I guess it's just part of the geeky mindset that when seeing something complicated, the discussion turns immediately to its flaws.
But for just a second, I'd like to appreciate how *freaking awesome* it is that GPL app like Open Office exists. Sure it has problems, but it's also an incredibly hard space to work in. The Microsoft monopoly is based very much on the office formats, and the dedication of Sun and the Open Office team to build this complex thing is creating all sorts of freedom for the rest of us. Microsoft knows this, and that's why they expended so much effort trying to mess up the formats... but it's not working, here we have a GPL tool that reads the newest Microsoft format.
It's pretty hard to function on the internet without some ability to deal with office documents. In fact, I suspect Open Office is creating more freedom and competition than Firefox. Writing a browser, strangely, is not *that* hard. I can think of ten or so browser projects, but only a few office suites.
Microsoft controls about 90% of the desktop/laptop operating system space. Just by common sense, that's a monopoly. It's been found to be a monopoly legally as well.
Having a monopoly (in the US) is legal, you just can't leverage it to gain advantage in some other domain, like say, Flash technology.
For example, Microsoft creating a Photoshop killer and marketing it normally is fine. Bundling it with every OS install is not ok.
Microsoft wanted Yahoo in order to force-push Silverlight in front of all those email users. HTML and javascript are terrible for Microsoft, since they were cross platform (witness Microsoft ignoring IE with its terrible implementation all those years, hoping HTML would just die off).
Microsoft wants to push their proprietary Silverlight "web" to retake the glorious control they had pre-internet vs. today's picture with some degree of cross platform support and EEK competition!
Flash has quite a lead, but Microsoft's control of 90% of the world's machines with push-force-update software is pretty darn handy.
Mod parent up! In particular, standards are about competition, since it allows you to pluck out one vendor and put in another... making the vendors to competing on price, features, etc. Microsoft seems to have forgotten how to compete by making technically superior products, and instead tries to just make the competition not happen.
I think the slashdot tribe, in its heart, loves open, bruising competition, with the best technology coming out on top. That's why microsoft has such a terrible rep on slashdot... they have just gotten accustomed to winning a market by cheating instead of competing, and that deeply irks at the slashdot/geek ethos. I realize it's foolish to try to summarize the slashdot tribe with one word, but I think "competition" is the closest.
Let's see, what do these all have in common: TCP/IP, SMTP (email), HTML, XML, JPEG,... oh that's right, they're all STANDARDS. The amazing growth and value of the Internet is entirely due to standards allowing many different pieces of software work together. Microsoft screws up standards as much as possible to try to lock their customers in. Also if you implement a standard you have to start competing. A monopoly is all about avoiding competition since your customers can't leave, so that's another reason Microsoft avoid standards. The Microsoft strategy is so patently lame and contemptuous of their customers, I expect it will blow up in their faces someday, especially as phrases like "locked in" and "proprietary format" penetrate the mind of the buying public. I think many normal people get that it's handy that the JPEGs coming out of their digital camera will work in many places/device since its a standard.
Actually the Chandler project (basically the source of the CalDAV standard) -- provides both a CalDAV server and clients for Windows, Mac, and Linux. See http://chandlerproject.org/ -- they just released their beta. I very much look forward to the day that calendaring benefits from competition and interoperability as we have with HTTP, SMTP, etc., as opposed to everyone being locked into Microsoft's weird/patented/no-competition calendaring island. Anyway, real open calendaring is just getting going, and CalDAV is a big part of that.
Rather than track everyone in the world, you could just have a reputation system based on email addresses. Anyone can make up a brand-new email address at any time, but then it won't have an impressive reputation. The main barrier to this is that currently, the From: on an email is easily forged, so the notion of reputation does not work. Domain Keys and SPF attack exactly this problem -- enabling a durable notion of reputation for a domain. If we could get a few more people to implement those, it's going to make a huge difference. We're so close... yet the default attitude seems to be to exclaim how hopeless it all is. Domain Keys Now! http://dkim.org/
Actually making your general purpose computer secure is quite hard. What this shows is that we should have a little USB doohicky with a little display and a couple buttons that we use to sign/authenticate important transactions. For the final step of the transaction, you look at the display on the device and enter your pin on the device to confirm the transaction. The hardware for this is not that expensive, but the banking system has such organizational inertia, it's going to take them years to clue in to just send/sell the customer a device. You just cannot "secure" the whole PC with the confidence that you can for the little USB device.
We'll all have them someday! You heard it here first! I am not a crazy old man!
Forgery is very much related to spam. Here's how it works:
spammy email -> spammer's source domain
Think of this as extra good data for Spamsieve and the RBLs to use. If you are a spammer, how are you going to send a million emails without associating them all with a spammy domain? A different domain for each email? For each 100,000 emails? And of course it'll be easy to give a bad spam score to either a domain that was registered with the last week or a domain for which the world has not seen valid email previously.
I think the DKIM people are setting a low expectation, but in fact this will be a HUGE step forward for spam elimination. The spam filters do a pretty good job now with really pretty crappy data. DKIM exposes the whole filtering stack (Spamsieve, RBL databases,...) to a source of far better data.
Parent is incorrect! DKIM accounts for forwarding and other use cases... exactly the cases for which SPF has problems. Seriously.. do you think the IETF working group works on this thing for years, and doesn't think of some case you thought of from "A quick read of the RFC" ?
DKIM in conjunction with SPF and client filtering has a real chance to make Spam be not such a problem. It enables reputation system for senders, and Spammers will show up in such a system in a pretty obvious way. It will make far better data available for the spam ranking systems, so however well they work now... well they'll have much better (and unforgeable) data to use.
The Sun phone is interesting because JavaFX is open and open source, while Flash and Microsoft Silverlight are both very proprietary. I would love a rich-UI environment for phones devices etc. that is a standard and has an open source implementation. That said, JavaFX has to prove it actually works well (I've seen Java run great on teeny devices, so I know it's possible... probably by just dedicating a meg of ram to it)... ultimately the openness of the platform is potentially a real game-changing feature compared to Adobe/Microsoft lock-in.
Hellooooo everyone, the issue is not that historically the JVM takes forever to load or that Flash can be annoying. The key trend is that AJAX/Html are hitting a limit, which leads to all this recent energy about Flash/Flex as a better way to construct rich UIs, witness Microsoft's Silverlight.
Now both Flash and Silverlight are totally proprietary. That's a huge problem. If one of them were to "win" and get a zillion developers... well gee it seems historically this has led the winning vendor treat us all like crap. That's a real bummer when you have this expensive time investment in your website, but it's locked in to some vendor's intellectual property. The only other open rich alternative -- SVG + Javascript -- appears a bit dead.
So what's neat about this announcement, is that it's a Flash workalike that's OPEN SOURCE. If it were to "win"... that would be awesome. Even if the technology is just ok, the openness would make it worthwhile. Just think... it could work properly on Linux and phones and what have you. This is very much like what happened with HTML originally. Just an ok spec, but the openness catalyzed all sorts of growth and competition.
Another way this could work out is that it bluffs Adobe into opening up Flash, which I figure would be just as good an outcome. The key is to not be stuck developing your expensive web app, but with some vendor controlling the underlying technology.
The problem with buying software is that it puts the vendor in the sad position of adding stupid features and witholding bug-fixes in a hope of getting you to upgrade. Really you just wanted the old version with a few bugfixes. With rental, they can keep a small crew keeping it up to date, and we get to send them $20 a year or whatever to keep it running.
Indeed, customer-annoying moves like changing the file format seem much a feature of the sell-once model. With the rental model, they just want to keep you happy with the software as it is.
Rather than getting annoyed at Dell or working to circumvent their technology, this would be a great time for a little transparency. Printer manufacturers should be able to sell whatever dorked combination of printer and ink, but the package should be clearly labeled ot indicate what sort of ink it takes and what the expected costs are.
"The market" can work things out if the consumers have the information. You can imagine a series of printers that cost more but take "commodity" ink being popular with some people while some people like the "minimum up-front cost" Dell/Lexmark path.
Nutrition labels, car MPG labels, appliance Energy Star labels... these are all cases where a little government arm-twisting gives us a much more competitive and responsive market.
The theme is: use transparency to bring to the surface costs being shifted to the consumer. If they still choose to consume... so be it. This same great strategy is the basis of yesterday's article about
requirng labels on copy-restricted materials.
The Stanford CS Library has lots of docs of interest for beginning CS students (C, Perl, linked lists,...), and the Binky Pointer Video is a funny animated introduction to pointers
I teach CS at a university with lots of code-writing exams, so I'll try to speak up in their defense.
Problem solving with the computer is the most important thing, so we have lots of coding homeworks to exercise that material. That's fine, but it still leaves a place for code-writing questions.
A code-writing question is a very concentrated way for a student to show that they really understand a topic such as recursion or pointer manipulation. Ideally, the answer should be short, and obviously you don't want to grade on syntax or other superficial stuff. Forcing the student to visualize the data-structure and come up with solution code on their own shows off their understanding in a way that multiple-choice questions just don't get.
Here's a paraphrase of one of my favorite old exam problems from a languages course were we talked about low-level memory manipulation in C....
For the following function, 'a' points to a malloc allocation heap of the given length in bytes. The elements of a linked list (1, 2) may be allocated in the heap somewhere. Search the heap for the list, and if found return a pointer to the '1' element, or return NULL if not found. Assume that pointers and int are the same size, 'length' is a multiple of sizeof(int), NULL is represented by 0, and the list struct is not padded.
struct list { int data; struct list* next; }
struct list* find12(char* a, int length) { // your code here
get ready to have your insurance company jack your rates for going over 65mph.
The above comment is way too simplistic -- the black box allows the insurance to be priced more accurately. Higher prices for drivers more likely to cause accidents and cheaper prices for drivers less likely to cause accidents.
The black box could measure speed, abruptness of turning, stopping, and so on. Feed a pile of historical data into a correlator and figure out which behaviors are correlated with accidents and price the insurance accordingly. Of course this is not good news for the bad drivers, but it's great news for the good drivers.
The beauty of it is that it's largely self-selecting. Suppose I found the Transparent Insurance Co. and we use black boxes to set our rates. All the good drivers use us since our rates are lower for them. The bad drivers and privacy-at-all-costs people can keep using the no black-box insurance co, and of course their rates are higher, since they attract all the bad drivers.
I think it's going to be very hard for that prediction to stand up as the years pass. Partly it's the hardware getting faster, but a lot of it is the Hotspot/compile-on-the-fly technology getting better. At some point, Java is not interpreted at all. Java bytecode is just a distribution language -- by runtime it's compiled to native anyway.
Privacy is not always a good thing, and PayPal and ebay are great examples.
Would you hire a babbysitter who would not tell you their name? The more truth you know about someone, the more trust you can have to participate in a transaction with them.
Privacy is great for some things, but building a community of trust is not one of them. We need to move beyond the knee-jerk privacy=good mindset to a more balanced view seeing the costs and benefits of privacy and transparency for particular situations.
It seems like the earlier ebay fraud story is exactly the sort of thing that national id card would be handy for, even if it were voluntary
The problem with ebay transactions is that it is hard to be sure of the identity of the other party. It would be great to read an ebay ad that included the URL for the seller's national id, so you really knew who you were dealing with. Of course they could still find a way to lie, but we could make the lying more expensive.
Privacy has benefits, but costs as well. Sometimes a little transparency is just the thing.
Truth never damages a cause that is just -- Mohandas Gandhi
The basic assumption of the patent system goes something like this: companies invest resources to create valuable intellectual property. Because the new technology is better than what used to exist, the community is happy to pay a little more for it, and so the company can make back their investment.
However in post-Internet world, it seems to me that the value of a new technology to the community is split -- some of the value derives from the intellectual property created by the company, and some of the value is from the network effect from the participation of the community.
How much of the value of TCP/IP is due to TCP/IP's technical merit -- the quality of its framing algorithm etc, and how much is due to the network effect of everyone using TCP/IP? Or how about HTML -- a valuable technology, but hardly anything special as far as technical merit or intellectual propety goes. The value of HTML derives mostly from its network effect.
I think these examples show that technical merit is not that imporant. The technology needs to be competent, but that's it. What does matter is that the technology as widely adopted as possible. Put another way, the n^2 term in Metcalf's law swamps the contribution of technical merit.
Question:
What do you think the percentage are for value created in the technical merit and network effect for Internet technologies such TCP/IP and HTML? How does the W3C balance adopting new technologies vs. attempting to create the largest possible network effect?
What's happening is that the spammer is behaving like an ass, and so does not want to reveal their idenity -- they want "privacy" for their actions in this case. The forgery is just a symptom of their desire for privacy.
What's interesting is that this reverses the usual role of privacy in these discussions. Mostly privacy is regarded like fresh air or something -- the more the better. In reality, like most things privacy has many bad effects as good.
I look forward to the day I can program my mail system to only accept email from real signed identities -- i.e. no privacy for people sending me email. This sounds scary at first since the privacy==good thing is so conditioned, so you need to think about it a bit.
Slashdot Mirror
I guess it's just part of the geeky mindset that when seeing something complicated, the discussion turns immediately to its flaws.
... but it's not working, here we have a GPL tool that reads the newest Microsoft format.
But for just a second, I'd like to appreciate how *freaking awesome* it is that GPL app like Open Office exists. Sure it has problems, but it's also an incredibly hard space to work in. The Microsoft monopoly is based very much on the office formats, and the dedication of Sun and the Open Office team to build this complex thing is creating all sorts of freedom for the rest of us. Microsoft knows this, and that's why they expended so much effort trying to mess up the formats
It's pretty hard to function on the internet without some ability to deal with office documents. In fact, I suspect Open Office is creating more freedom and competition than Firefox. Writing a browser, strangely, is not *that* hard. I can think of ten or so browser projects, but only a few office suites.
Microsoft controls about 90% of the desktop/laptop operating system space. Just by common sense, that's a monopoly. It's been found to be a monopoly legally as well.
Having a monopoly (in the US) is legal, you just can't leverage it to gain advantage in some other domain, like say, Flash technology.
For example, Microsoft creating a Photoshop killer and marketing it normally is fine. Bundling it with every OS install is not ok.
Microsoft wanted Yahoo in order to force-push Silverlight in front of all those email users. HTML and javascript are terrible for Microsoft, since they were cross platform (witness Microsoft ignoring IE with its terrible implementation all those years, hoping HTML would just die off).
Microsoft wants to push their proprietary Silverlight "web" to retake the glorious control they had pre-internet vs. today's picture with some degree of cross platform support and EEK competition!
Flash has quite a lead, but Microsoft's control of 90% of the world's machines with push-force-update software is pretty darn handy.
Mod parent up! ... making the vendors to competing on price, features, etc. Microsoft seems to have forgotten how to compete by making technically superior products, and instead tries to just make the competition not happen.
... they have just gotten accustomed to winning a market by cheating instead of competing, and that deeply irks at the slashdot/geek ethos. I realize it's foolish to try to summarize the slashdot tribe with one word, but I think "competition" is the closest.
In particular, standards are about competition, since it allows you to pluck out one vendor and put in another
I think the slashdot tribe, in its heart, loves open, bruising competition, with the best technology coming out on top. That's why microsoft has such a terrible rep on slashdot
Let's see, what do these all have in common: TCP/IP, SMTP (email), HTML, XML, JPEG, ... oh that's right, they're all STANDARDS. The amazing growth and value of the Internet is entirely due to standards allowing many different pieces of software work together. Microsoft screws up standards as much as possible to try to lock their customers in. Also if you implement a standard you have to start competing. A monopoly is all about avoiding competition since your customers can't leave, so that's another reason Microsoft avoid standards. The Microsoft strategy is so patently lame and contemptuous of their customers, I expect it will blow up in their faces someday, especially as phrases like "locked in" and "proprietary format" penetrate the mind of the buying public. I think many normal people get that it's handy that the JPEGs coming out of their digital camera will work in many places/device since its a standard.
Actually the Chandler project (basically the source of the CalDAV standard) -- provides both a CalDAV server and clients for Windows, Mac, and Linux. See http://chandlerproject.org/ -- they just released their beta. I very much look forward to the day that calendaring benefits from competition and interoperability as we have with HTTP, SMTP, etc., as opposed to everyone being locked into Microsoft's weird/patented/no-competition calendaring island. Anyway, real open calendaring is just getting going, and CalDAV is a big part of that.
Rather than track everyone in the world, you could just have a reputation system based on email addresses. Anyone can make up a brand-new email address at any time, but then it won't have an impressive reputation. The main barrier to this is that currently, the From: on an email is easily forged, so the notion of reputation does not work. Domain Keys and SPF attack exactly this problem -- enabling a durable notion of reputation for a domain. If we could get a few more people to implement those, it's going to make a huge difference. We're so close ... yet the default attitude seems to be to exclaim how hopeless it all is. Domain Keys Now! http://dkim.org/
Actually making your general purpose computer secure is quite hard. What this shows is that we should have a little USB doohicky with a little display and a couple buttons that we use to sign/authenticate important transactions. For the final step of the transaction, you look at the display on the device and enter your pin on the device to confirm the transaction. The hardware for this is not that expensive, but the banking system has such organizational inertia, it's going to take them years to clue in to just send/sell the customer a device. You just cannot "secure" the whole PC with the confidence that you can for the little USB device.
We'll all have them someday! You heard it here first! I am not a crazy old man!
Forgery is very much related to spam. Here's how it works:
...) to a source of far better data.
spammy email -> spammer's source domain
Think of this as extra good data for Spamsieve and the RBLs to use. If you are a spammer, how are you going to send a million emails without associating them all with a spammy domain? A different domain for each email? For each 100,000 emails? And of course it'll be easy to give a bad spam score to either a domain that was registered with the last week or a domain for which the world has not seen valid email previously.
I think the DKIM people are setting a low expectation, but in fact this will be a HUGE step forward for spam elimination. The spam filters do a pretty good job now with really pretty crappy data. DKIM exposes the whole filtering stack (Spamsieve, RBL databases,
Parent is incorrect! DKIM accounts for forwarding and other use cases ... exactly the cases for which SPF has problems. Seriously .. do you think the IETF working group works on this thing for years, and doesn't think of some case you thought of from "A quick read of the RFC" ?
... well they'll have much better (and unforgeable) data to use.
DKIM in conjunction with SPF and client filtering has a real chance to make Spam be not such a problem. It enables reputation system for senders, and Spammers will show up in such a system in a pretty obvious way. It will make far better data available for the spam ranking systems, so however well they work now
The Sun phone is interesting because JavaFX is open and open source, while Flash and Microsoft Silverlight are both very proprietary. I would love a rich-UI environment for phones devices etc. that is a standard and has an open source implementation. That said, JavaFX has to prove it actually works well (I've seen Java run great on teeny devices, so I know it's possible ... probably by just dedicating a meg of ram to it)... ultimately the openness of the platform is potentially a real game-changing feature compared to Adobe/Microsoft lock-in.
Hellooooo everyone, the issue is not that historically the JVM takes forever to load or that Flash can be annoying. The key trend is that AJAX/Html are hitting a limit, which leads to all this recent energy about Flash/Flex as a better way to construct rich UIs, witness Microsoft's Silverlight.
... well gee it seems historically this has led the winning vendor treat us all like crap. That's a real bummer when you have this expensive time investment in your website, but it's locked in to some vendor's intellectual property. The only other open rich alternative -- SVG + Javascript -- appears a bit dead.
... that would be awesome. Even if the technology is just ok, the openness would make it worthwhile. Just think ... it could work properly on Linux and phones and what have you. This is very much like what happened with HTML originally. Just an ok spec, but the openness catalyzed all sorts of growth and competition.
Now both Flash and Silverlight are totally proprietary. That's a huge problem. If one of them were to "win" and get a zillion developers
So what's neat about this announcement, is that it's a Flash workalike that's OPEN SOURCE. If it were to "win"
Another way this could work out is that it bluffs Adobe into opening up Flash, which I figure would be just as good an outcome. The key is to not be stuck developing your expensive web app, but with some vendor controlling the underlying technology.
The problem with buying software is that it puts the vendor in the sad position of adding stupid features and witholding bug-fixes in a hope of getting you to upgrade. Really you just wanted the old version with a few bugfixes. With rental, they can keep a small crew keeping it up to date, and we get to send them $20 a year or whatever to keep it running.
Indeed, customer-annoying moves like changing the file format seem much a feature of the sell-once model. With the rental model, they just want to keep you happy with the software as it is.
"The market" can work things out if the consumers have the information. You can imagine a series of printers that cost more but take "commodity" ink being popular with some people while some people like the "minimum up-front cost" Dell/Lexmark path.
Nutrition labels, car MPG labels, appliance Energy Star labels ... these are all cases where a little government arm-twisting gives us a much more competitive and responsive market.
The theme is: use transparency to bring to the surface costs being shifted to the consumer. If they still choose to consume ... so be it. This same great strategy is the basis of yesterday's article about
requirng labels on copy-restricted materials.
The Stanford CS Library has lots of docs of interest for beginning CS students (C, Perl, linked lists, ...), and the Binky Pointer Video is a funny animated introduction to pointers
Problem solving with the computer is the most important thing, so we have lots of coding homeworks to exercise that material. That's fine, but it still leaves a place for code-writing questions.
A code-writing question is a very concentrated way for a student to show that they really understand a topic such as recursion or pointer manipulation. Ideally, the answer should be short, and obviously you don't want to grade on syntax or other superficial stuff. Forcing the student to visualize the data-structure and come up with solution code on their own shows off their understanding in a way that multiple-choice questions just don't get.
Here's a paraphrase of one of my favorite old exam problems from a languages course were we talked about low-level memory manipulation in C....
For the following function, 'a' points to a malloc allocation heap of the given length in bytes. The elements of a linked list (1, 2) may be allocated in the heap somewhere. Search the heap for the list, and if found return a pointer to the '1' element, or return NULL if not found. Assume that pointers and int are the same size, 'length' is a multiple of sizeof(int), NULL is represented by 0, and the list struct is not padded.
get ready to have your insurance company jack your rates for going over 65mph.
The above comment is way too simplistic -- the black box allows the insurance to be priced more accurately. Higher prices for drivers more likely to cause accidents and cheaper prices for drivers less likely to cause accidents.
The black box could measure speed, abruptness of turning, stopping, and so on. Feed a pile of historical data into a correlator and figure out which behaviors are correlated with accidents and price the insurance accordingly. Of course this is not good news for the bad drivers, but it's great news for the good drivers.
The beauty of it is that it's largely self-selecting. Suppose I found the Transparent Insurance Co. and we use black boxes to set our rates. All the good drivers use us since our rates are lower for them. The bad drivers and privacy-at-all-costs people can keep using the no black-box insurance co, and of course their rates are higher, since they attract all the bad drivers.
Quake [X] will never be written in Java.
I think it's going to be very hard for that prediction to stand up as the years pass. Partly it's the hardware getting faster, but a lot of it is the Hotspot/compile-on-the-fly technology getting better. At some point, Java is not interpreted at all. Java bytecode is just a distribution language -- by runtime it's compiled to native anyway.
Privacy is not always a good thing, and PayPal and ebay are great examples.
Would you hire a babbysitter who would not tell you their name? The more truth you know about someone, the more trust you can have to participate in a transaction with them.
Privacy is great for some things, but building a community of trust is not one of them. We need to move beyond the knee-jerk privacy=good mindset to a more balanced view seeing the costs and benefits of privacy and transparency for particular situations.
It seems like the earlier ebay fraud story is exactly the sort of thing that national id card would be handy for, even if it were voluntary
The problem with ebay transactions is that it is hard to be sure of the identity of the other party. It would be great to read an ebay ad that included the URL for the seller's national id, so you really knew who you were dealing with. Of course they could still find a way to lie, but we could make the lying more expensive.
Privacy has benefits, but costs as well. Sometimes a little transparency is just the thing.
Truth never damages a cause that is just -- Mohandas Gandhi
However in post-Internet world, it seems to me that the value of a new technology to the community is split -- some of the value derives from the intellectual property created by the company, and some of the value is from the network effect from the participation of the community.
How much of the value of TCP/IP is due to TCP/IP's technical merit -- the quality of its framing algorithm etc, and how much is due to the network effect of everyone using TCP/IP? Or how about HTML -- a valuable technology, but hardly anything special as far as technical merit or intellectual propety goes. The value of HTML derives mostly from its network effect.
I think these examples show that technical merit is not that imporant. The technology needs to be competent, but that's it. What does matter is that the technology as widely adopted as possible. Put another way, the n^2 term in Metcalf's law swamps the contribution of technical merit.
Question:
What do you think the percentage are for value created in the technical merit and network effect for Internet technologies such TCP/IP and HTML? How does the W3C balance adopting new technologies vs. attempting to create the largest possible network effect?
What's interesting is that this reverses the usual role of privacy in these discussions. Mostly privacy is regarded like fresh air or something -- the more the better. In reality, like most things privacy has many bad effects as good.
I look forward to the day I can program my mail system to only accept email from real signed identities -- i.e. no privacy for people sending me email. This sounds scary at first since the privacy==good thing is so conditioned, so you need to think about it a bit.
more slashdot on privacy vs. transparency
Nick