From: Sean Conner Date: 10:33 on 28 Feb 2008 Subject: So much hate, spread across so much software, it's incredible It all started some five hours ago with my own hateful blogging software, when the web interface (which is something I rarely use it's so loathsome, but that's my own fault) didn't work. It took me the better part of three and a half hours to track down the actual problem deep within Apache. It seems that the ErrorDocument directives I have break the basic Authentication directives I have. How, I don't know. All I know is that I wasted the better part of three hours tracking down *that* little problem, and thinking I found an actual bug in Apache (you don't say?) I thought I would mention it to the users@xxxxx.xxxxxx.xxx mailing list, despite the knowledge that the first thing out of *their* mouths would be "upgrade to 2.0.61^H2^H3/2.2.6^H7^H8 and try your plea again." Only when I sent my plea to the mailing list, only to get back: > Hi. This is the qmail-send program at apache.org. > I'm afraid I wasn't able to deliver your message to the following addresses. > This is a permanent error; I've given up. Sorry it didn't work out. > > <users@xxxxx.xxxxxx.xxx>: > Sorry, only subscribers may post (#5.7.2) That's odd. Then were do all these messages from users@xxxxx.xxxxxx.xxx that keep filling up my inbox come from? Obviously, I'm subscribed in *some* fasion to the list. Did my email client (mutt) puke and set the wrong from address? The users@xxxxx.xxxxxx.xxx mailing list is sending it to my "spc at conman.org" address, and yes, mutt sent it out as "spc at conman.org", so what is in fact, the problem? Maybe I'm half-unsubscribed? Who knows? Let me subscribe again. Okay, there's the confirmation, let me reply to that. *NOW* let me try sending in my plea. Great, now I get *TWO* copies from users@xxxxx.xxxxxx.xxx. Did I just spam the list twice? I sent a reply to that, apologizing for the double posting (maybe), when I get two replies from that. Okay, what the @#!$#@!#!$ is going on ... An hour later, and I have my answer, only I'm not exactly sure who to blame for this incredible mess. Is it: Me, for not noticing that I was subscribing to the list using not "spc at conman.org" but in fact "spc at brevard.conman.org"? Postfix, for setting a return path of "spc at brevard.conman.org" even though I told it to use a "myorigin" of "mydomain" and not "myhostname"? Ezlml (what Apache apparently uses for their mailing lists) for using the email address in the Return-Path: header and *NOT* the address that appears in the From: header? At this point, I'm pointing fingers and naming names at Ezlml. Why, oh, why, oh, why, would you use the Return-Path: and *NOT* the From: header? Grrrrrrrrrrrr ... -spc (When did this stuff get so difficult to debug?)
From: Sean Conner Date: 10:40 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible It was thus said that the Great Sean Conner once stated: > > Postfix, for setting a return path of "spc at brevard.conman.org" > even though I told it to use a "myorigin" of "mydomain" and not > "myhostname"? Oh, I almost forgot. I just learned that if you aren't running Postfix (all bazillion programs) and attempt to send mail (say, using the command "mail" from the command line), not only does Postfix not bother to queue the mail, it just silently drops it into the bit bucket (okay, not totally silently---I mean, it does log that it couldn't send the mail, but otherwise, no indication that it failed to send the mail). Grrrrrrrrrrrrr ... -spc (slowly realizing that email is totally borked ... )
From: Peter da Silva Date: 13:50 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-28, at 04:40, Sean Conner wrote: > -spc (slowly realizing that email is totally borked ... ) This is why I use qmail. It may be psychotic but at least it's paranoid.
From: Tony Finch Date: 15:07 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, 28 Feb 2008, Peter da Silva wrote: > > This is why I use qmail. It may be psychotic but at least it's paranoid. Boggle. I wouldn't call an MTA that accepts everything "paranoid". Also, it has been abandoned by its author and doesn't even compile on current unixes. Tony.
From: Peter da Silva Date: 16:07 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-28, at 09:07, Tony Finch wrote: > On Thu, 28 Feb 2008, Peter da Silva wrote: >> This is why I use qmail. It may be psychotic but at least it's >> paranoid. > > Boggle. I wouldn't call an MTA that accepts everything "paranoid". That's all part of the "psychotic" thing. OK, if you don't want to call it "paranoid" I'll go with "obsessive-compulsive". In mail software I believe that's a GOOD thing. > Also, it has been abandoned by its author and doesn't even compile > on current unixes. So long as I have the source I can make it compile on anything.
From: Gerry Lawrence Date: 16:16 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible Tony Finch wrote: > On Thu, 28 Feb 2008, Peter da Silva wrote: > >> This is why I use qmail. It may be psychotic but at least it's paranoid. >> > > Boggle. I wouldn't call an MTA that accepts everything "paranoid". It's default behavior does (as is spec'ed by that hateful RFC) but with qmail patches (and yes, that's the hateful state of its development, but as of 11/2007 all qmail will be in the public domain so development can finally more forward) you can change this -- always use the "fast fail" and REALRCPTTO patch -- doesn't accept unless it can deliver the mail. No bounces and no joe-job spam etc. > Also, > it has been abandoned by its author and doesn't even compile on current > unixes. > Gee, I guess I was hallucinating when I compiled it just last year.... SEE: netqmail-1.05.
From: Tony Finch Date: 16:24 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, 28 Feb 2008, Gerry Lawrence wrote: > Tony Finch wrote: > > > > > > > Boggle. I wouldn't call an MTA that accepts everything "paranoid". > > It's default behavior does (as is spec'ed by that hateful RFC) Where does it say that? > Gee, I guess I was hallucinating when I compiled it just last year.... > SEE: netqmail-1.05. If you run an unofficial fork then all bets are off, including DJB's security guarantee. (An MTA that's so broken it won't compile is of course much more secure than any other...) Tony.
From: Peter da Silva Date: 16:45 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-28, at 10:24, Tony Finch wrote: > If you run an unofficial fork then all bets are off, including > DJB's security guarantee. I'm sorry, I can't figure out what you mean by this. Why the hell should I care about what DJB guarantees or whether my code's "official" or not?
From: Gerry Lawrence Date: 17:01 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible Peter da Silva wrote: > On 2008-02-28, at 10:24, Tony Finch wrote: >> If you run an unofficial fork then all bets are off, including DJB's >> security guarantee. > > I'm sorry, I can't figure out what you mean by this. Why the hell > should I care about what DJB guarantees or whether my code's > "official" or not? > > By "does not compile" I guess he meant "does not compile without a guarantee from the author" -- so I guess that means NO software compiles.... I can understand hate for qmail. Qmail's history is unique -- it was written, then djb stopped. If qmail was any less than what it is, it would have died. But it was so good and people loved it so much they continued on the only way they could -- they patched it and it continues to live to this day. It's install is different from what people are used to, so it's weird. It does things differently, so it's hated. We don't care. It runs, it doesn't fail. And it absolutely will not stop delivering the mail. Once installed it gets out of your way and does its job. Once I discovered qmail I vowed to never touch a hated sendmail config again. Oh, an M4 config file -- now THAT will solve the problem!
From: Tony Finch Date: 17:54 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, 28 Feb 2008, Gerry Lawrence wrote: > > By "does not compile" I guess he meant "does not compile without a guarantee > from the author" -- so I guess that means NO software compiles.... No, I mean DJB's guaranteed version does not compile without patches. It has incorrect declarations for various libc internals such as errno. Tony.
From: Nicholas Clark Date: 18:00 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, Feb 28, 2008 at 05:54:30PM +0000, Tony Finch wrote: > On Thu, 28 Feb 2008, Gerry Lawrence wrote: > > > > By "does not compile" I guess he meant "does not compile without a guarantee > > from the author" -- so I guess that means NO software compiles.... > > No, I mean DJB's guaranteed version does not compile without patches. It > has incorrect declarations for various libc internals such as errno. Could it ever have compiled on any out of the box Unix variant ever? Did something he happened to once use have just the right level of sloppiness in libc to allow it to work? Nicholas Clark
From: Gerry Lawrence Date: 18:13 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible Nicholas Clark wrote: > On Thu, Feb 28, 2008 at 05:54:30PM +0000, Tony Finch wrote: > >> On Thu, 28 Feb 2008, Gerry Lawrence wrote: >> >>> By "does not compile" I guess he meant "does not compile without a guarantee >>> from the author" -- so I guess that means NO software compiles.... >>> >> No, I mean DJB's guaranteed version does not compile without patches. It >> has incorrect declarations for various libc internals such as errno. >> > > Could it ever have compiled on any out of the box Unix variant ever? > Did something he happened to once use have just the right level of sloppiness > in libc to allow it to work? > > Nicholas Clark > > The issue is the errno declarations that changed in some sorta gnu libc thing. The errno declarations are a simple change, and YES, it did compile with previous versions of gcc. Even now, it's possible to compile it without changing the code, with a: gcc -O2 -include /usr/include/errno.h There are a couple of bugs in qmail that you should patch for: See: http://www.thedjbway.org/qmail/qmail_at_eight.html
From: Peter da Silva Date: 18:41 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-28, at 12:13, Gerry Lawrence wrote: > The issue is the errno declarations that changed in some sorta gnu > libc thing. Oh, so by "does not compile without patches" means "does not compile on Linux without patches". All the world's a VAX of course.
From: Gerry Lawrence Date: 18:56 on 28 Feb 2008 Subject: qmail hate -- or love ( and the new KDE hate) Peter da Silva wrote: > On 2008-02-28, at 12:13, Gerry Lawrence wrote: >> The issue is the errno declarations that changed in some sorta gnu >> libc thing. > > Oh, so by "does not compile without patches" means "does not compile > on Linux without patches". > > All the world's a VAX of course. Oh I wish! Well, not really. Anyway, now that djb has announced that his past and future software is/will be in the public domain, maybe we'll see some improvement. Or maybe it will become a huge mess and destroy qmail, and then we can REALLY hate on qmail! I feel compelled to not post to this list with at least some hate, so here's something that's hateful: KDE3. SO MANY THINGS ARE BROKEN! konsole, to name one, went from being the best xterm, with good solid simple configs, reliable to a worthless buggy POS. Gone are the simple controls for fonts, size etc. It's cut and paste is broken (half the time unix-style select does not work, and I have not figured out why) SIgh. It's so much more hateful to see a good project gone bad then any other kind of hate.
From: Peter da Silva Date: 19:02 on 28 Feb 2008 Subject: Re: qmail hate -- or love ( and the new KDE hate) On 2008-02-28, at 12:56, Gerry Lawrence wrote: > KDE3. SO MANY THINGS ARE BROKEN! How come? I thought KDE was supposed to be the less-broken one.
From: Gerry Lawrence Date: 20:05 on 28 Feb 2008 Subject: Re: qmail hate -- or love ( and the new KDE hate) Peter da Silva wrote: > On 2008-02-28, at 12:56, Gerry Lawrence wrote: >> KDE3. SO MANY THINGS ARE BROKEN! > How come? I thought KDE was supposed to be the less-broken one. yeah -- kde2 apps are generally great -- k3b - awesome. ksnapshot for screenshots -- love it! But it seems like the new kde3 versions are slipping into hate.
From: Thomas Matelich Date: 22:03 on 28 Feb 2008 Subject: Re: qmail hate -- or love ( and the new KDE hate) ------=_Part_2762_26432524.1204236222867 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Thu, Feb 28, 2008 at 12:05 PM, Gerry Lawrence <gwlperl@xxxxx.xxx> wrote: > Peter da Silva wrote: > > On 2008-02-28, at 12:56, Gerry Lawrence wrote: > >> KDE3. SO MANY THINGS ARE BROKEN! > > How come? I thought KDE was supposed to be the less-broken one. > > > yeah -- kde2 apps are generally great -- k3b - awesome. ksnapshot for > screenshots -- love it! > > But it seems like the new kde3 versions are slipping into hate. > > > I think you've got an off-by-one error. And everyone knows this dot-oh release was a way to force things to progress rather than let kde4 stew in its own unreleased juices forever. If you really did just upgrade to kde3, wow. ------=_Part_2762_26432524.1204236222867 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Thu, Feb 28, 2008 at 12:05 PM, Gerry Lawrence <<a href="mailto:gwlperl@xxxxx.xxx">gwlperl@xxxxx.xxx</a>> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div><div></div><div class="Wj3C7c">Peter da Silva wrote:<br> > On 2008-02-28, at 12:56, Gerry Lawrence wrote:<br> >> KDE3. SO MANY THINGS ARE BROKEN!<br> > How come? I thought KDE was supposed to be the less-broken one.<br> <br> <br> </div></div>yeah -- kde2 apps are generally great -- k3b - awesome. ksnapshot for<br> screenshots -- love it!<br> <br> But it seems like the new kde3 versions are slipping into hate.<br> <br> <br> </blockquote></div><br>I think you've got an off-by-one error. And everyone knows this dot-oh release was a way to force things to progress rather than let kde4 stew in its own unreleased juices forever.<br><br>If you really did just upgrade to kde3, wow.<br> ------=_Part_2762_26432524.1204236222867--
From: Gerry Lawrence Date: 22:14 on 28 Feb 2008 Subject: Re: qmail hate -- or love ( and the new KDE hate) > I think you've got an off-by-one error. And everyone knows this > dot-oh release was a way to force things to progress rather than let > kde4 stew in its own unreleased juices forever. > > If you really did just upgrade to kde3, wow. yes, meant kde4 -- I was seething with so much hate for it that lapsed into this lists hate for version numbers I... ;-) Continuing on with my "No post without hate" policy -- emacs 22 -- WHAT THE F??? Space doesn't complete any more? I have to turn OFF the stupid splash screen? GRRRRRRRRRRR HATING.
From: Nicholas Clark Date: 22:19 on 28 Feb 2008 Subject: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Thu, Feb 28, 2008 at 05:14:01PM -0500, Gerry Lawrence wrote: > Continuing on with my "No post without hate" policy -- emacs 22 -- WHAT > THE F??? > > Space doesn't complete any more? I have to turn OFF the stupid splash > screen? GRRRRRRRRRRR > HATING. I felt that emacs jumped the shark when it decided that "Set face" deserved a simpler shortcut key than "goto line". One of these is important in a text editor. One isn't. Ooh. That was over 2 years ago: http://nick.hates-software.com/2005/11/18/4ebabb3e.html Nicholas Clark
From: Aristotle Pagaltzis Date: 22:02 on 01 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) * Nicholas Clark <nick@xxxx.xxx> [2008-02-28 23:30]: > I felt that emacs jumped the shark I thought emacs was *born* on the other side of the shark. Regards,
From: Peter da Silva Date: 23:21 on 01 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 2008-03-01, at 16:02, Aristotle Pagaltzis wrote: > * Nicholas Clark <nick@xxxx.xxx> [2008-02-28 23:30]: >> I felt that emacs jumped the shark > > I thought emacs was *born* on the other side of the shark. Emacs is older than the shark. The shark was born on the wrong side of Emacs.
From: demerphq Date: 11:50 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 02/03/2008, Peter da Silva <peter@xxxxxxx.xxx> wrote: > On 2008-03-01, at 16:02, Aristotle Pagaltzis wrote: > > * Nicholas Clark <nick@xxxx.xxx> [2008-02-28 23:30]: > >> I felt that emacs jumped the shark > > > > > I thought emacs was *born* on the other side of the shark. > > > Emacs is older than the shark. The shark was born on the wrong side > of Emacs. *Nix editors are laughingly hateful across the board. Lots of features, too bad about the goddamned USER INTERFACE.
From: Joshua Juran Date: 12:33 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Mar 2, 2008, at 3:50 AM, demerphq wrote: > On 02/03/2008, Peter da Silva <peter@xxxxxxx.xxx> wrote: >> On 2008-03-01, at 16:02, Aristotle Pagaltzis wrote: >>> * Nicholas Clark <nick@xxxx.xxx> [2008-02-28 23:30]: >>>> I felt that emacs jumped the shark >>> >> >>> I thought emacs was *born* on the other side of the shark. >> >> >> Emacs is older than the shark. The shark was born on the wrong side >> of Emacs. > > *Nix editors are laughingly hateful across the board. Lots of > features, too bad about the goddamned USER INTERFACE. How else do you make it work on a vt220? It's not like everybody has a fancy terminal with a mouse, bitmapped display, and network stack. Josh
From: demerphq Date: 12:59 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 02/03/2008, Joshua Juran <jjuran@xxxxx.xxx> wrote: > On Mar 2, 2008, at 3:50 AM, demerphq wrote: > > > On 02/03/2008, Peter da Silva <peter@xxxxxxx.xxx> wrote: > >> On 2008-03-01, at 16:02, Aristotle Pagaltzis wrote: > >>> * Nicholas Clark <nick@xxxx.xxx> [2008-02-28 23:30]: > >>>> I felt that emacs jumped the shark > >>> > >> > >>> I thought emacs was *born* on the other side of the shark. > >> > >> > >> Emacs is older than the shark. The shark was born on the wrong side > >> of Emacs. > > > > *Nix editors are laughingly hateful across the board. Lots of > > features, too bad about the goddamned USER INTERFACE. > > > How else do you make it work on a vt220? It's not like everybody has > a fancy terminal with a mouse, bitmapped display, and network stack. You mean that its impossible to figure out a way to make it not suck when you do?
From: Joshua Juran Date: 13:46 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Mar 2, 2008, at 4:59 AM, demerphq wrote: > On 02/03/2008, Joshua Juran <jjuran@xxxxx.xxx> wrote: >> On Mar 2, 2008, at 3:50 AM, demerphq wrote: >> >>> *Nix editors are laughingly hateful across the board. Lots of >>> features, too bad about the goddamned USER INTERFACE. >> >> >> How else do you make it work on a vt220? It's not like everybody has >> a fancy terminal with a mouse, bitmapped display, and network stack. > > You mean that its impossible to figure out a way to make it not suck > when you do? That was sarcasm -- *everyone* has access to said 'smart' terminals these days. (Anyone still using a dumb terminal must be assumed to prefer it.) But somehow, it's mysteriously difficult to make remote file editing not suck. Or maybe nobody cares enough. Obligatory hate: I tried setting up a local HTTP server with a CGI script that invokes my preferred editor, forwarded the port over ssh, and then using an 'editor' that sends the file over HTTP and receives it back. It basically works, except that ssh port forwarding sucks because it wants me to pick the port number. I don't give a rat's ass which port it uses, but it thoughtfully makes me pick one so that it's my fault when it conflicts with either one chosen by hypothetical other users on the remote system, or very real other local systems I'm connecting from. If ssh could just pick a free port and then set an environment variable saying which one it was, that would be very cool, except of course for allowing ANYONE ON THE REMOTE SYSTEM to connect to my forwarded port. WTF!? How about forwarding to a unix- domain socket? Am I really the first person to think of this? Yes, it's my first rant and I'm hating on ssh. Because the most poignant hate stems from the software I love the most. Josh
From: Andy Armstrong Date: 14:29 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 2 Mar 2008, at 13:46, Joshua Juran wrote: > I tried setting up a local HTTP server with a CGI script that > invokes my preferred editor, forwarded the port over ssh, and then > using an 'editor' that sends the file over HTTP and receives it > back. It basically works, except that ssh port forwarding sucks > because it wants me to pick the port number. I don't give a rat's > ass which port it uses, but it thoughtfully makes me pick one so > that it's my fault when it conflicts with either one chosen by > hypothetical other users on the remote system, or very real other > local systems I'm connecting from. If ssh could just pick a free > port and then set an environment variable saying which one it was, > that would be very cool, except of course for allowing ANYONE ON THE > REMOTE SYSTEM to connect to my forwarded port. WTF!? How about > forwarding to a unix-domain socket? Am I really the first person to > think of this? Doesn't a little script that does rsync, edit, rsync cut it?
From: Joshua Juran Date: 15:05 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Mar 2, 2008, at 6:29 AM, Andy Armstrong wrote: > On 2 Mar 2008, at 13:46, Joshua Juran wrote: >> I tried setting up a local HTTP server with a CGI script that >> invokes my preferred editor, forwarded the port over ssh, and then >> using an 'editor' that sends the file over HTTP and receives it >> back. It basically works, except that ssh port forwarding sucks >> because it wants me to pick the port number. I don't give a rat's >> ass which port it uses, but it thoughtfully makes me pick one so >> that it's my fault when it conflicts with either one chosen by >> hypothetical other users on the remote system, or very real other >> local systems I'm connecting from. If ssh could just pick a free >> port and then set an environment variable saying which one it was, >> that would be very cool, except of course for allowing ANYONE ON >> THE REMOTE SYSTEM to connect to my forwarded port. WTF!? How >> about forwarding to a unix-domain socket? Am I really the first >> person to think of this? > > > Doesn't a little script that does rsync, edit, rsync cut it? Assuming you mean from the local system: No. That fails to reuse not only the context implicit in the remote shell (foreign host, current directory), but also the perfectly good already-established ssh tunnel. If I don't have authorized keys set up I'll even have to enter my password again. Josh
From: Aristotle Pagaltzis Date: 17:23 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) * Joshua Juran <jjuran@xxxxx.xxx> [2008-03-02 16:20]: > On Mar 2, 2008, at 6:29 AM, Andy Armstrong wrote: >> Doesn't a little script that does rsync, edit, rsync cut it? > > Assuming you mean from the local system: No. That fails to > reuse not only the context implicit in the remote shell > (foreign host, current directory), but also the perfectly good > already-established ssh tunnel. Given a sufficiently recent OpenSSH, you can at least get it to reuse the existing SSH connection. Doesn't fix the other problems of course. Regards,
From: Daniel Pittman
Date: 23:57 on 02 Mar 2008
Subject: Re: emacs hate
Aristotle Pagaltzis <pagaltzis@xxx.xx> writes:
> * Joshua Juran <jjuran@xxxxx.xxx> [2008-03-02 16:20]:
>> On Mar 2, 2008, at 6:29 AM, Andy Armstrong wrote:
>>> Doesn't a little script that does rsync, edit, rsync cut it?
>>
>> Assuming you mean from the local system: No. That fails to
>> reuse not only the context implicit in the remote shell
>> (foreign host, current directory), but also the perfectly good
>> already-established ssh tunnel.
>
> Given a sufficiently recent OpenSSH, you can at least get it to
> reuse the existing SSH connection. Doesn't fix the other problems
> of course.
...or so you might think. Except that OpenSSH will happily kill all
your other sessions if you terminate the master session, or fail to act
on port forwarding requests on slave sessions, or randomly not forward
X, or...
Then again, it might not. These "features" come and go. So, in a few
years when the process works sensibly and all it might be valuable.
Until then you can look forward to stabbing the developers who released
such a half-arsed implementation of the idea to the world.
Daniel
From: Phil Pennock Date: 22:19 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 2008-03-02 at 05:46 -0800, Joshua Juran wrote: > If ssh could just pick a free port and > then set an environment variable saying which one it was, that would be > very cool, except of course for allowing ANYONE ON THE REMOTE SYSTEM to > connect to my forwarded port. WTF!? How about forwarding to a unix-domain > socket? Am I really the first person to think of this? Hack-around: after picking the port, tell the host's packet filter to only allow outbound connections to 127.0.0.1 on that port by processes running as whichever users you want to have access. It's foul and fragile and dependent upon you having "127.0.0.1 connections only from local-host" strong end-system enforcement filters; otherwise an attacker could use a host without 127.0.0.1 configured and a manual arp-table entry pointing at your host and connect, say, 192.0.2.42<->127.0.0.1; Unix-domain socket permissions would be much cleaner. pf will let you restrict which user can make outbound connections to a particular ip/port (or can listen on a port for connections) and I believe that iptables has that support nowadays too. -Phil
From: Joshua Juran Date: 02:32 on 03 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Mar 2, 2008, at 2:19 PM, Phil Pennock wrote: > On 2008-03-02 at 05:46 -0800, Joshua Juran wrote: >> If ssh could just pick a free >> port and >> then set an environment variable saying which one it was, that >> would be >> very cool, except of course for allowing ANYONE ON THE REMOTE >> SYSTEM to >> connect to my forwarded port. WTF!? How about forwarding to a >> unix-domain >> socket? Am I really the first person to think of this? > > Hack-around: after picking the port, tell the host's packet filter to > only allow outbound connections to 127.0.0.1 on that port by processes > running as whichever users you want to have access. It's foul and > fragile and dependent upon you having "127.0.0.1 connections only from > local-host" strong end-system enforcement filters; otherwise an > attacker > could use a host without 127.0.0.1 configured and a manual arp-table > entry pointing at your host and connect, say, 192.0.2.42<->127.0.0.1; > Unix-domain socket permissions would be much cleaner. In other words, upon implementing this I'd immediately hate it. Thanks for validating the use of Unix-domain sockets, though. > pf will let you restrict which user can make outbound connections to a > particular ip/port (or can listen on a port for connections) and I > believe that iptables has that support nowadays too. You mean I have to be root? I guess I'll just manually assign port numbers. Or use nano. Josh
From: Michael G Schwern Date: 00:29 on 03 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) Joshua Juran wrote: > But somehow, it's mysteriously difficult to make remote > file editing not suck. Or maybe nobody cares enough. FWIW, emacs TRAMP has a slight positive pressure. I use it in rsync mode. http://www.emacswiki.org/cgi-bin/wiki/TrampMode That said, where's my user-space remote filesystem that can detect remote changes? It's 2008 people!
From: Daniel Pittman
Date: 01:57 on 03 Mar 2008
Subject: Re: emacs hate
Michael G Schwern <schwern@xxxxx.xxx> writes:
> Joshua Juran wrote:
>
>> But somehow, it's mysteriously difficult to make remote file editing not
>> suck. Or maybe nobody cares enough.
>
> FWIW, emacs TRAMP has a slight positive pressure.
It did. The bastards have been "improving" it since I last worked on
it, so you can't use multi-hop editing in a sane way any longer, and so
forth.
At least I don't have to use it much these days, which is pleasant.
> That said, where's my user-space remote filesystem that can detect
> remote changes? It's 2008 people!
Lost behind the wall of inefficiency, somewhere out in space...
Daniel
From: Aaron J. Grier Date: 22:11 on 03 Mar 2008 Subject: remote editing (was Re: emacs hate) On Sun, Mar 02, 2008 at 05:46:48AM -0800, Joshua Juran wrote: > *everyone* has access to said 'smart' terminals these days. (Anyone > still using a dumb terminal must be assumed to prefer it.) But > somehow, it's mysteriously difficult to make remote file editing not > suck. Or maybe nobody cares enough. how do you want to disassemble the problem? remotely accessing a running editor has existing hateful partial-solutions (X, screen, remote desktop). a local editor accessing a remote file has existing partial-hateful solutions (SMB, NFS, FTP, HTTP). allowing the "remote" bits of file editing to be put into an editor would result in a veritable plethora of hateful one-per-editor "remote file access" implementations, which I'm sure emacs and vim have already implemented.
From: Joshua Juran Date: 04:01 on 04 Mar 2008 Subject: Re: remote editing (was Re: emacs hate) On Mar 3, 2008, at 2:11 PM, Aaron J. Grier wrote: > On Sun, Mar 02, 2008 at 05:46:48AM -0800, Joshua Juran wrote: >> *everyone* has access to said 'smart' terminals these days. (Anyone >> still using a dumb terminal must be assumed to prefer it.) But >> somehow, it's mysteriously difficult to make remote file editing not >> suck. Or maybe nobody cares enough. > > how do you want to disassemble the problem? remotely accessing a > running editor has existing hateful partial-solutions (X, screen, > remote > desktop). a local editor accessing a remote file has existing > partial-hateful solutions (SMB, NFS, FTP, HTTP). allowing the > "remote" > bits of file editing to be put into an editor would result in a > veritable plethora of hateful one-per-editor "remote file access" > implementations, which I'm sure emacs and vim have already > implemented. Since the originating hate is the need to run a tty-based editor (all of which are hateful, as well as the limitation in the first place), my approach is to replace the remote-shell-invoked editor with another program that spawns an editing session locally. The problem with X11 is that you have to have the X11 client installed on the remote host (and that it limits you to X11 clients, which are all hateful to me). My suggestion is to have a single wrapper program implement editor semantics which talks to remote- editing client plumbing installed on the remote host, which talks to remote-editing server plumbing on the local host, which invokes my editor of choice, which now is not limited to X11 clients. It's easily implemented over HTTP, and works over ssh if you manually forward the port and trust all remote users. Partial source code for your hating pleasure: <http:// lamp.cvs.sourceforge.net/lamp/jTools/local-edit-client/> Josh
From: Michael G Schwern
Date: 16:23 on 05 Mar 2008
Subject: Re: remote editing (was Re: emacs hate)
Aaron J. Grier wrote:
> On Sun, Mar 02, 2008 at 05:46:48AM -0800, Joshua Juran wrote:
>> *everyone* has access to said 'smart' terminals these days. (Anyone
>> still using a dumb terminal must be assumed to prefer it.) But
>> somehow, it's mysteriously difficult to make remote file editing not
>> suck. Or maybe nobody cares enough.
>=20
> how do you want to disassemble the problem? remotely accessing a
> running editor has existing hateful partial-solutions (X, screen, remot=
e
> desktop). a local editor accessing a remote file has existing
> partial-hateful solutions (SMB, NFS, FTP, HTTP). allowing the "remote"
> bits of file editing to be put into an editor would result in a
> veritable plethora of hateful one-per-editor "remote file access"
> implementations, which I'm sure emacs and vim have already implemented.
That's exactly what I hate about collaborative editors, like SubEthaEdit,=
=20
which is just the next level up from remote editing.
There is Sobby, a daemon to coordinate collaborative editing. Since the=20
protocol is public, any editor can (potentially) talk to Sobby and thus e=
dit=20
documents. Unfortunately, only Gobby speaks the protocol and Gobby is ga=
rbage.
--=20
7. Not allowed to add =E2=80=9CIn accordance with the prophesy=E2=80=9D t=
o the end of
answers I give to a question an officer asks me.
-- The 213 Things Skippy Is No Longer Allowed To Do In The U.S. Army
http://skippyslist.com/list/
From: orc (david parsons)
Date: 17:13 on 02 Mar 2008
Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate))
In article <5A834E35-B787-46D0-AA59-BD5F8C7DEECD@xxxxx.xxx>,
Joshua Juran <jjuran@xxxxx.xxx> wrote:
>On Mar 2, 2008, at 3:50 AM, demerphq wrote:
>> *Nix editors are laughingly hateful across the board. Lots of
>> features, too bad about the goddamned USER INTERFACE.
>How else do you make it work on a vt220? It's not like everybody has
>a fancy terminal with a mouse, bitmapped display, and network stack.
I'd be happier if more developers were stuck with vt100s,
because there's too damn much software out there that just
won't do it without having a gui to blat things out on.
Nothing is as bad as the hatefully gui-centric view of plan9,
but having bad hands makes it *really* fun trying to deal
with modern PC software.
I [heart] the user interface of my *nix editor, because it
doesn't require that I dart my fingers all over the keyboard
(or fight with the twiddle-stick on my keyboard) to do even
the most rudimentary editing chore.
____
david parsons \bi/ Die, Crystal Reports, die!
\/
From: Aristotle Pagaltzis Date: 15:01 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) * demerphq <demerphq@xxxxx.xxx> [2008-03-02 13:00]: > *Nix editors are laughingly hateful across the board. Lots of > features, too bad about the goddamned USER INTERFACE. Win32 editors are *so* much better. Regards,
From: demerphq Date: 15:17 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 02/03/2008, Aristotle Pagaltzis <pagaltzis@xxx.xx> wrote: > * demerphq <demerphq@xxxxx.xxx> [2008-03-02 13:00]: > > > *Nix editors are laughingly hateful across the board. Lots of > > features, too bad about the goddamned USER INTERFACE. > > > Win32 editors are *so* much better. > Win32 editors are laughingly hateful across the board. Nice user interface, too bad about the goddamned FEATURES. I'm not prejudiced, I find them both hateful. Yves
From: Peter da Silva Date: 16:16 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 2008-03-02, at 05:50, demerphq wrote: > *Nix editors are laughingly hateful across the board. Lots of > features, too bad about the goddamned USER INTERFACE. What's an example of an editor that isn't hateful? Because I've never run across an editor on any other system that didn't suck by comparison with "vi", and most are more hateful than "ed".
From: demerphq Date: 16:39 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On 02/03/2008, Peter da Silva <peter@xxxxxxx.xxx> wrote: > On 2008-03-02, at 05:50, demerphq wrote: > > *Nix editors are laughingly hateful across the board. Lots of > > features, too bad about the goddamned USER INTERFACE. > > > What's an example of an editor that isn't hateful? To do that I would have to find one that wasn't implemented as software. > Because I've never run across an editor on any other system that > didn't suck by comparison with "vi", and most are more hateful than > "ed". As far as user interfaces go 'vi' is *really* high up on my hate list, so we probably aren't going to see eye to eye on this one. :-) yves
From: Abigail Date: 17:23 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) --9jxsPFA5p3P2qPhR Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Mar 02, 2008 at 05:39:33PM +0100, demerphq wrote: > On 02/03/2008, Peter da Silva <peter@xxxxxxx.xxx> wrote: > > On 2008-03-02, at 05:50, demerphq wrote: > > > *Nix editors are laughingly hateful across the board. Lots of > > > features, too bad about the goddamned USER INTERFACE. > > > > > > What's an example of an editor that isn't hateful? >=20 > To do that I would have to find one that wasn't implemented as software. Well, there's always the "twiddling with a magnet" way of editing. > > Because I've never run across an editor on any other system that > > didn't suck by comparison with "vi", and most are more hateful than > > "ed". >=20 > As far as user interfaces go 'vi' is *really* high up on my hate list, > so we probably aren't going to see eye to eye on this one. :-) My biggest hate regarding 'vi' are those wannabee-Un^WLinux distros that launch some editor when you type 'vi' on the command line, but which isn't vi at all. Abigail --9jxsPFA5p3P2qPhR Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFHyuKJBOh7Ggo6rasRAuMoAJ9QUKsH1p1YnNqezigZClw62DSWOgCffhZG zG+gvYAnbqgcOAZ+QmbetBU= =FOV1 -----END PGP SIGNATURE----- --9jxsPFA5p3P2qPhR--
From: Ann Barcomb Date: 17:30 on 02 Mar 2008 Subject: vi love (was Re: emacs hate) On Sun, 2 Mar 2008, Abigail wrote: > My biggest hate regarding 'vi' are those wannabee-Un^WLinux distros > that launch some editor when you type 'vi' on the command line, but > which isn't vi at all. No kidding. I have to keep a vimrc file handy to get it to behave like real vi. I was starting to think I was the only person in the world who would rather have the original. - Ann
From: Peter da Silva Date: 19:09 on 02 Mar 2008 Subject: Re: vi love (was Re: emacs hate) On 2008-03-02, at 11:30, Ann Barcomb wrote: > On Sun, 2 Mar 2008, Abigail wrote: >> My biggest hate regarding 'vi' are those wannabee-Un^WLinux distros >> that launch some editor when you type 'vi' on the command line, but >> which isn't vi at all. > > No kidding. I have to keep a vimrc file handy to get it to behave > like > real vi. I was starting to think I was the only person in the world > who would rather have the original. No, you're not. I keep an nvi.tar.gz around for the same reason.
From: Aristotle Pagaltzis Date: 19:17 on 02 Mar 2008 Subject: Re: vi love (was Re: emacs hate) * Ann Barcomb <ann@xxxxxxxxx.xxx> [2008-03-02 18:35]: > I was starting to think I was the only person in the world who > would rather have the original. http://ex-vi.sourceforge.net/ at least indicates you're not. Regards,
From: Abigail Date: 21:16 on 02 Mar 2008 Subject: Re: vi love (was Re: emacs hate) --SkvwRMAIpAhPCcCJ Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Mar 02, 2008 at 06:30:01PM +0100, Ann Barcomb wrote: > On Sun, 2 Mar 2008, Abigail wrote: >=20 > >My biggest hate regarding 'vi' are those wannabee-Un^WLinux distros > >that launch some editor when you type 'vi' on the command line, but > >which isn't vi at all. >=20 > No kidding. I have to keep a vimrc file handy to get it to behave like > real vi. I was starting to think I was the only person in the world > who would rather have the original. Well, it's not so much I prefer vi over something else (in fact, my editor of choice is a vi-clone (but not vim)). It's the fact that if I type 'vi' at the prompt, I want 'vi'. Not something that's proud of not being vi. If you call a bird a duck, you better make sure it walks and talks like one. Abigail --SkvwRMAIpAhPCcCJ Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFHyxkdBOh7Ggo6rasRAr04AJ9FhvbhmIpUVj6U8KtTuP/t1qF/TgCdGaCK qSmMPvv5wYn08LbtOe31V10= =EBWh -----END PGP SIGNATURE----- --SkvwRMAIpAhPCcCJ--
From: Andy Dougherty Date: 02:16 on 03 Mar 2008 Subject: Re: vi love (was Re: emacs hate) On Sun, 2 Mar 2008, Abigail wrote: > Well, it's not so much I prefer vi over something else (in fact, my editor > of choice is a vi-clone (but not vim)). It's the fact that if I type 'vi' > at the prompt, I want 'vi'. Not something that's proud of not being vi. I've always thought that's what "vim" stood for, "vi-Imposter".
From: Sean Conner Date: 19:05 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) It was thus said that the Great Peter da Silva once stated: > On 2008-03-02, at 05:50, demerphq wrote: > >*Nix editors are laughingly hateful across the board. Lots of > >features, too bad about the goddamned USER INTERFACE. > > What's an example of an editor that isn't hateful? For me, it's IBM's PE v1.0. Don't laugh. Outside of these three limitations: 1. Since it was written for MS-DOS 1.x, it doesn't understand directories. 2. It will truncate any line longer than 255 characters in length. 3. Lines *have* to end with CRLF (in that order) or you will end up with a single line that's 255 characters in length (see 2). I have yet to find a bug (incredible for a 1.0 release). It's small (40k executable), programmable (in that, for all the actions it could support, you can easily bind to any key combination), can edit multiple files at once, and can handle files of any size (as long as that size fits into the remaining RAM of the PC). Heck, I'm still tempted to set up a virtual DOS environment under Linux *just* to run that editor. -spc (Hey, wait! 40k small?! When did that happen?!)
From: Yoz Grahame Date: 21:51 on 02 Mar 2008 Subject: Re: emacs hate (was Re: qmail hate -- or love ( and the new KDE hate)) On Sun, Mar 2, 2008 at 8:16 AM, Peter da Silva <peter@xxxxxxx.xxx> wrote: > On 2008-03-02, at 05:50, demerphq wrote: > > *Nix editors are laughingly hateful across the board. Lots of > > features, too bad about the goddamned USER INTERFACE. > > What's an example of an editor that isn't hateful? I was going to post a hesitant, somewhat embarrassed defense of ActiveState's Komodo (which, despite being an IDE and hence a bigger hate magnet than an NRA convention, mostly makes me happy by staying out of my way and giving me enough useful features that, when I go looking, I'm more likely to find something new and pleasantly surprising rather than a hole where a desired feature should be, but that said it's still a very long way from stable and based on top of XULRunner of all things, so I'm just going to leave that where it is and quietly disown this paragraph while creeping away) but then I remembered which list I'm on, so: Eclipse. Whenever I think of Eclipse, I think of a friend's First Rule Of Text Editors, which goes something like: A decent editor is one which lets you just drag a file in and start editing it. Eclipse gets this so devastatingly wrong that I think it must be on purpose. Every time I try it, which is a good couple of years after the previous time and my accelerating senility has masked any scars, by the time I go from installation (which deserves several threads' worth of hate itself) to Actually Editing A God-Damned File Like What I Wanted to Do In The First Place, I am propelled more by teeth-gritted determination not to let this bastard thing defeat me than whatever it was that motivated me in the first place. But eventually Eclipse's "No STOP EDITING and look at all my bloody FEATURES you PHILISTINE" will win and I'll get the magnet out. -- Yoz
From: Tony Finch Date: 19:09 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, 28 Feb 2008, Gerry Lawrence wrote: > > The issue is the errno declarations that changed in some sorta gnu libc thing. You mean "in every unix that supports multi-threading". > There are a couple of bugs in qmail that you should patch for: But surely qmail is perfect! Tony.
From: Peter da Silva Date: 19:21 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-28, at 13:09, Tony Finch wrote: > You mean "in every unix that supports multi-threading". You mean "in every UNIX that doesn't support non-threaded apps". Threads are not the only, or even the best, model for dealing with a concurrent environment. > But surely qmail is perfect! No, just psychotic, paranoid, and obsessive-compulsive. It's a *dry* hate.
From: Tony Finch Date: 15:10 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, 28 Feb 2008, Peter da Silva wrote: > > You mean "in every UNIX that doesn't support non-threaded apps". A thread-compatible errno doesn't imply a non-thread-non-compatible errno. Of course, the fact that errno is a global variable is hateful, and the fact that they had to implement a disgusting hack to make it work in threaded code is just one indication of that. Tony.
From: Peter da Silva Date: 16:15 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible Compounding the hatefulness of "extern int errno" by breaking existing non-threaded code is doubly hateful.
From: Gerry Lawrence Date: 20:10 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible > But surely qmail is perfect! > NOTHING, is perfect, qmail is not perfect, and no one ever said it was. - it's just really, really good. And far better than any of the alternatives. Oh wait: I have to post some hate.... Microsoft hate is like shooting fish in a barrel, but here goes. HATE: microsoft "apps" that can not be moved or iconified, and you can't get to the desktop underneath it. Infuriating. An example of this is the windows updater. Oh, and the windows updater that gets 60% of the way through the updates then wants to update explorer and STOPS because you need to answer some silly questions. (Only happens if you go directly to the update web page and start from there). (I don't use windows but frequently get asked to work on friends computers with windows. I don't know why they ask the unix guy, but what the heck.)
From: David Cantrell Date: 11:11 on 01 Mar 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Thu, Feb 28, 2008 at 06:00:40PM +0000, Nicholas Clark wrote: > On Thu, Feb 28, 2008 at 05:54:30PM +0000, Tony Finch wrote: > > No, I mean DJB's guaranteed version does not compile without patches. It > > has incorrect declarations for various libc internals such as errno. > Could it ever have compiled on any out of the box Unix variant ever? > Did something he happened to once use have just the right level of sloppiness > in libc to allow it to work? He actually uses libc? I thought he didn't trust it.
From: Jarkko Hietaniemi Date: 13:09 on 01 Mar 2008 Subject: Re: So much hate, spread across so much software, it's incredible David Cantrell wrote: > On Thu, Feb 28, 2008 at 06:00:40PM +0000, Nicholas Clark wrote: >> On Thu, Feb 28, 2008 at 05:54:30PM +0000, Tony Finch wrote: >>> No, I mean DJB's guaranteed version does not compile without patches. It >>> has incorrect declarations for various libc internals such as errno. >> Could it ever have compiled on any out of the box Unix variant ever? >> Did something he happened to once use have just the right level of sloppiness >> in libc to allow it to work? > > He actually uses libc? I thought he didn't trust it. No, you are confusing him with Ulrich Drepper. >
From: Gerry Lawrence Date: 16:54 on 28 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible >> It's default behavior does (as is spec'ed by that hateful RFC) >> > > Where does it say that? > > I'm not sure where I read that and now can't find it, so I was wrong. > If you run an unofficial fork then all bets are off, including DJB's > security guarantee. Oh I see you use exim..... Hey hate qmail all you like, a lot of it's deserved. And since this is software hate, not love, I'll hate on software I love -- qmail. I discovered just the other day that running multiple qmail servers on a machine with multiple IP addresses - qmail wants to treat them all as local, even though it's been told to specifically run on one and only one IP address. Fortunately there's a facility to easily correct this: smtproutes but I'll be darned if anyone has documented this behavior.
From: Sean Conner Date: 20:24 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible It was thus said that the Great Sean Conner once stated: > > Ezlml [my mistake---it's supposed to be ezmlm] (what Apache > apparently uses for their mailing lists) for using the email address > in the Return-Path: header and *NOT* the address that appears in the > From: header? So, can anyone here explain the rather hateful practice of ezmlm of ignoring the From: header and using the Return-Path: header? I've read Dan Bernstein's justification for this [1], but while the individual words are in English, together the words don't make sense. -spc (And, as usual, the Apache crowd is ignoring my issue that started this whole mess ... ) [1] http://www.imc.org/ietf-822/mail-archive/msg00149.html
From: Gerry Lawrence Date: 20:57 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible Sean Conner wrote: > So, can anyone here explain the rather hateful practice of ezmlm > ofignoring the From: header and using the Return-Path: header? I've > read DanBernstein's justification for this [1], but while the > individual words arein English, together the words don't make sense. Huh? Um, I don't think ezmlm messes with reply-to by default. YOU can mess with reply-to (see below) and many list owners do so -- http://www.ezmlm.org/faq/FAQ-9.html 9.8 Setting ``Reply-To: list@host''. http://www.ezmlm.org/faq/FAQ-9.html This is not recommended, since it leads to dissemination via the list of messages returned from bad auto-responders and MTAs. Also, it may lead to public replies to the list where personal replies were intended. In addition, the original ``Reply-To:'' header is lost. If you do want to add a reply-to list header, put ``reply-to'' into *DIR/headerremove*, and ``Reply-To: |list@xxxx.xxx|'' into *DIR/headeradd*.
From: Sean Conner Date: 21:13 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible It was thus said that the Great Gerry Lawrence once stated: > Sean Conner wrote: > >So, can anyone here explain the rather hateful practice of ezmlm > >ofignoring the From: header and using the Return-Path: header? I've > >read DanBernstein's justification for this [1], but while the > >individual words arein English, together the words don't make sense. > > Huh? Um, I don't think ezmlm messes with reply-to by default. YOU can > mess with reply-to (see below) and many list owners do so -- It's not munging with Reply-To:. I subscribed to the list. The From: header contained "spc at conman.org". The Return-Path: contained "spc at brevard.conman.org". Guess which address ezmlm used to subscribe me to? -spc (*THAT'S* what I'm talking about!)
From: Denny Date: 21:21 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Fri, 2008-02-29 at 15:57 -0500, Gerry Lawrence wrote: > YOU can mess with reply-to (see below) and many list owners do so [...] > This is not recommended, since [...] it may lead > to public replies to the list where personal replies were intended. And which happens more often - replying to the list about a list message, or replying to one person about a list message? Anyway, this tangential hate has been done to death all over the 'net many many times... google for 'reply-to munging considered harmful' and 'reply-to munging considered useful'. Next time, start an emacs/vi argument - they're funnier. :) *wastes precious seconds fixing incorrect address in To: field, sends*
From: Phil Pennock Date: 22:23 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-29 at 15:24 -0500, Sean Conner wrote: > So, can anyone here explain the rather hateful practice of ezmlm of > ignoring the From: header and using the Return-Path: header? SMTP Envelope Sender (Return-Path:) is guaranteed singleton, whilst the From: header is permitted to contain multiple addresses? Not that anyone sane ever does that. It normally just doesn't matter, and if you're sending out one fixed up but not the other then you probably have a bug your end (even though it's harder to diagnose because of the choice they made)? It's easier to implement closed mailing-lists with MTA support if you can determine membership based upon the SMTP Envelope Sender? You might be able, if original sender isn't logged in any headers, to have distinct addresses with the From: being the public face and the SMTP Envelope Sender being your subscription address and that way keep the subscription address private and not have random nutters unsubscribe you from mailing-lists with a single faked mail. Now, getting any common client to support that is another matter entirely. Your argument that From: should be used makes some sense, but for most people with sane setups it doesn't make any sense and if you're looking at "supporting everything that should be supported" then not constraining From: headers wins out. Mostly. Really, they're both equally spoofable so they are both flawed. But I've not seen PGP-signature-authenticated-posting lists. So we suck it up and deal with broken software because it mostly kind of works. -Phil
From: Tony Finch Date: 23:08 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Fri, 29 Feb 2008, Phil Pennock wrote: > > Your argument that From: should be used makes some sense, but for most > people with sane setups it doesn't make any sense and if you're looking > at "supporting everything that should be supported" then not > constraining From: headers wins out. Mostly. In practice it's easier for people to control their From: line than their return path, so keying off From: causes fewer problems for mailing list posting restrictions. Tony.
From: Phil Pennock Date: 23:58 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On 2008-02-29 at 23:08 +0000, Tony Finch wrote: > On Fri, 29 Feb 2008, Phil Pennock wrote: > > > > Your argument that From: should be used makes some sense, but for most > > people with sane setups it doesn't make any sense and if you're looking > > at "supporting everything that should be supported" then not > > constraining From: headers wins out. Mostly. > > In practice it's easier for people to control their From: line than their > return path, so keying off From: causes fewer problems for mailing list > posting restrictions. Aargh, slipped and used wrong word, sorry. I meant: "for most people with sane setups it doesn't make any *difference*". *sigh* -Phil
From: Tony Finch Date: 23:06 on 29 Feb 2008 Subject: Re: So much hate, spread across so much software, it's incredible On Fri, 29 Feb 2008, Sean Conner wrote: > > So, can anyone here explain the rather hateful practice of ezmlm of > ignoring the From: header and using the Return-Path: header? DJB thinks parsing is too hard, so his mailing list manager looks at the envelope and ignores the message data because that's easier. He also thinks that his mailing list manager is allowed to use VERP but normal users are not. Tony.
Generated at 10:25 on 16 Apr 2008 by mariachi