer it.
评论 #9554078 未加载
评论 #9554511 未加载
LordHumungous大约 10 年前
It&#x27;s cool, but it kind of reminds me of Ok Cupid back in the day when they would ask you 200 questions about your personality, and then it would match you with people who gave similar answers. I went on dates with a couple people who were 99% matches, expecting love at first sight, only to find that in real life we didn&#x27;t click at all. It seems that there are many dimensions to human interaction that are not easily captured by these kinds of quizzes.
Nadya大约 10 年前
Took quite a while due to the site crashing several times, but I finished it.<p>I think of it as an interesting experiment - even if nothing ends up coming from it.<p>My largest concern is there are about 3 questions where I was torn between two answers. I&#x27;m thinking I should create email addresses to &quot;cover every base&quot; by alternating my answers on those 3 questions in every permutation. On the other hand, I would feel a bit bad for using up some of the &quot;freebies&quot; that legitimate people might miss out on and have to pay the small fee.<p>@ColinWright<p>&gt;8^8 should be thought of less as a scientific black box, but as a friend who claims to know someone you&#x27;ll hit it off with, and wants the two of you to get together. It might be wrong, but it could very well be right.
评论 #9553429 未加载
hvm大约 10 年前
Yeah, the site is painfully slow. A site with a multiple choice test should handle thousands of users at a time with a tiny server. How is this thing made?
评论 #9554296 未加载
评论 #9553587 未加载
评论 #9553531 未加载
评论 #9553947 未加载
karmakaze大约 10 年前
On too many of the questions there were multiple choices which seemed equally true thus my results would be largely arbitrary.
评论 #9553174 未加载
ColinWright大约 10 年前
<p><pre><code> To prevent abuse, we need to ensure that everyone taking the 8^8 test is a human and not a bot. Click on all of the icons below that represent animals. </code></pre> And then the site is so slow it only shows me 5 of 7 images, the other two failing to load.<p>Good one.<p>Don&#x27;t get me wrong, it&#x27;s a cute idea, although I have to say that I&#x27;m not sure I&#x27;d get along with someone who thinks exactly as I do.<p><i>Edit: I&#x27;ve passed the spam test and started the &quot;test&quot; - most of the answers are &quot;none of the above&quot; or &quot;any one of these 4&quot;. The usual frustrating experience. I mean:</i><p><pre><code> THE KEY TO BEING SUCCESSFUL IN LIFE IS ... </code></pre> <i>How should I know?</i>
评论 #9553054 未加载
评论 #9553081 未加载
lips大约 10 年前
At a party you... ...listen in fear at the sounds of people who enjoy other people, observe small material details as though you were an anthropologist encountering alien beings, depend on searching for beer to consume time, and occasionally find someone who is amused by your off-kilter humor, but who ultimately is not really anyone you&#x27;re likely to form a meaningful sober bond with?<p>Oh, apparently I have to actually be remotely ok in social settings. Seriously, what the fuck.
评论 #9555143 未加载
评论 #9567706 未加载
logfromblammo大约 10 年前
I can&#x27;t speak on behalf of anyone else, but I&#x27;m pretty sure I wouldn&#x27;t want someone just like me as a best friend.<p>To assemble an effective RPG party, you need multiple roles, like warrior, wizard, cleric, and rogue. Effective RL social circles work the same way. I&#x27;m more of an interjecting quipsniper or a technical sidebarbarian, and I really rely on other people to carry the majority of a conversation. If I were to hang out with a copy of myself, there would be no conversation to add zingers, counterpoints, and trivia to.<p>There&#x27;s only so much of that crap that other people can take before it gets annoying, so there&#x27;s not much use in adding two to the same party, unless they&#x27;re scripted, like Crow and Tom Servo.<p>Also, I have a theory of conversation that keys on conversational coefficients. If a group of people are having a conversation, you add together their coefficients. If the sum is one, you have a very natural, comfortable conversation. If it is less than one, you experience some uncomfortable lulls. If it is greater than one, some people get interrupted, can&#x27;t finish sharing their thoughts, or are excluded. If the coefficient approaches two, separate simultaneous threads of conversation will form, and participants will spontaneously rearrange or split themselves between conversations so as to make each one have a coefficient sum as close to one as is possible.<p>People don&#x27;t have a fixed coefficient. They can adjust it within a certain range, that is somewhat dependent on atmosphere and subject matter. For instance, a lecturer who can teach an entire class without losing the attention of the audience can stretch up to 1.0 for conversations, but probably only for that one topic. Someone who has trouble yielding conversational priority may have a lower limit somewhere above 0.5. I suspect that most people can easily handle a range from 0.2 to 0.5. But I max out at probably a 0.4, on very few topics, so my &quot;soul mate&quot; would need to be <i>much</i> more talkative than I am, not at the same level, because it would take at least three of me to have a good conversation.
flurpitude大约 10 年前
9,945 people have taken the test according to the stats at the bottom, and the site&#x27;s painfully slow. Definitely needs some performance tweaking.
评论 #9553301 未加载
评论 #9553577 未加载
ams6110大约 10 年前
Nice way to build a profile of a bunch of email addresses which will then be quite valuable for targeted marketing campaigns.
评论 #9554884 未加载
JasonFruit大约 10 年前
Sadly, my counterpart can&#x27;t take the test either, because neither of us is running javascript.
chrisra大约 10 年前
I don&#x27;t like the idea of finding a soulmate, like there&#x27;s only one person who you can be truly happy with.<p>You find someone, and then work your guts out to be lovable and to love, and you become soulmates.
dash2大约 10 年前
There is a widespread idea that the way to find true love is to spend all your effort searching for one optimal partner.<p>Here is an alternative algorithm: spend some effort finding a good enough partner, and at least as much effort building a good relationship with him or her.<p>I like this site: it shows the widespread idea in its purest form. Kind of a reductio ad absurdum.
评论 #9554505 未加载
6502nerdface大约 10 年前
The site operator may like to know that the site is probably blocked by many corporate proxies; certainly by mine, which categorizes it as<p><pre><code> URL Category: &quot;Malicious Outbound Data&#x2F;Botnets&quot; </code></pre> I assume because the domain name looks suspicious to a classifier that&#x27;s been trained on evil domain names.
评论 #9553638 未加载
beloch大约 10 年前
&quot;If you take the 8^8 test right now all our services will be 100% free for you. This offer is good only for those taking the 8^8 test before we reach the milestone of 1,000,000 tests taken. Thereafter, to cover server costs, there will be a modest fee for connecting with matches. &quot;<p>--------<p>I&#x27;m usually pretty pessimistic about the motives of websites like these. For example, this site seems like a great way to get a huge database of marketing info associated with email addresses. Nowhere on the main page or in their faq does it say they won&#x27;t be selling this data, so they probably will. Use your spam-address if you&#x27;re going to do this!<p>That being said, someone just like me also wouldn&#x27;t pay for a service that&#x27;s likely going to sign them up for spam. Given that payment will be required <i>long</i> before the odds favour a match for me showing up, I think I&#x27;ll just skip it.
dheera大约 10 年前
&quot;Click on all of the icons below that represent animals.&quot; Anyone else notice that only the animals are simply-connected 2-manifolds? A bot could easily solve this.
评论 #9555801 未加载
colinbartlett大约 10 年前
I hope they fix their email regex. This kind of thing infuriates me because this really is the email I use daily: <a href="http:&#x2F;&#x2F;imgur.com&#x2F;F9a2XeG" rel="nofollow">http:&#x2F;&#x2F;imgur.com&#x2F;F9a2XeG</a>
cmstoken大约 10 年前
Completely underwhelming. It&#x27;s just a bunch of random personality questions that give no real insight about the person.
XYEaQMZJvS大约 10 年前
I really like the concept, but I find the implementation disappointing.<p>I know it sounds dumb, but I think of myself as being a rather odd person, and not just like, collecting human skulls odd. It&#x27;s hard to explain, but I don&#x27;t feel that most people would get me, or like me once they got to know me. I&#x27;d love to find someone who&#x27;s just like me (sometimes), so when I first clicked, I was a little excited, perhaps naively so. I was disappointed to find that it was just eight multiple-choice questions, many of which I had a hard time answering. There was no nuance.
评论 #9554208 未加载
abustamam大约 10 年前
I wish there were a way to see what your results were. I took the test this evening; it&#x27;d be neat to retake it in a week and see if my answers match the ones I provided tonight.
iyn大约 10 年前
I love the idea! The &quot;non-optimal&quot; (location &amp; time based) way of finding the people you spend your time with can be quite frustrating, if you think about it. I&#x27;m happy to have a great people around me, but why not meet more interesting people?<p>But, as others already pointed out, there are some problems with the current version:<p>* one could prefer multiple answers on some questions. I think that the test should allow for multiple answers and use ML&#x2F;statistics to find close&#x2F;closest matches. Using True&#x2F;False (exact match) may not be the best. The &quot;workaroud&quot; would be to take several tests with all the permutations, but IMO that is a design flaw.<p>* the questions may not be the best. This is an interesting problem - how can we pick questions&#x2F;attributes that would be a good &quot;model&quot; of human personality&#x2F;mind&#x2F;psychology?<p>* such test need to reach a lot of people and it may take a long time. Therefore, it needs to be done _right_ from the beginning, so that all the people take the same test.<p>Again - I love the concept and I&#x27;m glad that somebody it taking the time and effort to build this. IMO this has the potential to influence who you spend your time&#x2F;life with - so it&#x27;s very important to do this well.
robert-wallis大约 10 年前
I was thinking about how fast it would be to find matches, so I wrote a little C program that runs in about 1&#x2F;7th a second: <a href="https:&#x2F;&#x2F;gist.github.com&#x2F;robert-wallis&#x2F;b7faf94976b153923fde" rel="nofollow">https:&#x2F;&#x2F;gist.github.com&#x2F;robert-wallis&#x2F;b7faf94976b153923fde</a><p>It generates a random number, then tries 8^8 times to generate a new random number and find a match.
SCHiM大约 10 年前
I love the idea! Since I personally believe that every person has a &#x27;perfect match&#x27;. To those who don&#x27;t, consider the following:<p>1) You either do or don&#x27;t get along with certain people.<p>2) This is determined by how they act, look and your dispositions to those actions and looks.<p>3) Given these terms for quantifying the amount of &#x27;like&#x27; you feel towards a person, it follows that there are combinations of act&#x2F;look that you absolutely loathe or love.<p>4) There exists a combination of act&#x2F;look that is your perfect match, the combination that you love the most, ie. Your soulmate.<p>What does not follow is that you would like someone who, as the site puts it, is very much like you. I&#x27;m quite sure I couldn&#x27;t stand myself in certain situations. Maybe your soulmate is actually someone who is entirely unlike you in certain aspects?<p>EDIT:<p>Also it&#x27;s likely, at least in my case, that some of the questions don&#x27;t have the options you&#x27;d pick. I guess it kinda proves that 8 options over 8 questions can&#x27;t be used to arbitrarily identity a person.
评论 #9553913 未加载
评论 #9553978 未加载
stuaxo大约 10 年前
Don&#x27;t you have to kill your clone ?<p>I liked this until they said &quot;soulmate&quot; it works much better as a weird sort of atsy project.
评论 #9553842 未加载
brohee大约 10 年前
The FAQ has some funny nuggets...<p>&quot;Psychology is a science&quot; - Anyone following Retractation Watch may beg to differ...<p>&quot;8^8 is blocked outside of countries where English is both the official language and the most common vernacular.&quot; - Could take it from France, no VPN involved...
Codhisattva大约 10 年前
These questions are so heavily stilted towards white male start up culture it&#x27;s laughable.
评论 #9556260 未加载
ExpiredLink大约 10 年前
Slightly off topic, why so many links to &#x27;juvenile&#x27; questions on HN?
Diti大约 10 年前
&quot;8^8 is not yet available in your country at this time. Hopefully a translated version will be available soon.&quot; Wait, what? I don&#x27;t need a translation, I just want to give that website a go.
tempestn大约 10 年前
FYI, email address validation doesn&#x27;t support new gTLDs.
hobarrera大约 10 年前
I find that the options to the answers are ridiculous limited:<p>Question 1: &quot;AT A PARTY YOU...&quot; Missing answer: &quot;I don&#x27;t go to parties&quot;.<p>The same applies to most of them. It seems like the author got some friends to write down possible answer, but didn&#x27;t assume that there are people way more different than themselves. I can&#x27;t answer honestly to most of them, so I can see how it&#x27;ll fail horrible to find someone else like myself.
Glyptodon大约 10 年前
Some of the questions need more choices. Didn&#x27;t feel happy with any of the answers for some of the questions.
radikalus大约 10 年前
After 5 attempts, I made it. Out of 17.5k test takers, no matches yet? Birthday paradox be damned.
评论 #9554956 未加载
robocat大约 10 年前
Sampling bias: I actually don&#x27;t want to find me (I already know where to find men similar enough to me).<p>Complements: for many traits, I prefer someone who has a strength where I have a weakness. I like opposing opinions as they help me work out my wrongnesses.
评论 #9554277 未加载
Dansvidania大约 10 年前
I wander if this could be from the guy asking how to spend 100k on AWS.
edem大约 10 年前
At the time of writing there are 15099 filled tests so producing a match (assuming that all 8 answers must be the same) is highly unlikely (0,08%) if I assume even distribution.
评论 #9555373 未加载
frevd大约 10 年前
Lol, fail - I don&#x27;t understand the question:<p>#2 &quot;REMOVING WHICH OF THE FOLLOWING WOULD MAKE YOU A COMPLETELY DIFFERENT PERSON?&quot;<p>I&#x27;m not a native speaker -- what am I supposed to do here?
评论 #9555651 未加载
sergiotapia大约 10 年前
How can a website this simple be so slow? I don&#x27;t get it.
vannevar大约 10 年前
What if my soulmate is a roomful of monkeys on typewriters? It seems a fallacy to attribute coincident answers with any sort of real compatibility.
facetube大约 10 年前
If someone cloned me, I&#x27;m pretty sure one of us would murder the other one. Similar != compatible when it comes to humans.
rpwverheij大约 10 年前
nice idea. I stopped at the first question though. The answer that would correctly describe my average situation at a party involves a combination of 3 of the mentioned answers and probably 3 others that were not mentioned, so I doubt a test with 8 questions with each 8 answers can match me properly to someone that is very much like me.
cbaker大约 10 年前
Why would you want to meet someone exactly like yourself? You already know you. I&#x27;d rather meet people different from me.
评论 #9553898 未加载
yramagicman大约 10 年前
Obligatory XKCD, though it&#x27;s not a comic.<p><a href="https:&#x2F;&#x2F;what-if.xkcd.com&#x2F;9&#x2F;" rel="nofollow">https:&#x2F;&#x2F;what-if.xkcd.com&#x2F;9&#x2F;</a>
timkofu将近 10 年前
We can read and write English just fine in Kenya. No translation needed.
EGreg大约 10 年前
OkCupid does this
nutate大约 10 年前
wait... soulmates?
Dewie3大约 10 年前
If there was a <i>twin</i> of me out there... we wouldn&#x27;t get along anyway.