![]() |
|
Home | Help | Status | Forums | Glossary | Account
|
log in |
Message boards : Number Crunching : Some more piracy (I mean privacy)
Author | Message |
---|---|
Assuming the crew cannot post to the Wish List anymore, comments may be directed here. | |
ID: 4811 | Rating: 0 | rate:
![]() ![]() ![]() | |
Yep, I'm not allowed to post a reply on the Wish List now. | |
ID: 4813 | Rating: 0 | rate:
![]() ![]() ![]() | |
Yep, I'm not allowed to post a reply on the Wish List now. Same here (tried it to see if the different rankings might be in effect since I dropped down to "Seaman"). | |
ID: 4817 | Rating: 0 | rate:
![]() ![]() ![]() | |
Yep, I'm not allowed to post a reply on the Wish List now. Okay, now thread creation is completely locked down also. Interestingly, when logged out I can still see the original post in the wish list, but not any replies? | |
ID: 4821 | Rating: 0 | rate:
![]() ![]() ![]() | |
Since comments to the privacy issue should be posted here, here are my comments. | |
ID: 4828 | Rating: 0 | rate:
![]() ![]() ![]() | |
As there seems to be issues dealing with underage students and their identity, is there anyway to just not show the student names (and possible signatures) on the posts unless your logged in? That is certainly a good place to start, even if it does not solve the whole problem. And is should be relatively easy to implement, once I dig in to it. It's easy to tell if the viewer is not logged in, and then we would not show the sig. How much of the author block would be shown, if any at all? Perhaps a big shadow for the avatar? And just the userid to distinguish the person? Or is even that too much? There is also some concern about who can see what even internally. An administrator at one school may not like the idea that students at another school can see information about his/her students. But I'd like to facilitate collaboration between students at different schools, and working with someone you only know as a shadow and a number is not as useful as working with someone you know a little about, even if it's a screen name and avatar. So another idea I'm toying with is having both real names and screen names, and both a real picture ("headshot") and an avatar. So there are now 3 levels of information about the person. Then we need some way of determining what to show to who. If it's a fixed, closed policy, then it's not very useful. (eg. no student can see anything about a student at another school - hence it's hard for them to collaborate). It would be better if it could be configured to allow for different policies. But then it gets complicated, and you have to be sure there are not any loopholes or back doors. But not showing personal information to someone who is not logged in may be the best place to start, so I'll likely try it out. Thanks. ____________ -- Eric Myers "Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats | |
ID: 4829 | Rating: 0 | rate:
![]() ![]() ![]() | |
So another idea I'm toying with is having both real names and screen names, and both a real picture ("headshot") and an avatar. So there are now 3 levels of information about the person. Then we need some way of determining what to show to who. If it's a fixed, closed policy, then it's not very useful. (eg. no student can see anything about a student at another school - hence it's hard for them to collaborate). It would be better if it could be configured to allow for different policies. But then it gets complicated, and you have to be sure there are not any loopholes or back doors. How I see this working is as follows (a rough idea): A data base field would need to be added to differentiate the schools with a unique school code. That unique school code would be applied to each student who registers so that they are coded to their school and cross linked to the topics that they can see. Students would only see and post threads to the topics that are assigned the that school code as well as only seeing their school information. Sub adminisrators could create topics only for their school code. The Master Administrator would be able to assign multiple codes, through the administration screen, to the topics to allow multiple schools to participate if needed which would allow veiwing for multiple assigned schools. So for example, sub administrators from school "A" could create topics only available to students of school "A" and sub administrators from school "B" could create topics available only to students of school "B". If there was a project that involved multiple schools, then the master account administrator could assign the topic to both school "A" and "B". By only allowing the master administrator access to cross link, it reduces the problem of privacy issues between the schools and limits who is able to make the changes. It does sound like this would be a coding nightmare to address the concern however but it limits who can allow people from various schools to see each other's information. ____________ ![]() | |
ID: 4832 | Rating: 0 | rate:
![]() ![]() ![]() | |
I would propose a slightly modified structure based on this one. I could envision that the actual educational process would be more topic-based rather than school based. Thus, a structure that allows sub-admins to sign up entire schools for specific topics might need to be added as a layer (perhaps in a way that sub-admins submit proposed topics that only the master admins can create?). I am also curious if the issue of student-age/minor status has been considered. Many high school-aged students in the US have the legal right to allow others to view private material (all 18 year-olds and some 17 and 16 depending on the state). I wonder if this adds yet another layer (or at least database field) to the mix? | |
ID: 4834 | Rating: 0 | rate:
![]() ![]() ![]() | |
Right now I'm using the BOINC teams to represent the schools. This fails if a teacher or student attends more than one school (as I did in high school), but let's put that aside for now. | |
ID: 4837 | Rating: 0 | rate:
![]() ![]() ![]() | |
The big question is this. How much personal information is going to be required and of what type? | |
ID: 4841 | Rating: 0 | rate:
![]() ![]() ![]() | |
The big question is this. How much personal information is going to be required and of what type? All I want to show is basically what we have here, but with real names, a real picture, your school instead of your team, and your state in place of your country (it's a US based project, at least for now). The point being that this facilitates remote collaboration with someone you have not met. A little bit of personalizing information adds to the experience. It doesn't have to be a lot, just enough to distinguish one person from another by a little bit. Of course as we are demonstrating now that can also work even if people don't know their "real" names, so it may be that a "screen-name" and an avatar are enough. My thinking though was that for a school activity using your real name is more appropriate, and using a screen-name for school sounds sketchy. Maybe that's backwards. ____________ -- Eric Myers "Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats | |
ID: 4842 | Rating: 0 | rate:
![]() ![]() ![]() | |
I would propose a slightly modified structure based on this one. I could envision that the actual educational process would be more topic-based rather than school based. This is sort of what's happening, and sort of a part of the problem. The original project which started this is [[w:QuarkNet]], in which schools maintain a cosmic ray detector, upload their data to a central server, and then can download data from all detectors to do interesting investigations. It's sort of a YouTube for cosmic ray data, and it's a great idea. So now the folks who started QuarkNet have invited other physics experiments to join in and do something similar. Instead of a distributed network of detectors, the ATLAS and CMS experiments at CERN and the STAR experiment at Brookhaven will allow students to download some of their data and use that for investigations. LIGO will make data available from environmental sensors like seismometers and weather stations. So there would certainly be a variety of topics, and schools could participate in one or several of them. But the complication I am worrying about is that different schools might have different ideas about how much identifying information can be shown about a student, even to other schools. It may be that each school can only see participants from their own school. It may be that students can only interact with their lab partner and the teacher and nobody else. (I think this is how QuarkNet is set up now, but don't quote me on it.) The thing is, I've seen from BOINC projects how valuable it can be to allow for wider remote collaboration. Einstein@Home is a great example of that, for many reasons, but even what goes on here on this ship shows what is possible (people working together to figure something out). I'd like to find a way to make that a part of the whole package that schools can use. But the privacy issue is going to be important, and in the end maybe it's a show-stopper. I hope not. ____________ -- Eric Myers "Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats | |
ID: 4847 | Rating: 0 | rate:
![]() ![]() ![]() | |
The thing is, I've seen from BOINC projects how valuable it can be to allow for wider remote collaboration. Einstein@Home is a great example of that, for many reasons, but even what goes on here on this ship shows what is possible (people working together to figure something out). I'd like to find a way to make that a part of the whole package that schools can use. But the privacy issue is going to be important, and in the end maybe it's a show-stopper. I hope not. I agree with you completely that the remote collaboration that is characteristic of BOINC projects would defintely add to the educational experience of such projects. Perhaps there can be a lessoned learned from social science research here that might help with privavcy issues. Alternative ID's (or pseudonyms, etc. which seem to be one issue that you are focused on) are one way that anonymity or confidentiality are maintained in epidemilogic data, but others are sometimes used. The one that I think might apply best here is the use of aggregation to eliminate deductive identification in such datasets (e.g., only allowing reports to go down to the county level rather than census blocks or tracts). For the problem here, one might envision a combination of the pseudonym method and aggregation. For example, if you required a public pseudonym (and avatar) from users in addition to the actual names and pictures as you have suggested in earlier posts and then combined this with some actual public geographic location then you might gain the best of both worlds: privacy of student data and better collaborative efforts as participants can place the pseudonym with a real, though general, location (e.g., perhaps as specific as the participating school, or more generally the county or state location). | |
ID: 4849 | Rating: 0 | rate:
![]() ![]() ![]() | |
To be honest, I'd keep it at a bare minimum and for options like avatars and signatures, leave it up to each individual school to decide if they will allow the items to be displayed or if they want to leave them turned off. | |
ID: 4860 | Rating: 0 | rate:
![]() ![]() ![]() | |
Message boards : Number Crunching : Some more piracy (I mean privacy)
Home | Help | Status | Forums | Glossary | Account
|