aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/docs/ClientHamr
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--docs/ClientHamr/WebBrowsers.txt74
1 files changed, 74 insertions, 0 deletions
diff --git a/docs/ClientHamr/WebBrowsers.txt b/docs/ClientHamr/WebBrowsers.txt
new file mode 100644
index 0000000..82e8a05
--- /dev/null
+++ b/docs/ClientHamr/WebBrowsers.txt
@@ -0,0 +1,74 @@
1Web browsers are a fact of life, both in and outside of the virtual
2world. Inside the world, SL people love to play youtube videos on their
3virtual big screen TVs to show their friends. Business people want to
4gather around a web browser at meetings. MOAP is a thing. Other social
5media is a thing, most of it web based, and people might want to share a
6tweet by showing their virtual laptop to their virtual friends in the
7same virtual room. Outside of the world, SL and OS based grids have
8always used a web page for their front screen in viewers. Some of the
9tools are web based.
10
11So we need a web browser that can run on meshes and prims, as well as in
12windows (separate or internal windows). The problems are (legion) ...
13
14
15Web rant
16--------
17
18Damn the web sucks.
19
20This bit will be filled in next time I feel like ranting about the web,
21but damn it sucks really badly.
22
23Bloated, slow, horrid, inconsistent, over complicated, ... , it sucks.
24
25
26Less of a web rant
27------------------
28
29OK, so in typical SL based viewers, the web browser component makes up
30about one third of the entire viewer package. At first, when Java in
31the web browser was more of a thing, matrix-RAD tried to work with the
32web browser. Now we are (hopefully) designing the next big thing to
33replace the web, so we get to do things differently. So now we can
34(hopefully) treat the web as a second class citizen that we begrudgingly
35support as a legacy thing. Note that we are proudly using HTTP as the
36bulk transport agent, it's the web content standards that suck.
37
38If matrix-RAD was any indication, I can replace everything that a fully
39bloated web browser can do, in something the size of a typical web page
40of today, and STILL be better! lol
41
42
43Actual plans
44------------
45
46So I think we could use multi flavours of web browser.
47
48First of all, try to use and encourage the matrix-RAD style stuff as the
49UI of choice for in world stuff. It should work across the network
50transparently as matrix-RAD does. This was my HTML replacement that
51worked within web pages. Now we can dispense with the web pages all
52together. The original matrix-RAD included a plan for a translation
53layer, any web browser that didn't like Java could get a "legacy" simple
54web page version. We could do something similar, or go up a step and
55translate Lua into ECMAScript.
56
57Include something small like Dillo or NetSurf as the built in web
58browser. It should be able to spread itself across in world objects,
59meshes and prims, as well as run in internal or external windows.
60
61As the next step up, an optional module could replace Dillo / NetSurf
62with a bigger browser component, based on one of the main ones, uzbl for
63instance. Hopefully it can be spread across an in world object somehow.
64
65Finally, the user can always decide to show any given URL, or already
66displayed web page, on an external web browser of their choice, with
67their default browser being the top choice. Which naturally should be
68able to use the in world matrix-RAD type stuff via the translation layer
69(uzbl should be able to as well, NetSurf maybe, if I remember they where
70adding a ECMAScript engine, or had already? Dillo I don't think does
71ECMAScript at all).
72
73
74