Re: Nw application development
Alex Russell <[email protected]> Sat, 13 Sep 2003 18:00:01 -0500
| Newsgroups | gmane.comp.windows.devel.netwindows |
|---|---|
| Organization | burstlib.org |
| Message-ID | <[email protected]> |
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Saturday 13 September 2003 6:18 am, Christian Boulanger wrote: > > Note that this is incremental change, and not > > invasive at all > > unless you decide to write a pure netWindows app (like > > Christian's). > > ... Yes, I begin to notice that my approach is quite different > from what I understand the developers of NW themselves focus on... Actually, it's not. As you outline below, there seem to be two "classes" of NW apps emerging (incrmental usage, or full-bore NW apps), both of which I believe deserve first-class support. Hence our support of both inline and programatic constructors on almost every widget in the toolkit. > There seem to be two different directions one can go with NW: > > 1) The page-centered approach of richt and diverse websites, which > uses mainly inline-constructors and loads pages into windows as > needed. This approach also support the idea of "degradable widgets" > for browsers that cannot display nw widgets correctly. > > 2) The data-centered approach, which loads the complete application > and a lot of the data at startup, and then works from there. This > approach uses mainly programmatic constructors and relies on > structured, mostly uniform data queried and sent by the server. In > this setting, "degradability" is of course impossible to achieve. > > This is what I am trying to do... But the problem might be that I > need to do almost everything programmatically, from populating > tables with data to constructing forms. I don't know how well nw > supports this approach at this point. As well as any other DHTML toolkit you're going to use (if not better). Under the hood of the "inline" constructors is almost always a "regular" constructor and a bunch of method calls on the widget that just hide the programatic approach in a declarative style. Doing things programaticaly in NW is always going to be a bit more verbose and difficult, but will ultimately be more flexible. I have a STRONG desire to see both incrmental and ground-up style applications be successful in using NW to get where they're going. > I know that I want to stick with this approach because I don't want > to have to load each bibliographic reference from the server when I > want to do something with it - I want to have them in memory on the > browser, and only have server interaction whenever something gets > changed.... That's my idea of a real web application - to avoid > roundtrips at all costs.... ( Another reason is that I really like > javascript programming now as much as I begin to dislike > PHP-style). NW is designed, in part, to play to this kind of app design. Your app isn't out of the scope or conception of what I imagine NW to be good for, but rather one of the first publicly-available apps that's using the ground-up approach. That's all = ) Regards - -- Alex Russell [email protected] BD10 7AFC 87F6 63F9 1691 83FA 9884 3A15 AFC9 61B7 [email protected] F687 1964 1EF6 453E 9BD0 5148 A15D 1D43 AB92 9A46 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2-rc1-SuSE (GNU/Linux) iD8DBQE/Y6FxmIQ6Fa/JYbcRAl8oAJ0cmJ04Ujll1AHXBN29TkmSujCa3gCfXB++ vc4RwzYoXfKDARYfFMC86yI= =MXnt -----END PGP SIGNATURE----- _______________________________________________ The netWindows developers list: [email protected] http://netwindows.org/mailman/listinfo/devel_netwindows.org