explain xkcd - User contributions [en]

Talk:1425: Tasks

2024-03-13T21:01:12Z

172.70.86.179:

the source of title text maybe is Szeliski, ''Computer Vision: Algorithms and Applications'' (2010), p. 10. --[[User:Valepert|valepert]] ([[User talk:Valepert|talk]]) 06:59, 24 September 2014 (UTC)

[http://www.wired.com/2012/06/google-x-neural-network/ Google’s Artificial Brain Learns to Find Cat Videos] might be useful as a description of the problem [[Special:Contributions/108.162.250.219|108.162.250.219]] 08:34, 24 September 2014 (UTC)

:Sorry for editing your comment but external links have different syntax that internal links so it wasn't working. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 11:21, 24 September 2014 (UTC)

Nice Superman joke there, Pudder! --[[Special:Contributions/141.101.99.49|141.101.99.49]] 10:26, 24 September 2014 (UTC)
:It had been removed in an edit, so I shoehorned in back in :P --[[User:Pudder|Pudder]] ([[User talk:Pudder|talk]]) 12:25, 24 September 2014 (UTC)
Isn't there an xkcd where the estimate of 5 years of work is equivalent to "might take forever?" [[User:Rtanenbaum|Rtanenbaum]] ([[User talk:Rtanenbaum|talk]]) 13:16, 24 September 2014 (UTC)
:I'm pretty sure you're refering to 678. [[Special:Contributions/173.245.52.132|173.245.52.132]] 15:00, 25 September 2014 (UTC)

The link in the description is to a document by {{w|Seymour Papert}} and the [http://books.google.com/books/about/The_Summer_Vision_Project.html?id=qOh7NwAACAAJ book] on the project is also by Papert. Is there any contemporary evidence that it was actually Minsky who assigned the project? I think he just got interested in it later. 14:17, 24 September 2014 (UTC)

[http://xkcd.com/678/ 678: Researcher Translation] is probably what you're thinking of, Rtanenbaum. [[User:Ndgeek|Ndgeek]] ([[User talk:Ndgeek|talk]]) 17:44, 24 September 2014 (UTC)

Is it possible that Randall's selection of bird rather than another naturally occurring object is a subtle reference to the Birdsnap app (http://engineering.columbia.edu/it-crow-or-raven-new-birdsnap-app-will-tell-you-0) which has solved some of the aspects of this problem? [[Special:Contributions/173.245.48.137|173.245.48.137]] 22:02, 27 September 2014 (UTC)

Hopefully I can add that this also seems to make reference to the U.S. Forest Service intention to make everyone have a permit to take pics, etc., in national parks. https://www.yahoo.com/travel/dont-take-that-picture-the-u-s-forest-service-might-98484656432.html {{unsigned ip|108.162.216.21}}

Post the picture to an online forum, say it's a bird, if it's not everyone will correct you as per http://xkcd.com/386/, so scrape forum and if there's a lot of attention it's not a bird, if there isn't much attention it probably is a bird. [[Special:Contributions/141.101.99.78|141.101.99.78]] 23:06, 3 October 2014 (UTC)

A dev team at Flickr took this comic as a challenge, and set up a PoC at http://parkorbird.flickr.com/ (that seems to work fairly well). --[[Special:Contributions/108.162.210.135|108.162.210.135]] 20:08, 20 October 2014 (UTC)
:I was duly impressed. It doesn't recognize big bird very well, though. ;) [[User:Suspender guy|Suspender guy]] ([[User talk:Suspender guy|talk]]) 20:26, 17 February 2016 (UTC)

A 'picture of a bird' from a CS perspective is a reverse engineering problem. The picture is a 2 dimensional rendering of a 3-dimensional world and a 'bird' is a 3-dimensional object. It takes years for the mind of a newborn human to be able to recognize a majority of objects based on their 'first look' at a stereoscopic (two-eyes) image presented by their visual cortex. The software equivalency of this would be to create a 3 dimensional representation of objects and create a linear-algebra algorithm that can define the statistical probability that any given shape is within a certain degree of exclusion a matrix representation of the target shape (area) of the 3 dimensional object (bird) based on distance (using spacial reconstruction). It's not impossible, it's just really really hard. - nerd answer {{unsigned ip|173.245.54.166}}
:To be honest I don't think it is impossible to replicate any function of human intelligence and mental capacity on a computer system. It just requires sufficient processing ability, appropriate hardware, and of course, an understanding of how humans do it in the first place. -Pennpenn [[Special:Contributions/108.162.250.162|108.162.250.162]] 03:29, 17 September 2015 (UTC)

Or just give Google a little less than two years, and they'll make [https://cloud.google.com/vision/ Google Cloud Vision API] for you [[User:Gpk|Gpk]] ([[User talk:Gpk|talk]]) 20:39, 13 June 2016 (UTC)

I read somewhere that when you ask CS/IT specialist for a probable ETA for solving an interesting problem, you need to multiply the given time to the ETA by 4 and take the next larger unit (a minute becomes 4 hours, an hour becomes 4 days etc.). Can't find the source of that though. [[Special:Contributions/141.101.70.229|141.101.70.229]] 15:47, 12 September 2016 (UTC)

== GIS being "easy" ==

All these years later, I still struggle with the classification of "are we in a national park" as "easy"..

It 'only' requires a functioning GPS-system. A military super-project, whose [https://nation.time.com/2012/05/21/how-much-does-gps-cost/ initial setup cost 12 billion], still costs ~2 million a day, and whose principles of operation depend on [https://skeptics.stackexchange.com/questions/20230/does-gps-use-general-relativity both special '''and''' general relativity] for correctness.
And that's ''before'' we add the record-keeping and (internet?)logistics involved with providing each phone an accurate GIS-database. The OpenStreetMap (most likely free/gratis source of this type of data, for a cheap app) is a massive undertaking!

(sarcasm on) GIS-lookup sure is easy! Only took a minor Manhattan-project, a literal Einstein, and an army of internet volunteers to solve!(sarcasm off)

(I'm leaving out mobile internet access while in said National Parks (Telecom operators are among the wealthiest companies in the world, and those phone-towers-disguised-like-trees don't come cheap...), because the App would probably be shipped with a hardcoded park-database, not do live queries.)

-- Jules @ [[Special:Contributions/162.158.91.77|162.158.91.77]] 08:13, 18 May 2020 (UTC)

:This is about implementation of something existing, not inventing it from scratch. The use of the word "app" implies, that this comic is happening in the smartphone area, so GPS on phones should be a regular thing. --[[User:Lupo|Lupo]] ([[User talk:Lupo|talk]]) 09:57, 18 May 2020 (UTC)

:: "app" sets the real-world context, but the punchline is about the comparative hardness '''in CS'''.
:: For the pragmatic app-developer, "previously solved" equals "easy"; for a doctorate in computational theory, it doesn't :-)
:: -- Jules @ [[Special:Contributions/162.158.91.77|162.158.91.77]] 12:16, 18 May 2020 (UTC)
:::That might be true, but this comic is definitely about developing an app, so it doesn't matter if GPS involves complicated hard- and software setups outside of the app or not. And unless you focus on the theoretical work also for a computer scientist, it is easy to use established GPS methods. --[[User:Lupo|Lupo]] ([[User talk:Lupo|talk]]) 12:45, 18 May 2020 (UTC)

now deep learning is common you not need research team and five years anymore

:And it took about five years since the comic was posted to get to that point...

This comic is referring to doing a GIS lookup which is a glorified sql Query which has nothing to do with GPS and the the USGS spatial data a GIS database is commonly populated with is not derived from GPS information anyway. A GIS Lookup IS easy. Gathering the spatial data is difficult, though as previously mentioned its already widely and freely available for use. --[[User:PlatterMonkous|PlatterMonkous]]
:The GIS data is being looked at to determine if GPS-derived metadata lies within one of its boundaries, surely? Without GPS, the query has no sensible question to ask.
:(Then again, none of my own pictures have that sort of EXIF information. Either they're taken on a 'dumb' digital camera, that doesn't have inbuilt GPS, or even ''if'' they're done via my GPS+Camera-equipped tablet (rare) I've likely not allowed the one to be fed data that the other one knows. If it's even turned on. But the comic scenario clearly assumes otherwise.) [[Special:Contributions/172.70.86.62|172.70.86.62]] 20:21, 10 November 2022 (UTC)

I think that, like in a number of other comic explanations, the explanation of the humour has given way to the technical exposition of the situation. What made me chuckle about this comic is the old adage in the software industry that: ''It takes 5% of the time to implement the first 95%, but 95% of time to complete the last 5%''. Even when experienced programmers correctly identify the difficult element of a problem and attempt to compensate for that difficulty in their implementation schedules, they can still be ''wildly'' off the mark. In this case, 60 years and counting... [[Special:Contributions/172.71.94.28|172.71.94.28]] 10:55, 9 August 2023 (UTC)

The year is 2024, and there is an ArsTechnica article about the problem in this comic now being solved. https://arstechnica.com/information-technology/2024/01/famous-xkcd-comic-comes-full-circle-with-ai-bird-identifying-binoculars/ [[Special:Contributions/172.70.210.41|172.70.210.41]] 22:24, 17 January 2024 (UTC)
:Updated to add credit where credit is due, the research team in our reality that created the technology is the Cornell Lab of Ornithology in the Center for Avian Population Studies and Macaulay Library: https://ebird.org/about/staff -- No doubt if Ponytail was the lead the staff could've done it in half the time! [[Special:Contributions/162.158.90.24|162.158.90.24]] 22:32, 17 January 2024 (UTC)

The rewrite of the explanation is really long overdue! If it's hard to start from scratch, someone could request GPT-4 to make a plan on how to update the article, and then use the ideas for inspiration {{unsigned ip|162.158.151.152|20:44, 13 March 2024}}
:If you think it could do with a rewrite, rewrite it yourself. (Stay away from GPT, though, still likely to give an unsatisfactory explanation. If not downright hallucinating.) [[Special:Contributions/172.70.86.179|172.70.86.179]] 21:01, 13 March 2024 (UTC)

2846: Daylight Saving Choice

2023-10-31T10:52:19Z

172.70.86.179: /* Explanation */

{{comic
| number = 2846
| date = October 25, 2023
| title = Daylight Saving Choice
| image = daylight_saving_choice_2x.png
| imagesize = 231x386px
| noexpand = true
| titletext = I average out the spring and fall changes and just set my clocks 39 minutes ahead year-round.
}}

==Explanation==
{{incomplete|Created by TWO CLOCKS SET ONE HOUR APART - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Daylight saving time}} (DST) is a practice best known for changing the clock one hour ahead for approximately half the year, typically from spring to autumn. Countries nearer the equator do not see significant changes in daylength between winter and summer and so have rarely had a reason to follow this practice. Many countries which used to follow this practice no longer do,{{Actual citation needed}} and a few now follow year-round DST - however summer-only DST is still used in North America, Europe, and parts of South America, Oceania, Africa and Asia.

Within countries that still follow this practice, there are frequent arguments (mostly during the 2-3 days surrounding the clock change) over the pros and cons of it. [[Black Hat]] is suggesting that everyone should observe or ignore daylight saving time based on their personal opinion. While it might put an end to the arguments (although this itself is debatable) it would clearly cause disharmonious time. This would eventually break the population into at least three categories: those who do not follow daylight saving changes and choose to remain on "daylight" time year round; those who do not follow and choose to stay with "non-daylight" time year round; and those who readily switch to daylight saving time during the prescribed period. There would probably also be a further 'group' who choose to change their clocks on an arbitrary date and time that suits them. So, some people might think it's 8:00 while others think it's 9:00, or vice-versa, but the relative number of people who believe it is each time would shift throughout the year. This would lead to many scheduling errors, delays, and other mistakes, resulting in widespread inconvenience and harm.{{Actual citation needed}}

The joke here is that, while most options in life can be left to individual choice,{{actual citation needed}} clock time is only fully useful if everyone involved agrees on what it means. There may also be a humourous reference to the confusion already often caused around this time when countries do not all begin or end DST on the same date, for example in scheduling calls or online meetings between Europe and North America in the week after publication of this comic.

There are known incidents in which an actual application of Black Hat's proposal
[https://www.sandiegouniontribune.com/opinion/the-conversation/sdut-daylight-saving-time-sunday-2015mar07-htmlstory.html rendered a terrorist plot void]. One of them is a [https://darwinawards.com/darwin/darwin1999-38.html 1999 Darwin Award Winner]

This comic was posted 4 days before the end of 2023’s daylight saving time in most European countries, and 11 days before the end of 2023's daylight saving time in most of North America. If the proposal is actually instituted at this time, those in the Northern Hemisphere who do not like the fuss of changing their clocks would ''remain'' on DST (as {{w|Sunshine Protection Act|has been actually proposed}}), yet those who are happy with it will fall back to non-DST over the winter months. Presumably, unless anyone changes their minds over the 'winter' period, everyone would actually be back in sync for future 'summer's.

However, the rule (as spoken by Black Hat – not known for being imprecise, or ''unintentionally'' misleading) does not restrict people to merely choosing whether the daylight offset is personally used during DST periods. It instead seems to ''impel'' them to undertake (or not) the statutary changes according to personal convictions, perhaps contrary to what their convictions actually desire. It is left open-ended ("From now on...") if people from ''both'' mindsets can arbitrarily change their minds in the future. If they can, and act accordingly, this time next year there could be people on three different 'summertime' offsets: zero (change now, but not change later), +1 (steadfast change/no change) and +2 (don't change now, but shift forward in spring). Beyond next year's "fall back" date, there could be people on -1 (fall back, don't spring on, fall back ''further'') and each full year beyond may add additionally positive/negative extremes of offset by those who periodically change their inclinations to only obey ''one'' of the relative imperatives, and a potential {{w|Galton board|standard distribution}} of everyone else between. All this could just be a badly worded explanation of the policy, or even in the wording of the legislation behind it, but the presence of Black Hat at the lectern probably indicates that he fully expects and ''intends'' such a boding and expanding chaos.

The title text suggests splitting the difference by using a constant offset which is the average of the daylight saving offset across days of the year. We do not know if in this system Randall would change his clock for leap year.

==Transcript==
:[Black Hat is speaking at a lectern, flanked by Ponytail and Hairy.]
:Black Hat: From now on, everyone who likes daylight saving time should change their clocks, and everyone who doesn't, shouldn't.

:[Caption below the panel:]
:The government finally decides to put an end to all the arguments.

{{comic discussion}}
[[Category:Daylight saving time]]
[[Category:Comics featuring Black Hat]]
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Hairy]]

Talk:2833: Lying

2023-09-27T05:53:11Z

172.70.86.179:

Why the heck is the image so biiiiig? {{unsigned ip|172.69.135.23|03:54, 26 September 2023}}
:well, looks like he accidentally published the source file for the comic... [[Special:Contributions/141.101.100.205|141.101.100.205]] 04:06, 26 September 2023 (UTC)
::Good. People should be using HD monitors by now. (EDIT: I didn't realize it was 8k, but still, don't most browsers let you resize images anyway?) [[Special:Contributions/172.70.126.188|172.70.126.188]] 08:41, 26 September 2023 (UTC)
:::Why do I have to lug around an HD monitor as well as my smart-tablet?
::::Why doesn't your smart-tablet have an HD screen? My phone is 2960 x 1440 and it still fits in my pocket. [[Special:Contributions/172.70.127.168|172.70.127.168]] 00:29, 27 September 2023 (UTC)
:::::Nobody said HD ''screen''. It is the suggested additional monitor that is impractical. Including a suitable power supply and of course the appropriate dongles/etc. [[Special:Contributions/172.71.242.63|172.71.242.63]] 01:53, 27 September 2023 (UTC)
:::(Actually, it's not width that's the problem, for me, but height. As using in portrait orientation makes text too small for browsing, and I hate sites that 'mobile optimise' assuming I'll turn my device that way.)
::::Most mobile browsers have a button for "show desktop version". I just hate when people assume vertical videos are okay. Even when I'm using my phone, landscape is better. [[Special:Contributions/172.70.100.65|172.70.100.65]] 00:30, 27 September 2023 (UTC)
:::::Do you know how well the "show desktop version" ''doesn't'' work. It frequently doesn't switch away from the mobile-URL (e.g. "en.m.wikipedia.org" to "en.wikipedia.org", or "/FOO-mobile.htm" to "/FOO.htm") and you still have to make that sort of change manually to the address-bar to fuly jailbreak out of the hard-coded "useful" mobile functionality (which may come back once it forgets your preference and gets redirected to use them all again), and I'm pretty sure that the places that do this detection invisibly (fingerprinting the browser used) to make the changes behind the scenes don't all fully honour the browser setting, perhaps just get over-riden by some more insistent indicator that the site-author has encoded later in the decesion-stack (e.g., at least for me, I ''cannot'' get the non-mobile layout of https://cyclingtimetrials.org.uk/find-events to display on this device, for whatever reason). [[Special:Contributions/141.101.98.174|141.101.98.174]] 01:11, 27 September 2023 (UTC)
:::Note that especially wide images play merry-hell with the rest of the page (when it breaks out of the pixel-limits assumed), and an image that's twice as large (in both dimensions) could be quadruple the data (depending upon image compression ratios), which has data/bandwidth/etc issues that not everyone can easily suck up and laugh off, even in this post dial-up era. It ought to be best not to assume that the best quality image is the 'best' or desired, although that ship has long sailed. [[Special:Contributions/172.71.242.83|172.71.242.83]] 09:08, 26 September 2023 (UTC)
::::Unless I'm misunderstanding, the image is now in the usual size range (590x887 in this case). -- [[User:Dtgriscom|Dtgriscom]] ([[User talk:Dtgriscom|talk]]) 20:32, 26 September 2023 (UTC)
:::::Something changed, yes. I haven't yet investigated if it was softcoded ("imagewidth=" in the template, or whatever it is, but perhaps still serving the full image to scale-down) or the uploaded image reuploaded in the 'friendlier' pixel-sizes. And for me it doesn't matter, but obviously the exact solution may be just as important for others, so I hope it isn't just the fudged one. (I may go and check, though I currently have no power to do anything about it.) [[Special:Contributions/141.101.98.174|141.101.98.174]] 01:11, 27 September 2023 (UTC)
:::::Addendum: Just checked, and it was ''softcoded too large'', originally, with "imagesize = 4422x6653px" for some unknown reason. I don't know why the uploadbot set it like this (incorrect metadata/sidedata at source?) but that definitely 'broke' the original page.
:::::Until just now I had only seen the slightly later version or the page where the image now merely strained the page-size (from a revised imagesize=, obviously to deal with the page-rescaling issue), then the latest where it's given the more usual one (but is still the exact same source _2x image behind the scenes).
:::::From all the talk about this, I had assumed that the image (like it sometimes has been) was the 'compositing' version not yet cleaned up by Randall to remove onion-skin 'planning' layers, etc. Instead, we were just given an upscaled correct version, but of course people were still able to dwell upon the details exposed by zooming in. [[Special:Contributions/172.71.178.219|172.71.178.219]] 01:29, 27 September 2023 (UTC)
::::::It was reuploaded to be normal sized later. http://web.archive.org/web/20230926034049/https://imgs.xkcd.com/comics/lying.png [[Special:Contributions/141.101.100.194|141.101.100.194]] 05:20, 27 September 2023 (UTC)
:::::::The initially uploaded explainxkcd version [https://www.explainxkcd.com/wiki/index.php?title=2833:_Lying&oldid=324385 as seen dynamically mis-scaled here] has not been changed, though. Which is different from my original assumption taken from the chatter about it.
:::::::If (perhaps, just looking at the URL above) it's that the _2x was normal _2x but the '1x' was mis-sized (we've seen that happen before) and/or the rough/not-yet-finalised working copy, then this might need better explaining. That'd explain why the _2x was given larger default dimensions. (I think the 'bot currently grabs the _2x but 'suggests' the non-2x size... I could then see how this would have happened.) [[Special:Contributions/172.70.86.179|172.70.86.179]] 05:53, 27 September 2023 (UTC)

The rhythm of the title text calls to mind Spock's words to Kirk as he's dying at the end of Wrath of Khan: "I have been, and always shall be, your friend." This can't be a coincidence. {{unsigned ip|172.70.210.182|08:06, 26 September 2023}}

I'm the same. Can't stand these games. I have a hard enough time with jokes that people refuse to explain; if I have to _intentionally_ mislead people, who know my tics to start, where's the line? What's real, what's fake, what's important, what's just another joke?

I need to lighten up? No, the world needs to be comprehensible. I can't just choose to know what's real and what isn't. Other people can very easily make it clear to me, if they so desire.
[[Special:Contributions/162.158.2.39|162.158.2.39]] 06:35, 26 September 2023 (UTC)
:i, uh, good for you. [[user talk:lettherebedarklight|youtu.be/miLcaqq2Zpk]] 07:04, 26 September 2023 (UTC)
:Simple answer: Don't play those games.
:Non-simple answer: Learn how to play those games with your advanced hyper-analytical abilities being used to your advantage (or as a "non-optional social convention") in which the reality is the game you're in and thus you are fulfilling the role of your existence.
:Meta-answer: I think you're a Joker (winning condition is to be lynched). Or maybe one of a number of other player-types which demands that you play differently from either mainstrean Mafia or vanilla Villager. Which, in a four-player game (very short of practical assignments!) makes it a bastard-setup of some sort. (Rather than single mafia/werewolf and all the rest vanilla village, or ''possibly'' one cop/special-role of some kind.) [[Special:Contributions/172.71.242.83|172.71.242.83]] 09:08, 26 September 2023 (UTC)

How do you even play Mafia with four people? Under standard rules, you have one moderator (Alice), one mafioso (Bob) and two civilians (Charlotte and Dave). Bob kills Dave in the first night, then there are only one mafioso and one civilian left, and the mafia wins, game over. Does anybody know a mod that would make it work with so few players? [[User:Comsmomf|Comsmomf]] ([[User talk:Comsmomf|talk]]) 11:16, 26 September 2023 (UTC)
:It's a Mafia-''style'' game, apparently. It could be moderatorless (or with a playing-controller who manages the gameplay outwith whatever role they have) and streamlined.
:How about four (playing?)cards, one designating the villain. Cards are dealt randomly an all close their eyes for the night phase, only the villain gets to open them and 'tap out' their choice of villager (some subtle way, like a feather on a stick ((not shown in comic!)) that everyone has, that can be used at will to silently reach anyone - reverse the stick to tap the centre of the table to signal completion?).
:Because that leaves little logic for the Town to follow, just blindly moving towards "lynch or lose" by luck, you can afford (maybe) a Cop role (from a different card) who then operates after that, at 'night'. They use ''their'' stick to ask a given player to thumbs up/down their status, with the and/or you even could do a Blocker that way (tell a victim to ignore being night-killed), etc. Or even mix things up with a role-giving role, whatever you need to balance play in the right way. 'Dead' night-role players could just 'tap completion' without having done anything, if there's no lynch-reveal (beyond town/scum, if even that before the sole scum announces this game is over and they won/lost).
:It'd have to be on the honour-system, and I could see mistakes and accidental reveals, but the post-mortem of a few such games might suggest refinements and precautions that haven't occured to me.
:...though it'd be easier (even 'player led') with maybe a minimum of six participants (could afford to have two villains, subtly gesturing ideas to each other, and a more complete set of "power townies" than just the one. [[Special:Contributions/172.71.134.192|172.71.134.192]] 12:51, 26 September 2023 (UTC)

I guess I just assumed every comic out there by anyone was printed with Comic Sans but when I was skimming through the source image I noticed the letters are unique and he hand-writes them. [[Special:Contributions/172.70.179.43|172.70.179.43]] 12:09, 26 September 2023 (UTC)

: Yes that's always been noticable from the kerning. His habit of tucking the left-hand vertical of an "H" under a preceding "T" - and shortening the left side of a "T" after an "I" (to pick just two examples)...is not something that any automated text rendering system that I'm aware of can produce. --[[Special:Contributions/172.71.167.135|172.71.167.135]] 12:25, 26 September 2023 (UTC)
::Also, his O's are quite noticably different from one another. Perfect repetition of loop sizes is a dead giveaway of a "handwritten" font. [[Special:Contributions/172.69.247.44|172.69.247.44]] 14:22, 26 September 2023 (UTC)