Wikipedia:Reference desk/Archives/Computing/2010 December 28

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.

December 28

How to cause coaxial cable interference?

Illegal, potentially lethal, and not possible anyway.

I googled this for an hour and got no answers. Let's say there's two rooms right next to each other. Room A has a coxial cable plugged into a TV for cable TV Room B has a coxial cable that's not plugged into anything. The cables have go along so the central core wire inside the coaxial cables between each room has some connection between them (as well as to the cable company and elsewhere). Okay so what can be done to the central core wire in Room B that would cause interference in Room A? This is purely hypothetical for school. And would only alternating current cause interference or would direct current cause it too? Are you ready for IPv6? (talk) 04:48, 28 December 2010 (UTC) Coaxial cable has a grounded sheath that is supposed to suppress any interference, but any electrical activity, especially electric motors where there is some sparking, or faulty fluorescent fittings, or switching on or off of AC or DC currents (especially with an inefficient switch that does not make a clean break) is likely to cause an electromagnetic field that will produce some induced current in the cable to the TV. It is possible that the interference coupling between the cables is capacitative if the shielding is not earthed in the unused cable. Interference like this is very difficult to predict. Partial prevention can be achieved by efficient earthing, but a sparking switch running an inductive load can be a serious interference problem over a considerable distance. Are you aiming to cause, or to prevent the interference? Dbfirs 09:17, 28 December 2010 (UTC) I'm trying to cause interference. I tried putting the inner core wire into AC voltage and it just blew a fuse. Does the AC/DC voltage need to switch on/off all the time or can I just leave it switched on and that would interfere? I'd think with DC, the voltage would just stay at the tip and that's it. But if I don't make a circuit, maybe I might have a better chance, like if I put a positive or negative end to the inner core wire and see if that does something (which would more likely cause interference, the positive or negative end?) Are you ready for IPv6? (talk) 20:14, 28 December 2010 (UTC) Before we give advice on sabotage, could you reassure us about the purpose of the interference, and also about the exact configuration of the cables. It would obviously not be wise to connect any voltage at all to a wire connected to either a TV or a cable company's equipment. If the wires run inside the same sheath, then the same applies. Dbfirs 20:55, 28 December 2010 (UTC) I'm doing this for school. Are you ready for IPv6? (talk) 21:11, 28 December 2010 (UTC) Odds are, there is no direct electrical connection between the conductors on the 2 cables, so anything you put on the B conductor will not necessarily get to A. If this question is a hypothetical for a class, then I've just given you enough to determine your answer. If this is for anything else, be advised that intentionally causing interference to communications systems (of which CATV is one) is very often illegal. ArakunemTalk 21:33, 28 December 2010 (UTC) The fact that you managed to blow a mains fuse suggests that you are not sufficiently aware of the exact configuration to safely perform experiments, so please stop. Your idea of a DC voltage "staying on the tip" ignores capacitance. I was about to suggest a safer and possibly more effective setup, but in view of legal and safety issues, I suggest that this discussion should be closed. Dbfirs 22:32, 28 December 2010 (UTC) Agreed, and I will be doing so momentarily. IPV6, the fact that you said you blew a fuse when you hooked up the mains suggests that this is not "purely hypothetical for school" as you claimed. You cannot treat mains AC the same as CATV AC - Mains AC contains life-threatening voltage, and if that fuse had not blown, either you, or the guy you seem to be trying to mess with, very well may have learned that first hand. Furthermore, you don't seem to understand how CATV signals are distributed, coupled, and split. What you're trying to do is simply not possible without very expensive equipment specially designed for the purpose, and the way you're trying to do it is dangerous to life and property. ArakunemTalk 00:36, 29 December 2010 (UTC)

Considering there's consensus on the reference desks to freely give out suicide information, I don't really see how closing this as "potentially lethal" is justified. Pretty much everything is "potentially lethal" in some way - doesn't mean we shouldn't talk about it. But... the OP went from "hypothetical for school" to actually doing it, so I guess there is a problem there. Anyway, probably the most effective way to cause interference of a coaxial cable is to simply cut it. 82.44.55.25 (talk) 10:43, 29 December 2010 (UTC)

ole

is there a list of the nCmdID supported by IE to pass to IOleCommandTarget::Exec Method? t.i.a.--83.103.117.254 (talk) 10:55, 28 December 2010 (UTC)

Best lossless compression for lots of similar photos?

1) Having recently used VLC to successfully get a high-quality frame photo from an avi video file, would it be possible to do the reverse and losslessly compress many similar photos into some sort of video file? The benefits of compressing into a video file would be that as well as compression within each image, you could also benefit from compression between frames as well. Each frame could be viewed with VLC using the "e" hotkey to advance by one frame. (Do not know if there is any hotkey to go back by one frame). I am rather a fanatic about seeing detail and not losing information - I'm only interested in lossless compression.

2) Considering individual photos, pngcrush can apparantly reduce the size of a png file considerably. Would you get much or any compression by converting jpg or jpeg files to png and using pngcrush?

3) Is there any photo viewer that can be used to view images within a compressed file, such as zip, without having to uncompress them all beforehand? Thanks 92.24.187.63 (talk) 12:31, 28 December 2010 (UTC)

To answer your questions sequentially:

1 - read about inter-frame coding - it's possible, but because it involves lossy-compression, if your frames are not very similar, you will either distort them significantly or fail to compress them (depending on the particular codec and settings you choose).

2 - That depends entirely on the current quality and size of the JPEG files, as well as their image-content. The PNG standard officially uses lossless compression; and pngcrush uses lossless compression as well; but you could conceivably perform a lossy compression step first (such as performing a lossy conversion to JPEG, then use PNG to compress the resulting (different-from-original) image. It is unlikely that this will reduce the file size beyond what the JPEG lossy method already achieved; one interesting result of JPEG and other compression systems is that they typically produce bitstreams which are statistically distributed in such a way that they compress very poorly. (Lossless compression requires certain statistical patterns in the input data in order to be effective).

3 - That depends entirely on the archive file. ZIP files can support indexing; and ZIP files are not solid archives, meaning that it is possible to easily uncompress a single file from the middle (without needing to seek or decompress through the entire archive). You can use a smart zip file viewer, such as WinZip or WinRAR, or some of the free software tools like Ark (actually, have a look at Comparison of file archivers). Most can seamlessly extract a single file; my archive manager (File Roller) integrates with my Gnome desktop manager to present archive files as a regular part of my file-system, albeit with a bit slower performance. (In fact, slower or faster, depending on the particular file. For some files, disk access is so much faster, the decompression time is negligible, and the entire operation is faster than accessing an identical uncompressed file). Nimur (talk) 15:03, 28 December 2010 (UTC)

1) Lossless inter-frame coding is possible in principle, but I think you'd get very little benefit from it, if any, on natural photos. On artificially generated images you might see an improvement.

2) Converting JPEG files to PNG or any other lossless format will almost certainly make them much larger. There are a few exceptions, like images that are a single solid color. Also, converting JPEG images to PNG won't make them any more lossless. The loss happens during compression and/or decompression, and extra conversion steps can only make things worse.

3) There are some image viewers with special support for viewing images inside compressed archives. Those files sometimes use the extension CBZ or CBR. See Comic Book Archive file and Comparison of image viewers. But typical archive compression algorithms won't reduce the size of typical JPEG/PNG files by very much. Recent versions of WinZip have a special JPEG compression mode that can reduce file sizes by 20% or so, but I think this is not supported by any of the image viewers, or any free software (?). -- BenRG (talk) 06:44, 29 December 2010 (UTC)

Facebook

Hello and Merry Christmas,

I have a question. I recently sent a message on Facebook to somebody who in the past was not friendly to me. I stayed polite and said what I thought of that person. The person said to me that I was going to receive news. The person wants to complain against me. He or She does not have my name and doesn't know me because I have used a pseudonym. Can this person know who I really am? I did not give my name because this person targeted me in the pass and could target me more by knowing me. It wouldn't change anything. I do sometimes ask myself if there is still a freedom of speach. Thank you. —Preceding unsigned comment added by 81.241.130.118 (talk) 12:35, 28 December 2010 (UTC)

If you are in danger from that person, you should seek help elsewhere (such as your local police department); this is the computer reference desk and it is unlikely we can give you useful advice for dealing with your social situation. Regarding the technical side, Facebook exposes a lot of personal data in a lot of unintuitive ways. Depending on your privacy settings and whether you have 3rd-party Facebook "applications" installed, you may have unwittingly exposed any of the following: your location (where you were when you posted the message); your social network graph (the list of friends/contacts); a recent history of your activities on Facebook (and in some cases, other websites you have visited); your computer type (operating system, web browser, and so on); and possibly even your IP address (though Facebook makes a fleeting effort to mask this, 3rd-party applications can trivially determine that). Here is the Facebook Privacy Policy - understand that this is the legal document that expresses Facebook's official policy. However, it doesn't account for serious potential privacy-leaks that are illegal according to its terms - such as any 3rd-party application that violates the terms of service. (In my estimation, those are your biggest source of worry from a technical standpoint). So, if you have created a fresh Facebook account with a completely false name, with no friends, no applications, and connected from a public terminal (like a library or internet-cafe), you are essentially anonymous, in the sense that your constructed identity can not be decisively matched to you as a person. But, it sounds like your conversations with your acquaintance may betray your identity - as is typical in computer network security, the weakest link is the human-actions you took. Did your conversation expose identifiable details (or even raise hints) about your identity? You know what you said; you have to evaluate whether you exposed yourself through conversation. Does your friend have access to your real life identity? Only as much as you make available, Mr. 81.241. Nimur (talk) 13:36, 28 December 2010 (UTC)

Also, you might want to peruse Freedom of speech in the United States for an understanding as to why the First Amendment to the United States Constitution has nothing to do with you being able to tell someone off with complete impunity. Asking "if there is still a freedom of speach" in this context is ridiculous. Matt Deres (talk) 20:09, 28 December 2010 (UTC)

SE search vs flight search

Why does the latter take much longer? Both are DB searches, and airline companies and travel portales should have the means to build a descent DB, with fast access, but any time I try to perform a search for a flight, it takes much longer than the typical 1.x seconds of Google.Quest09 (talk) 14:15, 28 December 2010 (UTC)

Google's performance capabilities are due to incredibly expensive investments in technology. Where-ever you live (in the United States and Europe, at least), you are not far from a complete clone of Google's database - they have purchased redundant infrastructure, long-haul network connections, and millions of servers. Expert analyses currently estimate that a single Google query by one user requires as many as 10,000 Google servers. (source: Energy limits force complete rethinking of processors, programming, Kunle Olukotun) (a source which is ironically only available in paper form even months after its release). Airlines are not willing to spend that amount of money to speed your queries up by a few milliseconds. I am still baffled why Google is willing to spend that amount of money on network gear; the usual flimsy response is "advertising revenue." Nimur (talk) 14:46, 28 December 2010 (UTC)

It probably comes down to indexing. If you want to pick a particular DB record out a million records and it's indexed on a unique field, it can likely be returned in a tiny fraction of a second using that field for the search. On the other hand, if you do a search on a field or combination of fields for which no index exists, that could involve reading every record and checking to see if it matches your search criteria. This could take a million times longer (or more, if multiple DB tables are involved). StuRat (talk) 16:48, 28 December 2010 (UTC)

And why wouldn't you have the flights indexed, if you know that people will use your database for querying flights? I suppose the OP mean sites like Opodo where you search for the cheapest flight between day 1 and day 2 and place A and place B.Mr.K. (talk) 01:58, 29 December 2010 (UTC)

That's right. I'd bet they are indexed under flight number, but searches using a range of non-indexed parameters (like price, departure airport, arrival airport, and date) have to use a brute-force approach, instead. StuRat (talk) 07:07, 29 December 2010 (UTC)

Domain name and server space advice please

Happy holidays Wikipedians.

I'm a bit of a web development noob, and am about to ask some questions that may appear absurdly simple, but please bear with me.

I have spent the last few weeks developing a news/e-zine type site offline using Macromedia Dreamweaver. I'm pretty happy with it now and I would like to make it live, but I have a handful of questions:

• Should I buy the domain name or rent the server space first?
• When I buy the domain name, what happens (How do I prove I own it when renting server space)?
• When I rent the server space, what happens (Do I get some sort of login details, do I just upload the folders and files that I have developed off-line...)?
• When I have bought the domain name, rented the server space, and made the site live - How do I edit it (Does this vary from server space providers, or is there a set of standard practices. Do I sync Dreamweaver with my on-server site - If so, how)?
• Finally - Any good tips for reasonably priced, but reliable server space providers?

I've had a quick look through the Comp. Ref. Desk archives, but didn't really find any suitable threads, but if you know of one (I'm sure some exist), a simple link to that will suffice. Also, if anybody is aware of any decent online sources that provide a noob introduction to the areas I have mentioned, please provide a link.

Now, as many of you guffaw at the noob-ness of my questions (Glad I made you laugh over the holidays), I hope you find the time to drop in a few useful links or offer some tips.

Cheers guys and gals and happy hols, Darigan (talk) 15:00, 28 December 2010 (UTC)

1. It doesn't matter what order you buy them in. I would get the hosting first, because often a domain name registration is thrown in free with that through the hosting companies.

2. If you buy a domain name through a third party registrar (e.g. GoDaddy or whomever), you have the ability to "point" it at a server. To do that, you get some DNS numbers from the registrar (they'll provide you with them) and plug them into the right place on the server's configuration screen (pretty straightforward) and like magic it'll all work correctly. It's not very hard though the first time you do it, it takes some clicking around, but the hosting and domain services I've used make it pretty straightforward. The trick is you have to change the settings on both ends — that's how you prove "ownership." I may be mixing some things up here; it's been awhile since I've done this. But rest assured, it is not difficult.

3. Usually server hosts have rather sophisticated config screens that let you set up FTP accounts, manage e-mail addresses, change settings regarding PHP, set up virtual domains, add password protected areas, check your bandwidth usage, etc. It's not usually just a set of login details, these days. At least not the ones I've used.

4. You update the site using FTP (the uploading), or you can set up Dreamweaver to sync up with it. I don't know the details of how to do that with Dreamweaver, but it's surely under its FTP/server settings. Once you have the ability to set up an FTP account on the server, it shouldn't be too hard to make Dreamweaver automatically sync.

5. Opinions will vary on good hosts. Two very common "cheap" hosts include Bluehost and Dreamhost. They will run you something like $10 a month; less if you buy a few years in advance. I've used Bluehost for years with great results. They are almost never down, I've never come close to hitting any kind of bandwidth cap, and the few times I've had customer service issues they resolved them very quickly. But again, experience will vary, as any Googling will reveal. Some people on here swear by more expensive servers. If I were running a site where my income depended on it having constant uptime, I'd probably use something more expensive. (If I were running a major e-business, I would buy my own servers, of course.) There are some who swear by free servers; I find those to be a bit too unreliable and feature-poor, even for the little sites I make. It kind of depends what the site will be. If you have modest needs, check out Bluehost and Dreamhost. They are big companies, widely used, reasonably cheap, feature-rich, and won't disappear tomorrow. --Mr.98 (talk) 16:47, 28 December 2010 (UTC)

This is all good advice. I would add that you should be sure, if someone is giving you a domain for free, that you own the domain and not them. If there's a dispute, and you're the admin contact, you can always take your ball and go home, or at least to another provider. If you're not, you're screwed. Marnanel (talk) 16:55, 28 December 2010 (UTC)

Thanyou Mr.98 & Marnanel - I think that (as being a website virgin) I am a little bit nervous at the mo. Mr.98 - You make it sound *reasonably* straightforward, although, thanks for the words of warning Marnanel. I think that I will do this on Friday GMT (Thats when I get paid), and I'll give you guys/gals a heads up when its all ready. Thanks again Mr.98 & Marnanel, Darigan (talk) 17:08, 28 December 2010 (UTC)

It is truly not that hard anymore. I think maybe 10-15 years ago things were more difficult, required more technical knowledge, and was more of a "wild west" in terms of policies and services, but things have become settled at this point, at least with the big hosting companies. Feel free to post other questions here as they come up. --Mr.98 (talk) 18:10, 28 December 2010 (UTC)

Here's a question from the archives about domain registering and ownership, has some interesting information you might find useful 82.44.55.25 (talk) 18:02, 28 December 2010 (UTC)

HP Laptop Recovery

I bought an HP G42-355TU laptop yesterday. HP requires its users to create datat recovery disks after buying a new lap. I started creating the recovery disks today (which required me to burn the data into 4 dvds). But, after finishing the 2nd disk, there was a power cut in our area, and my lap displayed a message saying that it can only create a recovery disk when the lap is plugged into an ac source. So I had to quit the burning. Is there any way to resume the burning from where I left off?? Also, if I create the recovery disks, and then partition my drives, will the disks still work? There are only 2 partitions in my lap as of now, C and D (D is the backup drive with about 15 gb of soace alloted only for HP recovery data). What happens if I delete the backp partition?? Thanks in advance. 110.225.131.166 (talk) 15:11, 28 December 2010 (UTC)

I have an HP laptop, and went through the same process. Not sure about the power outage and resuming burning the DVDs (can you simply start over? Or is it not letting you do even that?), but yes, if both the disks and the partition are created, both will still work. It's simply (from what I understand) a way to ensure that you'll have some way of recovering when you need to. If you delete the recovery partition, you'll simply have that much more free space on your hard drive in a different drive letter (unless you adjust with Computer Management), but I'm not completely sure you *can* delete the recovery partition. Good luck. Kingsfold (Quack quack!) 19:29, 28 December 2010 (UTC)

Thanks. HP won't let me continue from where I left off. But yeah. I can start over. Although, there is a message that says it won't allow me to create the recovery disks more than once. But, since I didnt finish burning the entire thing last time, it'll let me start over. I guess I'll delete the recovery partition, once my disks are done... Thanks for all your help Kingsfold. Cheers, and a very happy new year in advance!! 27.57.223.66 (talk) 08:34, 29 December 2010 (UTC)

I'd be extremely hesitant to delete the recovery partition. Disk space these days is extremely cheap, and if the recovery partition is... what... 5-10 GB of space (if that), it seems worth it to me to keep. As an alternative, you might consider moving that much other "stuff" off of your hard drive onto DVDs or an external drive, or even a thumb drive; and leaving the recovery partition intact. But, you know, follow your bliss. Kingsfold (Quack quack!) 19:00, 3 January 2011 (UTC)

Which one should I buy...?

I want to purchase a android phone .... and I have choosen Spice M300 and Micromax A 60.. both are indian phone... can anyone suggest me which one should I buy...? or is there any other phone available in the range of Rs. 8000 in india.... My required features are ; 3G, wifi, gps, android, capacitive touch.. —Preceding unsigned comment added by 220.225.96.217 (talk) 20:26, 28 December 2010 (UTC)

Null caps lock

I have an HP keyboard, model no. KG-0851. I would love to tun caps lock permanently off, so that the key simply does not work. I tried going to control panel, then to keyboards, but the only thing in there is a speed tab, which has things like blink rate, and a hardware tab, which has in in a list of my keyboard devices (just this one) and properties, which then provides tabs for "general" driver" and "details". In General, I can click on "change settings" but all I get in there is a message: "this device is working properly". Anyone have a way for me to do something here about turning the caps lock off? The way this keyboard is set up, it is constantly being unintentionally put on, to my great frustration).--71.183.175.213 (talk) 23:29, 28 December 2010 (UTC)

Which OS? It's easy in Linux under X11 (read "man xmodmap", or, in recent distributions, use the appropriate config tool). --Stephan Schulz (talk) 23:39, 28 December 2010 (UTC)

Windows 7; sorry should have thought to include that.--71.183.175.213 (talk) 23:52, 28 December 2010 (UTC)

Some people physically remove the key. It can be remapped to another key with registry changes but I haven't tried it and it may be risky if you don't know what you are doing or you use bad software to do it. See for example http://www.howtogeek.com/howto/windows-vista/disable-caps-lock-key-in-windows-vista/. If you hold down Num Lock for 5 seconds then you may get an option to turn on a sound signal when you press Caps Lock or Num Lock. This should warn you so you can press it again. PrimeHunter (talk) 00:15, 29 December 2010 (UTC)

Thank you, thank you, thank you! Done. Caps lock is now just another shift key. This will save me many, many hours of time over the long haul.--71.183.175.213 (talk) 00:58, 29 December 2010 (UTC)

Thanks seconded! Always more to learn than one has had time for. 87.81.230.195 (talk) 13:04, 30 December 2010 (UTC)