<div dir="ltr">2008/10/7 Arun Raghavan <span dir="ltr"><<a href="mailto:arunisgod@gmail.com">arunisgod@gmail.com</a>></span><br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> On Tue, Oct 7, 2008 at 5:35 PM, Mikkel Kamstrup Erlandsen<br> <<a href="mailto:mikkel.kamstrup@gmail.com">mikkel.kamstrup@gmail.com</a>> wrote:<br> [...]<br> <div class="Ih2E3d">> Yes, it makes sense to keep the breakage between 1.0 and 2.0 as<br> > minimal as possible.<br> <br> </div>My main thoughts on this ("borrowed", as you can no doubt see):<br> <br> 1) Release early, release often: get the spec out and let people have<br> something stable to use<br> <br> 2) Make one to throw away: it's okay if some things need to change<br> radically over the API after the first version. If comaptibility can<br> remain, that's good, but let's deal with that particular problem when<br> we get to it. As Mikkel says, if there is a compelling reason to use<br> the newer API, clients will get (re)written switch to it. And there<br> are always ways to make that transition less painful.</blockquote><div><br>I understand this as agreement to some degree...<br><br>Based on the responses here I'll try and add the proposal where SearchChanged signal passes along the relevant content- and source categories. Then when we have it written down it might be easier to discuss.<br> <br>Regarding the ontology the most parts of the PIM components will have to be removed before 1.0 as they really can't be made to make sense. Perhaps Evgeny would care to elaborate :-) ?<br></div></div><br clear="all"> Link-by-id will be part of 1.0 unless we discover some problems. Without the nested queries (2.0 material) it will be of limited use though. Still useful though. I'll write up a page for this asap so we can get it reviewed before 1.0.<br> <br>My hope is that we can roll 1.0 around Nov. 1st.<br><br>-- <br>Cheers,<br>Mikkel<br> </div>