Wikipedia:Reference desk/Archives/Computing/2008 November 12

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.

November 12

Reducing restrictive NAT on wireless router

Both my PS3 and Xbox360 report that my Wireless b/g/n router is using a type of NAT (type 3) that could cause problems connecting to online games and chat. How do I change my router settings so it plays nice with my game consoles? --69.151.187.196 (talk) 02:10, 12 November 2008 (UTC)

It could be that you need to forward ports and/or enable UPnP on your router. --Rixxin 23:37, 12 November 2008 (UTC)

Yeah, if your router has it, enable UPnP. It will save you many pains, as it essentially allows devices to forward ports as they wish. 24.76.161.28 (talk) 09:11, 13 November 2008 (UTC)

If it works semi-properly. UPnP is notorius for being shoddily implemented in a myriad of different ways. Washii (talk) 00:42, 14 November 2008 (UTC)

It doesn't appear that Airport Extreme base stations support UPnP directly. Now what? --69.151.187.196 (talk) 05:49, 15 November 2008 (UTC)

Dell Inspiron 1501 Powerpack issues

If you're curious, I'm writing this on my brother's computer.

I've had my computer and its powerpack for over a year now. Today, I unplugged the computer from the powerpack and moved it into the dining room for a change of location. when it told me that battery was low, I brought the powerpack in and replugged it into the wall and the computer. I continued working, for a long enough period of time that the powerpack must have been working to keep the computer running. At the end of the session, I unplugged the computer from the powerpack and brought it back into my room. I hooked it up onto my printer to print my work. Realizing that that computer should be plugged in, I walked back into the dining room, brought back the powerpack, and plugged it into the wall and the computer. While it was printing, I left the room, and did not return until nightfall, a few minutes ago. I noticed that the printing job was unfinished, because the printer had run out of paper. Also, the computer was off. when I pushed the button to turn it on, it beeped three times and told me "WARNING! BATTERY IS CRITICALLY LOW. PRESS F1 TO CONTINUE." I did so, but there was not enough juice, and the computer soon shut itself back off. It was fully plugged in and everything, so I wasn't sure why it wasn't working. I moved the powerpack and computer into the dining room and tried it in the socket there. It still didn't work. I tried the socket in the kitchen. It still didn't work. In the past, whenever the powerpack was plugged into the wall, a little green light would turn on. These times, no light turned on. What's wrong, and is there a way to fix it? Alternately, if not, can I still get the rest of the printing job from the printer (I've refilled it with sufficient paper to finish the job). — Preceding unsigned comment added by 70.179.52.204 (talk) 03:14, 12 November 2008

Depending on the size of the print job and assuming you haven't powered-off the printer, the rest of your document could be in the printer's memory. If an indicator light is orange or flashing, press the button to print whatever is still in the memory.

I'm assuming you have made sure the powerpack is connected properly (the powercord on my Dell is pretty stiff to insert in the powerpack). IIRC the "little green light" is on the PC and not the powerpack itself, so it could be a faulty powerpack (replacable from Dell or try eBay), or it could be faulty circuits inside your PC (not really replacable - sorry). If it's still under warranty, call Dell tech support - I was pretty pleased with their warranty repair service only a couple of days ago. Astronaut (talk) 06:49, 12 November 2008 (UTC)

IE6 "Active content" warnings, javascript permissions, blah

IE6 is "busting my balls" as they say. I am on a deadline to develop a product that will ship as HTML files on a CD that people can use (think of it as an online brochure, but off-line and more complicated).

The product uses extensive Javascript. It is mandatory for its functionality because it is offline. It has a javascript search engine and tree browser and all sorts of things.

This is fine in every browser I've used it on except IE6, which is still a bit too popular to ignore. IE6 has decided that it basically won't let the user execute javascript from locally hosted pages without lots of big yellow warnings that say "Internet Explorer has restricted this file from showing active content that could access your computer". Personally I'm dubious that Javascript alone can "access your computer" in a dangerous way, and think basically blocking all Javascript is throwing the baby out with the bathwater, but anyway. That's the message.

I've tried using the "saved from url" trick that I've seen on some boards. It does not seem to be working on my testing setup (XP SP2 with IE6 on a virtual drive). For the life of me I cannot get IE to run this product correctly (Firefox is fine, Safari is fine, etc.).

Any suggestions? What are my options? Sometimes it doesn't even show the little yellow bar, it just lists it as an "error" that my page is trying to use Javascript. Really, really unhelpful. --98.217.8.46 (talk) 03:15, 12 November 2008 (UTC)

Upgrade to IE7. Here's the link for you to download: http://www.microsoft.com/downloads/details.aspx?FamilyID=9ae91ebe-3385-447c-8a30-081805b2f90b&DisplayLang=en . Personally though, I'd stick to Firefox. Seems more reliable in my opinion. --Crackthewhip775 (talk) 23:13, 12 November 2008 (UTC)

The whole problem is that s/he's not the end user and so has no control over what browser will be used to access it. — Matt Eason ^{(Talk • Contribs)} 23:32, 12 November 2008 (UTC)

This page has some good ideas: http://www.winxptutor.com/lmzunlock.htm . You could change the extension to .hta or you could add change the registry settings to allow active content to run by default using a .reg file or the reg command.--Areateeth34 (talk) 23:41, 12 November 2008 (UTC)

Ship a copy of FireFox on the CD. It's free - it works. Write in big, friendly letters: "If IE6 sucks for you - use the free copy of FireFox that we enclosed for your viewing pleasure". That way you have complete control - if your software works great with that exact version of that exact browser - then any kinds of 'support' issues you get that relate to browser incompatibilities can be fended off with "Well, you can always run the copy of FireFox we enclosed on the CD."...obviously you should still seek compatibility with as many browsers as possible - but there comes a point with some of the more ancient ones where you have to give up and get on with life. You could even dive into the Firefox source code and 'hard-wire' the thing to have Javascript always enabled and your 'front page' HTML file be the default page on startup - then make that version of FireFox autostart when they stick the CD in the drive. SteveBaker (talk) 00:34, 13 November 2008 (UTC)

There's really no need to do that. You can actually make a special browser inside Visual Studio using the WebBrowser control. You add your own buttons for back and forward and an address bar. But I'd just make the user run a batch file modifying the registry or rename the HTML files.--Areateeth34 (talk) 01:34, 13 November 2008 (UTC)

You could just put in a browsercheck thing, and if they have ie6, then make the page tell them to use the included Firefox. flaminglawyer^c 04:41, 13 November 2008 (UTC)

Sharing the Program Counter

I have been studying for my PhD exam and I hit a question: "Can a thread share its program counter with its parent?" I know the answer is "no", but it got me thinking... Is it possible in any current popular operating system to have a thread create a pointer to the PC register? If so, sharing the pointer with the parent would be trivial. I can create a pointer to the PC register in non-threaded environments (I remember doing it on the Commodore 64). I haven't had any reason to try on any modern computer, so I don't know if it is possible. -- kainaw™ 04:16, 12 November 2008 (UTC)

Really the only way to share the program counter is in a 'SIMD' type of machine - the GPU in your graphics card for example. But in any case - you can't take the address of a register...that's meaningless...so a "pointer to the program counter" doesn't make a whole lot of sense in a real, physical computer (maybe in a simulated one you could do that). You couldn't take the address of the PC register on the Commodore 64 - it was a 6502 CPU and there is no such mechanism. Perhaps the "program counter" used by the BASIC interpreter would have been shareable...because it's just an "emulation" of a notional BASIC-CPU. SteveBaker (talk) 05:30, 12 November 2008 (UTC)

I agree. I think it may be possible in a virtualized machine where the CPU is virtualized also. That gets the registers into memory. Basically like the C64, which I mentioned. The current step (program counter) is all in memory. However, I just realized that I may have interpretted the question wrong. I think it is asking if a thread and parent can use the same program counter. That is obviously wrong since a thread can go off working one thing while the parent can continue working on something else. Of course, it could be read as asking if they can use the same register to store their program counter in when they are processing. That is correct - a single CPU will use the same register for all processes. -- kainaw™ 11:58, 12 November 2008 (UTC)

Well, in a 'real' computer (not a 'virtual' computer or a 'simulated' computer) - the program counter is a single, unique specialised piece of hardware - it tells the computer where the next instruction is to be fetched from. Because the PC changes with every instruction executed - two threads using the same PC would have to run on different CPU's (different cores - at least) and they'd have to run in perfect lock-step. That's not a feature that x86-style hardware is likely to be able to do. One thread runs an instruction and bumps the PC to point at the next instruction - but the other thread probably wasn't executing at the time - so nothing much happens until the PC has been bumped a few more times. Then the other thread might run - but now the PC has moved by (say) a few dozen instructions - that thread would 'miss' the instructions that the other thread had executed. The entire concept is so utterly meaningless as to be almost beyond my ability to articulate the problem! It can really only apply to a SIMD machine where all of the separate 'cores' or 'CPUs' are locked together so they all execute the same instructions (from the same location in memory) in perfect synchrony. In a virtual machine (like the Java virtual machine) - it would be possible to make two threads run in this perfect lock-step - but I don't quite see why that would be useful. Worse still - unless you have the kind of specialised hardware that SIMD machines have - things like conditional execution is going to be very tricky - if both threads say "if ( x > 10 )..." and one thread has x==9 and the other x==11 - then one needs to have the program counter step into the 'then' part - while the other has to have it go into the 'else' part. At this point, sharing the program counter completely breaks any kind of rational programming paradigm. In SIMD machines (such as the shader processor in your GPU) all of the processor threads execute both the 'then' and 'else' clauses - and the hardware simply write-protects the memory and registers in any of the threads that are running instructions that they are not supposed to be executing because of the failure of a conditional test. SteveBaker (talk) 00:06, 13 November 2008 (UTC)

Copy Protection

I have an old CD called TriplePlayPlus!Chinese that I dug out this evening and tried to run. It installs fine on Windows 98, but when I try to run it, the computer locks up. According to the task manager, something called CDCops ver 1.27B #3747 is Not responding. I've tried this several times in different ways, rebooting every time since there's no other way to get it moving again. I've also uninstalled and reinstalled it, with no success. Any ideas how I can get around this thing? It can't be the copy protection itself, since this is a legally purchased CD and it accepted the access code when I was installing it. Black Carrot (talk) 07:31, 12 November 2008 (UTC)

If you are running it on a newer version of Windows that might be the problem. The copy protection software may not be compatible with anything but Windows 98. The wonders of copy protection! Rilak (talk) 10:51, 12 November 2008 (UTC)

"It can't be the copy protection itself"—oh, if only it were true. In fact, the odds are it is the copy protection software, put in a new operating system, wigging out and preventing you, the guy who bought it, from using what you bought.

If you're really devoted to using it, I recommend using it with VirtualBox and a copy of 98, if you have one of those lying around. Otherwise... take this as a clear example of why DRM technology is actually extremely problematic for the honest consumer. --140.247.41.106 (talk) 16:54, 12 November 2008 (UTC)

I'm running it on 98, I said so before. On XP it didn't even install. Is there a way to bypass the copy protection without hurting the software? Black Carrot (talk) 17:56, 12 November 2008 (UTC)

Yes: Find a crack for the copy protection. Keep in mind that 1) doing so may not be legal where you live, and 2) cracks sometimes include viruses or other malware. --Carnildo (talk) 22:10, 13 November 2008 (UTC)

Merging accounts on AOL

Is it possible to merge two accounts on AOL, so that messages will at least appear in the same IM window? Ideally, the merge would be transparent, so a person IMing me on one account would not know about the other. I say this because I have two accounts, one for real life, the other for the internet where I'm sometimes wary of giving out real life information. And I want status messages/etc. to appear on both. 68.63.171.112 (talk) 08:25, 12 November 2008 (UTC)

Yeah. There should be a button (on the aim website or on the AIM client, I don't remember) that says Link accounts or something like that. After you do it, when you're on the AIM window, you can see both the acc's at once. I did it, so I know it works (or at least it used to work). And, like you said, I could message you on one acc and have no idea that the other is linked, or that you even have another acc. flaminglawyer^c 04:22, 13 November 2008 (UTC)

Yep. It took me a good 10 seconds, but I found a link. (sign in before clicking) flaminglawyer^c 04:25, 13 November 2008 (UTC)

Flash not cached

This page at "jedreport.com" starts off with embedded Flash from CBS. I find this chunk of a news show interesting and want to save it for future reference. (What else should I do with my monster hard drive?) However, the Firefox downloading add-on "DownloadHelper" doesn't see it, and more surprisingly it doesn't turn up in the cache: no item anywhere near big enough is in the cache directory as viewed with a file manager, and also it doesn't show up when I view "about:cache?device=disk" (as recommended in this page).

It occurred to me that the cache might just not be big enough at 120MB, so I tried increasing it to 300MB -- whereupon Firefox seemed to stop caching anything. So I brought the size back down again.

I wonder why I can't grab this Flash thing in the regular way (why/how it can not be in the cache), and what other way I can grab it. (I can choose among Firefox, Konqueror, Opera, Safari, Shiira, and even MSIE.) Tips? -- Tama1988 (talk) 08:53, 12 November 2008 (UTC)

Random guess, but try using the download manager of your choice on this link: http://www.cbs.com/thunder/swf30can10cbsnews/rcpHolderCbs-3-4x3.swf. I haven't tried it myself though. Rilak (talk) 10:46, 12 November 2008 (UTC)

That file is surely just the "player", not the video itself. The variables of the OBJECT no doubt point to whatever the FLV video itself is: "http%3A%2F%2Fwww%2Ecbsnews%2Ecom%2Fvideo%2Fwatch%2F%3Fid%3D4586892n&partner=cbssports&vert=News&autoPlayVid=false&releaseURL=http://release.theplatform.com/content.select?pid=4riNJ7D8_f6jKqW60ntgY356PanH9YNB&name=cbsPlayer&allowScriptAccess=always&wmode=transparent&embedded=y&scale=noscale&rv=n&salign=tl". Feel free to disentagle if you are interested.... ---140.247.41.106 (talk) 16:57, 12 November 2008 (UTC)

there are shareware/freeware utilities that will grab and save a flash file. however, as everybody suggests, in this case it's absolutely likely that the flash is just the player of a video file. installing the newer versions of real player installs a popup in flash window players which sticks up a tab when you mouse over it, saying "download this video to real player" which works very well, if it's a real player video, which they almost all seem to be on youtube, etc. Gzuckier (talk) 18:02, 12 November 2008 (UTC)

Thank you all for your efforts to help. I got the impression that this URL (an XML file) might be the key. It's got obvious links to "DoubleClick" crapola, and among this also the link "rtmp://cp21928.edgefcs.net/ondemand/?auth=dbEdwdHa5apd2dBbyaCaqaPcHcEbnbnaCbj-bjg5G8-h0-buy-Gc zMjUog&aifp=v001&slist=video/_!/CBS_Production/882/79/60_Obama_1109%3Cbreak%3Evideo/_!/CBS_Production/8 82/79/60_Obama_1109.flv", which looks promising. (I've inserted two spaces in this to avoid the need for horizontal scrolling.) However, Mozilla doesn't know what to do about any "rtmp" protocol, and anyway there are entities in that URL [?] that resolve to angle brackets around "break", so only the last part, "video/_!/CBS_Production/8 82/79/60_Obama_1109.flv" seems to matter. I tried splicing that to form "http://release.theplatform.com/video/_!/CBS_Production/8%2082/79/60_Obama_1109.flv" but that brings a 404 message. I'm pretty sure that theplatform.com is the host, though, thanks to the mumbo-jumbo on that domain's top page, mumbo-jumbo that I only dimly understand but that seems relevant to video hosting for big-bucks corporations.

Some software like this sounds good; I wonder if there's an equivalent for some other OS. Time permitting, I'll explore all the pages [gulp] of this long but useful-looking thread later today. Tama1988 (talk) 03:31, 13 November 2008 (UTC)

Having enabled Flash, and yeah, my link was the wrong one... Has anyone noticed that the video has an embed option with a link? I wasn't able to copy the link (Javascript problem on my side perhaps), but it might be worth a quick look. Rilak (talk) 08:39, 13 November 2008 (UTC)

This video has me stumped. I suspect that, given enough time, I could work out precisely what it is that I want, but I also suspect that since the hosting site (surely theplatform.com) takes such trouble to obfuscate it would also do referrer sniffing (is it called?) and tell me that no such file was available.

Maybe out of pure ignorance, I infer that webhosts mark material as either cacheable or non-cacheable -- how? in the http header? -- and that web browsers obediently comply. If this is so, to me it goes against the spirit of the web as I know it. Consider CSS: even if your browser knows what to do with CSS you can tell it to ignore CSS or to apply your own stylesheet instead or in preference. By analogy, I'd expect that browsers would have options for cacheing, with "cache as directed" as default but also "cache nothing" and "cache nothing". Actually some browsers do have that second option (aka "private browsing" or whatever); I wonder why none has the third.

Anyway, I'm on the lookout for something like "Replay Media Catcher" ("Runs on any PC with Windows 2000, XP, Server 2003 or Vista") for some other OS. -- Tama1988 (talk) 04:16, 14 November 2008 (UTC)

Aha, there's a Firefox add-on called BetterCache that promises to do just what I want. I'm experimenting with it now, Tama1988 (talk) 04:56, 14 November 2008 (UTC)

And the experiment failed. Sniff. Tama1988 (talk) 10:07, 14 November 2008 (UTC)

Columns in HTML tables

header 1
row 1, cell 1		row 1, cell 2		row 1, cell 3
row 2, cell 1			row 2, cell 2

Hi, if you want a table to have three even cells in the first row and two even cells in the second row, do you have to use colspan like in the example above? Is there a better way? Any help will be much appreciated. Thank you. --Kjoonlee 10:23, 12 November 2008 (UTC)

Colspan is the way to go. Use CSS to set widths of each cell to get them exactly the way you want them. An alternative is to embed tables. Make the top row and bottom row a single cell. Inside the top row, add a table with three cells. Inside the bottom row, add a table with 2 cells. You may run into issues with cellspacing and cellpadding, but playing around with them will eventually get you what you want. -- kainaw™ 12:00, 12 November 2008 (UTC)

Yeah, I'd take Kainaw's second suggestion unless you have a specific reason to not use nested tables. Set the spacing and padding to zero (at least when you start) and then make changes as you see fit. COLSPAN always gets me into trouble, but YMMV. ;-/ Matt Deres (talk) 17:44, 12 November 2008 (UTC)

Or you could set a colwidth="33%" thing, which works in wikitext and HTML. flaminglawyer^c 04:36, 13 November 2008 (UTC)

linking iphone to a different itunes account

What happens if, after linking an iphone to your itunes account, you sell it, and the buyer registers it to his itunes account? Does it wipe any data? thanks, 202.89.166.179 (talk) 13:20, 12 November 2008 (UTC)

I would like to delete my Wikipedia URL

Madam or Sir:

Please let me know how I may delete my URL. // *****It's a wikipedia URL. I have admin access. **********//

Thank you,

Bill Motsenbocker removed —Preceding unsigned comment added by 130.184.211.125 (talk) 17:51, 12 November 2008 (UTC)

Which URL do you need to delete and from where do you need it deleted? Astronaut (talk) 18:21, 12 November 2008 (UTC)

PS: These help and ref desk pages are highly visible across the internet and putting personal contact details on your comments is just inviting various spam-bots, crawler-bots and so on to send you even more junk. I have removed your details to protect your privacy.

Are you trying to ask how to delete an article on Wikipedia? See WP:Deletion process. -- kainaw™ 20:17, 12 November 2008 (UTC)

Are you perhaps asking how to get your user account deleted? If so, see WP:Right to vanish. All you need to do is to put {{db-user}} on the top of your user page and eventually, your account will be deleted. (It seems odd that someone with admin rights wouldn't already know these things). This question was asked (somewhat differently) on the Wikipedia helpdesk - and referred here because it didn't seem (then) to be a Wikipedia question...O think it really is a Help-desk issue - but our OP needs to explain in more detail - and more clearly. SteveBaker (talk) 23:45, 12 November 2008 (UTC)

The helpdesk sent him here. APL (talk) 02:56, 13 November 2008 (UTC)

That is just because "delete the URL" makes no sense. -- kainaw™ 13:02, 13 November 2008 (UTC)