All the Perl that's Practical to Extract and Report
use Perl Log In
WebDragon (1204)
(email not shown publicly)
http://www.webdragon.net/
Macintosh owner/user since 1987
Perl hacker since 2000
Linux (Redhat/Fedora) user since 2001
Pandora Awakens
[this was spurred on by Pudge's journal posting and in the end, I decided to make it a new journal entry rather than a reply, so as to elicit more comment.]
Gentlemen, the time has come.
The time is in fact long since past, but intertia and whatnot.
It's plain, clear, and obvious to me that some people simply cannot be trusted not to abuse an open standard. (i.e one of openness and trust)
Thus we see spammers exploiting open relay servers, and the like.
It seems to me that the problem is no longer the spammers or the companies that hire them.
The problem is and remains the seriously outdated trustful mail transport system we all know as e-mail. It is no longer viable as a medium of trusted communication with individuals and organizations electronically. It is time to scrap it.
It is time to scrap it completely, and replace it with something more secure, with tighter and more stringent standards.
And then simply *IMPLEMENT* it as widespread as possible. And leave behind ANYONE not switching to the new system.
It will not be compatible with existing e-mail software or transport agents. All of these will need to be rewritten. WHY? It's obvious if you think about it.
It is the ONLY way to get some lazy admins to implement security and secured communication between client and server. Leave them in the dust, if they do not. I have been pushing comcast for a *long* time, to have encrypted connection and security between the user and the mailserver, PARTICULARLY due to the party-line nature of comcast cablemodem networking. They "can't be arsed" in the vernacular, and it's a source of extreme irritation.
Something must be done, and it is no longer this race to keep up that we and the worldwide mailservers are slowly LOSING because of innefective standards and enforcement. Plus a zillion homebrewed solutions that don't combat the problem itself but merely enact a holding action against abusers of the problem.
registry, without which you CANNOT send e-mail. once registered, abuse gets you REMOVED. (maybe I'm dreaming, maybe I'm not) without registry how do we know you're a trusted user? "everyone gets one."
Maybe I'm just whistling in the dark here, but THINK about it for a while. CAN we replace the e-mail system entirely with something better? It's obvious from the configuration nightmare that sendmail has been, from the growing spam problem, forged addresses, forged headers, HTML e-mail that hogs bandwidth in what used to be a text-only medium including web-bugs to identify you to spammers, viruses (remember when it was IMPOSSIBLE to get viruses from e-mail and the whole AOL meme virus thing? and I mean IMPOSSIBLE. Not without actually opening an attachment. simply viewing a text e-mail CANNOT give you a virus.) and trojans that send e-mail from people that aren't really the owners/users of the computer, open relays being exploited, that something has to give. Yes there are other MTA's out there, but it's still the same old e-mail with the same old problems.
Maybe I'm opening pandora's box here, but without doing so, where's my hope for a better system than what we have now ?
I'm *tired* of this crap.
This is the sound of Mail Transport Agents *crashing* in a completely non-computer-related fashion. Not with a bug, but with me finally unzipping my lip and screaming, BY GOD I've Had ENOUGH!
"Hallelujah! Where's the Tylenol?"
an endless discussion thread about what features to include.
-- Jarkko Hietaniemi
OK, how do we start? (Score:1)
Re:OK, how do we start? (Score:1)
last one first; no, it doesn't start in the middle of a thought. "Registry" is the thought. Maybe I should have bulletted it.
on the questions you raised:
1a yes
1b yes
2a Abuse the system
2b To Be Determined
3 A more efficient and effective system of having each user with a "registered" address. A better PGP ? jury's still out on this. Registry database?
4. good questions. I
A Simple Refinement (Score:1)
While you're at it, perhaps you can take five minutes and create a better English than English? This language can be so ambiguous sometimes. It'd be nice to send e-mail that can't be misunderstood.
Seriously, the answer to any given problem is exceedingly rarely "throw it all away and start over from scratch", especially if you have years and years of history.
Re:A Simple Refinement (Score:1)
The problem we're facing now, though is how to get it there, and how to eliminate abuse of the system. The actual content is up to the individual users, and their varying communication abilities.
I understand where you're coming from, but sometimes the history has to take a beating so we can learn from our mistakes.
All too often I've stripped down a Perl program I've written in order to rewrite it from scratch doing t
Re:A Simple Refinement (Score:1)
Come back when your Perl program is some thirty years old and is being used by millions of people billions of times a day. Seriously. You're playing with a paper boats in a Navy drydock with that comparison.
Consider XHTML. It wasn't a complete rewrite of HTML 4. It was a refinement. How many people are using it? Consider SOAP. It was a complete rethinking of RPC over the Internet. How many people immediately dropped CORBA or document-based XML or XML-RPC for SOAP? (And do you want to be associate
Re:A Simple Refinement (Score:1)
However, consider this: that this is not a refinement or a rewrite idea but something completely new and different.
When you look at it from this perspective, it sheds a whole new light on the problem. A completely different solution is required to solve the modern problems that the current solution is incapable of doing more than sandbagging a
Re:A Simple Refinement (Score:1)
I would count that as a rewrite. I think you and I are at an impasse now. I agree that the current system is flawed in that it makes it trivial to spam. I disagree that throwing away everything of the current system is the right idea not only are there plenty of good things you'd lose, but it'll take absolutely forever to get it adopted widely.