hmmm. is this a paradox or is it "apples and oranges" (mutually exclusivity)? i understand the disconnect but i thought it interesting nonetheless. advocates of open source coding are likely strong supporters of privacy rights. why is transparency called for in one arena and not the other? discuss.
This is a good question, and well phrased. Probably I should think a little more before responding, but I don't have the time for that today (or any day it would seem, if you're familiar with my writing.) Basically, I think the difference is between advocating a certain position, and legislating one. No open source advocate would support a law mandating that all software must be developed using the open source model. The argument is more that some systems (especially lower level software, like operating systems, and programming languages, and web service software) are best developed in an open manner. Transparency is useful in these cases because so much is built on top of this low level software (so much built by so many different people) that it works out best to have these foundations be accessible to all developers equally. This transparency allows for interoperability between higher level systems (for instance, Internet Explorer and Netscape Navigator can both be used to view web pages, because the underlying standards that make the web are open.) These higher level systems however, don't necessarily have to be open. Netscapes code is now open, Explorer's code is not; few people care about this (in fact, Explorer is considered the better browser, even by most slashdoter's.) On the other hand, Apache (web server software) is open while Microsofts Internet Information Server is closed, and people do seem to care about this (i.e., Apache rules the market.) The reason is that server software is low level enough that you need to integrate many other products "on top" of it, and it's easier to do this if you can see what's going on down there in the foundation. Similarly with operating systems, technical people love linux because if something isn't working right (and with an OS, this usually means that you can't get some other piece of hardware or software to work right with the OS ) you can get in there and figure out what's wrong (and fix it.) With Windows or the Mac OS you are stuck having to harass a major company to do the work for you (because only they can get into the guts to do it.) But nobody minds that Microsoft Flight Simulator is not open source. It sits on top of all the other technology, so it's not necessary to be able to integrate other components on top of it (although another game - Quake - has been open sourced with the amazing result of building legions of fanatical fans who do integrate their own graphic worlds on top of the open Quake engine.) I guess it comes down to whether or not the software in question needs to be extended (enhanced) in different ways by different people. If it does, then an open source approach is probably the best way to get the job done.
Sticking with the low level / high level system split, we could say that the sort of strong privacy laws desired by the techno-libertarian-open sourcers deal with high level information. We want strong laws to protect our personal information. But at the low level, we want the same transparency as in low level computer systems. One example of this is what we call "science." Scientific research is "open" in the same sense that low level software development (linux, apache, sendmail, perl, ect...) is open: because peer review assures us of the highest possible truth level, and since so much depends on these low level systems, we have to all make sure that they are right.Another example is law. Our laws obviously have to be open or we would be at the mercy (and whim) of those who did have access to them.Personal data is a different story. It doesn't form the foundation for other structures. There's no reason for it to be open, and many good reasons for it not to be. So, for instance, the open source advocate would want strong laws protecting their video rental history data (because this has been seriously abused by government officials,) but would want laws supporting open access to data such as the human genome (not to mention the laws of gravity.) This type of fundamental information should be open because so many other things have to be built on top of it. But nothing has to be build on top of my medical history data (except perhaps a decision to exclude me from future insurance coverage due to the negative outcome of some financial equations.) So, to sum up: information should be open if it is fundamental to constucting necessary higher level systems. If not, then it shouldn't be anyone else's business.
Taking a cue from Dave's very tactfully posed question, I'll tack on this paragraph about how, of course, it's not always so cut and dry, and determining the line between what is fundamental and what is "on top" can be difficult or even impossible. Still, I think this low level / high level split can be useful in thinking about this.
|
- dave 4-03-2000 2:22 pm
This is a good question, and well phrased. Probably I should think a little more before responding, but I don't have the time for that today (or any day it would seem, if you're familiar with my writing.) Basically, I think the difference is between advocating a certain position, and legislating one. No open source advocate would support a law mandating that all software must be developed using the open source model. The argument is more that some systems (especially lower level software, like operating systems, and programming languages, and web service software) are best developed in an open manner. Transparency is useful in these cases because so much is built on top of this low level software (so much built by so many different people) that it works out best to have these foundations be accessible to all developers equally. This transparency allows for interoperability between higher level systems (for instance, Internet Explorer and Netscape Navigator can both be used to view web pages, because the underlying standards that make the web are open.) These higher level systems however, don't necessarily have to be open. Netscapes code is now open, Explorer's code is not; few people care about this (in fact, Explorer is considered the better browser, even by most slashdoter's.) On the other hand, Apache (web server software) is open while Microsofts Internet Information Server is closed, and people do seem to care about this (i.e., Apache rules the market.) The reason is that server software is low level enough that you need to integrate many other products "on top" of it, and it's easier to do this if you can see what's going on down there in the foundation. Similarly with operating systems, technical people love linux because if something isn't working right (and with an OS, this usually means that you can't get some other piece of hardware or software to work right with the OS ) you can get in there and figure out what's wrong (and fix it.) With Windows or the Mac OS you are stuck having to harass a major company to do the work for you (because only they can get into the guts to do it.) But nobody minds that Microsoft Flight Simulator is not open source. It sits on top of all the other technology, so it's not necessary to be able to integrate other components on top of it (although another game - Quake - has been open sourced with the amazing result of building legions of fanatical fans who do integrate their own graphic worlds on top of the open Quake engine.) I guess it comes down to whether or not the software in question needs to be extended (enhanced) in different ways by different people. If it does, then an open source approach is probably the best way to get the job done.
Sticking with the low level / high level system split, we could say that the sort of strong privacy laws desired by the techno-libertarian-open sourcers deal with high level information. We want strong laws to protect our personal information. But at the low level, we want the same transparency as in low level computer systems. One example of this is what we call "science." Scientific research is "open" in the same sense that low level software development (linux, apache, sendmail, perl, ect...) is open: because peer review assures us of the highest possible truth level, and since so much depends on these low level systems, we have to all make sure that they are right.Another example is law. Our laws obviously have to be open or we would be at the mercy (and whim) of those who did have access to them.Personal data is a different story. It doesn't form the foundation for other structures. There's no reason for it to be open, and many good reasons for it not to be. So, for instance, the open source advocate would want strong laws protecting their video rental history data (because this has been seriously abused by government officials,) but would want laws supporting open access to data such as the human genome (not to mention the laws of gravity.) This type of fundamental information should be open because so many other things have to be built on top of it. But nothing has to be build on top of my medical history data (except perhaps a decision to exclude me from future insurance coverage due to the negative outcome of some financial equations.) So, to sum up: information should be open if it is fundamental to constucting necessary higher level systems. If not, then it shouldn't be anyone else's business.
Taking a cue from Dave's very tactfully posed question, I'll tack on this paragraph about how, of course, it's not always so cut and dry, and determining the line between what is fundamental and what is "on top" can be difficult or even impossible. Still, I think this low level / high level split can be useful in thinking about this.
- jim 4-03-2000 3:47 pm [add a comment]