Slashdot Mirror
Spam is Dead
Vainglorious Coward writes "Two years on from Bill Gates' promise to eradicate spam, an article in The Observer claims that spam has passed its peak and is now declining. Is it just me that hasn't noticed this?" I got almost a third more spam in 05 than 04. I guess I exist outside the bell curve on this one.
As soon as 2006 hit, my gmail account started getting spam. I have gotten 7 today alone. Argh.
Click Click Bloody Click PANCAKES!
Gates and co. would have to have an effective monopoly on email traffic for that to work. (Which might have been conceivable before the advent of Gmail, by the way.)
My spam peaked early 2004 with about 30,000 mails per stuffing not only my inbox, but also my DSL connection. I had a "catch all" option on several dozen domains and most of the spam I received was addressed to non existing mail boxes. Due to my local spam filters very efficient handling of the problem I only started to worry about the situation when downloading all the spam started to take hours and my provider complained about the daily traffic.
The problem with the non existent mail addresses became a large one sometimes in 2003, when enough people had some kind of spam filtering that deflected most of the usual spam. I guess that sometime in 2004 even the last catch all rules have been disabled, so that today simply guessing email addresses will gain nothing for the spammer.
So maybe spam has not really peaked, but there are simply different waves of spam techniques. Some of them rely on mass, others on tricking the filters. We may simply be in a "smart spam phase". A lot of the spam that reaches me today shows the message as a picture instead of text and I have not yet figured out why thunderbird will display those pictures, since I disabled this.
But the article is right in spam becoming something like a background noise. I still have to manually mark about 100 mails per day as spam, but I got very fast in recognizing it and it only takes a few seconds. I'm always astonished if I meet friends whose email address have not been public for more than a decade and who are very annoyed if one or to mails per week pass their spam filter. To me it is like complaining about banner ads. It's just an unavoidable part of the internet ecosystem, like mosquitos.
Chriss
--
memomo.net - brush up your German, French, Spanish or Italian - online and free
memomo: free web based language trainer DE-EN-ES-FR-IT
> You are opening yourself up to massive spam if you make so much as a single post
> there
I post to Usenet all the time, using an unobfuscated email address and I don't get any spam, so either what you're saying is no longer true, or gmail's spam protection kicks ass!
I wouldn't buy from someone whose corporate address is run out of a free web-based email service. If a company can't bother to have its own domain name or is blatant about using free-mail for its official corporate functions, then they appear no better than the stereotypical pimply kid running a business out of his parents' basement. I don't know how many customers bypass you because of this amateurish idea, but I know I would.
Why not register a domain and have all the corporate email addresses just forward everything to the corresponding gmail account? (you say you gave up your domain, so I'm guessing you're all joeblow@gmail.com). It would look a lot more professional. Domains are so cheap these days that there's not really an excuse to continue to use amateur-appearing tactics. You'd probably be better off with this compromise, or a similar one.
i am a soviet space shuttle
With Gmail, and I think most of the other webmail services (Hotmail, Yahoo, etc.), roughly all spam is sent to the spam folder, and I never have to look at it. So how is spam doing any significant "damage" to email? The average person probably wouldn't be annoyed by having a spam filter.
I make websites and stuff. Buy one.
I didn't even notice that TFA had ads. My Mozilla AdBlock filters are pretty minimal, too:
*.falkag.net/*
http://adserver./
*.atdmt.com/*
*.indieclick.com/*
http://adsrvr./
*.burstnet.com/*
*.tribalfusion.*
*.doubleclick.net*
*.loanweb.com*
*/ad.asp?*
*/ads/*
*/sponsors.*
*/advertise/*
*/adimage.php?*
*googlesyndication.com*
*personals.yahoo.com*
*/banners/*
http://ads./
*.valueclick.com/*
*.chitika.net/*
*/bannerads/*
*/marketing/*
*.adrevolver.com/*
*&adspace=*
24 filters, and I don't see more than 10 or 15 ads a DAY. I can't beat Yahoo, though, because they store their ads right in with the pictures for news articles and stuff. Keenspot uses the same dirty trick; I can't read some Keenspot comics without having to see Keenspot ads.
My own gmail account remains Free and Clear; I actually got one spam message ever on it, and I've had it for quite awhile now (and get quite a few e-mails and even subscribe to a few yahoo groups via it). And it's not like my e-mail address is that obscure, just my own first name followed by two other letters (and then the @gmail.com, naturally). The same could be said of my ISP e-mail address, or my university e-mail, or my hotmail/msn address, or even better my yahoo mail address which I fling around willy-nilly to sign up for things or whatnot whenver they require an e-mail address. And yet none of those e-mail addresses, all of which (except for my Uni one) I use astonishingly frequently and throw around all over the place, get any spam. Whatsoever. None. Except for that one gmail one (which ruined my perfect record, grr).
Note, also, that I turned off spam protection in hotmail, turned it off in yahoo mail, have none for my ISP one or my Uni one (both would only mark e-mail as spam instead of blocking it anyways, so I would know), and etc. Considering how high the signal-to-noise ration is, the possibility for false-positives understandibly outweighs the miniscule spam concerns I would have.
So what the hell am I doing right that most people seem to be doing wrong?
First off, none of my addresses are entirely intuitive or plain. No numbers even, nothing other than pure letters, but nothing that would show up unmodified in a wordlist or namelist (not even with good ol' "two random letters at the end of the string"). My sister has a gmail address of the same length as mine, but gets literally hundreds of spam messages every single day. The difference is that hers is her last name, while mine is my first name with two letters from my last; so hers is likely to show up in wordlists. That seems to be the kicker.
Meanwhile, my yahoo address seems to attest to the idea that signing up for things online won't get you spam, BUT the things I sign up for are message boards at places like BeyondUnreal.com or the official The Trews webboard or maybe to view some newpaper online (for those amnesiac days that I don't remember about BugMeNot). So nothing particularily sketchy.
In other words, as long as a person is relatively smart about how they handle their e-mail, they should be fine, 'tis my theory. This theory is not without major flaws, though, I'll admit. And furthermore, sometimes a person just wants a specific e-mail address, and it sucks then that it might just doom them to spam.
And further going down the questionable route of using my own personal experience as a scientific study, seeing as I had no spam until that one message, it would look something like this, starting arbitrarily in 2000:
2000 - 0%
2001 - 0%
2002 - 0%
2003 - 0%
2004 - 0%
2005 - 100% OMFG 2005 IS TEH SPAM APOCALYPSE
2006 - 0% (so far...)
So, in other words, I can prove anyone right. Parent? Sure, spam has
increased DRAMATICALLY in the last while. Naysayers? Bah, spam isn't
a problem! Etc. Ah, subjectivity.
I remember sigs. Oh, a simpler time!
Unless you work for Google (in which case you should have mentioned it before you started this thread) that is almost certainly a violation of their Terms of Use.
Mathematics is made of 50 percent formulas, 50 percent proofs, and 50 percent imagination.
...no: there is no technological fix to ingenious asocial behavior.
Yes there is. It is called a gun.
And its application is a bullet to the head of the anti-social person given by the governmental authorities of the day. The anti-social person can no longer affect society and can no longer by pass any methods intended to keep him in check.
But of course there is a major moral problem with my suggestion and should never be taken as advice.
I'm just stating the theoretical situation in which technology trumps social behavior. Obviously, its an extreme and we don't want to be going around shooting spammers (even though I'm sure some of you want to) but eventually given enough technology you can prevent everything.
Or rather what I am saying is that all social and political problems can be solved with technology. It just depends on your application of the technology and how far you are willing to go with the application. I'll take a bit of annoyance with my freedoms though.
"I am the king of the Romans, and am superior to rules of grammar!"
-Sigismund, Holy Roman Emperor (1368-1437)
This couldn't be handled at the client's end reliably because that would defeat the whole purpose (not to mention being a target of all those SpyWare vendors) -- in order to prevent bandwidth waste, it would have to be handled by the server.
For this to work, servers would have to indicate the going rate for messages (either by size, number of recipients, number of messages, etc.), and then the sending system would have to either accept it and actually transfer funds before sending the message, or just abort the transaction. The sender could choose how much they want to pay for this "ePostage" before sending it, and then the server could handle it automatically.
The main problems I forsee with such a system are eMail lists (as someone else already pointed out), and automatically generated eMails from other services (free or otherwise) that the user has signed up for. Why should Google AdSense or PayPal or eBay have to pay to notify me that my contact information is invalid, for example (I'm sure a skilled con-artist can see obvious ways to exploit something like this)? And do the users also deserve a share of this income, or just the ISP?
In addition to that, a few technical matters will need to be resolved before anyone can start thinking about even implementing such a system:
0. A new protocol to replace SMTP will be needed (it's not appropriate in my view to add this to SMTP, which is based on a trusted model rather than a costed/financial model). The protocol could be exactly the same as SMTP, but with one additional step inserted immediately after the "HELO/EHLO" stage in order to reduce development overhead for everyone.
1. Automated micropayment transfer protocols will need to be available to these new mail servers, and high-volume servers will need to be set up by the various providers of these financial services. Features will need to be able to handle currency exchange in a simple manner. Dispute procedures will need to be very, VERY well thought out.
2. The potential for criminals to launder large amounts of money by setting high rates or just claiming high volume when it doesn't exist (and both sides indicating this to be correct) in order to facilitate transfers between one another would be of great concern to government and military organizations aiming to impede the funding of so-claimed enemies (e.g., mafia, terrorist groups, trade blocked nations, etc.).
3. Micropayment service providers will likely compete on such things as percentages (e.g., they keep 0.05% of each micropayment to help cover their costs), various service charges (including fees for dispute resolution), usage fees, monthly service fees, etc. Banks are well-known for these types of tactics, and these micropayment providers will likely earn the same notariety.
In the end it will all just end up being very expensive and time-consuming, and I suspect that people will simply abandon it in favour or reverting to SMTP again in order to save money.
It's an interesting pipe dream, but I don't see how it will catch on in our current global economic climate given the current costs of doing business.
The Lumber Cartel, local 42 (Canadian branch)
British Columbia, Canada
Based upon my logs and those of two other machines that I do mail admin for, I'm not seeing that at all. If anything, there are more infected Winboxen out there than ever before, spewing tons of trash, and it's usually the Russians, Soloway, or some mysterious spammer hosted in a block of Chinese servers, all sending via these compromised Winboxen. If anything, my numbers are down at home, though that's because I can be a bit more restrictive about my firewall rules. Spamassassin is doing a very good job at filtering a large majority of this drek.
1) Take Spamassassin
2) Make it work in Japanese
3) ???
4) Profit.
Help poke pirates in the eyepatch, arr.
I have been system administrating several large scale email servers with around 50,000 users or so in total. During the "spam peak" we would have over 400 spam emails a minute being marked which was around 60% of the total email volume through that period. Now we are seeing around 60 emails a minute with more users and domain names on the system than before. However statistics are not everything. If we look more closely at the stats we see that while we would have an average of 400 emails per minute as spam it would peak up to several thousand a minute at times and sometimes it would be less than 20-30 spam emails a minute. While now we are almost flat lining at around 55-65 spam messages which means its not as big a drop as would have originally expected but it is still a drop. One of the issues we also note is that many of the cable providers are now blocking port 25 which was traditionally a large percentage of the traffic spam on our service.