|
|
 |
|
What about safety?... |
||
What about safety?...
The safety aspects of a gridclub are complex and important. As a simple starting point we must assume three things:
- Firstly that there are adults who would seek to use the intimacy of a contributory internet system to prey on children.
- Secondly that some of those adults will be parents and even teachers.
- Thirdly that no technology is foolproof.
Inevitably this leads us to a complex set of solutions, some public, some not publicly explained, to achieve three clear aims:
- Firstly to help children, parents and teachers understand that their own vigilance, common sense and sound social behaviours are a protection in "cyberspace" just as they are at the bus stop, the home, the cinema or on the pavement;
- Secondly to have in place some adequate "early warning" mechanisms that flag obviously inappropriate material quickly so that its author can be called to account;
- Finally to have in place, ideally, some server based monitoring of textual contexts to seek inappropriate behaviours even where they are covert.
The first of these can be achieved by a simple set of guidelines of the sort used by many projects - these reinforce common sense advice from other areas of social life. For example:
People at the other end of an email are not always what they seem. Someone claiming to be "she" and 14 could be "he" and 40. This may not matter but think carefully before giving out personal information, or developing a relationship.
If you come across messages that are deliberately provocative, obscene, racist, illegal, pornographic, threatening or that simply make you feel uncomfortable, do not respond to them. In doing so you would be opening a dialogue with the person who posted the message.
...and this advice can form the basis of a good dialogue between children and parents leading to a heightened critical awareness.
The second is again relatively straightforward. As long as all contributions are from an identity that can be clearly "audited" (ie the project knows with precision who it is) then the chain of responsibility is all that needs addressing (and this can be complex when children's work both outside school and inside school is the subject of scrutiny). Pre-moderation of contributions can work, but only if delays are minimised at every opportunity - any delay at the "point of contribution" is significantly demotivating.
However, with this second "solution" everything fails if the identity of the contributor is uncertain - and this means protection of (and careful establishment of good habits to protect) passwords as well as the initial allocation of identities. Immediately this rules out on-line registration for example and places a cost on the project with a central mechanism for registration and checking, with a helpline (passwords) and so on.
The third "solution" requires a substantial server capability - context searching is not a task for a desktop PC or even a small school server. It is not simply pattern matching (much more than just looking for phrases like "and don't tell anyone" for example) and requires considerable processor power. Ideally a workng relationship with the national police agency would lead to the server being "fed" with instances of past abuse to help it to judge what behaviours to look for. Many projects manage without this last safety solution, of course.
What this means for a GridClub is expense if safety is to be implemented adequately and political danger if it is not. Although alerting users to the fact that no technology can provide a perfect protection is helpful, having mechanisms that work well when things do go wrong is an absolute requirement too.The comfort is that with a contributory project and a safe audit trail through identities, experience from around the world is that abuse is minimal by children; the danger we are guarding against therefore is adult abuse of children using the GridClub as their conduit of communication.
return to contents page