I very much want to see this in, but I'm less sure of where it should go. I'm sure that where there is actual data, any references should go in the trailing field rather then in the storables, as references (which might, for example, be a large image or diagram) would give away information about the contents.
The issue is, I guess, that it means a lot of extra work for the network to carry a trailing field if one is not really necessary, so having one just to fill it with one line of meta-data is a bit of shame. However, it might also make sense to carry other content information with such references, so maybe it will be necessary to have a trailing field anyways. Considering that it needs to contain an unhashed key, and that there is no need for the server to see it at all, I think my vote is for trailing field anyways. If the performance of this is an issue, we will have to deal with that. On Wed, 19 Apr 2000, hal at finney.org wrote: <> > That sounds fine to me. It is good to have a resolution. > > Do you think it would be premature to create a meta-data field that > means "redirection", aka "alias", aka "pointer data". The idea is that > the content of an entry with this meta-data is a pointer to another > document in Freenet. Clients, upon retrieving data of this type, > would automatically and transparently re-do a fetch operation, using > the retrieved data as the key. The change to support this would only be > in the client, not in the nodes. > > This would be useful already so that you could insert the same data under > several different keys into Freenet, without duplicating it. This would > improve guessability while we are still figuring out how to do searching. > Plus it would be used in the future for metadata which is separate from > main data and whose content points at the main data item. > > Is it too early to do this since we have not resolved all the issues? > Or is it a step in the right direction? > > If we do implement this, should it be as a storeable header field? > Or is that revealing too much, and so it should be in this new meta-data > prefix block in the file data? > > Hal > > _______________________________________________ > Freenet-dev mailing list > Freenet-dev at lists.sourceforge.net > http://lists.sourceforge.net/mailman/listinfo/freenet-dev -- Oskar Sandberg md98-osa at nada.kth.se #!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj $/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1 lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/) _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
