<div class="gmail_quote">Hello Travis,<br><br>On Mon, Jun 21, 2010 at 18:16, Travis Reitter <span dir="ltr"><<a href="mailto:travis.reitter@collabora.co.uk">travis.reitter@collabora.co.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> Really, EDS isn't ideal, but it's gotten a lot better in the last year<br> or two, and the people working on it now seem to have a good idea of<br> what needs to be done to improve it.<br></blockquote><div><br>i see, well EDS improvements might help on a legacy tip..<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> <div class="im"> > What's the issue with using plain files from your point of view (if<br> > there is any)?<br> <br> </div>You could do that, but it wouldn't surprise me if it's a lot slower than<br> e-d-s. Eg, if you've got 300 contacts, and you want to search, you'll<br> have to open 300 files, read them, marshal them into your in-memory<br> structure, then start the actual search.<br></blockquote><div><br>for 2000 files with 1kB each that would be 2MB, yes, i can see that being an annoyance on slow machines.. I see your point here, too.<br> </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> I think a database really makes sense in the context of contacts.<br> They're really optimized for retrieving and slicing data, which address<br> books spend a lot of their time doing.<font color="#888888"><br></font></blockquote><div><br>exactly. Maybe Tracker can do this for you?<br>It's fresher and looks like a useful candidate to me..<br></div></div>