Wikipedia:Reference desk/Archives/Computing/2007 November 14

= November 14 =

Firefox cache
I posted this thinking it was a wikipedia issue. But I'm convinced now that it isn't, so I've moved my question here. Can anyone think of a reason why Firefox's cache would stop "working". I mean, files appear to be being cached, but everytime I hit the back button, the old page reloads (or if I click a link, things like images and stylesheets are reloaded, even though they should be cached). I'm running Firefox 2.0.0.9 and this problem doesn't occur in Opera, nor on another computer I have running Firefox.-Andrew c [talk] 14:43, 14 November 2007 (UTC)


 * This could be a long shot, but you might want to check the value for browser.cache.check_doc_frequency in the about:config screen. If it is set to 1, that means it will reload a page every time.   The other valid values are 0 = Once per session, 2 = Never, and 3 = When appropriate/automatically.  3 is the default.  I have no idea how FF determines "when appropriate".  -- LarryMac  | Talk  15:26, 14 November 2007 (UTC)


 * Good idea. But it has to be something else. It was set for 3, the default. I changed it to 0, and the pages would still reload when I hit back. At this point, I think I'm going to try to contact Firefox support. Thanks for your help though.-Andrew c [talk] 14:48, 15 November 2007 (UTC)

How did the spammers find my gmail account?
I have a gmail account whose address I've never published. I use it strictly as an anti-spam filter for another address (which I also keep unpublished). I include a (slightly edited) copy of the spam headers below. Any thoughts on how the spammers managed to find this gmail address and if there's anything I can do to stop this from happening in the future?

Delivered-To: REDACTED@gmail.com Received: by 10.78.164.8 with SMTP id m8cs45882hue; Tue, 13 Nov 2007 21:02:02 -0800 (PST) Received: by 10.70.72.11 with SMTP id u11mr2554614wxa.1195016518126; Tue, 13 Nov 2007 21:01:58 -0800 (PST) Return-Path:  Received: from mail.com ([59.92.80.99]) by mx.google.com with SMTP id h20si474712wxd.2007.11.13.21.01.11; Tue, 13 Nov 2007 21:01:58 -0800 (PST) Received-SPF: neutral (google.com: 59.92.80.99 is neither permitted nor denied by domain of kimala_nair@yahoo.com) client-ip=59.92.80.99; Authentication-Results: mx.google.com; spf=neutral (google.com: 59.92.80.99 is neither permitted nor denied by domain of kimala_nair@yahoo.com) smtp.mail=kimala_nair@yahoo.com Message-Id: <473a8146.1486460a.600b.3e78SMTPIN_ADDED@mx.google.com> Reply-To:  From: "Jonathan Mesmar"  Subject: Get over 4000 TV Stations for a small one-time fee! Date: Wed, 14 Nov 2007 10:31:56 +0530 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1251" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000

- Donald Hosek 16:08, 14 November 2007 (UTC)
 * If the removed part is an everyday, regular word then it may simply have been guessed, especially if the spammer was SMTPing themselves and guessing starts of e-mails.  Lanfear's Bane |  t  16:20, 14 November 2007 (UTC)
 * I have some gmail addresses from its early history that are single words, and they get pummeled by spam -- ⁪ffroth 16:58, 14 November 2007 (UTC)


 * Spammers have botnets that can send billions of spams out every day. They can easily send spam to a@gmail.com, b@gmail.com ... aabc@gmail.com, aabd@gmail.com ... kainaw34@gmail.com, kainaw35@gmail.com ... so you can see that they don't "know" your gmail address.  They are sending spam to every possible gmail.com address. --  k a i n a w &trade; 17:41, 14 November 2007 (UTC)
 * Damn, that means that they're going through in excess of 109 pure alphabetic combinations just to get to me. At the moment it's a fairly small volume, so I'll live with it (I think I probably average less than one per day and google's spam filter catches it), but it makes me feel a bit better knowing that it's a brute force attack (although a little annoyed that Google doesn't do a better job against protecting against that). Donald Hosek 18:05, 14 November 2007 (UTC)


 * It is difficult to stop. Consider yourself a mail host.  You have mail coming in from thousands of computers located all over the world (OK - so most are in the U.S. - but we're being theoretical).  The mail comes in designed specifically to get past all your known filters - changing the subject and message just enough to look unique.  You can't block by a block of IP addresses, because this is being sent by infected computers all over the Internet.  You can't block by subject line because it keeps changing.  You can't block by the message because it keeps changing too.  When it comes down to it, it isn't the spammers who should be put against the wall and shot.  It is the people who feel it is perfectly fine to infect their computers with free music trading garbage, turning their PC into another spamming bot.  Stop people from turning their computers into bots and you'll stop the botnets. --  k a i n a w &trade; 19:22, 14 November 2007 (UTC)
 * It's completely ridiculous for spammers to go to jail- for what? Sending too many emails? As if that's not exactly what the infrastructure is made for.. same with malware authors- people are voluntarily running the code on their machines, the writers of the code aren't responsible for any damage -_- -- ⁪ffroth 21:19, 14 November 2007 (UTC)
 * They absolutely should (and indeed sometimes do) go to jail for doing it. In the US at least they are violating the CAN-SPAM Act of 2003 and costing us billions a year in networking gear to carry email - of which 90% is not needed nor asked for.  Worse still, as responsible businesses are too smart to advertise that way, Spam is almost always something disreputable.  Sending 10 year old girls emails about making their penises longer and what the consequences of this might be is certainly prosecutable in many jurisdictions under indecency laws.  So you are 100% wrong.  If we could find a better way to enforce these laws, Spammers would and most certainly should go to jail.  To the extent that CAN-SPAM has been used, people have gone to jail. SteveBaker 00:57, 15 November 2007 (UTC)
 * *rolls eyes* It's just email. It costs billions a year because mail servers are designed to serve indiscriminately- a mind-bogglingly stupid idea. Almost as stupid as enacting a law specifically to prevent sending too much email.. 100 years in the future the local newspaper will feature a "hilarious laws from the last century that are still technically in effect" column, and some kid will ask "what was 'eee maol' mommy, and why did they spend so much time regulating it instead of solving real problems?" It's astonishing that email has lasted this long; it should either go peer-to-peer or be a small network of databases run by Google, yahoo, ISPs, etc, instead of using the internet to let SMTPs contact mail servers. Anyway, it's not like it's costing "American taxpayers" billions per year- business is paying these costs and still staying profitable, so it's not crippling to the internet -- ⁪ffroth 01:25, 15 November 2007 (UTC)
 * Umm, "businesses" don't pay for my bandwidth, I do. People are sending messages to my e-mail server at my expense without my wanting them to do so (in fact, with my explicitly wanting them to not do so). They go to great lengths to circumvent my trying to get them to stop. I'm opposed to the death penalty. But I'd make an exception for spammers. There does need to be a better e-mail system, but pretending that spam isn't a problem because "businesses" can absorb these costs is absurd. Donald Hosek 06:42, 15 November 2007 (UTC)
 * What's really absurd is killing people for sending too much email! -- ⁪ffroth 17:13, 15 November 2007 (UTC)


 * It's astoundingly naive to say "business pays for it". Where do the businesses get their money?  It's from goods and services that they make or sell.  They either sell them to other businesses or to us.  If they have to pay more for their internet access - then that puts up their operating costs - which means that we have to pay more for our homes/cars/food/service/utilities - so in the end we definitely pay for Spam.  In fact, because businesses have to make a percentage markup on their products, we pay not only for the cost of the spam but also for the business profit margin on the cost of the spam - so it's actually more expensive if they pay for it than if we do directly.  SteveBaker 18:46, 15 November 2007 (UTC)
 * You don't have to pay for their products- it's not like taxation where you have to pay. I don't know, I'm just saying that is the businesses' problem, and the government should butt out. If businesses want to run open mail servers that just indiscriminately accept email without cryptographic signatures, then they have to deal with the inevitable spam. They're just asking for it. And they don't have to pay for the spam-handling infrastructure if they don't want to. -- ⁪ffroth 22:30, 15 November 2007 (UTC)


 * Um, malware is usually not strictly voluntary (don't confuse ignorance with consent). And I think it's a little silly to suggest that people who write code that lets people do really nasty things to other people's property aren't in any way responsible. (If they should be legally culpable is a different question, a subset of responsibility. I'm not necessarily arguing that—there are times when they should and times when they aren't.) People with specialized knowledge are always somewhat responsible for how that is used, whether they are engineers, scientists, computer programmers, whatever. Obviously the people who actually use the tools take up the brunt of the responsibility, but the person who put them out there to use certainly is in an important way responsible as well. Tools don't self-construct. --24.147.86.187 22:55, 14 November 2007 (UTC)


 * Personally I think the spam issue by itself is a bit overrated. OK, it's a little annoying, but honestly, it's not that hard to just delete it. OK, the net amount of resources it uses are a lot, but on an individual level it doesn't amount to much. Botnets are, though, a major problem—they make a lot of things other than spam possible, such as distributed infrastructure attacks, which is a recipe for bad news. But spam itself — who cares? It's one of the many small annoyances that come with any new technology, and not half as bad as some of the annoyances that come with other technologies (e.g. I consider almost all of the negative side effects of automobiles—pollution, accidents, noise, gridlock, suburban sprawl, oil dependency—to make spam look like a laughable problem). Anyway, if you really hate spam, the real perpetrators are the guys who cooked up SMTP and made it such an easy protocol to hijack. It's basically an ideal system for spam by design. --24.147.86.187 22:55, 14 November 2007 (UTC)


 * If it takes me (say) 2 minutes per day to erase junk mail from my inbox at work, that's 1/30th of an hour or 1/240th of an 8 hour day. So around 0.4% of the labour costs relating to staff having email use comes from the cost of deleting the email (not including the cost of extra network bandwidth, etc).  I think that's actually a huge underestimate because a lot of people actually read through each message before deleting it and/or are confused by why they are getting it and so forth.  But even 0.4% actually a huge amount when multiplied over the entire economy.  As for the idea that the SMTP designers are at fault, would you consider it not to be an offense to steal someones car if they leave it unlocked?  That's what you're saying.  Then there is the problem of younger children and the sheer volume of obscene shit that comes from these idiots...the only defense for caring parents is to keep kids from using email until they are old enough to understand this crap.  My elderly mother got very excited about email and being able to send photos of grandchildren back and forth and getting in touch with her old friends...then she started getting all of the usual junk mail and was so horrified that she turned off the computer and won't turn it on again.  Spam is a terrible thing and I'm greatly disappointed that so many people here have not thought through the issues to even the slightest degree - I didn't appreciate that there were so many shallow thinkers here. (sigh) SteveBaker 18:46, 15 November 2007 (UTC)
 * If there's a machine designed to send mail, using it more than you should isn't equivalent to stealing a car if it's unlocked. And I'm not saying the SMTP designers are at fault- they could never have anticipated that the internet wouldn't always be so trusting a place. But the mail protocols are outdated and anyone still implementing them just has to deal with the inevitable consequences. I'd give an example, but this very case is really the best example- suppose you set up a mail server that accepts email from anyone on the internet. You are literally inviting anyone on the internet to send you as much mail as they want, because that's what the protocol allows. There's no moral code in protocols- if the server will accept it, and it works, then you're just using the system just as it was designed. If the protocol is stupid and allows massive transmission of unsolicited messages, then get rid of the protocol, don't try to control how people use it. -- ⁪ffroth 22:36, 15 November 2007 (UTC)
 * Since we are now talking about SPAM in general, I often wonder exactly who responds to these spam messages. Is there REALLY such a market for penis enlargement and erectile dysfunction drugs? Obviously only a tiny tiny fraction of people respond, but I am still amazed that there is money to be made. I bet the people that respond to spam are the same that buy HeadOn. Its really disheartening to think that sizable number people can be duped so easily. -- Diletante 00:24, 16 November 2007 (UTC)

Name of a data structure
For one project I'm working on, I've (re-?)invented a datastructure, and I was wondering if there was a name or existing literature about it. Basically, it is a list of values and indices where they start, sort of like a sparse list except the previous value continues until the next value starts. At the moment I'm calling it an 'index list'.

If we have this binding:

ex = index_list((0, 'f'), (1, 'o'), (3, 's'))

Then, as_joined_string(ex) -> "foos", but as_normal_list(ex).index(20) -> 's'. Does anyone have a more specific name than 'index list' for this? —Preceding unsigned comment added by 79.75.201.99 (talk) 19:44, 14 November 2007 (UTC)


 * I don't understand the as_normal_list part, but representing a string as a substring with a repeat-count is just run-length encoding. --Sean 20:04, 14 November 2007 (UTC)


 * The number isn't exactly a repeat count though - it's the start index at which the letter starts repeating. The example we're being given isn't very good - so here is another:

ex = index_list((0, 'a'), (2, 'r'), (5, 'g'), (9, 'h') )


 * Which yields "aarrrggggh" ...followed by an infinite number of h's or something. If it were runlength encoded, it would be:

ex = runlength_encoded_list((2, 'a'), (3, 'r'), (4, 'g'), (1, 'h') )


 * There aren't names for all possible tiny variations on a standard data structure - so who cares? Just invent one.  This is something like a cross between a classic 1D sparse matrix and classic 1D run-length encoding. Programmers are inventing weird hybrid data structures all the time - we don't usually feel the need to give them names. SteveBaker 21:02, 14 November 2007 (UTC)

Dedicated Laptop Video RAM
I'm a little confused about video memory in laptops. A fellow at a local computer place told me that the video hardware installed in laptops never have their own RAM, and that they take RAM from the main system's RAM (like most on-board video in PC motherboards). Browsing around a few computer outlets I noticed some of the laptops have the video card advertised with "dedicated RAM". So does this guarantee that the video card has its own RAM (and doesn't go through the same bus as system RAM accesses do)? Or does it simply mean that the RAM it grabs from the system is fixed at all times? Thanks. --Silvaran 20:47, 14 November 2007 (UTC)


 * He probably shouldn't have said "never". Most laptops have fairly weak graphics capabilities.  But, there are exceptions.  I can't imagine by "dedicated RAM" they mean anything other than what you're describing.  My laptop has 128 Mb of video ram, very separate from main system RAM, on a PCIe bus.  So, yes, they do exist. Friday (talk) 20:52, 14 November 2007 (UTC)


 * My laptop has a discrete graphics card (low end, but even so it's a heat beast) with 128MB of video memory. That tech is wrong. Go with an integrated graphics card though- you should have a desktop computer for gaming, and a dedicated graphics card in your laptop will cost you weight, heat, and battery life (some of those negated). -- ⁪ffroth 21:15, 14 November 2007 (UTC)


 * Obviously Froth is not a fan of gaming in the tub... I got used to my laptop and would never consider a desktop ever again.  Of course, I do my "heavy" gaming on a console and save the Mac for strategy games and whatnot.  Friday (talk) 21:18, 14 November 2007 (UTC)
 * I'm not a fan of consoles either- if you can afford a decent computer there's no reason to buy a dedicated machine just for games. You'll get much better graphics and a FAR larger library of games (like 2 decades worth of backward compatibility + console emulators + make your own) out of a PC. Also I'm not enthusiatic about spending so much money on a computer locked down with DRM and signed code, to the point where you can't even control your own machine without an illegal mod chip. -- ⁪ffroth 23:13, 14 November 2007 (UTC)