Wikipedia:Reference desk/Archives/Computing/2012 July 11

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 11

Reporting underage nudity on a webforum

The Reference Desks cannot give legal advice
The following discussion has been closed. Please do not modify it.
So I am a former member of a webforum where members can (and a few do) post pictures of their own genitals. However, it should be noted that the website is not pornographic in nature. Shortly before I quit the membership, I noticed that a certain user had posted pictures of his genitals. I vaguely recalled that user giving his age as 16 in an earlier posting on the same website. When I notified the admin (and owner) of the website, he shrugged it off as being irrelevant. I then took it upon myself to sift through the poster's history and found the post where he did indeed give his age as 16. Again, I notified the owner of the forum, but he only edited the other user's post, removing the indication of his age. He left the uploaded pictures live on the forum. So my questions are: Am I obliged to report this to the authorities? And where would I report the incident? The website is located in the United States. --87.79.225.108 (talk) 00:07, 11 July 2012 (UTC) What you are talking about is a variation on a good Samaritan law, requiring you to report what you believe to be a crime. Such laws are rarely enforced, even where they exist. Check with a lawyer in your location to find out if your state has, and enforces, such a law, as Wikipedia doesn't offer legal advice. StuRat (talk) 00:55, 11 July 2012 (UTC) Maybe he is now 17. Not your business. Kittybrewster ☎ 01:04, 11 July 2012 (UTC) What difference does 16 or 17 make? Underage is underage, no? --87.79.225.108 (talk) 01:42, 11 July 2012 (UTC) IIRC underage nudity is not illegal (at least not in most of the USA). You're presumably thinking of child pornography... even though you've already stated it's not pornographic, and that he presumably took the photograph himself, which likely rules out the exploitative nature of child pornography (why it tends to be illegal). ¦ Reisio (talk) 01:33, 11 July 2012 (UTC) The pics were self-taken and self-uploaded and are not "titillating" in any way. Nevertheless, knowingly hosting nude picures of underaged persons is not a problem in the United States? That... surprises me, to say the very least. Nevermind then, I guess. Good point btw. --87.79.225.108 (talk) 01:42, 11 July 2012 (UTC) I wasn't asking for legal advice. Also, it is very aggressive and uncalled for to retroactively label an already concluded discussion as illegitimate. Then again, I have come to expect no better from the editor who did so. --87.78.4.112 (talk) 11:24, 11 July 2012 (UTC) Sorry, but your question was clearly asking about legal advise. And since people here don't know 1. if the web-site was indeed in the US, 2. what kind of nudity it was, and 3. the relevance of you being in Germany, where people are much more tolerant towards this kind of things, so no one here can answer it. OsmanRF34 (talk) 12:39, 11 July 2012 (UTC) Except for the three of us that answered him. :p ¦ Reisio (talk) 18:14, 11 July 2012 (UTC) You three didn't provide any valid information, btw. I hope any OP does not rely on random information from random strangers when it comes down to solve legal or medical problems. And the question is not even relevant to this RD~on the top of that. OsmanRF34 (talk) 20:14, 11 July 2012 (UTC) …and yet he seemed so satisfied. ¦ Reisio (talk) 21:46, 11 July 2012 (UTC) The RD doesn't provide medical or legal advise, and that's independent of the fact whether people would be happy getting some advise here or not. It's simply a bad idea due to several reasons. OsmanRF34 (talk) 22:24, 11 July 2012 (UTC) I am confused. Does how to report a possible crime count as legal advice? I think there is a difference in some countries between under-age nudity and under-age porn or the Coppertone and diaper commercial people may be in big trouble. I would say that a non-911 call to the local police may answer it or just google 'report child porn' for a 1-800 number or site in your area.--Canoe1967 (talk) 21:03, 11 July 2012 (UTC) In other words, the best legal advice we can give is "go get some legal advice." BigNate37(T) 22:24, 11 July 2012 (UTC) Deciding whether it was a crime is a legal question. The OP asked: "Am I obliged to report this to the authorities?" Answering this is legal advise. "And where would I report the incident? The website is located in the United States." That's also legal advise and it's not as simple as it looks like. Just imagine that he has an obligation to report it, but reports it at the wrong country or wrong place. OsmanRF34 (talk) 22:24, 11 July 2012 (UTC)

How do I upgrade to IPv6?

I've tried ipconfig /release and /renew on the cmd prompt, but my old IP keeps boomeranging back. I wonder if it also involves changing my modem/router hardware in order to be able to get IPv6.

Also, there's a gentleman named "Cholby" on a popular chatting network who claims to be able to borrow fresh IPs from anywhere around the world. What would I need to download and use in order to borrow said IPs just as well?

Thanks. --70.179.170.114 (talk) 02:44, 11 July 2012 (UTC)

Do you have an IPv6 address? If not, you will need to get one either from an ISP or a tunnel broker. Looie496 (talk) 03:12, 11 July 2012 (UTC)

With all due respect, if you don't even really know what 'upgrading' to IPv6 means, trust me, you don't have a reason to "upgrade". In fact, I can't think of a single valid reason for any home user to be on IPv6 yet. For 1, your computer probably gets an IP address from your router, so there's zero reason to use IPv6 there, (unless you have several billion devices in your house using IP Addresses). Secondly, your router is assigned an IP address from your ISP, whatever IP address they assign to your router is really their business not yours, as long as it works. To actually get it working, you would probably at least need to spend a few hundred dollars on a router that supports it, and it would be literally throwing money away. Vespine (talk) 06:05, 11 July 2012 (UTC)

To get IPv6 via a tunnel broker or even just 6to4 or Teredo you probably have to spend zero money although you do have to know what you're doing. Also there are several advantages to having IPv6 which I mentioned before. Note even having more then one device does mean each device can get a fully routable IP address which they can't get if their ISP only provides a single IPv4 address (of course with IPv4 exhaustion, some people aren't going to get even a single fully routable IPv4 as already sometimes happens in parts of Asia) requiring you to use NAT or similar. Of course whether you really want that, particularly since your fully routable address is IPv6 may be questionable, but the point is a billion devices doesn't come in to it. (Yes you can probably get multiple IPv4 address if you're willing to pay, but that's beside the point.) Note that I'm explicitly not saying, I think the OP will actually notice ay advantage to getting IPv6, as I've remarked before since their primary reason appears to be their mistaken belief we and others aren't going to notice who they are, there's no point. And if they are literally being banned, they should not expect IPv6 will do much then annoy whoever is banning them, possibly to the extent they complain to the tunnel broker (if they're using one). Nil Einne (talk) 10:38, 11 July 2012 (UTC)

To give an example, I spent zero dollars getting this IP, solely using my existing setup. (I'm not counting time, data, energy or anything like that of course.) 2001:4428:220:0:511C:9FEA:CB5:4477 (talk) 10:46, 11 July 2012 (UTC)

Actually most recent hardware already supports ipv6 and has done for quite a while, but your ISP probably doesn't. 2002:5CE9:401A:0:0:0:5CE9:401A (talk) 09:52, 11 July 2012 (UTC)

Questionable although it depends on what you mean by recent and 'a while'. As I noted a few weeks ago with this source [1], a lot of CPE particularly ADSLx ones still don't support even native IPv6. See also [2]. Even some routers don't support IPv6. (I mean heck, even pfSense is only now adding IPv6 and that's based on BSD and uses commodity hardware.) However I do agree the first problem is likely to be their ISP. Nil Einne (talk) 10:38, 11 July 2012 (UTC)

OP notes, <quote>"...there's a gentleman named "Cholby" on a popular chatting network who claims to be able to borrow fresh IPs from anywhere around the world..."</quote>

Which chat network? URL, please. The above makes me think Cholby and company are up to no good, as in reanimating zombie IPv4 space to provide hosting for fraudulent schemes. If so, not someone who merits help. --Elvey (talk) 17:04, 12 July 2012 (UTC)

That would be Freenode. Cholby claims to be a helicopter pilot from Boston. --70.179.170.114 (talk) 23:27, 12 July 2012 (UTC)

You can ask your isp to enable ipv6 access for you. usually they will assign a whole subnet to you. as they did here. then you have lots more than just one ip. but for most use it will be useless because ipv4 and ipv6 are totally seperate networks. like two seperate versions of the internet. but to get started with it, do the ipv6 certification and exam thing at http://tunnelbroker.net/ . it is free and a lot of fun. and you will learn a lot about ipv6. they also assign a subnet to you there so you can use your pc's at home to set up mailserver, dns and whatever.. --helohe (talk) 23:08, 15 July 2012 (UTC)

I need to find a great Desktop sidebar with which to replace Google Desktop.

Google made the mistake of discontinuing support for Google Desktop Sidebar. It was GREAT the way it was.

Since I can't find the "Currencies" gadget anymore on it, I need to find a viable replacement that gets updated regularly. It must work on Windows 7, and be as good as Google Desktop Sidebar used to be, if not better. Please offer suggestions. Thank you. --70.179.170.114 (talk) 08:50, 11 July 2012 (UTC)

Is this FSK signal Manchester encoded?

Spectrogram : http://i.imgur.com/I6J7t.png
Cleaner version of the top signal in the previous picture made by tone detection with the Goertzel algorithm : http://i.imgur.com/N1Ah7.png

I've been trying to figure out how to decode this for a while now. I suspect that it is Manchester encoded, but my attempts at trying to decode it have failed (although I must admit that I barely know what I'm doing.)

Is it Manchester encoded or is it something else? CoilSpring (talk) 09:01, 11 July 2012 (UTC)

It doesn't look like Manchester code to me; Manchester code spends at most one clock period at one voltage level. AFAIK Manchester coding generally isn't used in short bursts either, because a phase-locked loop is used to recover the clock, and that would need a continuous signal.

What this looks like to me is a simple unipolar code with 4-character frames, with the first two characters being the header. The header essentially tells the receiver "listen up, I'm about to send you some data". Because there is generally a lot of noise in wireless digital communication, the receiver needs a clear signal to prevent it from reading in too much random gibberish, and listening for a known header is one way to do that. Once it picks up such a header, it knows to expect a number of valid data characters in some known time-frame. Looking at this signal, I think it might be an inverted UART signal; the first bit of each character would be a start bit, then there'd be 8 data bits, and a stop bit of opposite polarity to the start bit and equal polarity to the idle level.--Link ^(t•c•m) 15:38, 11 July 2012 (UTC)

I still haven't been able to decode it to anything that makes sense, but I think you put me on the right track. Thanks! CoilSpring (talk) 05:04, 13 July 2012 (UTC)

Microoperation mix for matrix multiplication

Are any figures available on the mix of integer vs. floating-point microoperations in floating-point matrix multiplication (preferably LINPACK) on a reasonably modern CISC CPU (preferably x86-64)? How many integer uops arise from the addressing modes? NeonMerlin 10:14, 11 July 2012 (UTC)

Programming languages little differences

Why are some programming languages more limiting than others? For example, if you declare a list, why do some programming languages make you declare the type of the elements or only admit elements of the same kind? Or why declare the type of an element at all? If the compiler meet "a = 13" couldn't it simply be configured to know that 13 is an integer? OsmanRF34 (talk) 11:15, 11 July 2012 (UTC)

The main reasons for limitations are safety and efficiency. Loosely typed languages tend to have more programming errors and they are often hard to find. For example, the operation element[m] + element[n] may only work as intended if the given elements are numbers at the time. It can be hard or impossible to determine whether this will always be the case. If they have other types then you may get a run-time error, or a random looking result adding the binary representations of the elements. Strictly typed languages may have more compile time errors but these can be seen and fixed by the programmer before the program is released. Efficiency is also a major reason. Compilers can usually generate more efficient code if the types are known at compile time and don't have to be tested at run-time. If the compiler knows the type then it knows how much space it uses and which machine code will work on it. See more at Type system and Type safety. PrimeHunter (talk) 12:33, 11 July 2012 (UTC)

(ec)there are many reasons.

• On reason is because it may be the aesthetic choice of the implementer.

• Another important reason is to prevent weird bugs. When the programmer is forced to explicitly state what he is doing,

• • (a) implicit assumptions/guesswork of the computer will not conflict with the programmers expectations of the behavior. For example, in python < 3.0 dividing two integers will give the nearest integer as a result. This will cause a lot of unexpected errors as it is easy to forget when writing a big program.

a=1

b=3

fraction=a/b

print fraction

• • (b) implicit assumptions/guesswork by other programmers reading the code will also be reduced. If you are adding some code to a sofware and want to use a function/data structure written by someone else Its essential that you are sure that the code does what you think it does.

• Finally, high level languages are used when the convenience and clarity is important. (e.g. a big webserver where the code needs to be concise) In situations where you need low level languages, you also need to have a good idea of what data types and memory management schemes to use and what the exact instructions are in the final binary. For example, in the case of the linux kernel because everything needs to be under fine control, C is used with its manual memory allocation instead of using automatic C++ classes. In many cases, the absence of ambiguity and guessing can make a big performance impact.

Staticd (talk) 12:35, 11 July 2012 (UTC)

Articles exist for Strong and weak typing, Strong typing, and Weak typing. In a perfect encyclopedia, that would answer all your questions. However those three articles have been in a sorry state for some time, littered with merge recommendations and maintenance templates. Perhaps this discussion will bring them some much-needed visibility, and failing that they may still be somewhat informative for OP's question. BigNate37_(T) 22:33, 11 July 2012 (UTC)

Putting the above another way, the computer needs to know what kind of element it is so it knows how to store it in the memory correctly, what can be done with it, how to make it play well with other elements. Somebody needs to tell it what to do. Either the compiler can guess — which weakly typed languages do — or you can tell it yourself. The downside to the compiler guessing is what has been described above: it can guess wrong, or it can do it in a really inefficient way. The up-side is of course there's a lot less work for the programmer to do. Personally I much prefer not having to dimension my arrays completely or commit each element to a specific type; for the little types of programs I do, where efficiency isn't important and the errors are minor, this usually works out fine. But the more complexity your program has, and the more important the speed/memory requirements are, the less high-level you're going to want to be. Hence little scripts for webpages can go with a super weakly-typed language like PHP, but your 3-D video games are written in variants of C. You can be a purist and say that only the more low-level languages are suitable, but for lots of applications, it doesn't matter too much, and simplicity wins the day for most people, whether the purists like it or not. --Mr.98 (talk) 15:08, 12 July 2012 (UTC)

The "strong"/"weak" distinction doesn't mean a whole lot. Static typing and dynamic typing are decent article subsections, and pretty well-understood terms, at least. Another relevant article is type inference; it is, in fact, possible for a statically-typed language to determine types without input from the programmer, which might be what the the OP was asking about. However, type inference has two major disadvantages: type errors can be harder to understand, because the programmer has been "keeping their intentions secret" from the compiler, and type inference tends to stop working as you add features to the type system. (Note that I'm biased: I primarily work with a dynamically-typed language and a statically-typed language with very limited type inference.) Paul (Stansifer) 18:13, 12 July 2012 (UTC)

PrimeHunter gave a very good answers (others probably did too, but I haven't read them. I'd like to add this: In my opinion, with a strongly typed language, the compiler works for you. Otherwise, you work for the computer - doing work that the compiler could do for you. Bubba73 ^{You talkin' to me?} 04:27, 13 July 2012 (UTC)

Thanks for the answers so far. I gave the example of type, but I'm sure that there are other things that make a language like PHP easier than C. Can anyone cite more differences of things that make your life simple, but the the programs less efficient and more difficult to debut? OsmanRF34 (talk) 22:52, 13 July 2012 (UTC)

Does a smart phone need NFC enabled for reading QR code?

Or any smartphone can read QR code as long as I install the right apps? — Preceding unsigned comment added by 123.16.249.5 (talk) 13:13, 11 July 2012 (UTC)

It's a whole separate thing. QR reading requires a camera, which almost all smartphones have. NFC requires NFC hardware in the phone, which few have. Jim.henderson (talk) 13:22, 11 July 2012 (UTC)