Wikipedia:Reference desk/Archives/Computing/2009 October 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.

October 28

Amazon's Cloud RDS is new and different because... ?

From a ZDNet newsletter I received:

Amazon launches relational database service: Think MySQL in the cloud

Amazon on Tuesday launched a public beta of a service dubbed the Amazon Relational Database Service RDS. The main appeal: Allow customers to operate and scale database clusters while leaving pesky tasks like patching and administration to Amazon Web Services.

I can't read the article thanks to my company firewall, but I'm having a hard time visualizing any difference between this and just buying a hosting solution from another company. Nobody using a hosting provider has to actually deal with database software updates and patches, that's all part of the host's responsibilities... and no one big enough to need in-house hosting is going to want to suddenly drop it all for external hosting. So why all the fuss? 218.25.32.210 (talk) 00:48, 28 October 2009 (UTC)

If it works really well, you get the benefits (and demerits) of cloud computing; in particular it scales linearly with your datasize and your traffic, and it (ideally) makes geographically disperse replication mostly invisible. The downside is that your data is now in someone else's premises and under their control (although whether that's not equally true for regular offsite hosting of a db like mysql I don't know), which can present technical, legal, and perhaps commercial vulnerabilities. -- Finlay McWalter • Talk 01:06, 28 October 2009 (UTC)

Mmm... my read of the article (which is short and doesn't have a lot of technical details) is as you have it. Amazon is hosting off-site DB. They have optimized the DB (though how you do that well without knowing what the data is, I don't know). They have it on fast machines. They will take care of the DB. You just build your code to reference Amazon.

I can see ups and downs to it. In-house databases are often pretty sad and poorly maintained (I suspect about 25% of the stories on DailyWTF are about in-house database admins). Much of the daily upkeep could indeed be handled off-site. But setting the thing up in the first place is 90% of the trouble, and I doubt Amazon is going to do that for you. And you get all the downsides of hosting off-site—being dependent on their security, their servers working well, their uptime, etc. I guess it comes down to the price. Is it cheaper than hiring a full-time database admin, hosting the servers, and staying on top of that? If it is, then you can see the tradeoff. If it's not... then, yeah, no way. --98.217.71.237 (talk) 01:26, 28 October 2009 (UTC)

Youtube

Is there any way I can prevent people from seeing my youtube favorites?Accdude92 (talk to me!) (sign) 02:27, 28 October 2009 (UTC)

With the new layout, you can choose what you wish to display on your channel as such. – Latics ^Talk! 03:30, 28 October 2009 (UTC)

PSP Memory Sticks

The PSP video game system works with "Memory Stick Duo" and "Memory Stick Pro Duo" cards. What's the maximum size memory stick that works with an original PSP? Googling seems to yield only speculation by the unwashed masses, and the Sony website now merely shills the newest PSPs; Sony's "Support" section for the original PSP hasn't been updated since 4GB cards were the largest that were available. Comet Tuttle (talk) 05:17, 28 October 2009 (UTC)

I bought a PSP on launch day, and I've used a 8GB card. If that is useful information to you. Chris M. (talk) 20:10, 31 October 2009 (UTC)

Maximum iPhone acceleration?

What is the maximum g-force an iPhone can be subjected to? Sagittarian Milky Way (talk) 05:34, 28 October 2009 (UTC)

Do you want it to still work after being violently accelerated? That is survive the drop? Graeme Bartlett (talk) 11:52, 28 October 2009 (UTC)

Yes. Though (at least for case integrity) orientation is greatly important. Is there some solder/part or something necessary for use that could get worked loose or something from being repeatedly sub-damaging dropped? (One of my screen corners is now raised (a whole millimeter), which doesn't affect it one bit). And maybe being dropped on something that "gives" a few mm could even have a different way of failure than identical acceleration on something hard? Sagittarian Milky Way (talk) 15:05, 28 October 2009 (UTC)

html help

How can I make the grey line displayed on User_talk:Penubag/Sandbox3 go right below the title, not off-screen? I want it positioned on the right but don't want to define coordinates so when the browser is resized, it remains on the far most right. (I'm using firefox 3.5 on the monobook skin. Feel free to edit that page) penubag (sock account) 07:22, 28 October 2009 (UTC)

You mean something like this?

<div style="width:900px; margin: 39px 0 0 0; border:1px solid #AAAAAA; border-left-width:0px; 
border-top-width:0px; -moz-border-radius-bottomleft:17px; padding:0px 17px; font-size:100%; float:right; text-align:right;">Text</div>

--antilived^{T | C | G} 07:57, 28 October 2009 (UTC)

Only if it can be positioned at the top of the page as well using position:absolute;top:0px or equivalent. In other words, I want it aligned on the right and on the top of the page. Is this possible? penubag (sock account) 08:07, 28 October 2009 (UTC)

But position:absolute;top:0px doesn't put the box at the top of the page, but the top of the parent element. While for simpler sites it often is <html> it simply isn't the case on Wikipedia. An alternative is to use position:fixed instead, but that will scroll when the user scrolls, which may or may not be desirable. Example Code:

<div style="width:500px; margin: 0; border:1px solid #AAA; background:#FFF; position:fixed; z-index:100; top:0px; right:0px; 
border-left-width:0px; border-top-width:0px; -moz-border-radius-bottomleft:17px; padding:0px 17px; font-size:100%; text-align:right;">Text</div>

--antilived^{T | C | G} 08:43, 28 October 2009 (UTC)

how User_talk:Penubag/Sandbox3 is set up now makes the element go to the top where I want it. I don't want position:fixed 's effects here. So do you think it is impossible to make that element float to the right and stay on the top with Wikipedia?-- penubag  (talk) 09:09, 28 October 2009 (UTC)

That's how the standard say it should work. You can always do top: -xx px but that's really inelegant and will probably break between styles. --antilived^{T | C | G} 09:19, 28 October 2009 (UTC)

...or every time Wikipedia has a new announcement --antilived^{T | C | G} 09:30, 28 October 2009 (UTC)

(edit conflict)your recent edits look great! That's what I was kind of aiming for except it will break when Wikipedia notices on the top. Also, it's too bad it doesn't fit right in the Vector skin. Thanks a lot antilived, it's funny how something that seems so simple is impossible. -- penubag  (talk) 09:33, 28 October 2009 (UTC)

It's simple if you can control the semantic layout of the site itself (eg. make the parent of your div <body>). Maybe you can insert the element via javascript, but that only applies to your browser. --antilived^{T | C | G} 09:45, 28 October 2009 (UTC)

What is gmail doing on my computer?

I just noticed that when I try to follow a link in an email on a gmail account by selecting it and hitting 'Enter', I get a message

Grrr! A popup blocker may be preventing Gmail from opening the page. If you have a popup blocker, try disabling it to open the window.

However, when I click the link with the mouse, it works just fine. Firefox then gives a message that it blocked a popup, as it's supposed to. (I'd like that message to go away too, but that's a different issue.) I googled (!) that gmail-message and all I get is talk about how to de-activate popup-blockers. However, I like my popup-blocker. What I want is that message to go away. Opening that page is something Firefox is supposed to do, not gmail, right? So what is gmail doing on my computer? And why would a popup blocker prevent a page being opened anyway? Is Google participating in some sort of anti-popup-blocker scam? That's the only thing I can think of. But what scares me more is that Google can influence the behaviour of my browser. DirkvdM (talk) 08:08, 28 October 2009 (UTC)

Google isn't doing anything nefarious here. They're helpfully trying to fall back on other ways of opening the link despite your popup blocker, and as you've noticed, they sometimes succeed, but it depends on how you select the link.

You can disable your popup blocker on Gmail without disabling it everywhere. That's all they're asking you to do. And you don't have to listen to them, but then it becomes your problem when Gmail's interface doesn't work as intended.

Gmail works by using a technology called JavaScript that lets them modify certain aspects of the way your browser acts. This is the reason you can click messages and buttons in Gmail and have them do things without loading a whole new Web page every time you click. They have to change what it means to click on a link, or else it wouldn't be a very good e-mail client.

The Web is much more than documents and links these days. Interacting with a modern Web site does, in fact, require giving Web sites you visit some control over your browser. You're allowed to deny them that control as well, by turning off JavaScript, but the tradeoff is that you will get a much crappier interface for Gmail, and many other web sites won't work at all. You will be choosing to send your Web experience back to the 90s. rspεεr (talk) 08:25, 28 October 2009 (UTC)

If nineties technology works where 'naughties' technology doesn't, then of course I prefer the nineties technology. The fact that something is newer doesn't mean it's better (a common misconception). I know I can selectively disable the popup blocker. But I don't want to because I don't want any popups. They're a nuisance with no function whatsoever. And anyway, if I disable the blocker on gmail, then it will be disabled on all links in all my emails. Luckily there's the workaround of clicking the link, which does work. But that's not how it should be. So if this technology doesn't work properly yet, then it should be fixed before it gets implemented. I don't want to be a beta tester for a commercial company. DirkvdM (talk) 09:14, 28 October 2009 (UTC)

If you selectively disable popups for just mail.google.com, you will not get pop-ups. Gmail does not have or cause popups. Your popup blocker is merely interfering with a non-popup feature of Gmail. And if a link brings you to a different page that does cause popups, your popup blocker will kick in again, since the popups aren't coming from mail.google.com. "Naughties" gmail works just fine, in fact better than "nineties" technology. It is faster, more secure, and more aesthetically pleasing, to name a few. —Akrabbim^talk 13:20, 28 October 2009 (UTC)

I don't follow. If I selectively disable popups for just gmail, I will get popups there. Or do you mean I still don't get them elsewhere? Of course, that's what 'selectively' means. :)

If a gmail feature is disabled by a popup blocker, to the extent that I can't even follow a link, then that is a bug that should be fixed and in the meantime I should be able to disable that feature, but I don't even know what feature it is, let alone if it is useful to me. In any case, gmail should not disable the ability of Firefox to follow a link.

Which email provider is more aesthetically pleasing is very much a matter of taste. I still prefer Operamail, but the free version is horribly out of date in terms of storage space. DirkvdM (talk) 17:49, 28 October 2009 (UTC)

Gmail opens external links in new windows so you can still have your email open while you browse that new page - this is desirable behaviour, IMO. Your popup blocker should allow that as long as it can tell that you requested the page to open - it seems that your popup blocked can only tell that if you click on the link rather than selecting it and hitting enter. This may be something to do with how gmail handles keyboard shortcuts - do you have gmail's email shortcuts enabled? If so, try disabling then and see if that fixes it (you should still be able to open pages by selecting links and hitting enter - that is a Firefox shortcut rather than a gmail one). If it does and you are happy not to use gmail's shortcuts, then the problem is solved. If it does and you want to keep the shortcuts, you should report it to Google as a bug. If it doesn't fix it, then I don't know what the problem is - you should probably report it to Google too, at least you'll have ruled out one possible cause for them. --Tango (talk) 18:01, 28 October 2009 (UTC)

Popups are a characteristic of specific websites. Gmail does not have them, so if you disable your popup blocker for gmail, you still won't get popups from it. Other websites will still be blocked from causing popups. —Akrabbim^talk 20:16, 28 October 2009 (UTC)

Ah, ok, I misunderstood. Makes sense, I'll do that. Thanks. DirkvdM (talk) 09:18, 29 October 2009 (UTC)

Keyboard shortcuts are set to 'off', so that's not it. Opening a page by hitting enter is not a specific Firefox shortcut, but a very general one. Any browser does this. Including a file browser (which is almost the same thing); if I select a file and hit enter that activates that file, and in the case of an html file, it gets displayed (in another browser window). So Google is not following the standard here (which I consider a bad thing - standards are extremely helpful in making life easier).

I tried contacting Google before, but couldn't find an email address. That is usually at the bottom of the page or linked to at the top of the page. But not here. Again, they don't follow the standards. Of course I could try google@gmail.com. :) Actually, I just did, sending a test-mail first, referring to this thread to keep it short (in case the address doesn't work). DirkvdM (talk) 09:18, 29 October 2009 (UTC)

No, that address doesn't work. It's a short email, yet I received a message saying that it could not be delivered because "Storage quota exceeded". Of course Google has this address, so I assume they don't want to be contacted. I can imagine they get many mails, but that can hardly be a reason not to have any email address at all. Especilly weird for an email provider. DirkvdM (talk) 12:16, 29 October 2009 (UTC)

It's probably either a dummy account created by them to stop other people from using the address or it's marked as invalid and can't be used, but since they don't publish it the error is really only yours in assuming that is the address to use. I'm sure they do have email addresses, but more and more companies don't publish their external addresses and insist on you using web contact forms instead. Google is no different and if you try and use their help/contact pages describing your situation at the point the wizards can't help you any further they'll offer you a chance to "report it" to Google and that's where you can fill in the forms. The lack of publishing an email address is partly because of the spam problem and unwanted emails, but also because a lot of questions people ask can be answered by reading the help pages on the website so by forcing them to do that it seriously cuts down on the amount of mail they have to answer and frees up a lot more time. ZX81 ^talk 12:55, 29 October 2009 (UTC)

Google reads my email!

The above thread reminds me of another, more serious, issue I have with gmail. I often notice that gmail ads are related to the content of the email. In other words, Google reads my email. Whatever happened to "Don't be evil"? DirkvdM (talk) 08:12, 28 October 2009 (UTC)

Computers read your e-mail all the time. That's how it gets to other computers. And targeted advertising is Google's sole source of profit. There wouldn't be a company there to not-be-evil without it. rspεεr (talk) 08:25, 28 October 2009 (UTC)

If it's bothering you, you might want to try this firefox addon. It looks promising. -- penubag  (talk) 08:32, 28 October 2009 (UTC)

That addon deals with the more serious problem of Web sites sharing what they know about your browsing habits with other Web sites. It won't stop Google from doing stuff with the e-mail you specifically put into Google and told it to do stuff with. But then, I did just remember a silly way to disable targeted ads for some e-mail threads -- if you mention certain recent tragic events in your messages, it'll turn off the targeted ads so as to not offend people. rspεεr (talk) 08:43, 28 October 2009 (UTC)

If you want to ensure that no entity except your intended recipients ever reads your e-mail, not even a computer, you do have a way to do this -- you can encrypt your e-mails with public key cryptography. However, this will take some effort from both you and the recipients. Also, they will have to decode the message in order to read it, and if they do so at all, they are almost certain to do so using -- wait for it -- a computer. If you're worried about anthropomorphized computers reading your e-mail, you just can't win. rspεεr (talk) 08:39, 28 October 2009 (UTC)

Computers don't have to read emails to pass them on. They only have to pass on the packages, just like a mailman does with paper mail, without registering what's inside them. And I'm sorry, but the notion that there wouldn't be any (commercial) companies without advertising is complete nonsense. (Mind you, as ads go, I like the Google ads because they are so modest - I have even been known to click one, something I normally never do with ads (the ones that get through the mazes of Adblock, that is).)

Encrypting mail would be too impracitcal, since most people wouldn't know what to do with them, even if I told them. And them using their own computer is of course not the issue. What I expect from a mail company is to deliver my mail without reading it. I know that with email it's a lot easier to do that unnoticed, but that doesn't absolve them from the moral obligation not to to it. I wonder if they couldn't (and shouldn't) be sued for this. I assume there's no people reading my email, just a computer picking out key words. But that is already a form of reading, albeit low key. And I know that 'James Bond' can read my emails. There's little I can do about that. What worries me is that Google does this so openly, as if this is normal behaviour. This opens the door to a further invasion of our privacy, step by step leading us to a rather undesirable society. Big Brother is already watching us. But this is a whole lot worse than cameras in the (public) streets. This is an unseen 'camera' in our private lives. DirkvdM (talk) 09:42, 28 October 2009 (UTC)

"What worries me is that Google does this so openly, as if this is normal behaviour." It already is normal behavior. http://www.nbc.com/ has 9 web trackers on their website! And most websites have them as well. I'm talking about web trackers like Google Analytics, Google Adsense, Omniture, DoubleClick, Lotame, Tacoda, and Zedo. I've caught gmail sending information to Google Adsense on me a few times. If you want to view the trackers on an individual page, there's an addon for that. -- penubag  (talk) 09:58, 28 October 2009 (UTC)

I meant 'normal' in the sense of 'acceptable', not 'common'. (eg, if it is common to evade taxes, does that then make it normal?)

We don't have an article on web trackers, but I understand that those keep track of where I surf. That is something completely different, because it's in the public domain. Just like the the cameras in the streets that I mentioned. Mail and email, however, are in the private domain, especially the content. I find it really disturbing that Google does this so openly, which is an indication that they assume people don't mind their mail being read. As you indeed seem to do. Maybe I should raise this issue on a legal or political forum.

Another issue is what could be done about it. Email-providers operate internationally, so which legal court should deal with this? DirkvdM (talk) 12:01, 28 October 2009 (UTC)

Oh, and thanks for the tip about the Ghostery add-on. After installing it I visited nbc, and indeed there is a long list. Then I tried my bookmarks. About half don't use them, including Google, BBC, Wikipedia, Openstreetmap, Internic, Espacenet (patent bureau) and xs4all (my internet provider). But gmail and (surprisingly) operamail both use Google Analytics. And quite a few others too, including my newspaper, NRC, which is considered the most 'respectable' Dutch newspaper. And linuxquestions even has four of them! This really surprises me. I didn't expect these on .org sites, especially not open source sites (most of which indeed don't use web trackers, so this one is the odd one out).

Funny, I just gave a list of my bookmarks. :) But then that was my own choice, and few get to see them, which makes all the difference. DirkvdM (talk) 12:32, 28 October 2009 (UTC)

Yes, Gmail uses targeted advertising regarding e-mail content. That means it has little scripts that search for words and tries to think up what kind of ads go with them. That doesn't mean that a live person ever reads your e-mail, or that your security is compromised (e.g., that the ads can be traced back to specific content in your specific account). It does signal something different, though, as I'm pretty sure no other mail service (at least not one of this size) used the content of the e-mail as the basis of its targeted advertising (previously it would be based on information you put in, or targeted based on web browsing, etc.). Is this a problem? In a larger sense, it feels like a problem, but that feeling is actually you just realizing something that has been true all along: your mail host has access to your mail content, that your mail content is actually just floating around largely without your control before it gets to you, and that you are not really very happy with this. If this bothers you, at the moment your primary way to change it is to use a different mail service. Gmail is free, you are not paying for it, and have very little say with them other than being able to leave.

I doubt there are legal remedies. There is as far as I can tell no real violation of privacy, just a perceived one. It would be a very fine legal line, though, to ask whether a script that picks a few words at random out of messages and uses them for ads is violating any privacy, assuming no information leaks out. No other human being sees your message content. In any case, when signing up for the service, you no doubt agreed they could do this—read the fine print of their agreement, I am 100% sure it is in there.

As for what jurisdiction, it's doesn't really matter for international companies like this—you could file a suit anywhere they had a branch, I imagine. Best pick a place with actual legal protections of privacy though (that is, not the USA, where their headquarters are located, where privacy has very little protection). This is not legal advice, though. --Mr.98 (talk) 13:03, 28 October 2009 (UTC)

Here is a link to Gmail's privacy policy. Everything that they do that we have mentioned here is out in the open for you to read and be informed about. If you object to their conditions, you can cancel your account and pay for a more private and more secure mailserver. This is not evil corporate practice, as you mentioned before. Evil would be secretly breaking these conditions that they have laid out, which I think is beyond the scope of this thread. —Akrabbim^talk 13:26, 28 October 2009 (UTC)

Indeed, in the privacy policy there is a reference to 'the content of messages', but see my next post.

I also read there that they block spam. How can they know if I consider something to be spam? I never receive spam (anymore - I learned that the hard way), so there is only a danger there that they block something I do want to read. DirkvdM (talk) 18:26, 28 October 2009 (UTC)

Of course a mailprovider can look into my mail, as well as any other computer that it passes through on the way. Just like any plain old mail deliverer. The point is that that is immoral, and I assume illegal just about anywhere. Like I said, I understand that it's just a computer looking through my mail. And that computer can't understand what it is reading. Yet! But if we allow this then what else will be allowed? And of course computers get ever 'cleverer' and in the future it might be difficult to say where the dividing line is between a stupid computer and interpretation. The difference between data (key words) and information (reading) is not so clear and it will get ever less clear with the advent of intelligent computers. So we should draw a line (the same line that has always been there for mail deliverers) and say that no-one but the recipient should be allowed to look at the contents of email.

If the fine print (also an outdated notion in the world of browsers :) ) in the agreement states that gmail is allowed to read my mail, then the agreement is illegal. You can't make something legal by letting someone sign a statement that they agree with the illegal act. DirkvdM (talk) 18:26, 28 October 2009 (UTC)

It is not illegal at all. Email is not regulated the same as US mail is. There are no laws protecting your privacy from the company that is providing you with the service. You should be thankful that they offer such a nice privacy policy, and such tame, colorless advertisments. As for the spam, they have algorithms to be able to tell if something is coming from a spambot, as well as identifying spam-like attributes in the text. There are certain identifying characteristics, and Gmail is pretty good at what it lets through, and what it blocks. See also spam blocker. —Akrabbim^talk 20:12, 28 October 2009 (UTC)

If I were a judge, and I'm not, I would ask, "why expect mail privacy?" The answer is pretty straightforward, and leads back to the beginning of the postal system: if people are constantly able to see what you write, then you constrain what you write. If you expect your secrets will be known by others, you constrain them. And so forth. And secrets, of course, need not be illegal or illicit—they can just be something that one doesn't want other human beings to know, for whatever reason. We agree? Nobody cares if a dog knows your secrets—it's not the same thing. OK—is a script more like a person or a dog? At the moment, it's a dog. Definitely. All it can do is mechanically call up a table of possible ads and look (very cleverly) at your content and decide what ad might work best for you. Is that a privacy issue? If it's a dog, not really. It's not "reading" at all. There is zero comprehension involved. What if the machine gets smarter? Now we are getting precariously close to the Chinese room argument... but still, it's hard to create a sense of real violation unless we're talking about something that's actually sentient, that can actually interact with our world like a free agent. We're nowhere near that stage. Even if we were, your targeted advertising would be the least of our general social and legal quandries. So I'm not convinced that there's any privacy violation going on here. If the information in your mail cannot "leak" to real people (which is a big "if", to be sure, and one worth taking seriously), then where's the violation of privacy?

I'm not saying that Gmail's thing isn't creepy. It sure is, to me. But I'm not sure it's really creepy because something bad is happening. It's creepy because it looks like something bad is happening—it looks like someone is reading your mail. But no-"body" is—it is just a "dumb" bot.

As for legality... varies by jurisdiction. In the US, as I understand, you have no inherent, legal privacy over e-mail. Nada. There is no law that gives you that. You send something over someone else's servers, and they look at it—fair game, legally. It is not treated as postal mail. (There's an argument that this is a good thing, considering that e-mail is totally privately run for the most part. What would our e-mail systems look like today if the USPS had taken over e-mail responsibilities in the 1970s? Probably bad.) In any case, it's a free system that you signed up for, agreed to a license that lawyers no doubt inspected carefully, in a world with plenty of easy alternatives, and you continue to use just because you choose to, and there are no consequences for stopping your service other than slight inconvenience... aka, not terribly compelling, as far as claims to standing go.

As for spam... my understanding is that Gmail uses a combination of things to try and figure out what is spam. I'm not sure they actually have released what they do (as it would make it easier to defeat), but it's pretty straightforward. They look for common spam patterns (there are only so many iterations of spelling "VIAGRA" with numbers, after all), and look for things that other users have already labeled as "spam", and then subtract from that indications that it might be "ham" (real mail)—if it is sent from someone who you have corresponded with previously, for example. It then assigns a likelihood that a given message is spam or not, and if it is "spam" under its assessment, it moves it into a "spam" folder that just sits there, out of your way but not deleted, indefinitely. That's provided so you can check periodically to see if it is doing the right thing. Go ahead and take a look! Sometimes it gets things wrong, but it is pretty good, as far as such things go.

Anyway... if you don't like Gmail's policy on these things... you can leave! There are plenty of other services that offer similar services. There are plenty of services that won't give you ads at all and won't screen your spam out. Feel free to exercise your right to choice! There is really little holding you back, other than the hassle of a one-time "change of address" e-mail. (You can then set up Gmail to forward things your new address to avoid missed messages as people switch over.) Happens all the time. Not a big deal. --Mr.98 (talk) 00:33, 29 October 2009 (UTC)

Above, DirkvdM asked, "How can they know if I consider something to be spam?" Whether a message is spam is not based on your opinion; it's based on the conduct of the sender. Spam is not "mail I don't like"; it's "mail that is sent out in bulk, by a sender who knows that people never asked for it" -- and usually sent via any of a number of criminal techniques, such as using computers infected with a spammer virus (a botnet).

Mail service providers can identify spammers based on a lot of factors: spammers send large numbers of messages from home computers they've broken into; they put fake headers on the messages; they operate out of known-bad or known-insecure networks; they use various tricks to try to hide their origins or to defeat simplistic anti-spam systems; they send the same message from dozens of different (compromised) computers; and so on. The Internet's biggest spammers are well-known to the anti-spam community -- see the Register of Known Spam Operations (ROKSO) for instance.

Of course, a large mail provider (or even a site with a few hundred users) can learn plenty about spam from their users' reports. In the case of Gmail, there are buttons you can use to report a message as spam; and you can look through your "Spam" folder to mark messages as not-spam that have been incorrectly detected as spam. But even without this data source, a mail provider can do a *lot* to discern spam just based on the conduct of the sender. --FOo (talk) 03:51, 29 October 2009 (UTC)

Ah, after some searching I found the spam box. It's under '6 more' (nicely hidden). Luckily it's empty, so they haven't made any mistakes yet. DirkvdM (talk) 10:08, 29 October 2009 (UTC)

Mr 98, if it isn't illegal, then there's little I can do about that. But whether it's immoral and should be illegal is a different matter. You allude to the free market, but that shouldn't be the basis for things that are essential for normal living (in the modern world). For example, roads are so essential that one can not have them all be in private hands and say "well, if you don't like that none of the roads are open to bicyles or you can't afford the price they ask (or whatever), then tough luck". Of course, long distance communication is not as essential as roads, but it comes close. So if there is no government institution that provides the service to make sure it is available to everyone, then there should be laws that make sure there is a decent provision of the commodity. Which means that all providers of long distance communication should abide to those rules. You can't say, well some of them do, so the rest can be absolved from that. The postman shouldn't open my mail, my telephone company shouldn't listen in on my telephone calls and my email (and internet) provider shouldn't look in my email. In whichever form. Rules have to be kept as general as possible, because you can't possibly predict what future developments there will be. You say we're nowhere near sentient computers. How do you know? And even if it is decades into the future, that doesn't alter the principle.

I don't know USPS (I assume that means US Postal System and that that is government-run), but the Dutch postal system worked perfectly well before it got privatised. The fact that one specific government service doens't function well is not a reason to abolish all government services but to improve the ones that don't function well.

As for switching an email address, the biggest hassle is not the change itself or even notifying everyone (and reminding them several times later on). That takes at most one hour of my time. But I would want a provider that doesn't read my email. So I should read the small print of all the providers. A daunting task. And it would severely limit my choice. And what if there is no such provider, suppose they all do this, does your argument then still hold? This is the reason that they all should behave, to make sure that they don't all misbehave.

In a free market you can't say "you can do this, but you can't". That would create unfair competition. DirkvdM (talk) 10:08, 29 October 2009 (UTC)

I'm still at a loss as to why it is immoral. They are completely upfront about whatever information they use, and insistent that none of their processes violate your privacy and expose personal information. They don't take anything more than keywords, just so that they can keep a business running and provide you will a high-quality free service. I would be surprised if regular mail doesn't go through at least some kind of scan to detect harmful materials or bombs or something. That doesn't mean that they are looking inside your packages and finding out what's inside, beyond a simple "Dangerous: Y/N". —Akrabbim^talk 12:05, 29 October 2009 (UTC)

Let me take that last remark in my last post a little further. Suppose that unfair competition were resolved by saying that all communication companies can tap your communication. So telephone companies have computers that listen in on your conversations and when you have one about cameras then next day you get (e)mail from camera manufacturers. And mail companies hold your letters against the light to read them and stamp relevant ads on the envelope. That would even things out. Does that sound ok to you? DirkvdM (talk) 13:02, 29 October 2009 (UTC)

No, it doesn't sound ok, but then it doesn't happen either so I don't see how it's relevant? You're trying to compare Google's targeted adverts against a fictional 1984 type situation. Are you trying to suggest that by allowing Google to do this then it opens the doors to future things like this? I really can't see that being a catalyst, but everyone is welcome to their own opinions and beliefs! ZX81 ^talk 13:20, 29 October 2009 (UTC)

Not exactly 1984, because that is the government following people's every move. This is a private company. I'm not sure which is scarier. DirkvdM (talk) 19:46, 29 October 2009 (UTC)

(edit conflict) I've been reading this thread since it started and I'm sorry, but I don't think you understand the basics of how email works. You say you're concerned about the illegal/immoral actions of an automated computer reading your email, but I'd be more concerned about the fact that your email is completely unencrypted and to get to it's destination it will pass through multiple unrelated nodes, any of which *could* read or even copy it the email if they wanted. I'm not saying that's something you should just feel fine about, but it is the nature of how email works and if you're not happy with this you need to use encryption as written above. Also as previously mentioned without reading the body of the message the automated systems have no way of verifying it's not a scam/spam/virus email and handling it accordingly. If you want an email provider that doesn't do any filtering (very unlikely these days since the vast majority of customers want it!) then you would have to change to someone else or run your email yourself. As it's been pointed out above, Google are completely upfront about their email policies and you are free to move and go to someone else if you are unhappy with them. Saying you don't want to read all the small print because it's a "daunting task", I don't really see how you can argue then. If a company is telling you the in's and out's of how it operates and how it will handle your privacy and you can't be bothered to read it, then do you really have a right to complain if it later doesn't work the way you want/expect it to? If you don't want to use Gmail for whatever reason... don't. Probably the best solution to your problem is to purchase a Vanity domain and host that somewhere which will be your email address for life (or at least until you get so much spam you decide you need to change it). If you don't agree with someones policies, you'd then be able to pick up and move it to anywhere you want or even host it yourself if you feel technically inclined enough. Sure you'd have to tell everyone your new email address, but at least it's only the once. The only drawback is it isn't free, but it isn't an expensive cost (from my perspective anyway), considering the options it gives you. ZX81 ^talk 13:13, 29 October 2009 (UTC)

Yes, the OP still has not really specified what is immoral about it. The private vs. public debate is really somewhat off-topic (there are arguments either way). Most pertinent here is nobody is really "reading" the mail—it is a brainless robot that scans the text and finds matching ads. There is no actual privacy violation if no other living human being has access to the information. It feels like a privacy violation, because it looks like a human is reading it, but there is no comprehension whatsoever. It is equivalent to having a dog read your mail. --Mr.98 (talk) 14:54, 29 October 2009 (UTC)

Presumably, very precise statistics are being kept on what advertisements are shown, and in what markets. Certainly humans look at those. These are (one would hope) not connected to any particular user's account, but it's the equivalent of Google scanning their database and producing reports on what percentage of private emails talk about certain subjects. You may or may not consider that a privacy violation. APL (talk) 15:30, 29 October 2009 (UTC)

It's hard to come up with that being a privacy problem when looking at the aggregate. "Oh, 83.5% of people talk about their dogs and send photos of their kids." It's hard to construct a legitimate argument, in my opinion, that considers that a violation of personal privacy. It certainly should not lead to what we would consider the negative effects of privacy violation (liability, embarrassment, self-conditioning, etc.). --Mr.98 (talk) 19:20, 31 October 2009 (UTC)

The argument that I can switch providers keeps coming up, but that's not the issue. Also, having to read all the small print of all the providers was just a practical argument and not the issue either. My argument is that it is immoral and should therefore be illegal.

The comparison with other communication companies is not about a fear that that might happen in the future but that it is illegal for them. So it should be illegal for all communication companies. At least I assume pretty much everywhere it is illegal for mail companies to read the mail that passes through their hands and for telephone companies to listen in on phone calls. I said that allowing email providers to do this would constitute unfair competition, because that way they can generate income their competitors can't. But that's just a practical argument. The real argument is that it is immoral to scan people's private communication. I shouldn't have to come up with arguments for that because it is elementary (dear Watson). It goes against the grain.

Let me turn this around. Anyone who says it isn't immoral and should be legal for email providers should also find it ok for telephone and mail companies to do this. ZX81, you say that doesn't sound ok, so how can you say it's ok for email providers to do this? It's exactly the same thing. Your argument that that doesn't happen is circumstantial. And note that it doesn't happen because it's illegal. DirkvdM (talk) 18:32, 29 October 2009 (UTC)

Believe it or not, the Wikipedia Reference Desk does not have the authority to outlaw targeted advertising in email. APL (talk) 18:49, 29 October 2009 (UTC)

Really? Oh dear! :) DirkvdM (talk) 18:54, 29 October 2009 (UTC)

You've consistently dodged the question of what makes it "immoral", and the distinction being made between data mining and wiretapping. In the end, that is what it comes down to. If you cannot articulate why you think no distinction should be made, and will not engage with the distinctions that have been suggested, then it is really not worth discussing further, because you are not actually engaging the relevant issue. I'm not sure comparing to telephone companies works very well, but yes, if telephone companies somehow were paid for with targeted advertising, and they could do anonymous data mining of conversations that didn't involve any actual humans listening to the conversation, then I don't think that would be a privacy violation, no. (The idea of a telephone spitting ads at you is of course abhorrent, but that is a different question altogether.) --Mr.98 (talk) 19:20, 31 October 2009 (UTC)

By signing on to Google, you agree to this targetted advertising. If you don't like it, don't use Google, but remember that all e-mail is more like postcards than letters, and more like using a phone in a public bar than your private phone at home. All e-mail can be read by the technicians who look after the servers, though usually they will be too busy to bother, just like your postman will be too busy to read your postcards. Dbfirs 10:52, 3 November 2009 (UTC)

Visual Basic Flow Control Problem!

Hello. I have been programming with vb, but somehow, using the following function always results in the "Else" part! What is wrong? The Successor of Physics 10:38, 28 October 2009 (UTC)

Public Shared Function SetVal(Of outtype)(ByVal msg As String, ByVal def As String)
        Dim val As Object = InputBox(String.Format("Enter the new value for {0}", msg), "Set Value", def)
        If TypeOf val Is outtype Then
            Return CType(val, outtype)
        Else
            MsgBox("Sorry, unexpected input was recieved. This program will terminate immediatly", 16, "ERROR")
            End
        End If

What exactly are you trying to achieve? I suspect you're using "If TypeOf val Is outtype" to test where the user's input is an integer (for example). This won't work because InputBox returns a String - you can't use the TypeOf test to see whether a String could be converted to an Integer; you need to use Integer.TryParse or similar functions for this. (Also you misspelled "received" and "immediately".) AndrewWTaylor (talk) 12:23, 28 October 2009 (UTC)

I was trying to use the "If TypeOf val Is outtype" as an equivalent to the c# code "if (val is outtype) {}"The Successor of Physics 12:35, 28 October 2009 (UTC)

The syntax is correct, but "TypeOf val is outtype" will never be true unless outtype is String. To repeat my question, what are you trying to achieve? (Come to think of it, terminating the program immediately is rather a drastic response to incorrect user input.) AndrewWTaylor (talk) 12:44, 28 October 2009 (UTC)

plz help fft plots

Moved from Science desk by Falconus^{p t c}

i want to know something about this matlab code.what is it all about. i found it in image processings.

L = n_img;%No. of frames available for signal computation
NFFT = 2^nextpow2(n_img); % Next power of 2 from length of y
Y = fft(intensity_fluct,NFFT)/L;
f = Fs/2*linspace(0,1,NFFT/2);
plot(f,2*abs(Y(1:NFFT/2)))

yours.. SCI-hunter (talk) —Preceding undated comment added 10:47, 28 October 2009 (UTC).

I'm not a Matlab expert, but all it seems to be doing (or trying to do) is taking the data in the vector intensity_fluct and applying an FFT to it. It then plots the magnitude FFT output on the Y axis with the X axis calibrated from zero to half the sampling frequency. So it's basically graphing the different frequencies making up the input signal intensity_fluct. It seems a slightly over-complex way of doing it. --Lesleyhood (talk) 12:36, 28 October 2009 (UTC)

Thanks for previous answer. what is the significance of sampling frequence here.is it the (frequency/frame) i assign when i cut the frames from a video?if not then what is it.please help urgently. ~~ —Preceding unsigned comment added by SCI-hunter (talk • contribs) 04:56, 29 October 2009 (UTC)

Broken laptop due to wrong charger

I have a broken laptop. The defect was caused by a replacement battery charger, that was wrongly polarized. That means negative pole was connected where positive pole should connect and positive was connected where negative should be. The laptop can still work on battery power, since it is not heavily damaged, but it is not possible to load the battery, so it will not work for long. I suspect the laptop have a fuse that went off when electricity came in the wrong place. How difficult is it to repair this kind of thing? Is it enough to open it, replace it and start working again? Can it be done at home? --80.32.35.58 (talk) 11:39, 28 October 2009 (UTC)

A fuse would be quite easy to replace, but you may have blown a few power transistors or DC-DC convert chips and not be worth repairing. Graeme Bartlett (talk) 11:46, 28 October 2009 (UTC)

Well, I read somewhere else that laptops have a surge protection. I don´t know if mine has this, but perhaps there is a way of resetting it... --80.32.35.58 (talk) 11:48, 28 October 2009 (UTC)

Usually a surge protector will sacrifice itself to save whatever it is protecting. In that case, you will have to replace that component. If it was just a breaker, then I suppose it is just a switch. I don't know, however, because I have never owned a laptop and I have no idea what they come with. Falconus^{p t c} 13:54, 28 October 2009 (UTC)

(I am the OP.). Astonishing the laptop came back to life after I pressed a reset button, waited for some hours and started without the battery as they told me in another site. However, it still has a problem recognizing the battery. Speculative, I would say that the surge protection gets hot and cuts the power supply. After that it needs time to cool down to work again. Does anyone know what happened? It is an old ASUS, by the way, and I am using the same replacement battery charger (with right polarization this time). 80.58.205.37 (talk) 15:57, 28 October 2009 (UTC)

Best way to see all machines connected to my wireless router?

Resolved

I have cable Internet and a Linksys WRT54G wireless router. What is the easiest way to see all the machines on the wireless network? We have an XP laptop, a Mac laptop, an HP all-in-one, a Playstation 3 and a Tivo. I tried looking on the web admin page for the router (http://192.168.1.1) but couldn't find anything. Mike R (talk) 15:40, 28 October 2009 (UTC)

What do you mean by "see"? If you just want to know which machines are connected, nmap does it: nmap 192.168.1.1-255 and once that's done arp

If you need to actually browse the contents of those machines, you'd need to set up the relevant SMB/NFS shares accordingly. -- Finlay McWalter • Talk 15:47, 28 October 2009 (UTC)

According to the manual, you can get a "Wireless Client MAC List" by following the procedure on pp. 51–52. It's accessed through the MAC security controls, so I can see how it would be hard to find. -- Coneslayer (talk) 15:51, 28 October 2009 (UTC)

Thank you both; that was what I needed. Mike R (talk) 16:01, 28 October 2009 (UTC)

Home Network Question

I've just installed Ubuntu on an old PC I had lying around and decided to link it to my Vista rustbucket. No problems at all and the network is up and running perfectly - too perfectly, though, because I have set permissions for certain folders on my vista and the Ubuntu computer is now having no problems looking round at virtually everything else on the vista computer. Now, this is a problem, so I would like to password protect certain work folders on my desktop on the vista without archiving, zipping, compressing or doing anything else to them that would cause them to be slightly troublesome to work with (they are on the desktop for a reason!) from the vista computer. Is there any way to do this with Home Premium? Also, I have noticed that only my Ubuntu computer (and of course the Vista computer) show up on 'networks', and not my brother's XP or my mother's XP, which are all in the same house here. Why is that? Sorry, two questions here.... --KageTora - SPQW - (影虎) (talk) 16:21, 28 October 2009 (UTC)

I'm still trying to figure out the automation of the home network myself. When I replaced my router, Windows 7 auto-deleted home network 2 (what happened to 1??) and created home network 3... then rediscovered all my XP PC's. If it helps, each XP PC belonged to "Workgroup" when I had those setup under XP networking. Each XP PC also has an alternate IP setup, such as 192.1.1.5 (under TCP/IP properties). If you're battling, delete it and get vista to rediscover your network. As for the unix question, not sure, but maybe investigate linking them via NFS shares (or is it LFS in linux?) instead of a direct "root" share. Sandman30s (talk) 21:23, 28 October 2009 (UTC)

Hotmail doesn't work any better

At a college where the person helping me suggested I get Hotmail, the computers now have Firefox. I just happen to use Hotmail there, though it doesn't serve that much of a purpose. Various problems have caused me never to use Hotmail for composing. Excite, which has far more problems, gives me a pop-up plain-text work space which I can't lose as long as I don't click on the Red X. It can never be used to actually SEND the email, at least not the way I do it. That may be a question for a different day.

On Hotmail one would expect to go to the page to sign in and then to the inbox. That does happen. It used to take two tries, and still does with Yahoo.

If one goes to an email in that inbox and then follows a link, clicking "Back" should eventually bring you back to the inbox. Nope. Back to the sign-in page. You never see the inbox. You have to click on the inbox page because "Forward" will send you back to that email.

Of course, every tie you click on "Back" after looking at an email, it sends you back to that email after a brief glimpse at the inbox. Sometimes you can click on "Back" again to get back to the inbox.

And if you go to a different folder ... well, that's even worse. I generally do that at another library with Explorer because I have to double-check the sent folder in case an eamil I sent myself doesn't go through. You don't dare go anywhere else or you'll never find your way back.

I just wonder why this bizarre logic goes into the software for these email accounts.Vchimpanzee · talk · contributions · 17:44, 28 October 2009 (UTC)

Who knows? You might get lucky and have someone from Microsoft User Support come along and answer your question, but in the meantime, it may be helpful to contact them here directly. --KageTora - SPQW - (影虎) (talk) 17:53, 28 October 2009 (UTC)

Thanks, I may do that. I just get so frustrated with those people. Regardless of the email address, they take a year to do anything and they always come up with some excuse like "we can't duplicate the problem".Vchimpanzee · talk · contributions · 18:35, 28 October 2009 (UTC)

I am not at all surprised by the "we are unable to duplicate the problem" response, by the way. I have never seen the behavior that you have been complaining about. Comet Tuttle (talk) 21:43, 28 October 2009 (UTC)

You seem obsessed with the "back" button in your many similar lengthy questions about email. Why not just click on "inbox" instead? 78.144.248.102 (talk) 10:26, 30 October 2009 (UTC)

Well, that's what I do now, but if I want to go "back" to something I want to be able to do it. If I can't remember where I was, then how else will I get "back"?Vchimpanzee · talk · contributions · 20:38, 30 October 2009 (UTC)

c program

i need a simple code in c language in c,for the work of scienctific calculator,which takes 2 oparands and one oparator .for single four function calculator i made ,but for scienctific calculator i cant .plz give me a program for this,thanks210.212.20.75 (talk) 18:13, 28 October 2009 (UTC)

yacc and lex will generate you c code fo parsing your calculators input, the blanks should be easy to fill in. Even if you need to make it in pure c yourself, the bison (gnu yacc) manual could give you ideas. (hint hint, there's a calculator in the examples section). --194.197.235.240 (talk) 19:04, 28 October 2009 (UTC)

This sounds like a homework problem. What is it that you are having trouble with? Is it with getting input? Is it with parsing the input? Is it with printing the results? Is it a complete failure to understand what a C program is and how to compile it? Without understanding the problem, it is impossible to assist in a solution. -- kainaw™ 19:35, 28 October 2009 (UTC)

Someone asked this exact same question just a couple of days ago (here: Wikipedia:Reference_desk/Computing#program_of_c). The answer is really too complex for a Reference Desk question. However, read my (LONG!) response there - perhaps it'll be of some help. SteveBaker (talk) 00:43, 29 October 2009 (UTC)

Skype

One of my friends said skype was going bankrupt and that skype was shutting down, is this true?Accdude92 (talk to me!) (sign) 19:10, 28 October 2009 (UTC)

ebay paid $2.6B for Skype in 2005; it wrote the value down to $1.7B, but sold Silver Lake Partners 65% of it last month, the transaction putting the value of the whole company at $2.75B. So no. -- Finlay McWalter • Talk 19:25, 28 October 2009 (UTC)

What's the best VNC-like alternative?

I want to disconnect my Ubuntu machine from its monitor and just control it over the network from my various Windows machines. What's the best RealVNC-like program to do this? Is it still RealVNC itself, or is there some new hotness? Comet Tuttle (talk) 21:41, 28 October 2009 (UTC)

NX might be worth a look, though I've never tried it. -- BenRG (talk) 22:05, 28 October 2009 (UTC)

Ubuntu comes with its own VNC server, which you can enable under System/Preference/Remote Desktop. --antilived^{T | C | G} 04:00, 29 October 2009 (UTC)

What's wrong with RealVNC? --164.67.235.128 (talk) 08:42, 29 October 2009 (UTC)

Are we talking about what VNC server to run on the Ubuntu machine, what VNC client to run on the Windows machines or both? Nil Einne (talk) 08:36, 1 November 2009 (UTC)