You still don't get folders that are filtered server side, so your filtered mail is filtered, or flexible ability to create an infinite number of addresses. I can hand out:
spam.slashdot@brong.fastmail.fm
and not only easily filter the spam when it comes in, but if I have a folder 'INBOX.Spam' it will automatically file in there, and I can even create a subfolder 'INBOX.Spam.Slashdot' and it will file there instead. It makes handing your address to untrusted organisations a lot safer and easier - and checking what they did with it likewise.
Still - it's having the same folder structure in both your local client and the web interface that's the main benefit - you don't realise how much more powerful it is until you've used it, and then you never want to look back. You can move or delete a message locally and that operation is repeated on the server. You can reply and the web interface will have the replied flag marked as well.
It's the difference between managing your email twice in two different locations and having it all properly centralised.
If you want control over your email (and no evil search engine companies leveraging their giant database of email for nefarious purposes, natch) you're better off with something like FastMail. The free offering isn't quite so fancy, but the paid options rock.
In particular, our highest fee paying accounts now get to send and receive up to 50MB of attachments with an email, and that's a full 50MiB (including room for encoding in the Postfix limits)
Our interface is more designed around the IMAP protocol than Gmail, since that's what we use internally - and we offer (optionally) encrypted IMAP for everyone and encrypted POP & Auth SMTP for all paying users.
You also get a web site and file storage space which you can access directly from emails to attach or detach files, etc... but I'm not going to detail all the features here - just point out that the big names don't always offer the best features.
I'm pretty sure I emailed the author of the module at the time with a test-case - but it's over a year ago and at a previous job where I didn't get a copy of my sent email or work area when I left (it wasn't on the best of terms...) so I can't check.
Last time I tried the Perl YAML module I could generate a pathological perl data structure (strings designed to look suspiciously like bits of YAML) and corrupt the output sufficiently that it didn't parse back into the same data structure.
This was a bit over a year ago.
I'm sorry, but I'm just not interested in using a format where I can't rely on it being clean enough to even pass printable text cleanly through a conversion and back again. Get back to me when you've got a format which isn't a crock of shit.
Go with OO.o or Abi Word or something that doesn't explode when you put an MS Word file into it
And here we see the problems - users of Wordperfect who are only transferring files to other users of Wordperfect don't need to put MS Word files in - indeed that fact that they can't allows them (if powerful enough) to put pressure on others to use word processing power that plays well with Wordperfect instead.
Seatbelts are a good thing but they're not a cure-all.
And at least here in Australia, fitting full 5 point racing harnesses is illegal because it encourages people to speed, or something - never mind that they're a whole lot safer because they share the force out more evenly. Ditto roll cages.
It's amazing the sort of crap racing car drivers survive, and it's because they have real safety gear in those cars.
Given that my employer has offices in Australia and servers in the US of A but owns two.uk domains (I know that plenty well enough from having to renew them recently - what's with.uk domain registrars and complex manual processes for single year only registration, sheesh).
Why even bother testing unpatched Solaris when Sun specifically tells you to patch your boxes? It's like never changing your car's oil and then complaining that it breaks down too often.
That would be the same reason that people don't upgrade to Microsoft's latest security patches on a whole range of embedded systems or machines with a piece of shitware from $CRAP_VENDOR which is only certified to work on a paricular service pack level (if you're lucky it might even work with the security patches installed, but your up shit creek if it doesn't)
And if you're lucky, your crapware vendor might release an updated version which works with the latest patches from your OS vendor from 6 months ago... if you're lucky.
Research has established that the intent of the Signer's of our Constitution
I wonder what their intend regarding apostropies was?
Anyway, I'm not an American - as least, not any more. I lived there for 6 months recently. Glad to be back in Australia thanks very much - I don't think the Signer's of the Constitution intended for non-paper-trail electronic voting machines either.
What if amount of happieness is not a number, but a function of time, that the function does not have a limit at infinity (until the end of your life). What if the total amount of happieness is basically the integral of this time function. What if, this function is *not* integratable (cymbal crash).
Err... would you care to give me an example of a not integratable function.
Yep, I thought so.
It doesn't look anything like the sort of function that's bounded above by the sort of instant rapture that causes you to die instantly from the sheer joy of it (give me some poetic licence here, we are talking theoretical intanglibles after all, and they're devilish tricky) and below by the maximum unhappiness that would cause you to die instantly of despair.
Furthermore, it doesn't move immediately from one to the other - it takes your brain a certain amoun t of time to absorb events and respond by shifting its happiness index, so there aren't even instant discontinuities (something that makes things hard to differentiate, but doesn't hurt integratability btw).
Finally it's bounded at the start within a year before your birth (let's give those twinkles in your parents' eyes the benefit of the doubt and go outside the range here) and the point of your death.
Ok, so if you believe in reincarnation then it's a bit harder, but you can still keep score just for the single life, assuming no spillover between lives.
Looks like whatever function you throw on there you can integrate to sufficient precision by reading your happiness quotient every [delta] microseconds and summing them all up across the span of your life.
we need to stay mathematical, and examine things with pure logic, which doesn't allow for lax parameters such as "optimal happieness"...
Sure it does. All you need is a function which takes (a state of the universe, a time_space_fuzzy_pinpointer_thing, the identify of one person_soul_whatsit) and returns a scalar. I don't know what that function is, but one presumes that a God-like critter would, and you sort of need one for this game to work anyway.
Let me get this straight - it can be proved that you
a) created a plausible deniability capable link; and
b) intentionally released the key to said link so that someone else could impersonate you later.
Frequently all that's needed is the fact that you communicated with somebody for evidence - not the specifics of what you said. Sure maybe you just called them up and did some heavy breathing down the line - there's no proof you actually _spoke_, but any jury in the world would convict you.
Of course you work around that by creating a new link every hour to the same person, and maybe or maybe not using it - but it still shows you're in communication with them. There's no way around that.
Nice idea, but don't think your child pornography dealing down this link is going to somehow get you off the hook.
Assuming that all the mates presented are terrible...
See also my response to one of the other replies. I did of course forget to add the 'stay single' datapoint, and the amount of happiness you would gain from that over the course of your life.
If that's higher than any of the 100 choices then your best option is to stay single.
aying the length of the vector can be ordered is changing the problem from an n-dimension space, to a uni-dimensional scalar space (the length of the vector).
That's perfectly true, however we're trying to find a uni-dimensional value "amount of happiness you will have with such person for the rest of your life"
My point though is that, maybe, just maybe, you can't be happy with someone all your life.
Ok, so we need to add another vector to our above set, which is your happiness with no partner. Lather, Rinse, Repeat. If it turns out that's the longest vector then by all means don't partner.
You said "Absolutely wrong" twice about my optimum person statement, but as I see it your argument basically boils down to "there is no method (omni-whatsit God providing details aside) of knowing which multi-dimensional person maps to the largest uni-dimensional life-happiness for you".
Sure, I agree. That doesn't mean that if you had perfect forward knowledge about the rest of your life and could conduct parallel universe tests with each of the 100 people that you met, and evaluate the sum happiness at the end of the tests, that my argument would be invalid.
In summary - sure you need to be able to predict the future with perfect accuracy for there to be any point, but if there was then it is possible to come up with a unidimensional happiness value.
You do raise a very good point about the happiness value of 'not partnered' being an important member of the set you need to examine. Consider it the 'reserve price' if you will.
Assign numeric values to categories like this:... and then scale them by how much each one is important to you, otherwise you'll find out quick enough that your model was flawed.
While you're at it, realise that what's most important to you now is not what's most important to you later in life, so integrate across the years.
In conclusion, you probably want different girls at different times in your life. Try convincing them to go along with that plan though - maybe you'd better move to Utah, but that has its own problems;)
What are the chances that you would recognize the person most optimal for you, given the impossibility of actually finding out, i.e. living the rest of your life with each of them and then comparing?
Um, none.
Then again, after 37 trial runs, you'd have a fairly good idea of what worked and what didn't - for you.... while she might be the most optimal for you, you might not be *her* best choice...
Given that we're already talking about the unmeasurable here, I rebut you by pointing out that this is a compatibility factor. The fact that you're not a good enough choice for her means she also isn't a good choice for you precisely because she _would_ dump you. Reductio ad blah blah blah.
So, this isn't an issue. By being the one that you'll be most happy spending the rest of your life with, she's already committed to being happy with you. Yay.
It seems mathematics isn't best way to find a mate! Who knew?
The fact that there exists an optimum person for this problem is not trivial. Personally, I think it's like a limit. It might be 0, 1 or +infinity. Then again, a function might not have a limit.
The fact that there exists an optimum person is pretty trivial to prove if you're willing to accept an axiom that no two people are identical.
Use the same proof you would use to show that of any random set of vectors in N dimensional space, with none of them having equal lengths to any other, there must be one which is the longest.
Now calculating that and being able to know at the time which one is "the one" out of those one hundred, that's a different story. As a line from a great song by an Aussie band goes: "she's one in a millon, so there's 6 more just in New South Wales".
Conclusion: theoretically (which is what maths is good at) it makes sense, but practically you can't tell.
Says me who married the first woman I was in a relationship with for more than a week - I guess I wasn't not very mathematical about it, or something.
The 1/100 chance that the last person is the optimal choice assumes there exists one optimal choice in the original batch of 100 in the first place.
As the cowards have said, of course it's the optimal of that 100.
Given that, there are many different ways of measuring optimal, but I think that given the question "choose your life mate" the optimal has to be "person who you will be most happy with for the rest of your life".
Ok, we have a comparison function. Now I don't have a clue how you can tell which one is optimum - and besides you're only going to reach number 100 (assuming they're the best) if the second best was in the first 37.
I'm also assuming that no two people are exactly identical, and hence no two people are going to be exactly as "enjoy rest of life with" as each other. I think that's a fair assumption with humans.
No, don't. Just imagine running your mp3 player in the background while you actually do something else on the device - then go read up on the difference between multiprocessing and multiprocessor and be elightened.
Have you tried using it? I tried against a postgres backend (to check out some databases I already had) and the interface isn't quite what I'd expect for a Filemaker or Access replacement.
You can't even see views in the database.
No way (that I can see) to store connection information for more than one backend server. Ok, that's not the end of the world. At least you can make it cache your connection details.
The interface is a little strange with database as a dropdown list in the toolbar, and also a tree-menu at the side but with only one parent database entry at once.
Did I say "no views" yet?
Still, I guess if you're starting from scratch it isn't so bad, and it does fit in reasonably in KDE. Needs some work on the sidebar treeview not overlapping the database windows though. Definitely looks more complete than the gnome offering.
These days I get paid to do it, I did the support for IBM Lotus Notes 6.5.1 on Wine.
Wow, if I didn't have you as friend already, I certainly would at that. Not that I'm running Notes 6.5.1 (still running Notes 5 - it does the job), but it's actually nicer running under Wine than on Windows for me, because ssh tunnels are a little easier, and Notes is pretty good about working over a TCP tunnel.
Still, only 2 weeks left on this job. I'll keep paying for Crossover though - even just for office it is nice, and IE, while dodgy, is better than nothing for the times you need it.
a) to be useful for anything involving third parties where you don't already have a trust relationship, this would need to be common/easy enough to get that other people already have software to support these things. That's not going to happen any time soon - it's a big enough change you may as well come up with an already secure email infrastructure [insert boilerplate "why your solution to spam is stupid" here].
b) 8 tokens per second? Puhleaze. I get that many emails through just one small server with 5 domains on it.
c) as the subject says. Zombies. In a world where thousands of low TC0 machines are sitting around running malware, it's piss-easy for the blackhat spammers to collect their 8 tokens/second by running POWer@home on their zombie farm.
BZZZZt. Strike three and you're out. Nice idea, but not practical.
Slashdot Mirror
You still don't get folders that are filtered server side, so your filtered mail is filtered, or flexible ability to create an infinite number of addresses. I can hand out:
spam.slashdot@brong.fastmail.fm
and not only easily filter the spam when it comes in, but if I have a folder 'INBOX.Spam' it will automatically file in there, and I can even create a subfolder 'INBOX.Spam.Slashdot' and it will file there instead. It makes handing your address to untrusted organisations a lot safer and easier - and checking what they did with it likewise.
Still - it's having the same folder structure in both your local client and the web interface that's the main benefit - you don't realise how much more powerful it is until you've used it, and then you never want to look back. You can move or delete a message locally and that operation is repeated on the server. You can reply and the web interface will have the replied flag marked as well.
It's the difference between managing your email twice in two different locations and having it all properly centralised.
Disclaimer: I work for FastMail.
If you want control over your email (and no evil search engine companies leveraging their giant database of email for nefarious purposes, natch) you're better off with something like FastMail. The free offering isn't quite so fancy, but the paid options rock.
In particular, our highest fee paying accounts now get to send and receive up to 50MB of attachments with an email, and that's a full 50MiB (including room for encoding in the Postfix limits)
Our interface is more designed around the IMAP protocol than Gmail, since that's what we use internally - and we offer (optionally) encrypted IMAP for everyone and encrypted POP & Auth SMTP for all paying users.
You also get a web site and file storage space which you can access directly from emails to attach or detach files, etc... but I'm not going to detail all the features here - just point out that the big names don't always offer the best features.
I'm pretty sure I emailed the author of the module at the time with a test-case - but it's over a year ago and at a previous job where I didn't get a copy of my sent email or work area when I left (it wasn't on the best of terms...) so I can't check.
Last time I tried the Perl YAML module I could generate a pathological perl data structure (strings designed to look suspiciously like bits of YAML) and corrupt the output sufficiently that it didn't parse back into the same data structure.
This was a bit over a year ago.
I'm sorry, but I'm just not interested in using a format where I can't rely on it being clean enough to even pass printable text cleanly through a conversion and back again. Get back to me when you've got a format which isn't a crock of shit.
Go with OO.o or Abi Word or something that doesn't explode when you put an MS Word file into it
And here we see the problems - users of Wordperfect who are only transferring files to other users of Wordperfect don't need to put MS Word files in - indeed that fact that they can't allows them (if powerful enough) to put pressure on others to use word processing power that plays well with Wordperfect instead.
Seatbelts are a good thing but they're not a cure-all.
And at least here in Australia, fitting full 5 point racing harnesses is illegal because it encourages people to speed, or something - never mind that they're a whole lot safer because they share the force out more evenly. Ditto roll cages.
It's amazing the sort of crap racing car drivers survive, and it's because they have real safety gear in those cars.
Given that my employer has offices in Australia and servers in the US of A but owns two .uk domains (I know that plenty well enough from having to renew them recently - what's with .uk domain registrars and complex manual processes for single year only registration, sheesh).
Parent of parent lives in fantasy world.
Why even bother testing unpatched Solaris when Sun specifically tells you to patch your boxes? It's like never changing your car's oil and then complaining that it breaks down too often.
That would be the same reason that people don't upgrade to Microsoft's latest security patches on a whole range of embedded systems or machines with a piece of shitware from $CRAP_VENDOR which is only certified to work on a paricular service pack level (if you're lucky it might even work with the security patches installed, but your up shit creek if it doesn't)
And if you're lucky, your crapware vendor might release an updated version which works with the latest patches from your OS vendor from 6 months ago... if you're lucky.
If I took a bullet to the head, it wouldn't give me much comfort to know I had a brother as 'backup' though
;p
No, but it might give some comfort to your wife
mono.net with a LAMP(ish) server build.
.NET Postgresql Client API that seems to work well.
What, Linux, Apache, Mono, Postgresql? It's good that there's a
Research has established that the intent of the Signer's of our Constitution
I wonder what their intend regarding apostropies was?
Anyway, I'm not an American - as least, not any more. I lived there for 6 months recently. Glad to be back in Australia thanks very much - I don't think the Signer's of the Constitution intended for non-paper-trail electronic voting machines either.
What if amount of happieness is not a number, but a function of time, that the function does not have a limit at infinity (until the end of your life). What if the total amount of happieness is basically the integral of this time function. What if, this function is *not* integratable (cymbal crash).
Err... would you care to give me an example of a not integratable function.
Yep, I thought so.
It doesn't look anything like the sort of function that's bounded above by the sort of instant rapture that causes you to die instantly from the sheer joy of it (give me some poetic licence here, we are talking theoretical intanglibles after all, and they're devilish tricky) and below by the maximum unhappiness that would cause you to die instantly of despair.
Furthermore, it doesn't move immediately from one to the other - it takes your brain a certain amoun t of time to absorb events and respond by shifting its happiness index, so there aren't even instant discontinuities (something that makes things hard to differentiate, but doesn't hurt integratability btw).
Finally it's bounded at the start within a year before your birth (let's give those twinkles in your parents' eyes the benefit of the doubt and go outside the range here) and the point of your death.
Ok, so if you believe in reincarnation then it's a bit harder, but you can still keep score just for the single life, assuming no spillover between lives.
Looks like whatever function you throw on there you can integrate to sufficient precision by reading your happiness quotient every [delta] microseconds and summing them all up across the span of your life.
we need to stay mathematical, and examine things with pure logic, which doesn't allow for lax parameters such as "optimal happieness"...
Sure it does. All you need is a function which takes (a state of the universe, a time_space_fuzzy_pinpointer_thing, the identify of one person_soul_whatsit) and returns a scalar. I don't know what that function is, but one presumes that a God-like critter would, and you sort of need one for this game to work anyway.
Let me get this straight - it can be proved that you
a) created a plausible deniability capable link; and
b) intentionally released the key to said link so that someone else could impersonate you later.
Frequently all that's needed is the fact that you communicated with somebody for evidence - not the specifics of what you said. Sure maybe you just called them up and did some heavy breathing down the line - there's no proof you actually _spoke_, but any jury in the world would convict you.
Of course you work around that by creating a new link every hour to the same person, and maybe or maybe not using it - but it still shows you're in communication with them. There's no way around that.
Nice idea, but don't think your child pornography dealing down this link is going to somehow get you off the hook.
Assuming that all the mates presented are terrible...
See also my response to one of the other replies. I did of course forget to add the 'stay single' datapoint, and the amount of happiness you would gain from that over the course of your life.
If that's higher than any of the 100 choices then your best option is to stay single.
aying the length of the vector can be ordered is changing the problem from an n-dimension space, to a uni-dimensional scalar space (the length of the vector).
That's perfectly true, however we're trying to find a uni-dimensional value "amount of happiness you will have with such person for the rest of your life"
My point though is that, maybe, just maybe, you can't be happy with someone all your life.
Ok, so we need to add another vector to our above set, which is your happiness with no partner. Lather, Rinse, Repeat. If it turns out that's the longest vector then by all means don't partner.
You said "Absolutely wrong" twice about my optimum person statement, but as I see it your argument basically boils down to "there is no method (omni-whatsit God providing details aside) of knowing which multi-dimensional person maps to the largest uni-dimensional life-happiness for you".
Sure, I agree. That doesn't mean that if you had perfect forward knowledge about the rest of your life and could conduct parallel universe tests with each of the 100 people that you met, and evaluate the sum happiness at the end of the tests, that my argument would be invalid.
In summary - sure you need to be able to predict the future with perfect accuracy for there to be any point, but if there was then it is possible to come up with a unidimensional happiness value.
You do raise a very good point about the happiness value of 'not partnered' being an important member of the set you need to examine. Consider it the 'reserve price' if you will.
Assign numeric values to categories like this: ... and then scale them by how much each one is important to you, otherwise you'll find out quick enough that your model was flawed.
;)
While you're at it, realise that what's most important to you now is not what's most important to you later in life, so integrate across the years.
In conclusion, you probably want different girls at different times in your life. Try convincing them to go along with that plan though - maybe you'd better move to Utah, but that has its own problems
What are the chances that you would recognize the person most optimal for you, given the impossibility of actually finding out, i.e. living the rest of your life with each of them and then comparing?
... while she might be the most optimal for you, you might not be *her* best choice ...
Um, none.
Then again, after 37 trial runs, you'd have a fairly good idea of what worked and what didn't - for you.
Given that we're already talking about the unmeasurable here, I rebut you by pointing out that this is a compatibility factor. The fact that you're not a good enough choice for her means she also isn't a good choice for you precisely because she _would_ dump you. Reductio ad blah blah blah.
So, this isn't an issue. By being the one that you'll be most happy spending the rest of your life with, she's already committed to being happy with you. Yay.
It seems mathematics isn't best way to find a mate! Who knew?
Yeah, no shit.
The fact that there exists an optimum person for this problem is not trivial. Personally, I think it's like a limit. It might be 0, 1 or +infinity. Then again, a function might not have a limit.
The fact that there exists an optimum person is pretty trivial to prove if you're willing to accept an axiom that no two people are identical.
Use the same proof you would use to show that of any random set of vectors in N dimensional space, with none of them having equal lengths to any other, there must be one which is the longest.
Now calculating that and being able to know at the time which one is "the one" out of those one hundred, that's a different story. As a line from a great song by an Aussie band goes: "she's one in a millon, so there's 6 more just in New South Wales".
Conclusion: theoretically (which is what maths is good at) it makes sense, but practically you can't tell.
Says me who married the first woman I was in a relationship with for more than a week - I guess I wasn't not very mathematical about it, or something.
The 1/100 chance that the last person is the optimal choice assumes there exists one optimal choice in the original batch of 100 in the first place.
As the cowards have said, of course it's the optimal of that 100.
Given that, there are many different ways of measuring optimal, but I think that given the question "choose your life mate" the optimal has to be "person who you will be most happy with for the rest of your life".
Ok, we have a comparison function. Now I don't have a clue how you can tell which one is optimum - and besides you're only going to reach number 100 (assuming they're the best) if the second best was in the first 37.
I'm also assuming that no two people are exactly identical, and hence no two people are going to be exactly as "enjoy rest of life with" as each other. I think that's a fair assumption with humans.
Imagine a beowulf...
No, don't. Just imagine running your mp3 player in the background while you actually do something else on the device - then go read up on the difference between multiprocessing and multiprocessor and be elightened.
Sheesh. I guess IHBT.
Have you tried using it? I tried against a postgres backend (to check out some databases I already had) and the interface isn't quite what I'd expect for a Filemaker or Access replacement.
You can't even see views in the database.
No way (that I can see) to store connection information for more than one backend server. Ok, that's not the end of the world. At least you can make it cache your connection details.
The interface is a little strange with database as a dropdown list in the toolbar, and also a tree-menu at the side but with only one parent database entry at once.
Did I say "no views" yet?
Still, I guess if you're starting from scratch it isn't so bad, and it does fit in reasonably in KDE. Needs some work on the sidebar treeview not overlapping the database windows though. Definitely looks more complete than the gnome offering.
I guess you missed the part where printing was the main reason to keep filemaker or something like it. From the website:
Development is at an early stage, so it's not a complete database solution yet. The following future functionality should complete it:
* Virtual view fields (e.g. 'show the current price of this product in this field').
* Printing
* Reports
Phew, for a second there I thought the second character might have been an N and wondered how you got modded so high.
These days I get paid to do it, I did the support for IBM Lotus Notes 6.5.1 on Wine.
Wow, if I didn't have you as friend already, I certainly would at that. Not that I'm running Notes 6.5.1 (still running Notes 5 - it does the job), but it's actually nicer running under Wine than on Windows for me, because ssh tunnels are a little easier, and Notes is pretty good about working over a TCP tunnel.
Still, only 2 weeks left on this job. I'll keep paying for Crossover though - even just for office it is nice, and IE, while dodgy, is better than nothing for the times you need it.
What a crock of a system. Let's see:
a) to be useful for anything involving third parties where you don't already have a trust relationship, this would need to be common/easy enough to get that other people already have software to support these things. That's not going to happen any time soon - it's a big enough change you may as well come up with an already secure email infrastructure [insert boilerplate "why your solution to spam is stupid" here].
b) 8 tokens per second? Puhleaze. I get that many emails through just one small server with 5 domains on it.
c) as the subject says. Zombies. In a world where thousands of low TC0 machines are sitting around running malware, it's piss-easy for the blackhat spammers to collect their 8 tokens/second by running POWer@home on their zombie farm.
BZZZZt. Strike three and you're out. Nice idea, but not practical.