Slashdot Mirror
7 Days In Email Hell
jfruhlinger writes "If you first went on line in the '90s, you probably remember a time when every e-mail you received was exciting, or at least relevant, and was worthy of your personal attention. One brave writer decided to take that approach to his present-day overflowing inbox. He read every email he received and dealt with them all, either by replying, filing, or unsubscribing. He even scanned his spam filter for false positives. It was a lot harder than he thought it would be."
I can't comment on other people, but the guy in the article is someone who has subscribed to over 50 newsletters that he doesn't want to read. In the article he complains about his poor personal management skills, insults people who don't agree with him politically, insults people who do agree with him politically, and complains.
What he doesn't do is explain why a common email management scheme is hell.
"First they came for the slanderers and i said nothing."
He's on too many mailing lists and has never filtered down the information he gets to something manageable.
I don't delete stuff from my inbox. If I've read it, that's fine, but it's perfectly acceptable for me to just search when I need something particular. In ancient times I used to make folders that were months (or years) when I got stuff, but that was an artificial structure, and not particularly useful.
For anything sketchy I just use http://spambox.us/. You can create a temporary email address that is forwarded to you're regular account and set it for deletion after a period of time (1 hour, 1 day, 1 week, etc.)
Gmail has given me the LEAST spam of the 3 big name providers (Google/Yahoo/Microsoft), including when I had my own e-mail server with spamassassin. Not sure what problems you have with Gmail, but false positive rate is minimal and I rarely get more than 50 -actual- spam messages a month. Rest is notifications/newsletters I actually signed up for, or work related.
If you use gmail, you can do something similar with a dash. For example, johndoe-newegg@gmail.com is the same account as johndoe@gmail.com. Doing things like johndoe-trash@gmail.com, and creating a filter on that destination address yields fantastic results. You can also add any number of periods to your email address, i.e. john.doe@gmail.com is the same as john.d.oe@gmail.com and johndoe@gmail.com.
I do the same thing with hyphens and Qmail. It's practically eliminated spam as a problem for nearly a decade. The only two problems I have are people (and businesses) that get freaked out seeing an email address like me-yourname@mydomain.com, and websites that want an email address to recover a login (if I can't figure out what address I made up for that particular site... I have semi-standards, but they don't always work 100%).
Ya know that is the one kind of "spam" I have to say I really enjoy. Opening my inbox and finding something I wanted ultra cheap, like that 1Tb Samsung I got for $35? that's nice. Or the "77 features of Windows 7" which actually pointed out a few tricks I'd never heard of (type PSR in the start search and you can record what you are doing as a step by step tutorial, real handy when i'm teaching someone how to use a complex program) which showed up last week? That's nice.
But to me the sweetest thing about email today is how damned nice the spam filters have gotten. i remember when false positives were high and you'd still get a bunch of "4er8al v1agra" bullshit, but now? I can't remember the last time I saw spam in my Yahoo or my Gmail.
So while I can understand why some my want to unsubscribe I like getting my parts cheap too much or learning cool tricks to give up my newsletters. As long as the webmail guys (thanks webmail guys) keep the spam filters rocking finding a couple of sales flyers and a newsletter or two is just a nice diversion.
ACs don't waste your time replying, your posts are never seen by me.
Relying on Baysian neural networks to seed the pseudo random logic
Bayesian networks are completely different from neural networks. And they're not used for "seeding" any "psuedorandom logic"; bayes nets and neural nets are just useful for classification tasks, such as "spam" or "not spam."