Not a bad idea IMO. It'll stop SPAM from within, and bring slightly closer to home for ISPs and open relays etc the question, "So why are you shipping in shedloads of illegal email?" and hopefully raise interest in providing a TECHNICAL solution to the problem which is what we have needed all along. Of course, they may be able to hide behind "common carrier", but until that is proven ISPs may just be sufficiently vulnerable to do something about SPAM.
Ok ok, forged headers etc. So what's so difficult about Server B getting mail from Server A for Server C and checking that From=Server A and To=Server C? (oversimplification I know, but I would have thought it would be a piece of cake to reject email with forged headers).
whose plonker are you pulling? you drive onto the ferry, cars occupy the whole width; the off-ramp guides you to the correct side of the road. It must take some doing to get onto the wrong side of the road (although most of Europe appears to have managed it:-P) Otherwise once you're incontinent all countries drive on the right so it isn't an issue.
you do know there is water between England and Europe, don't you? (Sorry, I meant "the rest of Europe") Oh mind you, there hasn't been a war there recently... Currently only hovercraft owners enjoy the privilege of driving from England to France.
The only accident I have had abroad was misjudging the passenger side when driving one of those weird foreign mirror-image cars where not quite everything is completely swapped around (accelerator isn't next to the door, etc.) It would be much easier, I think, if the foreign cars were completely mirrored. No problem at all driving my own car on the right-hand side of the road, although overtakes on single carriageways were a bit more interesting...
Why not? Do you really think the devil is a guy in a red jump suit with horns, a pointy tail and a three pronged fork? If one wants an image of evil, it doesn't matter if it's red jump suit man, Darth, Saddam or whatever, and why not use an image that has some meaning to people? Pointy fork and tail man is little more than a cartoon figure to most people nowadays; Darth Vader represents evil much more to me than someone who looks like he's had one too many vindaloos (although the Emperor would be an even better choice because he doesn't turn good in the end.)
Hmm, Guild of Spammers...sounds like something Terry Pratchett might have thought up for the next installment of Ankh-Morpork.
(Couldn't find "spam" in my Latatian dictionary, which also doesn't have a section on how to convert the infinitive to past tense, so "to cook pig" will have to do.)
Lots of replies of the form "But I always need my scrolly button so you must be stupid."
I have eight Logitech three button mice. I like them a lot. I reprogram the middle button to double-click, because I don't like RSI and I think the double-click idea really sucks. I use the keyboard 95% of the time and only reach for the mouse when some lazy application programmer couldn't be arsed to take the 5 microseconds needed to put a keyboard shortcut in for a specific function.
I hate the scrolly wheel. I don't want a scrolly wheel. Yes I know they can be clicked, but they are designed to be scrolled, not clicked, so the click spring is much stronger than the spring on the other two buttons. Besides, you can scroll with the keyboard (or at least you will be able to when Mozilla works out that the currently selected tab is the one that should have the keyboard focus, not the one that's just finished loading). I'm not saying YOU, dear Reader, need to, just that I do, and I don't want to be reaching for the mouse when I can move my fingers to the up/down keys.
So, I personally can guarantee a market of eight optical three button mice for Logitech when they come out at a decent price. Oh, I had a look at the MX700. It's a fucking air traffic control centre that needs at least 23 people to operate. OVERKILL, people! I want an OPTICAL version of my three button mouseman, that's all. And the MX700 costs about $90. I'm NOT buying eight of those. (if you're wondering, 8=3 at work+3 at home+2 spare. Yes, I really like them a lot. Every time I get a computer with a mouse with a clitoris I replace it with a 3BMM.) Three button mousemen are currently going for about $25 in the UK. I'd pay $40 for an optical version, cos I'm bored with cleaning dirty balls and rollers. But I'm not going to buy ANYTHING that has a scrolly wheel on; I'd rather stick with my current mice.
So if anyone's with me, mod me up, and someone pass a reference to this article to Logitech.
On my PalmPilot the rate is currently 100%. Most Losedows crippleware I have tried out can be replaced with GPL stuff; e.g. PSP->Gimp, UltraEdit->Emacs. Everything else I use is freeware or demoware that's actually useful in its demo form (trillian, proxomitron (yeah I know, Shonenware, but they don't have any mp3s for download and I don't buy CDs just to see if they're any good any more), treesize, ivt), or stuff for which we have a site licence or my job justifies a corporate licence (winzip, Visual Studio). Still looking for a free equivalent of ClipMate but I'll be buying it soon if I can't find one.
Proper try before you buy shareware is fairly dead. Authors only seem to release crippleware these days, incorrectly calling it shareware, figuring that they won't make any money if they release full software. I don't know generally how successful this approach is, but after Slashdot my second favourite site is that one that sounds a bit like AltaVista... And I *do* buy stuff after cracking it, and trying it out properly, although this often takes longer than the ridiculously short "trial period" most crippleware authors seem to think is sufficient. Stuff that expires on a particular date, set after intallation, has often expired before I get to try it. What I don't buy is stuff I can't crack, cos I can't fully try it out.
PSP - you used to be cool, man. Being proper shareware is what got you where you are today, and now you spit in your fans' eyes.
Anyone know of a shareware site that lists crippleware as such, and not calling "X Lite" (where Lite means crippled) proper shareware when it's only a thinly veiled marketing release?
"Note: The term planet was first used to distinguish those stars which have an apparent motion through the constellations from the fixed stars, which retain their relative places unchanged."
and
"Middle English, from Old French planete, from Late Latin planta, from Greek plants, variant of plans, plant-, from plansthai, to wander"
The most useful definition of "star" appears to be "Any of the celestial bodies visible at night from Earth as relatively stationary, usually twinkling points of light."
So our original definition of planet depends on our own perception. As our ability to perceive objects improves, we either need to accept the classification of new objects under the old naming scheme, or come up with a new naming scheme, which will have to be based on something that is not variable, for example mass, or size. The obvious problem here is that we are trying to come up with a new naming scheme but using the old names; that, surely, is what is ridiculous, not whether or not Pluto is a planet.
Ok, so suppose I run a tarpit, a spammer tries to spam me and spots the tarpit, backs off, and I don't get his spam. Where exactly was the flaw in this plan? This would appear to be better than the original aim as it means the spam doesn't arrive in the first place.
Ok, so Joe Sixpack still gets spam, but when he finds out all his geek friends have eliminated spam he'll soon start asking his ISP why he can't have this feature.
Brilliant stuff. Now where can I get my dream case from - one with all the drives and IO and switches on the SAME SIDE so I don't have to keep turning the damn thing 180 every time I want to fiddle with something?
> Email providers would set up peering relationships wherein they can share validated email addresses.
Which would work until the same sort of dodgy ISP who allows SPAM out in the first place also sets up a database of fake, validated, peered email addresses.
Also how are you going to validate someone? Would you get a message on your screen "someuser@someisp.com wants to be validated: yes/no". But you don't know who they are or what they have to say. So they send you a message: "someuser@someisp.com says 'GET A BIGGER PENIS 1-888-555-1234': validate? yes/no".
If you don't validate someone, do they remain unvalidated or are they permanently blacklisted? What if you spitefully, or accidentally, click "no" if the latter, and it isn't a spammer? If the former they can continue spamming indefinitely. If the latter the spammer simply attaches a serial number to the address and it then doesn't make any difference how you validate "spammer-65987623543@spamisp.com"
The validation can work on an individual basis - ASK, TMDA etc, but there are people you want to hear from that others don't and vice versa, so if those databases of (un)desired senders are shared, how can the contradictions be resolved?
Reply-To validation is cool, but who ultimately validates a Reply-To address except the originating ISP, which in the case of the dodgy spamming ISP simply autovalidates anything and you're still no better off.
You seem determined to call mey suggestions stupid, yet I don't see any helpful suggestions in your posts.
Given that four billion people can't share a meaningful twenty character namespace, how do you suggest this problem is resolved?
Why shouldn't a commercial business have a.com domain? Because you are an ISP, and the.net TLD was created for ISPs. I don't think there is an implication in the.net TLD that you are somehow non-profit.
Ok, so IPs aren't fixed. You're missing my point - which is to have a system that *is* fixed and that is big enough for four billion people to share. Such a system surely cannot be impossible?
And one I've never heard of, no disrespect intended. You are currently no Pratchett, Gibson or Tolkien.
When you are, and you want profit to increase, and you see falling sales and increasing downloads, it won't take more than a few seconds for you to start thinking "hey, they're stealing from my children, the terrorists!"
At the moment you're benefitting a lot from the free publicity. When this stops, the downloads will too. Or at least the complementary nature of the web offering will decrease in value.
So base stuff on unchangeable identifiers. We already have these - they're called IP addresses. Design with co-operation in mind, rather than lawsuits. Remember that a name is a shortcut to an IP address - your website is not telocity.com but 216.227.62.81 (according to "ping telocity.com"). Why is an ISP using a TLD that isn't.net? This kind of practice needs to be stamped out as it is part of the problem. You guys are *not* telocity.com - you are commercial, sure, but you are an ISP and you have your own TLD available, namely.net.
The namespace is limited so one of two things needs to happen. (1) the namespace is expanded until it is no longer limited (so you could be telocity.(state).net.us (assuming you're Yanks). Now you are protected by state laws that presumably exclude another ISP starting up in that state with the name Telocity, but wouldn't prevent a shoe shop called Telocity having a domain telocity.(same_state).co.us (again flawed, as the "co" MLD is limited but if the laws are taken into account the MLD (? medium level, as opposed to top level) can be split up accordingly.) (2) You have to share. Play nicely, etc. If you want to be known as telocity.com, that's fine until someone else wants to be known as telocity.com. You have no more (or less) right (or left (oops, too much Life of Brian)) to that name than the aforementioned shoeshop. You want to be phroggy@telocity.com? Fine, until someone at the shoeshop also wants that name. Then the options are (1) first come first served, which is a little unfair if the shoeshop can prove it predates you; (2) neither of you gets it (3) both of you get it, but for (3) to work you will need an expansion of the namespace. phroggy.shoes@telocity.com and phroggy.isp@telocity.com is one of many possibilities. phroggy@telocity.(state).us.net is another, and the other phroggy would be phroggy@telocity.(same_state).(MLD appropriate to shoe shops).us.
So you now argue that your business will break. But it was your decision to build your business on a flawed model, and when that flawed model is fixed, your business will also need fixing. I'm sure there were lots of corporate noses put out of joint when slavery was abolished, when cockfighting was banned etc; you can bet "what about the jobs that will be lost?" isn't a new argument. So build your business on an unbreakable model (no such thing, but some models are more sturdy models than others). Use globally unique identifiers. telocity.(state).us.net is one such ID if you must use names. IP addresses are others. Secondly allow for the name you choose to vary. If you want to be phroggy@telocity.com, be prepared for that to change when someone else wants to be known by the same ID. There is no way four billion people can comfortably share a namespace that is only 20 characters long, particularly if those names are meant to be meaningful.
What we need are two changes of mindset. First adherence to the rules needs enforcing one way or another. Why should an ISP known as Telocity be granted the name telocity.com, when telocity.net is perfectly ok, and uses a TLD directly intended for ISPs? This is ICANN's fault, really, just handing out domain names willy-nilly without any validation. Secondly the lawsuit mentality needs dropping in favour of the co-operation mentality. These problems can be solved if we work together. IMO all problems we have can be solved if we work together - poverty, famine, internet name clashes etc.
And costs a lot less. Company A wants a website - www.widgets.com, so they get a hosting company just as they do now and publish.
Now Company B argues that they have a claim on www.widgets.com. Ok, now ICANN puts their foot down and states that a domain name is NOT a trademark, and offers to host www.widgets.com for both companies, with links to their main web pages, and possibly some descriptive text indicating what the different possibilities are. Company B's $1000 can go towards relocating company A's pages onto another server.
so www.mcdonalds.com could end up as:
_macdonalds_ the burger chain _macdonalds_ the Scottish kilt maker _macdonalds_ the cafe in Lower Aldershot.
Each pays a minimal hosting fee to ICANN (because they're not hosting a vast amount of stuff - just a stack of hyperlinks)) and hosts their pages off somewhere else. It's always struck me as daft that someone thinks up a perfectly good name for a website, then some company comes along and stamps all over them just because there's a name clash.
I get no end of people calling in with a problem, claiming to know lots and the root cause isn't X, so we go full circle, prove it can't be anything else, and the customer ends up eating humble pie and having to accept that it was X.
If you provide an "I'm not stupid" line then everyone will end up calling it and you'll be no better off because everyone thinks they have a clue.
And are you HONESTLY stating that you never miss the basics? That you have never explained a problem to a colleague and mid-explanation spotted that missing semicolon, or use of = instead of ==? Support start by asking basic questions because of Occam's Razor, not because we all assume you're stupid. Troubleshooting is largely a process of elimination, and eliminating the simple stuff to start with is an essential step.
There's always the possibility that you are right of course. Give up your job, provide support that works that way and prove you are right to the rest of us by taking all our customers away from us. After all, what do we know? We've only been providing tech support for several decades.
Ok, so make it fairer by making Kramnik play blindfold, then he can use a mental representation of the board instead of having to do all that complicated image recognition stuff.
And even if you did make them evaporate they'd only condense somewhere else. The only way to get them out of the closed system would be to dump them somewhere nobody's stupid enough to go like the Moon...oh, hang on...
Or you could do what Janeway did with the Meylon freighter and send it into a nearby star.
If telephone charges were applied to property, you would have to buy the house AND rent it, then pay an additional charge every time you wanted to use a room, and it would be more expensive to use the same room at different times of the day, e.g. the bedroom would be twice as expensive at night as during the day.
How they've got away with it this long escapes me. Not only do you pay for the line to be installed and buy the phone, you then have to pay line rental and call charges. Is it just me or have I just paid for the same thing four times?
I get very few popup ads, and very little hassle from Javascript. I use Incontinent Exploder, with general security settings quite high (no javascript and other settings). Sites I want to see, and want to allow javascript for, I add to Trusted Sites list, https requirement switched off, and security settings much lower. This way if a site opens a popup, I know exactly which one it is - the one I just added to the trusted list, and can remove it. If it does open someone else's site, that site usually doesn't have javascript enabled and can't do it's dirty deeds = end of infinite porn site loops. Junkbuster takes care of cookies for me, and if I add a site to the trusted list, and it annoys me, I just add it to the Junkbuster block list as a reminder next time I visit that site. Junkbuster also blocks a lot of banner ads. If I want to see a site that uses javascript and it opens popups, I just block the popup site with Junkbuster and, ok I still get the popup but I don't get any of the crap in it! I still allow ads from some sites -/. of course, but not many.
My solution runs in 100 day cycles. The light indicates if someone has been in twice in one cycle. If you go in the room for the first time in that cycle, do nothing. If you go in the second or more, switch the light on if it is off (it shouldn't be off if this is time3+ as it will still be on from your time2). Whoever goes in on the 100th day knows if the light bulb is still off that everyone has been in once only, and declares freedom. If the light bulb is on, someone has been in at least twice, so at least one person hasn't been in yet.
Assumptions: the prisoners can count, the warden keeps a list of who's been in, the prisoners can't see the light and can't communicate with each other except on the meeting day and through the bulb, and of course that the bulb is maintained, or at least is operated by a toggle switch.
This is guaranteed to work the first time the warden lets everyone in the room in a cycle, and will continue as long as the warden duplicates people. But release was always dependent on the warden anyway. This is also tolerant of the possibility that the warden knows the plans. The idea of giving each person a number wouldn't get people out in minimum time as the warden could pick P2, P1, P4, P3, P6*, P5... on days 1,2,3,4,5,6 - this way everyone gets in but nobody gets in on their numbered day and the system fails. My way it doesn't matter what order people are picked in. (* NaN, a free man)
The only thing that bothers me about this is that it would seem to be possible to shorten it further by determining at what point the warden stopped duplicating people. e.g. can freedom be declared on day 106 at the end of the sequence 1, 2, 1, 2, 1, 2, (stop duplicating here), 1, 2, 3, 4, 5, 6...; with my system it can only be determined on day 200 at the end of the sequence 1, 2, 1, 2, 1, 2, 1, 2, 3, 4...99, 100, 1, 2,...95, 96. But if anyone switches the light off for any reason in a cycle then whoever is picked on day 100 and finds the light off will not know if the light has been on. Of course if he does know if the light has been on that changes the parameters, but one of my assumptions is that solitary confinement means precisely that - no communication of any type with the outside world (with the possible exception of the warden at meal and bog times).
Slashdot Mirror
Not a bad idea IMO. It'll stop SPAM from within, and bring slightly closer to home for ISPs and open relays etc the question, "So why are you shipping in shedloads of illegal email?" and hopefully raise interest in providing a TECHNICAL solution to the problem which is what we have needed all along. Of course, they may be able to hide behind "common carrier", but until that is proven ISPs may just be sufficiently vulnerable to do something about SPAM.
Ok ok, forged headers etc. So what's so difficult about Server B getting mail from Server A for Server C and checking that From=Server A and To=Server C? (oversimplification I know, but I would have thought it would be a piece of cake to reject email with forged headers).
whose plonker are you pulling? you drive onto the ferry, cars occupy the whole width; the off-ramp guides you to the correct side of the road. It must take some doing to get onto the wrong side of the road (although most of Europe appears to have managed it :-P) Otherwise once you're incontinent all countries drive on the right so it isn't an issue.
you do know there is water between England and Europe, don't you? (Sorry, I meant "the rest of Europe") Oh mind you, there hasn't been a war there recently... Currently only hovercraft owners enjoy the privilege of driving from England to France.
The only accident I have had abroad was misjudging the passenger side when driving one of those weird foreign mirror-image cars where not quite everything is completely swapped around (accelerator isn't next to the door, etc.) It would be much easier, I think, if the foreign cars were completely mirrored. No problem at all driving my own car on the right-hand side of the road, although overtakes on single carriageways were a bit more interesting...
Why not? Do you really think the devil is a guy in a red jump suit with horns, a pointy tail and a three pronged fork? If one wants an image of evil, it doesn't matter if it's red jump suit man, Darth, Saddam or whatever, and why not use an image that has some meaning to people? Pointy fork and tail man is little more than a cartoon figure to most people nowadays; Darth Vader represents evil much more to me than someone who looks like he's had one too many vindaloos (although the Emperor would be an even better choice because he doesn't turn good in the end.)
Hmm, Guild of Spammers...sounds like something Terry Pratchett might have thought up for the next installment of Ankh-Morpork.
(Couldn't find "spam" in my Latatian dictionary, which also doesn't have a section on how to convert the infinitive to past tense, so "to cook pig" will have to do.)
Lots of replies of the form "But I always need my scrolly button so you must be stupid."
I have eight Logitech three button mice. I like them a lot. I reprogram the middle button to double-click, because I don't like RSI and I think the double-click idea really sucks. I use the keyboard 95% of the time and only reach for the mouse when some lazy application programmer couldn't be arsed to take the 5 microseconds needed to put a keyboard shortcut in for a specific function.
I hate the scrolly wheel. I don't want a scrolly wheel. Yes I know they can be clicked, but they are designed to be scrolled, not clicked, so the click spring is much stronger than the spring on the other two buttons. Besides, you can scroll with the keyboard (or at least you will be able to when Mozilla works out that the currently selected tab is the one that should have the keyboard focus, not the one that's just finished loading). I'm not saying YOU, dear Reader, need to, just that I do, and I don't want to be reaching for the mouse when I can move my fingers to the up/down keys.
So, I personally can guarantee a market of eight optical three button mice for Logitech when they come out at a decent price. Oh, I had a look at the MX700. It's a fucking air traffic control centre that needs at least 23 people to operate. OVERKILL, people! I want an OPTICAL version of my three button mouseman, that's all. And the MX700 costs about $90. I'm NOT buying eight of those. (if you're wondering, 8=3 at work+3 at home+2 spare. Yes, I really like them a lot. Every time I get a computer with a mouse with a clitoris I replace it with a 3BMM.) Three button mousemen are currently going for about $25 in the UK. I'd pay $40 for an optical version, cos I'm bored with cleaning dirty balls and rollers. But I'm not going to buy ANYTHING that has a scrolly wheel on; I'd rather stick with my current mice.
So if anyone's with me, mod me up, and someone pass a reference to this article to Logitech.
On my PalmPilot the rate is currently 100%.
Most Losedows crippleware I have tried out can be replaced with GPL stuff; e.g. PSP->Gimp, UltraEdit->Emacs. Everything else I use is freeware or demoware that's actually useful in its demo form (trillian, proxomitron (yeah I know, Shonenware, but they don't have any mp3s for download and I don't buy CDs just to see if they're any good any more), treesize, ivt), or stuff for which we have a site licence or my job justifies a corporate licence (winzip, Visual Studio). Still looking for a free equivalent of ClipMate but I'll be buying it soon if I can't find one.
Proper try before you buy shareware is fairly dead. Authors only seem to release crippleware these days, incorrectly calling it shareware, figuring that they won't make any money if they release full software. I don't know generally how successful this approach is, but after Slashdot my second favourite site is that one that sounds a bit like AltaVista... And I *do* buy stuff after cracking it, and trying it out properly, although this often takes longer than the ridiculously short "trial period" most crippleware authors seem to think is sufficient. Stuff that expires on a particular date, set after intallation, has often expired before I get to try it. What I don't buy is stuff I can't crack, cos I can't fully try it out.
PSP - you used to be cool, man. Being proper shareware is what got you where you are today, and now you spit in your fans' eyes.
Anyone know of a shareware site that lists crippleware as such, and not calling "X Lite" (where Lite means crippled) proper shareware when it's only a thinly veiled marketing release?
Hotter sun = hotter earth.
Good job we pay these guys a fortune, that would never have occurred to me.
Disappointed
Lithium rhyme
not a haiku.
So called because it looks like a pair of boobies, and that's the first thing the boobonomers, er, sorry, astronomers said when they saw it?
from dictionary.com, of course.
"Note: The term planet was first used to distinguish those stars which have an apparent motion through the constellations from the fixed stars, which retain their relative places unchanged."
and
"Middle English, from Old French planete, from Late Latin planta, from Greek plants, variant of plans, plant-, from plansthai, to wander"
The most useful definition of "star" appears to be "Any of the celestial bodies visible at night from Earth as relatively stationary, usually twinkling points of light."
So our original definition of planet depends on our own perception. As our ability to perceive objects improves, we either need to accept the classification of new objects under the old naming scheme, or come up with a new naming scheme, which will have to be based on something that is not variable, for example mass, or size. The obvious problem here is that we are trying to come up with a new naming scheme but using the old names; that, surely, is what is ridiculous, not whether or not Pluto is a planet.
Ok I might be missing something here:
... on the horizontal axis.
. ..
> The `trivial' zeros occur at the points -2, -4, -6,
Ok, so the first non-trivial zero is where z=-2. Zeta then expands to:
1/1^2+1/2^2+1/3^2+1/4^2...or
1+1/4+1/9+1/16
So one plus a bunch of positive numbers equals zero does it? No wonder they're having a job proving it.
Ok, so suppose I run a tarpit, a spammer tries to spam me and spots the tarpit, backs off, and I don't get his spam. Where exactly was the flaw in this plan? This would appear to be better than the original aim as it means the spam doesn't arrive in the first place.
Ok, so Joe Sixpack still gets spam, but when he finds out all his geek friends have eliminated spam he'll soon start asking his ISP why he can't have this feature.
Brilliant stuff. Now where can I get my dream case from - one with all the drives and IO and switches on the SAME SIDE so I don't have to keep turning the damn thing 180 every time I want to fiddle with something?
> Email providers would set up peering relationships wherein they can share validated email addresses.
Which would work until the same sort of dodgy ISP who allows SPAM out in the first place also sets up a database of fake, validated, peered email addresses.
Also how are you going to validate someone? Would you get a message on your screen "someuser@someisp.com wants to be validated: yes/no". But you don't know who they are or what they have to say. So they send you a message: "someuser@someisp.com says 'GET A BIGGER PENIS 1-888-555-1234': validate? yes/no".
If you don't validate someone, do they remain unvalidated or are they permanently blacklisted? What if you spitefully, or accidentally, click "no" if the latter, and it isn't a spammer? If the former they can continue spamming indefinitely. If the latter the spammer simply attaches a serial number to the address and it then doesn't make any difference how you validate "spammer-65987623543@spamisp.com"
The validation can work on an individual basis - ASK, TMDA etc, but there are people you want to hear from that others don't and vice versa, so if those databases of (un)desired senders are shared, how can the contradictions be resolved?
Reply-To validation is cool, but who ultimately validates a Reply-To address except the originating ISP, which in the case of the dodgy spamming ISP simply autovalidates anything and you're still no better off.
You seem determined to call mey suggestions stupid, yet I don't see any helpful suggestions in your posts.
.com domain? Because you are an ISP, and the .net TLD was created for ISPs. I don't think there is an implication in the .net TLD that you are somehow non-profit.
Given that four billion people can't share a meaningful twenty character namespace, how do you suggest this problem is resolved?
Why shouldn't a commercial business have a
Ok, so IPs aren't fixed. You're missing my point - which is to have a system that *is* fixed and that is big enough for four billion people to share. Such a system surely cannot be impossible?
> I am an author.
And one I've never heard of, no disrespect intended. You are currently no Pratchett, Gibson or Tolkien.
When you are, and you want profit to increase, and you see falling sales and increasing downloads, it won't take more than a few seconds for you to start thinking "hey, they're stealing from my children, the terrorists!"
At the moment you're benefitting a lot from the free publicity. When this stops, the downloads will too. Or at least the complementary nature of the web offering will decrease in value.
So base stuff on unchangeable identifiers. We already have these - they're called IP addresses. Design with co-operation in mind, rather than lawsuits. Remember that a name is a shortcut to an IP address - your website is not telocity.com but 216.227.62.81 (according to "ping telocity.com"). Why is an ISP using a TLD that isn't .net? This kind of practice needs to be stamped out as it is part of the problem. You guys are *not* telocity.com - you are commercial, sure, but you are an ISP and you have your own TLD available, namely .net.
The namespace is limited so one of two things needs to happen. (1) the namespace is expanded until it is no longer limited (so you could be telocity.(state).net.us (assuming you're Yanks). Now you are protected by state laws that presumably exclude another ISP starting up in that state with the name Telocity, but wouldn't prevent a shoe shop called Telocity having a domain telocity.(same_state).co.us (again flawed, as the "co" MLD is limited but if the laws are taken into account the MLD (? medium level, as opposed to top level) can be split up accordingly.) (2) You have to share. Play nicely, etc. If you want to be known as telocity.com, that's fine until someone else wants to be known as telocity.com. You have no more (or less) right (or left (oops, too much Life of Brian)) to that name than the aforementioned shoeshop. You want to be phroggy@telocity.com? Fine, until someone at the shoeshop also wants that name. Then the options are (1) first come first served, which is a little unfair if the shoeshop can prove it predates you; (2) neither of you gets it (3) both of you get it, but for (3) to work you will need an expansion of the namespace. phroggy.shoes@telocity.com and phroggy.isp@telocity.com is one of many possibilities. phroggy@telocity.(state).us.net is another, and the other phroggy would be phroggy@telocity.(same_state).(MLD appropriate to shoe shops).us.
So you now argue that your business will break. But it was your decision to build your business on a flawed model, and when that flawed model is fixed, your business will also need fixing. I'm sure there were lots of corporate noses put out of joint when slavery was abolished, when cockfighting was banned etc; you can bet "what about the jobs that will be lost?" isn't a new argument. So build your business on an unbreakable model (no such thing, but some models are more sturdy models than others). Use globally unique identifiers. telocity.(state).us.net is one such ID if you must use names. IP addresses are others. Secondly allow for the name you choose to vary. If you want to be phroggy@telocity.com, be prepared for that to change when someone else wants to be known by the same ID. There is no way four billion people can comfortably share a namespace that is only 20 characters long, particularly if those names are meant to be meaningful.
What we need are two changes of mindset. First adherence to the rules needs enforcing one way or another. Why should an ISP known as Telocity be granted the name telocity.com, when telocity.net is perfectly ok, and uses a TLD directly intended for ISPs? This is ICANN's fault, really, just handing out domain names willy-nilly without any validation. Secondly the lawsuit mentality needs dropping in favour of the co-operation mentality. These problems can be solved if we work together. IMO all problems we have can be solved if we work together - poverty, famine, internet name clashes etc.
And costs a lot less. Company A wants a website - www.widgets.com, so they get a hosting company just as they do now and publish.
Now Company B argues that they have a claim on www.widgets.com. Ok, now ICANN puts their foot down and states that a domain name is NOT a trademark, and offers to host www.widgets.com for both companies, with links to their main web pages, and possibly some descriptive text indicating what the different possibilities are. Company B's $1000 can go towards relocating company A's pages onto another server.
so www.mcdonalds.com could end up as:
_macdonalds_ the burger chain
_macdonalds_ the Scottish kilt maker
_macdonalds_ the cafe in Lower Aldershot.
Each pays a minimal hosting fee to ICANN (because they're not hosting a vast amount of stuff - just a stack of hyperlinks)) and hosts their pages off somewhere else. It's always struck me as daft that someone thinks up a perfectly good name for a website, then some company comes along and stamps all over them just because there's a name clash.
A view from within Support...
So how does this work then?
I get no end of people calling in with a problem, claiming to know lots and the root cause isn't X, so we go full circle, prove it can't be anything else, and the customer ends up eating humble pie and having to accept that it was X.
If you provide an "I'm not stupid" line then everyone will end up calling it and you'll be no better off because everyone thinks they have a clue.
And are you HONESTLY stating that you never miss the basics? That you have never explained a problem to a colleague and mid-explanation spotted that missing semicolon, or use of = instead of ==? Support start by asking basic questions because of Occam's Razor, not because we all assume you're stupid. Troubleshooting is largely a process of elimination, and eliminating the simple stuff to start with is an essential step.
There's always the possibility that you are right of course. Give up your job, provide support that works that way and prove you are right to the rest of us by taking all our customers away from us. After all, what do we know? We've only been providing tech support for several decades.
Ok, so make it fairer by making Kramnik play blindfold, then he can use a mental representation of the board instead of having to do all that complicated image recognition stuff.
And even if you did make them evaporate they'd only condense somewhere else. The only way to get them out of the closed system would be to dump them somewhere nobody's stupid enough to go like the Moon...oh, hang on...
Or you could do what Janeway did with the Meylon freighter and send it into a nearby star.
They'd make a killing...oh, hang on...
If telephone charges were applied to property, you would have to buy the house AND rent it, then pay an additional charge every time you wanted to use a room, and it would be more expensive to use the same room at different times of the day, e.g. the bedroom would be twice as expensive at night as during the day.
How they've got away with it this long escapes me. Not only do you pay for the line to be installed and buy the phone, you then have to pay line rental and call charges. Is it just me or have I just paid for the same thing four times?
I get very few popup ads, and very little hassle from Javascript. I use Incontinent Exploder, with general security settings quite high (no javascript and other settings). Sites I want to see, and want to allow javascript for, I add to Trusted Sites list, https requirement switched off, and security settings much lower. This way if a site opens a popup, I know exactly which one it is - the one I just added to the trusted list, and can remove it. If it does open someone else's site, that site usually doesn't have javascript enabled and can't do it's dirty deeds = end of infinite porn site loops. Junkbuster takes care of cookies for me, and if I add a site to the trusted list, and it annoys me, I just add it to the Junkbuster block list as a reminder next time I visit that site. Junkbuster also blocks a lot of banner ads. If I want to see a site that uses javascript and it opens popups, I just block the popup site with Junkbuster and, ok I still get the popup but I don't get any of the crap in it! I still allow ads from some sites - /. of course, but not many.
My solution runs in 100 day cycles. The light indicates if someone has been in twice in one cycle. If you go in the room for the first time in that cycle, do nothing. If you go in the second or more, switch the light on if it is off (it shouldn't be off if this is time3+ as it will still be on from your time2). Whoever goes in on the 100th day knows if the light bulb is still off that everyone has been in once only, and declares freedom. If the light bulb is on, someone has been in at least twice, so at least one person hasn't been in yet.
Assumptions: the prisoners can count, the warden keeps a list of who's been in, the prisoners can't see the light and can't communicate with each other except on the meeting day and through the bulb, and of course that the bulb is maintained, or at least is operated by a toggle switch.
This is guaranteed to work the first time the warden lets everyone in the room in a cycle, and will continue as long as the warden duplicates people. But release was always dependent on the warden anyway. This is also tolerant of the possibility that the warden knows the plans. The idea of giving each person a number wouldn't get people out in minimum time as the warden could pick P2, P1, P4, P3, P6*, P5... on days 1,2,3,4,5,6 - this way everyone gets in but nobody gets in on their numbered day and the system fails. My way it doesn't matter what order people are picked in. (* NaN, a free man)
The only thing that bothers me about this is that it would seem to be possible to shorten it further by determining at what point the warden stopped duplicating people. e.g. can freedom be declared on day 106 at the end of the sequence 1, 2, 1, 2, 1, 2, (stop duplicating here), 1, 2, 3, 4, 5, 6...; with my system it can only be determined on day 200 at the end of the sequence 1, 2, 1, 2, 1, 2, 1, 2, 3, 4...99, 100, 1, 2,...95, 96. But if anyone switches the light off for any reason in a cycle then whoever is picked on day 100 and finds the light off will not know if the light has been on. Of course if he does know if the light has been on that changes the parameters, but one of my assumptions is that solitary confinement means precisely that - no communication of any type with the outside world (with the possible exception of the warden at meal and bog times).