Hi all, I definitely think the auto-searching specification mechinism should be moved out into the searches so I don't have to modify search.htm - I think that's scary :) and most search writers don't want to touch it either - they just want to make a search and have it work. It would also make it much more module - no hardcoded regexp in search.htm.
It sounds like we essentially need an easy way to match these auto-searching specifications with a particular search. To have DQSD used by the masses instead of just very computer savvy people, I think we're going to need more ui to help them do things like this and define aliases etc. Some of my friends who I've converted to DQSD when I tell them how to define an alias - their eyes glaze over and they can't do it. So my wishlist is the following (not necessarily in this order): 1) move to a Windows deskbar (which Kim's got a good start on - I've been using for quite a while) 2) move the auto-search specification hardcoded-ness out of search.htm 3) make ui for setting up aliases and setting variable values - even if it's something as simple as mozilla firefox's about:config page 4) make disabled searches not take up memory - need an easy way to enable/disable searches other than renaming or deleting the searches you don't want 5) make ui for finding/adding/removing searches 6) support Google Deskbar extensions - this is something I've been rolling around in my head as something that might be cool. You know that there are going to be more and more extensions (searches) written for Google and it'd be nice to be able to use those out of the box. It would involve finding those and starting the .NET runtime if available - which would be a fun technological problem as well :) All in all, I think DQSD is nearly perfect for me - I don't need all the ui as searches, preferences, and aliases are easy enough for me to setup but for the casual user some of those things are overwhelming. So it just depends on where we all want to head (as for bloat I don't think a couple of dialogs will take up that much size and DQSD is already amazingly small - we'd be better served to get rid of disabled searches in memory if we're worried about bloat). Just my thoughts, Brent --- Kim Gr�sman <[EMAIL PROTECTED]> wrote: > Hey Shawn, > > I agree with both of you :) > > I appreciate the auto-searches (esp. the calculator, > currency conversion and > URL detection), but I guess my fear is that the > behavior is going to be > ambiguous and/or surprising. > > > If I type a phone number I want the > > reverse lookup result. > > This is a prime example. I live in Sweden - I don't > care about the > Switchboard search in the slightest, so I've built > my own, localized version > which queries www.gulasidorna.se instead. Having > DQSD detect phone numbers > per default is just a waste for me, since it uses > the wrong search. > > > If I type in an IP address I'd like to > > return information about that address (and the DNS > IP tools > > do seem to be a better fit for it). > > I would expect it to launch as an URL. Different > people expect different > things. > > To me, this is just a variation on the aliases > problem, where "." means CNN > (I've had a lot of people ask me about why they end > up at cnn.com when they > search for .NET samples), and # means Switchboard > (again). > > I don't know, I guess we could improve and rework > the design of > auto-handlers, and ship with a minimal set > (preferably the ones I'm using > ;]). But ideally, I'd like to avoid even more > long-distance coupling between > .js files via global vars. I'm looking at this more > from a maintainer- than > a user perspective. > > > I looked around the DQSD and SF sites and don't > see a > > roadmap. Are there any milestones for releases > beyond the > > next one "to include support for XP-SP2? > > You know, I was hoping we could work something out > on the list, and save it > in written form. I have some crazy ideas I want to > try out for the next > version, and I think we should make it a major > release (as Brent suggested > earlier), so we can take the liberty to work through > just the kinds of > issues you're talking about. I think for a major > version change, we can also > afford to break old behavior we no longer care for. > > I'm not sure how much time I'll have in the coming > few days, but I thought > I'd start writing up a simple design doc that we can > work from. Your > concerns are certainly well-founded, I want to start > with defining the > goals. > > We should get a Wiki. > > - Kim > > > > ------------------------------------------------------- > The SF.Net email is sponsored by: Beat the > post-holiday blues > Get a FREE limited edition SourceForge.net t-shirt > from ThinkGeek. > It's fun and FREE -- well, > almost....http://www.thinkgeek.com/sfshirt > _______________________________________________ > Archive: > https://lists.sourceforge.net/lists/listinfo/dqsd-devel > __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ Archive: https://lists.sourceforge.net/lists/listinfo/dqsd-devel
