Sawood Alam

> With this conception, if a client has some other list of archives they want to use, the client should instead run its own aggregator service. Sending a list of...

> In fact, it may be interesting to have aggregators talk to each other so that they answer requests about which archives they know about, a start in some sort...

> I think having an aggregator be 'fixed' over a set of archives has other practical considerations.. for example, the aggregator may be optimized for access (reusable connections, etc..) to...

On the matter of performance optimization, I remember various pieces of information available in LANL's archive registry that tells the sort-order, pagination status and similar things about each archive that...

> I am not convinced that an aggregator w/o a registry would be all that useful, as users will need to get a list of archives from somewhere, and then...

@galsondor, MemGator makes this task of selecting what archives you want to aggregate from very easy, but you will have to provide a custom archives list file as a flag...

While Firefox does not support custom elements natively just yet (the work is in progress), the polyfill for the same is in a pretty good state. It does already support...

Since Mink is Chrome specific, which has excellent support of Web Components, we don't need Polymer for this (though we can use it if we want to). That will be...

Have the custom user-agent in [square brackets] if you plan to append extra info to the existing browser user-agent, otherwise something like the following will be a preferred way. `Mink:OS:ARCH:VERSION`

Because that's not commonly present in the existing browser user agents.