Wikipedia:Reference desk/Computing

Welcome to the computing section
of the Wikipedia reference desk.

skip to bottom

Select a section:

• Computing
• Entertainment
• Humanities
• Language
• Mathematics
• Science
• Miscellaneous
• Archives

Want a faster answer?

Main page: Help searching Wikipedia

   

How can I get my question answered?

• Select the section of the desk that best fits the general topic of your question (see the navigation column to the right).
• Post your question to only one section, providing a short header that gives the topic of your question.
• Type '~~~~' (that is, four tilde characters) at the end – this signs and dates your contribution so we know who wrote what and when.
• Don't post personal contact information – it will be removed. Any answers will be provided here.
• Please be as specific as possible, and include all relevant context – the usefulness of answers may depend on the context.
• Note:
  • We don't answer (and may remove) questions that require medical diagnosis or legal advice.
  • We don't answer requests for opinions, predictions or debate.
  • We don't do your homework for you, though we'll help you past the stuck point.
  • We don't conduct original research or provide a free source of ideas, but we'll help you find information you need.

Ready? Ask a new question!

How do I answer a question?

Main page: Wikipedia:Reference desk/Guidelines

• The best answers address the question directly, and back up facts with wikilinks and links to sources. Do not edit others' comments and do not give any medical or legal advice.

See also:

• Teahouse
• Help desk
• Village pump
• Help manual

June 16

AI

In the news nowadays, almost every new innovation in the use of computers, good or bad, seems to be dubbed "AI" -- things that even just a year ago would have been merely "people doing stuff with computers". Are all these so-called "AI" applications really qualitatively different from what went before, before everything was termed "AI", or is it just a case of everyone jumping on an "AI" bandwagon? 2A00:23C8:7B09:FA01:9425:E4FF:5FA8:86C8 (talk) 20:10, 16 June 2023 (UTC)

Inasmuch as these innovations are based on Large Language Models, Transformers, or any deep learning methods based on artificial neural networks, it is IMO fully justified to label them as applications of AI. A technology-independent criterion is, for a given cognitive task that humans are able to perform, whether we understand how humans can do that. If we understand it, we can describe it in the form of an algorithm. If we do not understand this but nevertheless make a machine perform the task reasonably well (or perhaps very well), it is "artificial intelligence".  --Lambiam 22:04, 16 June 2023 (UTC)

But most of the labelling of things as AI these days is done by people such as journalists with no special knowledge in the computing field, or by people with even fewer qualifications for writing stuff about stuff. Or even more worse, by people wanting to market a new product or feature. It IS an overused term. HiLo48 (talk) 02:48, 19 June 2023 (UTC)

I mean, of course there's an element of marketing, AI being the current hotness. But that said, Lambiam's definition is almost identical to mine ("it's AI if we don't understand how it works"), and computing meeting that definition has in a very short time come to dominate the landscape of almost everything you do other than on your personal CPU, and probably some of what you do do on your personal CPU. --Trovatore (talk) 03:14, 19 June 2023 (UTC)

That assumes it is possible to develop an AI that humans cannot understand. All current models of AI are understood. It may be very tedious to backtrack through something like a million-node neural network, but there is nothing that cannot be understood. A better definition, in my opinion, is simply a simulation of human intelligence. The field is very vast and encompasses everything from simple monotonic logic to machines that appear to think on their own (they don't, but it appears that they do). 97.82.165.112 (talk) 11:56, 19 June 2023 (UTC)

Oh, neural nets can be understood at the level of individual nodes, no debate there. What no one seems to understand is why they do what they do. We can follow each step, but we still don't understand why they work, not in the same way we understand, say, an algorithm in Introduction to Algorithms. Maybe it would have been clearer if I'd said "why" rather than "how". But in any case I disagree with you that these models are "understood". --Trovatore (talk) 17:52, 19 June 2023 (UTC)

Even very simple programs, having nothing to do with AI, can exhibit unexplainable behaviour. You can have a program that produces an infinite string of 0s and 1s, for example

01101011110101011110111010101101110101101011101101101010101110101111011110110101...

Now you wonder, why is a 0 always followed by a 1? Why are there no two 0s in a row? This question may prove unanswerable. You cannot even decide if occurrences of 00 are merely rare, or impossible. For any given 0 you can figure out the next symbol given a sufficiently long life and an inexhaustible supply of paper and pencils, but even though you then know why the 0 in position 211172672 was followed by a 1 in position 211172673, you still don't know why the pattern 00 is excluded in general, or even if it is actually excluded.  --Lambiam 08:28, 20 June 2023 (UTC)

The repeated statements that AI is not understood or explainable by humans is very wrong. The developers of AI products are human. They fully understand what they are doing. They fully understand the algorithms. They fully understand how the training sets affect the ouput. They fully understand how the PRNG algorithms affect the output. They can trace any form of output to the appropriate PRNG seed and training set source. There is nothing about AI that is beyond human comprehension. Now, we can separately discuss script kiddies. They don't understand the AI, but they don't understand anything at all. They are just slightly trained monkeys. But, I find it silly to use them as a basis for defining AI. The term itself is decades old and refers to a machine being used to mimic human intelligence. Nothing more. Nothing less. 12.116.29.106 (talk) 11:39, 20 June 2023 (UTC)

OK look, let me first admit something. I'm not a super-expert on machine learning. I work with some of them and understand it around the edges, but I am not myself such an expert. There are people who know a lot more about it than I do. It's possible you are one, and if so I invite you to educate me.

But so far you haven't engaged the issue. Your responses look like those of a glib reductionist who thinks that if we understand particle physics, then we understand hurricanes.

Test case: Suppose we train a neural net to classify the MNIST database of handwritten single digits. This can be done with very high precision and recall, and you get a classifier that can very accurately decipher most people's handwritten digits. The classifier, modulo the ordinary boilerplate code that adds up inputs to a node, applies the activation function, and passes it on, is just a bunch of weights.

Now, what is it specifically about those weights that recognizes a 4, and distinguishes it from an 8?

That's what I think is not well understood. Undoubtedly there are people who understand it a great deal better than I personally do, and maybe you are one of them. But so far you have not addressed this point. --Trovatore (talk) 17:10, 20 June 2023 (UTC)

That is almost exactly my area of work. So, I want to give a more complete answer, but I want to avoid a long rambling answer. I hope this will find a balance. The network is a series of weights. It is not difficult to go through the weights and identify which ones play a significant part in telling a 4 from an 8 and which don't provide much help at all. As an example from a different field, assume that you had the entire component diagram of an old tactical radar and controller system. They were published in multiple books of schematics, usually around 20 thick books. That is a lot. It is complicated. Nobody is expected to know exactly what the voltage at every point in the system is at any point in time. But, humans did understand it. If something broke, they could go through the circuitry, identify what wasn't working, and fix it. AI is no different. Humans designed the algorithms, the data structures, and the calculations. The end result may be too large to easily see everything all at once, but it isn't impossible to understand. If there is a problem - or even just a question about why something was produced - a human can go through the program and identify the problem and fix it. Some of my hardest problems are with probabilistic networks, which can be very large. I am asked why a single patient out of millions of patients in a population was identified as being diabetic when they aren't diabetic. I have to go through the AI program, step by step, and see how it weights every input and how those affect everything everything else until I map it to something that indicates diabetes. Then, I know what input is causing the issue and I can either manually fix the AI or I can use examples of that case to train the AI to stop making that mistake. I feel it is important to note that I don't trace through by drawing a big map on a whiteboard with numbers and formulas. I write computer programs that trace through and tell me what I want to know. So, I go from knowing how the AI works to writing a program to tell me exactly what the AI is doing at some point and how that affects other things the AI is doing. I also want to point out that this is very unrelated to the current GPT craze. That is just pattern matching. I give you the start of a sentence like "It looks like" and you use that to find a pattern of words that match it well. Autocomplete has been around for many years. GPT is just a different way of building the patterns to draw from, but the result is still just stringing along patterns that have a high probability of being correct. It is very good at exploiting rules that we probably know but don't pay close attention to. The way I like to explain it is that the GPT models are very good at stringing together words that follow the proper rules of the source language, but they have absolutely no clue what any of the words mean. 12.116.29.106 (talk) 17:37, 20 June 2023 (UTC)

That was very interesting. Thank you. --Trovatore (talk) 18:27, 20 June 2023 (UTC)

June 20

Locating a string from between identifiable characters

I've been using regular expression to match certain strings so long as they rest between two identifiable parts of a text. For example, the matched string in {{WPDENMARK|class=}} would be WPDENMARK, and it would be replaced with {{WikiProject Denmark|class=}} if and only if there are two curly brackets to the left of WPDENMARK and two curly brackets or a pipe to the right. I've been trying to write a regex that does this regardless of whether the curly brackets or the pipe are on completely different lines, or if there are characters between them (or even whitespaces), and this has been a problem. Is there a commonly used configuration for accomplishing this? Nythar (💬-🍀) 11:42, 20 June 2023 (UTC)

Being on separate lines is a limitation of the implementation of the regex that you are using. Either it handles newlines as a stop or it goes to the end of the text. I will ignore that as it is not pertinent and assume that the regex imlpementation you are using easily handles newline and return characters. So, you want {{. That is easy. Now, it might have whitespace. That is {{\s*. Depending on your implementation, it might be \s or \\s or [:SPACE:] or [[:SPACE:]] etc... Now, you want text. I am going to assume this text cannot contain a | or }. I would use {{\s*[^|}]+. But, it looks like you are really looking for WP at the beginning to replace with WikiProject. So, I would use {{\s*WP[^|}]+. Now, it can end with | or }}. So, you give it the two options: {{\s*WP[^|}]+(\||}}). You can see that I had to escape the | when it isn't inside [ ]. But, what if there is a space there... {{\s*WP[^|}]+\s*(\||}}). Now, it has been matched, but you need parenthesis to get the stuff that follows WP for your replacement. {{\s*WP([^|}]+)\s*(\||}}). The first match is what you want to keep. The second match is the ending, either | or }}. For my implementation, I use \1 for the first match and \2 for the second, so I replace it with {{WikiProject \1\2. Hopefully the editor doesn't really mess up the characters. I tried to place everything in nowiki tags.12.116.29.106 (talk) 13:01, 20 June 2023 (UTC)

Your suggestion is working well, but it matches the pipe and curly brackets. I was thinking of lookbehinds when I said I was writing "regex that does this regardless of whether the curly brackets or the pipe are on completely different lines." My idea would be a match of only WPDENMARK and nothing else using positive lookbehinds and lookaheads. I assumed this was possible. Do you happen to know of a way? Nythar (💬-🍀) 15:58, 20 June 2023 (UTC)

If you only want one or the other, you can use [|}] to match either a pipe or the first } of the pair of }s. that matches one and only one character. Place it in parenthesis if you want to know what it was for later use. If you give an example of input that doesn't work and what you expected, it is easier to identify exactly what you want. 12.116.29.106 (talk) 16:38, 20 June 2023 (UTC)

Talk:Marija Karan is an example. WikiProject Biography is on one line, while the other parameters are all on separate lines (so neither | or }} appear at the first line). In this case, I think a lookahead or lookbehind is needed, but for some reason it's quite difficult to write such a regex. I've been trying to take characters from both sides into account (using \n), but they're failing certain tests. Nythar (💬-🍀) 17:07, 20 June 2023 (UTC)

It appears that the issue is that regex is not parsing it as one text, but line by line. If it parsed it as one text, \s would capture the newline character as whitespace and then | would follow. That is an issue with the parser, not the regex expression. Is there an option in your parser to do two things: First, treat the text as one string, not an array of lines of strings. Second, match all occurences because there are two on that page. 12.116.29.106 (talk) 17:42, 20 June 2023 (UTC)

Dotall seems to be what I was missing. Regex is matching successfully now; hopefully it'll remain that way. Thank you. Nythar (💬-🍀) 11:36, 21 June 2023 (UTC)

June 22

What would happen if you dialed 911 on a 2G device in the United States in 2023?

I want to make a short video demonstrating how emergency call back mode worked on archaic CDMA devices in the United States, but without it actually going through to the PSAP and bothering call takers. To my knowledge, there are no 2G signals anymore where I live (Charlotte County, Florida). The device says “searching for signal”. Without considering the legal aspect (I know this forum doesn’t give legal advice; even if it did go through I’m fairly confident in my jurisdiction telling them it was an accidental dial would be sufficient to avoid legal trouble anyway because I’ve actually dialed by accident before), is this safe to do? 71.208.76.56 (talk) 01:46, 22 June 2023 (UTC)

It will depend on your carrier. T-Mobile and Cellcom have not yet shutdown 2G service. See [1] and 2G#Phase-out. Unclear what will happen if your phone has a different carrier. RudolfRed (talk) 04:00, 22 June 2023 (UTC)

Generally if your phone is still capable of connecting to a network whatever the carrier, it will for emergency calls. There should be some sign of this on the phone without needing to even try to dial out. Nil Einne (talk) 08:15, 22 June 2023 (UTC)

Does a computer read a 5 MB text document the same speed as a 5 MB image document?

If text document and image file are exactly the same amount of bits in 5 MB, it should be able to read and load the text document, and image file, at the same speed? But ChatGPT says no, and doesn't quite explain why. So if it isn't, which 1 is faster and why. 170.76.231.162 (talk) 17:45, 22 June 2023 (UTC).

What exactly do you mean by "read and load". If you are talking about a simple download from the internet to disk, then there will be no difference, they are both 5 MB of binary. If however you are talking about displaying the text or image, then it will depend upon the software you are using. For example, to load text into a text editor - fast; into a wordprocessor - far slower. Martin of Sheffield (talk) 20:06, 22 June 2023 (UTC)

I wrote a discussion of this but deleted it because I was getting out of my depth. Here's some thoughts, though:

• ChatGPT is full of shit, for a start, but it might be correct anyway by accident.
• There might be optimizations (in ... routers?) for the transfer of certain types of data over the internet. I vaguely remember something like this being proposed for text, back when the majority of traffic was text. Somebody observed that it would be possible to say, for instance, "right, this is an email" or maybe even "this is an email in English" and thus send it using a protocol (?) to compress it in a simple way based on things that are true of all emails, or all English. I don't think this plan was ever carried out. But there might be on-the-fly data compression for internet data? It seems reasonable: and then some kinds of data are more compressible that others, and text might be more compressible than images (which are probably already compressed).
• But there are things like bandwidth throttling and web caches to confuse matters. Maybe we should be thinking about reading from media like a drive, rather than from the internet.
• If the data is read in a stream, that might take several operations. Hard drives are going out of fashion, but accessing a hard disk always piles on the milliseconds and it's better to transfer all the data to memory in one go. Often though the program doesn't want to do that: instead it wants to interpret the data as it goes along and organize it into different memory locations, depending what it discovers. But probably the stream is taken care of by the system, which might transfer the whole file to memory anyway. If it doesn't, though, then reading a text file (such as configuration data for a game with a complicated world) could take longer than reading an image file of the same size.
• Which relates to the next point, the fine distinction between reading (just lumping the data straight into a block of memory) and loading (putting it in the right places in memory), not to mention decompressing and processing and displaying. If you include these things then there could be all sorts of reasons for the image to take longer than the text (maybe it's an svg file, those always seem to take a long time to render, although they can also be viewed as text in which case you can see them quickly!) ... or for text to take longer than an image (maybe you want to display it in an obscure font, and then your system has to do something complicated to load the font into memory).
• It also struck me that nobody ever looks at an entire 5 MB text document all at once, so the software might choose to read from disk only the part the user is currently looking at. I don't know if seeking around with a 5 MB file on disk and reading (say) 10k at a time would actually be any faster, though, than just reading the whole thing into 5 MB of memory.

 Card Zero  (talk) 23:36, 22 June 2023 (UTC)

Putting music on website help.

15 years ago, I devised this code.

<!--[if !IE]>--><embed src=" music file " height="0" width="0"></embed><!--<![endif]-->
<!--[if IE]><bgsound src=" music file " /><![endif]-->

Now, IE is out of the picture. And so, the above code only works on Google Chrome and not Firefox or Surf.

And, it only works on Google Chrome, if the page was clicked on, from a prior site. If you go to the page directly, the music will not work. (Which could be a good thing, what is this concept called?).

ChatGPT gave me this code, <audio type="audio/mpeg" autoplay="autoplay" src=" music file" controls>

Which is a workaround. But the autoplay="autoplay" part also only works for Google Chrome, not the others, but for the others, it does work if the user clicks on the play button, just that it auto-plays only on Chrome. 170.76.231.162 (talk) 18:23, 22 June 2023 (UTC).

Yes, Firefox blocks all media with sound from playing automatically, by default.  Card Zero  (talk) 20:05, 22 June 2023 (UTC)

Creating pictures and wallpapers online

First- All the digital wallpapers, how are they created?

Second- Online comics and graphic novels, are they directly created online or hand drawn first and then modified online?

Every artist have their own style. In hand drawing that is possible, but how is that done online? Ajay Sharma 2014. rxx (talk) 18:24, 22 June 2023 (UTC)

Artists use a variety of methods and techniques. Until not that long ago, almost all digitally available art was first created by hand on physical material (paper, canvas, ...) using traditional techniques and then scanned and possibly digitally retouched. Today, many digital artists use digital painting applications. These can be used for creating digital wallpaper and comics and graphic novels alike. Many other artists still prefer traditional techniques.  --Lambiam 21:51, 22 June 2023 (UTC)

Regarding webcomics, I found a (very old) description of a webcomic artist's workflow. (Everything Lambiam mentioned is involved, including both the traditional techniques and the digital painting.) For digital wallpapers, one (equally old) amusing technique is camera tossing. I found Clip Studio Paint (aka Manga Studio) as a comic-oriented graphics application which supports the use of a graphics tablet. (In fact it also runs on mobile operating systems, which means that you could plug a tablet into your tablet, if you really wanted to.)

I'm confused by your repeated use of the word "online". There are online applications such as Fotor, although these are generally inferior to software that you install on your computer (and they're even inferior to apps that you install on your device). There's Adobe Photoshop which is now delivered as Software as a service, although I suspect the only beneficiary of the service is Adobe: I don't see why the end users would want it this way.  Card Zero  (talk) 00:24, 23 June 2023 (UTC)

Looking for a verb describing the switching to a keyboard state which is valid only for the next keystroke

Imagine you are unable to press two keys simultaneously on your keyboard, but have access to a software that enables you to press the Shift key first, then pressing a letter key alone to get the uppercase letter.
This can be implemented in two ways:

1. You "switch" to the Shift state, i.e. you get uppercase letters until you invoke another function to "switch" back. (In fact, this is the way the CapsLock key works.)
2. You "???? (the verb looked for here)" to the Shift state, thus it is valid only for the neyt keystroke, i.e. the overnext keystoke and subsequent ones will yield lowercase letters without any need to "switch" back.

In fact, my question yields on a problem in ISO/IEC 9995-9, which is due to revision. There, special keyboard states (like "Greek" or "superscript") are addressed, and in the definition part of this standard is stated:

• 4.23 switch to <a group>: select a group or a mode which then stays in effect until another group or mode is selected, be it by switching to another group or selecting another reference group
• 4.12 latch to <a group> select a group in such a manner that only the next key actuation is affected, with the group reverting to the previously selected group (reference group) afterwards

In the ISO working group for this standard, we now have the request that we have to replace the verb "to latch" as the use defined in the standard is not compatible with the everyday meaning of this word.

For me, "to bounce" or "to leap" comes in mind, but at this time there is no native English speaker in the ISO working group, and we do not want to make an inappropriate decision again. Does someone have advice which verb is appropriate here? Thank you. -- Karl432 (talk) 21:16, 22 June 2023 (UTC)

Though I'm coming up with words for events which are triggered once, none of them clearly distinguish between events which are quickly over and then return to being as before, and events which are ongoing states. So for instance machinery can be primed for use (but the use is then ongoing, unless it's something like a firework display), a gun can be cocked and then triggered (but this isn't a verb you'd transfer to anything else), a trap can be set and then sprung (but "set" is very ambiguous, and anyway the sprung trap does not return to its previous state), and a trip switch can be latched (latching is mentioned in this section) and then tripped. Like a trap, a trip switch stays in its new state, so both of those metaphors are unsuitable. The most applicable out of all these is cock, so you might say "cock the group", but that's an uncommon verb and violates the principle of least surprise. I wonder what the best word (or phrase) is. I'm looking through Category:Switches for something that behaves in the same way, to see what verb it takes.

I think the preposition to in "latch to a group" is out of place, anyway. That sounds like "latch onto a group" and brings to mind an image of the user's fingers being tied to the keyboard with some sort of metal hook. If the verb is latch, it should be simply "latch a group", meaning that the group is prepared for triggering.

I've also found arm, which could apply to a torpedo, a land mine, or an alarm system.

Moving away from weapons and machinery, there's Accidental (music), which works the right way (only affects the next item ... well, the rest of the bar) but has no associated verb.

There's also deviate to and stray to, which suggest the change is temporary. Perhaps vary to, so that it sounds intentional.  Card Zero  (talk) 22:19, 22 June 2023 (UTC)

Does it need to be a one-word verb? Otherwise you might consider temporarily switch (or temp-switch). Or, if you wish, coin a new technical meaning of the verb twitch (⇪ Caps Lock is "switch to upper case"; ⇧ Shift is "twitch to upper case").

A remark about other terminology. The use of "group or mode" versus just "group" is slightly inconsistent. In 4.23 above, "group" in the phrase after the comma obviously is meant to include "mode", and the sense of "group" in 4.12 is almost certainly the same as "group or mode" in 4.23. Standards are often hard to interpret; a strictly consistent use of terms contributes a great deal to making them intelligible. If the effect of key actuations depends on the selected (reference) group, then the currently selected group defines (an aspect of) the current mode.  --Lambiam 22:28, 22 June 2023 (UTC)

Or "temp-select", to make it transitive. "Temp-select group 2" is fine, really. Ugly but functional.  Card Zero  (talk) 22:51, 22 June 2023 (UTC)

In Unicode something like that would probably be called a modifier. NadVolum (talk) 22:34, 22 June 2023 (UTC)