it's just an early sign that the copyright industries have realised that the pirates were right all along - copyright infringement isn't theft at all. it's actually torture and/or body mutilation.
in a few years, they'll realise that even that's not accurate - copyright infringement is murder.
his point is that you'd have to be a complete fucking cretin to click on ads, just as you would have had to be a complete fucking cretin to have voted for Bush.
not that voting for Bush's opponents in 2000 or 2004 would have been particularly appealing(*), but voting for the far greater of two evils is just fucking stupid...Bush will go down in history as the idiot puppet who destroyed what little was left of the american experiment.
and as for the current crop of republican candidates - i honestly can't understand how *anyone* sane could even consider voting for ant of them, or why their very presence as serious candidates for the US Presidency isn't regarded as a cause for deep shame and embarrassment by Americans. they all seem like complete lunatics. terrifying lunatics.
but then, i'm not an american and i think it's blindingly obvious that pretty nearly all of your politicians are either batshit insane or are the enthusiastic puppets of megacorps or both.
(*) Gore wasn't too bad but if he had won, he would have found (just like Obama) has that he wouldn't be allowed by the corporate owners of America to actually do anything major. So he wouldn't have been able to do much good, but at least he wouldn't have done anything actively evil....he certainly wouldn't have jump-started the whole war on terror bullshit or the invasion of Iraq (best excuse: "he tried to kill my daddy") and dragged america (and the conga-line of suckholes in the "coalition of the willing") into the state of permanent war and the Orwellian nightmare that we're living in right now.
And he wouldn't have let the banking scumbags and wall street thieves get away with the massive frauds they perpetrated over years that resulted in the economic crisis of 2007/2008.
So yeah, you would have had to be a complete fucking cretin to vote for Bush. Or for any of the current crop who promise only more of the same but worse and with an even larger dose of fucked-up religious insanity and/or libertarian lunacy layered on top.
there's no way that interest charges (even at credit card rates) on 331 pounds (600 - 269) over 24 months comes to anywhere near 294 (1133 - 839).
even 20% interest on 331 pounds adds up to 66 pounds in a year, or 5.50 pounds/month (plus compounded interest) - if you don't make ANY payments at all. significantly less if you have a functioning brain and do actually make payments. paying it off at 30 pounds per month would reduce the principle of the CC debt by 24.5 pounds per month at the start of the loan (and more as the debt reduces).
(do any CCs even charge 20% interest these days? last i checked, here in Australia they're all mostly around 11 to 14%....with anywhere from one to six months interest free on purchases. but lets assume 20% with no interest-free period because it's pretty much a worst-case scenario)
even if you can't afford to pay the 600 to buy the phone outright, you're STILL much better off buying it on credit, and paying it off over, say, 13 months (i.e. an extra 30 per month for the CC payment - 11 months for the 331 plus, say, two months to cover interest).
so, for the first 13 months you'll be paying 40 a month. after that, only 10 pounds per month. that adds up to 899 pounds (269 + 13*40 + 11*10). about 60 pounds worse off than buying the phone with cash, and about 234 pounds better off than buying it for 269 plus a 24 month contract.
even if you paid for the phone entirely by credit card (i.e. the full 600 pounds rather than paying 269 up front and 331 on credit), that just means you'll be paying 40 per month (30 for phone, 10 for usage) for about 24 months. or 960 pounds. i.e. adds only 60 pounds to the total cost (but if cash flow is a problem, you're better off doing this, so you don't have to pay 269 pounds up front).
of course, doing that takes a little bit of planning and self-discipline (i.e. pay off the CC at a rate that significantly reduces the principle rather than the we-want-you-to-be-in-debt-to-us-forever minimum payment recommended by CC companies). and some basic arithmetic. so most people won't do it. and stupid people can and will fuck this up.
also, of course, the interest and payment calculations work exactly the same whether the units are pounds or dollars or euros or anything else.
ps: i don't have a credit card and don't want one. i firmly believe that if you can't afford to pay cash for something then you can't afford it at all and shouldn't buy it (with the sole exception of a house - very few can afford to pay cash for a a house, yet housing is essential and you'd have to pay rent otherwise)...so i pay cash for almost everything, or use a debit card for convenience.
credit card debt sucks, especially for stupid or innumerate people. one of the few things that suck worse than CC debt is phone contracts.
Unfortunately, it's you that's making yourself look like a nutjob because you don't have a fucking clue what you're talking about.
The GPL does not prohibit commercialisation of derivative works. The GPL prohibits making them proprietary. Proprietary and Commercial are *not* the same thing, not even close. Nor are they dependant traits.
I'll repeat my original point since you seem to have missed it "The *ONLY* restriction that the GPL has that the BSD license doesn't is that it prevents you from restricting other people's freedom."
*ALL* of the extra clauses in the GPL vs the various BSD-like licenses are aimed at closing one loophole or another that could be used to restrict the freedom of anyone who receives the code.
That is not a freedom that most FOSS developers care about or want. A minority of FOSS developers do want it - good luck to them.
In short, like proprietary software if you want to use GPL code there's a price to pay. Unlike proprietary software, the price is not money or signing an NDA, the price is that you grant the same freedoms that you received to anyone you distribute the code or a derivative work to. If you don't want to pay that price, then don't distribute GPL-ed code or anything derived from it. Write your own, or find an alternative with a license acceptable to you.
the only freedom that BSD-like licenses grant you that the GPL doesn't is the freedom to restrict other people's freedom.
some FOSS developers think that that is a worthwhile and useful freedom for their project. Most don't.
It's certainly not a useful or desirable freedom from the POV of an end-user running software without the right to view, modify, or redistribute the source code.
Here's a much better method for optimising security on your smartphone or tablet:
DON'T INSTALL WORTHLESS SHIT
Apple's App Store and Google's Marketplace make Sturgeon's Law seem like hopelessly naive pollyanna-ish optimism. They each may have a few hundred thousand apps, but less than one in ten thousand or so are worth even looking at, let alone installing.
this is the 2nd or 3rd similar article i've seen in recent months.
it's starting to smell a lot like a publicity campaign to discredit science in general....wonder if these studies are being funded by oil companies or their shills, or neo-liberal think tanks with a vested interest in discrediting science in general and climate science in particular.
since i run my own mail server with just a handful of users, i only have to care about "power-users" or "techies."
piping messages is essential - software that tells me, on my own server, "you can't be trusted to do that" is not likely to endear itself to me. and if it's disallowed by sieve because of sieve's nature, that means sieve is not suitable for my needs.
procmail *should* be doing anything i want it to. including piping messages through formail for Message-ID dupe detection.
as i said before if one comes along that's better than procmail and sufficiently good that it's worth the hassle of switching, then i'll consider doing so. I haven't seen one yet....and it would have to be extremely good to be worthwhile.
and if i stay with procmail, i won't have to learn another mail-filtering language either:)
BTW, procmail is standard too - the nice thing about standards is that there are so many to chose from.
Last time i looked, sieve couldn't pipe a message (header or body or both) into an external program. and it doesn't even support regexp matching by default.
For me, both are required....and regexp matching is what I would consider to be a base-level core feature required of any mail filtering software.
I just had another quick look at it and noticed that it also seems to be lacking:
* the ability to detect duplicate messages by Message-Id:
* the ability to clone an incoming message so that it can be handled independently by two or more different rules.
* an equivalent to procmail's ^TO_, ^FROM_DAEMON, etc pre-defined regexps.
Sieve might be a nice option for someone with fairly basic mail-filtering needs and without 2+ decades of procmail use, but it doesn't offer a compelling reason for me to switch.
Courier's maildrop comes a lot closer to my needs....actually, if I wanted to, I could probably switch to it without any noticable loss of functionality. It would be a lot of work with no real benefit though.
as for 'sane syntax', I find procmailrc a lot easier to read and modify that what i've seen of sieve, which seems to add a lot of verbosity without sufficient features to justify it. I guess I just don't need yet another C-like language.
Mutt on the other hand, defaults to color, which is rather annoying
it's easy enough to write a monochrome.muttrc
you can probably find one on the net if you search.
and can't do local delivery but depends on the availability of a sendmail compatible MTA/MDA.
mutt shouldn't do local delivery - or do you routinely misconfigure systems so that mailboxes are world-writable? or do you want mutt to be setuid root or setgid mail?
mutt's job is to be an MUA, not an MTA or MDA.
there are numerous light-weight MTAs & MDAs that can be installed to handle local delivery...most of them just an apt-get install or yum install away.
e.g.
dma - lightweight mail transport agent masqmail - mail transport agent for intermittently connected hosts
even postfix or exim aren't too heavy to just install as a matter of course on most, if not all, machines. it's hard to imagine any machine that shouldn't have a local MTA, or wouldn't benefit from one. even a little openwrt router or similar embedded device should have one just so that it can send alert messages.
lots of things already depend on (or assume) the presence of an MTA anyway. cron, for example. so you need something to fill that job, whether it can handle local delivery (most MTAs) or not (nullmailer, and similar relay-only MTAs)
On multi-user systems, it's nice to be able to leave a message to other users, without depending on another mail delivery agent
echo "message" | mail -s "hi" username
an MTA is an (almost) essential service that can be used by many other tools to send mail (both local and remote), without each of them having to reinvent the wheel.
There is far better tool available now than procmail.
name one.
Procmail syntax sucks badly.
it's not perfect, nothing is. but it's adequate.
or is that just your way of saying "it's too hard! waaah!"
There is no good reason to use a poorly featured client like Mutt nowadays.
mutt is "poorly featured". that's amusing...in the same way that a small child saying something fundamentally wrong but earnestly and sincerely believing it is amusing.
mutt is the full-featured email client. with many more features than can be found in *any* GUI mail client.
GUI mail clients typically have only one big feature - that they have a GUI interface.
In exchange, they've sacrificed pretty nearly every other useful feature that a mail client should have, other than the absolute basics of read, write, reply, forward, delete, etc.
Thunderbird makes a halfway decent effort at getting some of those features back in a GUI client (and with the External Editor plugin even lets you use gvim or your favourite editor), but none of the others do.
It's not necessary, but it is useful. and yes, a different problem.
changing folders in mutt is a PITA. You have to close your current folder when you open the new one. To close it, you lose all your current context in that folder - you have to save or abandon any changes (e.g. deletions) you've made, and lose the tag status of any tagged messages. and similar small annoyances. and if they're large folders, they can take several seconds to save-and-close, and then more time to open the new one. even on a fast SSD.
by having multiple mutts running, each open on a different folder, you can avoid that. when you want to change folder, just press ^Z to suspend the current mutt and use the standard sh job control command 'fg' to bring another mutt to the foreground....instantly.
If you write a script to start multiple mutts like this, then you get them in the same order every time...and you very quickly learn, e.g., that 'fg 1' gets you your main inbox, 'fg 2', gets you your sent-mail folder, 'fg 3' gets you to mailing list 1, 'fg 4' gets you to mailing list 2, and so on.
and if i forget, i can type 'jobs' in the shell and get a list.
even better, if i run that mutt startup script inside screen then I can attach to it via ssh from home, work, or from anywhere in the world.
BTW, I have *MANY* more folders than just 20. 3287 in 169 different sub-directories at the moment, according to find:
$ find ~/Mail -type f | wc -l 3287 $ find ~/Mail -type d | wc -l 169
(btw, this is why i don't use the sidebar plugin for mutt. it's unbearably slow. literally unusable with that many folders it has to scan)
The 20 (actually, 24 at the moment) are just the ones I use most frequently, the ones I'm switching between all the time. mutt knows about the others via the usual configuration, like:
I deliberately exclude my archive mbox folders (i.e. those not named *.incoming) here because I can easily find them with mutt's folder browser anyway. the mailboxes line just gives me a quick pick list of all the mbox folders that procmail is filtering incoming mail into.
the point of email is communication. - communication with all readers, not just those who happen to be using the same software as you.
HTML in email is fucked up and stupid for the same reason that MS Word documents in email are fucked up and stupid, or why writing your email in LaTeX would also be fucked up and stupid.
plain text works with all mail clients. HTML doesn't. XML doesn't. LaTeX doesn't.
and when morons of your ilk switch to BLINGML 2015 and you're an old loser stuck on boring old XML or ancient HTML you might begin to understand why dropping backwards compatibility for the sake of following fads is cretinous.
And HTML isn't? It's practically XML.
HTML is NOT XML. it looks superficially similar but it's not even close to XML.
you can pipe html or xml easily enough. you can't can't do anything useful with that, though, unless you write an xml or html parser (or use an existing one). which magnifies a simple Q&D one-liner into a small-scale development project.
So I'll use a <pre> tag, for fuck's sake. Every email reader I've ever seen supports it.
you obviously haven't seen many, then. mutt doesn't. elm doesn't. pine doesn't. none of the non-html MUAs support it.
and if you're goint to argue that they'll just render the html tags as plain text in the body - well, they might. if your sending email client is broken and doesn't use the correct mime-type for the body attachment, they will. and even then <em>html</em> tags <blink>scattered>/blink> throughout the <i>text</i> make the text <blink>unreadable</blink>
3. <80 columns (72-78 is typical) is the standard for email.
Again, according to whom?
RFCs 822, 2822, and 5322 to start with.
you know, those unimportant little documents that define the standard for Internet Message Format.
I *am* the OP, and the reasons I cited above are most of why i think proportional fonts are ghastly for email.
BTW, proportional vs monospace is not what defines whether a font is ugly or legible of pleasant to read. There are ugly and eye-straining fonts in both proportional and non-proportional styles. There are also pleasant, legible, and eminently readable fonts in both.
some kinds of text, however, pretty much require a NP font to be rendered legibly. source code and log files for example. terminal windows for another - and for the same reason.
people still use procmail in 2012 for the same reason they still use screwdrivers or hammers or other basic tools.
for the job that they do, there is nothing better.
maybe someday there'll be something that does mail filtering & sorting as well as procmail. and maybe someday there'll even be something that does a better job than procmail. when that day arrives i'll evaluate whether it provides enough of an improvement to be worth the effort of switching to it.
ps: you're the one who sounds like a 'luddite dick'. demonstrating your technophobia kind of gives it away.
1. plain-text is readable by anything, and is easily grepped, piped, manipulated by standard tools, and otherwise *used*.
2. log file lines, programs, and tabulated data are all unreadable in proportional font.
3. 80 columns (72-78 is typical) is the standard for email.
the *only* valid excuse for including longer lines is when copy-pasting log lines or code fragments or similar. for discussion, reformat paragraphs with par (or fmt if you don't have par installed)
BTW, not all non-proportional fonts are as ugly as Courier. Deja-Vu Sans Mono is quite nice, for example.
if you're used to pine, it will take a day or two to get used to mutt's keybindings. it'll probably piss you off while you're learning them, because there are some subtly annoying differences.
after that, you'll be glad you did and you'll never look back.
mutt's searching and tagging features alone are worth the switch.
Slashdot Mirror
of course it isn't satire.
it's just an early sign that the copyright industries have realised that the pirates were right all along - copyright infringement isn't theft at all. it's actually torture and/or body mutilation.
in a few years, they'll realise that even that's not accurate - copyright infringement is murder.
"rigged badly"? you mean like the vote in Florida in Nov 2000?
later in his talk, he explained how War Is Peace, Freedom is Slavery, and Ignorance is Strength.
this shouldn't need explaining but...
his point is that you'd have to be a complete fucking cretin to click on ads, just as you would have had to be a complete fucking cretin to have voted for Bush.
not that voting for Bush's opponents in 2000 or 2004 would have been particularly appealing(*), but voting for the far greater of two evils is just fucking stupid...Bush will go down in history as the idiot puppet who destroyed what little was left of the american experiment.
and as for the current crop of republican candidates - i honestly can't understand how *anyone* sane could even consider voting for ant of them, or why their very presence as serious candidates for the US Presidency isn't regarded as a cause for deep shame and embarrassment by Americans. they all seem like complete lunatics. terrifying lunatics.
but then, i'm not an american and i think it's blindingly obvious that pretty nearly all of your politicians are either batshit insane or are the enthusiastic puppets of megacorps or both.
(*) Gore wasn't too bad but if he had won, he would have found (just like Obama) has that he wouldn't be allowed by the corporate owners of America to actually do anything major. So he wouldn't have been able to do much good, but at least he wouldn't have done anything actively evil....he certainly wouldn't have jump-started the whole war on terror bullshit or the invasion of Iraq (best excuse: "he tried to kill my daddy") and dragged america (and the conga-line of suckholes in the "coalition of the willing") into the state of permanent war and the Orwellian nightmare that we're living in right now.
And he wouldn't have let the banking scumbags and wall street thieves get away with the massive frauds they perpetrated over years that resulted in the economic crisis of 2007/2008.
So yeah, you would have had to be a complete fucking cretin to vote for Bush. Or for any of the current crop who promise only more of the same but worse and with an even larger dose of fucked-up religious insanity and/or libertarian lunacy layered on top.
there's no way that interest charges (even at credit card rates) on 331 pounds (600 - 269) over 24 months comes to anywhere near 294 (1133 - 839).
even 20% interest on 331 pounds adds up to 66 pounds in a year, or 5.50 pounds/month (plus compounded interest) - if you don't make ANY payments at all. significantly less if you have a functioning brain and do actually make payments. paying it off at 30 pounds per month would reduce the principle of the CC debt by 24.5 pounds per month at the start of the loan (and more as the debt reduces).
(do any CCs even charge 20% interest these days? last i checked, here in Australia they're all mostly around 11 to 14%....with anywhere from one to six months interest free on purchases. but lets assume 20% with no interest-free period because it's pretty much a worst-case scenario)
even if you can't afford to pay the 600 to buy the phone outright, you're STILL much better off buying it on credit, and paying it off over, say, 13 months (i.e. an extra 30 per month for the CC payment - 11 months for the 331 plus, say, two months to cover interest).
so, for the first 13 months you'll be paying 40 a month. after that, only 10 pounds per month. that adds up to 899 pounds (269 + 13*40 + 11*10). about 60 pounds worse off than buying the phone with cash, and about 234 pounds better off than buying it for 269 plus a 24 month contract.
even if you paid for the phone entirely by credit card (i.e. the full 600 pounds rather than paying 269 up front and 331 on credit), that just means you'll be paying 40 per month (30 for phone, 10 for usage) for about 24 months. or 960 pounds. i.e. adds only 60 pounds to the total cost (but if cash flow is a problem, you're better off doing this, so you don't have to pay 269 pounds up front).
of course, doing that takes a little bit of planning and self-discipline (i.e. pay off the CC at a rate that significantly reduces the principle rather than the we-want-you-to-be-in-debt-to-us-forever minimum payment recommended by CC companies). and some basic arithmetic. so most people won't do it. and stupid people can and will fuck this up.
also, of course, the interest and payment calculations work exactly the same whether the units are pounds or dollars or euros or anything else.
ps: i don't have a credit card and don't want one. i firmly believe that if you can't afford to pay cash for something then you can't afford it at all and shouldn't buy it (with the sole exception of a house - very few can afford to pay cash for a a house, yet housing is essential and you'd have to pay rent otherwise)...so i pay cash for almost everything, or use a debit card for convenience.
credit card debt sucks, especially for stupid or innumerate people. one of the few things that suck worse than CC debt is phone contracts.
Unfortunately, it's you that's making yourself look like a nutjob because you don't have a fucking clue what you're talking about.
The GPL does not prohibit commercialisation of derivative works. The GPL prohibits making them proprietary. Proprietary and Commercial are *not* the same thing, not even close. Nor are they dependant traits.
I'll repeat my original point since you seem to have missed it "The *ONLY* restriction that the GPL has that the BSD license doesn't is that it prevents you from restricting other people's freedom."
*ALL* of the extra clauses in the GPL vs the various BSD-like licenses are aimed at closing one loophole or another that could be used to restrict the freedom of anyone who receives the code.
That is not a freedom that most FOSS developers care about or want. A minority of FOSS developers do want it - good luck to them.
In short, like proprietary software if you want to use GPL code there's a price to pay. Unlike proprietary software, the price is not money or signing an NDA, the price is that you grant the same freedoms that you received to anyone you distribute the code or a derivative work to. If you don't want to pay that price, then don't distribute GPL-ed code or anything derived from it. Write your own, or find an alternative with a license acceptable to you.
the only freedom that BSD-like licenses grant you that the GPL doesn't is the freedom to restrict other people's freedom.
some FOSS developers think that that is a worthwhile and useful freedom for their project. Most don't.
It's certainly not a useful or desirable freedom from the POV of an end-user running software without the right to view, modify, or redistribute the source code.
Here's a much better method for optimising security on your smartphone or tablet:
DON'T INSTALL WORTHLESS SHIT
Apple's App Store and Google's Marketplace make Sturgeon's Law seem like hopelessly naive pollyanna-ish optimism. They each may have a few hundred thousand apps, but less than one in ten thousand or so are worth even looking at, let alone installing.
OTOH, a 500 ton asteroid in earth orbit is a space station / tiny moon that they own.
this is the 2nd or 3rd similar article i've seen in recent months.
it's starting to smell a lot like a publicity campaign to discredit science in general. ...wonder if these studies are being funded by oil companies or their shills, or neo-liberal think tanks with a vested interest in discrediting science in general and climate science in particular.
since i run my own mail server with just a handful of users, i only have to care about "power-users" or "techies."
piping messages is essential - software that tells me, on my own server, "you can't be trusted to do that" is not likely to endear itself to me. and if it's
disallowed by sieve because of sieve's nature, that means sieve is not suitable for my needs.
procmail *should* be doing anything i want it to. including piping messages through formail for Message-ID dupe detection.
as i said before if one comes along that's better than procmail and sufficiently good that it's worth the hassle of switching, then i'll consider doing so. I haven't seen one yet....and it would have to be extremely good to be worthwhile.
and if i stay with procmail, i won't have to learn another mail-filtering language either :)
BTW, procmail is standard too - the nice thing about standards is that there are so many to chose from.
Last time i looked, sieve couldn't pipe a message (header or body or both) into an external program. and it doesn't even support regexp matching by default.
For me, both are required....and regexp matching is what I would consider to be a base-level core feature required of any mail filtering software.
I just had another quick look at it and noticed that it also seems to be lacking:
* the ability to detect duplicate messages by Message-Id:
* the ability to clone an incoming message so that it can be handled independently by two or more different rules.
* an equivalent to procmail's ^TO_, ^FROM_DAEMON, etc pre-defined regexps.
Sieve might be a nice option for someone with fairly basic mail-filtering needs and without 2+ decades of procmail use, but it doesn't offer a compelling reason for me to switch.
Courier's maildrop comes a lot closer to my needs....actually, if I wanted to, I could probably switch to it without any noticable loss of functionality. It would be a lot of work with no real benefit though.
as for 'sane syntax', I find procmailrc a lot easier to read and modify that what i've seen of sieve, which seems to add a lot of verbosity without sufficient features to justify it. I guess I just don't need yet another C-like language.
maildrop is also, IMO, more readable than sieve.
then you're either an ignoramus, an arsehole, an idiot, or some combination of all three.
start cheering.
it's easy enough to write a monochrome .muttrc
you can probably find one on the net if you search.
mutt shouldn't do local delivery - or do you routinely misconfigure systems so that mailboxes are world-writable? or do you want mutt to be setuid root or setgid mail?
mutt's job is to be an MUA, not an MTA or MDA.
there are numerous light-weight MTAs & MDAs that can be installed to handle local delivery...most of them just an apt-get install or yum install away.
e.g.
dma - lightweight mail transport agent
masqmail - mail transport agent for intermittently connected hosts
even postfix or exim aren't too heavy to just install as a matter of course on most, if not all, machines. it's hard to imagine any machine that shouldn't have a local MTA, or wouldn't benefit from one. even a little openwrt router or similar embedded device should have one just so that it can send alert messages.
lots of things already depend on (or assume) the presence of an MTA anyway. cron, for example. so you need something to fill that job, whether it can handle local delivery (most MTAs) or not (nullmailer, and similar relay-only MTAs)
echo "message" | mail -s "hi" username
an MTA is an (almost) essential service that can be used by many other tools to send mail (both local and remote), without each of them having to reinvent the wheel.
name one.
it's not perfect, nothing is. but it's adequate.
or is that just your way of saying "it's too hard! waaah!"
mutt is "poorly featured". that's amusing...in the same way that a small child saying something fundamentally wrong but earnestly and sincerely believing it is amusing.
mutt is the full-featured email client. with many more features than can be found in *any* GUI mail client.
GUI mail clients typically have only one big feature - that they have a GUI interface.
In exchange, they've sacrificed pretty nearly every other useful feature that a mail client should have, other than the absolute basics of read, write, reply, forward, delete, etc.
Thunderbird makes a halfway decent effort at getting some of those features back in a GUI client (and with the External Editor plugin even lets you use gvim or your favourite editor), but none of the others do.
It's not necessary, but it is useful. and yes, a different problem.
changing folders in mutt is a PITA. You have to close your current folder when you open the new one. To close it, you lose all your current context in that folder - you have to save or abandon any changes (e.g. deletions) you've made, and lose the tag status of any tagged messages. and similar small annoyances. and if they're large folders, they can take several seconds to save-and-close, and then more time to open the new one. even on a fast SSD.
by having multiple mutts running, each open on a different folder, you can avoid that. when you want to change folder, just press ^Z to suspend the current mutt and use the standard sh job control command 'fg' to bring another mutt to the foreground....instantly.
If you write a script to start multiple mutts like this, then you get them in the same order every time...and you very quickly learn, e.g., that 'fg 1' gets you your main inbox, 'fg 2', gets you your sent-mail folder, 'fg 3' gets you to mailing list 1, 'fg 4' gets you to mailing list 2, and so on.
and if i forget, i can type 'jobs' in the shell and get a list.
even better, if i run that mutt startup script inside screen then I can attach to it via ssh from home, work, or from anywhere in the world.
BTW, I have *MANY* more folders than just 20. 3287 in 169 different sub-directories at the moment, according to find:
$ find ~/Mail -type f | wc -l
3287
$ find ~/Mail -type d | wc -l
169
(btw, this is why i don't use the sidebar plugin for mutt. it's unbearably slow. literally unusable with that many folders it has to scan)
The 20 (actually, 24 at the moment) are just the ones I use most frequently, the ones I'm switching between all the time. mutt knows about the others via the usual configuration, like:
mailboxes \
specific-named-mailboxes-here \
`find ~/Mail -name "*.incoming" | xargs echo `
I deliberately exclude my archive mbox folders (i.e. those not named *.incoming) here because I can easily find them with mutt's folder browser anyway. the mailboxes line just gives me a quick pick list of all the mbox folders that procmail is filtering incoming mail into.
no, you're just being stupid and short-sighted.
the point of email is communication. - communication with all readers, not just those who happen to be using the same software as you.
HTML in email is fucked up and stupid for the same reason that MS Word documents in email are fucked up and stupid, or why writing your email in LaTeX would also be fucked up and stupid.
plain text works with all mail clients. HTML doesn't. XML doesn't. LaTeX doesn't.
and when morons of your ilk switch to BLINGML 2015 and you're an old loser stuck on boring old XML or ancient HTML you might begin to understand why dropping backwards compatibility for the sake of following fads is cretinous.
HTML is NOT XML. it looks superficially similar but it's not even close to XML.
you can pipe html or xml easily enough. you can't can't do anything useful with that, though, unless you write an xml or html parser (or use an existing one). which magnifies a simple Q&D one-liner into a small-scale development project.
you obviously haven't seen many, then. mutt doesn't. elm doesn't. pine doesn't. none of the non-html MUAs support it.
and if you're goint to argue that they'll just render the html tags as plain text in the body - well, they might. if your sending email client is broken and doesn't use the correct mime-type for the body attachment, they will. and even then <em>html</em> tags <blink>scattered>/blink> throughout the <i>text</i> make the text <blink>unreadable</blink>
RFCs 822, 2822, and 5322 to start with.
you know, those unimportant little documents that define the standard for Internet Message Format.
here's a site that summarises and discusses them:
http://mailformat.dan.info/body/linelength.html
damn. html ate my <.
that should have been "< 80 columns".
I *am* the OP, and the reasons I cited above are most of why i think proportional fonts are ghastly for email.
BTW, proportional vs monospace is not what defines whether a font is ugly or legible of pleasant to read. There are ugly and eye-straining fonts in both proportional and non-proportional styles. There are also pleasant, legible, and eminently readable fonts in both.
some kinds of text, however, pretty much require a NP font to be rendered legibly. source code and log files for example. terminal windows for another - and for the same reason.
no, MRSA is a direct result of the moronic overuse of antibiotics.
people still use procmail in 2012 for the same reason they still use screwdrivers or hammers or other basic tools.
for the job that they do, there is nothing better.
maybe someday there'll be something that does mail filtering & sorting as well as procmail. and maybe someday there'll even be something that does a better job than procmail. when that day arrives i'll evaluate whether it provides enough of an improvement to be worth the effort of switching to it.
ps: you're the one who sounds like a 'luddite dick'. demonstrating your technophobia kind of gives it away.
it's about functionality, not fashion:
1. plain-text is readable by anything, and is easily grepped, piped, manipulated by standard tools, and otherwise *used*.
2. log file lines, programs, and tabulated data are all unreadable in proportional font.
3. 80 columns (72-78 is typical) is the standard for email.
the *only* valid excuse for including longer lines is when copy-pasting log lines or code fragments or similar. for discussion, reformat paragraphs with par (or fmt if you don't have par installed)
BTW, not all non-proportional fonts are as ugly as Courier. Deja-Vu Sans Mono is quite nice, for example.
yes.
if you're used to pine, it will take a day or two to get used to mutt's keybindings. it'll probably piss you off while you're learning them, because there are some subtly annoying differences.
after that, you'll be glad you did and you'll never look back.
mutt's searching and tagging features alone are worth the switch.
that's undoubtedly because you were too lenient on the top posters.