World: r3wp

the developer states that his technique gives better results than 
Cairo and AGG

Well, AGG is good enough imo. Our problems imo lay elsewhere - GUI 
incompletness, GUI look (skin).

... and even more deeply ... the world goes HTML5, what do we do 
about it? :-) (I know there is still place where R3/GUI will be usefull)

Petr, this is just a question of smooth anti-aliasing - very useful 
for scalable UIs - if the algorithm is easy to implement why not 
look at it?

as for the question re. HTML5, I'm not sure what the solution could 
be - maybe an embedded Webkit and a REBOL graphics dialect that generates 
JS?

and HTML markup, of course

the problem is that if somehow JavaScript would be given equal/significant 
standing in REBOL, maybe it's harder to make the case for needing 
REBOL in the first place

Maybe there could be an alternative graphics framework at some point 
that could include WebKit modified to allow for the embedded browser's 
renderer to execute both JS and a REBOL dialect, the value of REBOL 
could become evident by way of conciseness, integration with full 
REBOL (and all the benefits that would bring).

The morphic AA looks nicer though I wouldn't see any big difference 
if the pictures were shown separately to me. Also I haven't found 
any details..just couple of screenshot comparions with Inkscape. 
The author is not mentioning a word about preformance etc. As Pekr 
said AGG have already good quality and imo there will be 'slight' 
visual difference between Morphic unless you start comparing pixels.

Also don't forget there is not only antialiasing in play. AGG uses 
very vell written vector graphic 'framework' packed with features. 
All the new algorithms presented in recent years are still missing 
that part.

But I agree. If Morphic(or whatever else) will offer faster&better 
quality AA it should be possible to write custom rasterizer/renderer 
module(simmilar as the special Flash rasterizer) so it can be used 
inside AGG framework.

here's the thread where he announces the new project and he mentions 
performance to some degree:


http://forum.world.st/The-Morphic-3-project-td2239971.html#a2239971

re HTML5, JS vs Rebol etc. I personally don't want to use Rebol instead 
of JS in a browser. It would just make me feel Rebol is crap :). 
The problem is not in the language (JS is not so bad imo) but in 
the (X)HTML/CSS/DOM design and implementation. That's the real source 
of almost all the Web dev frustration IMHO.

I agree with the headache brought on by the pieces you mentioned, 
but how to get people to transition unless you make it easy for them 
to do so?

I suppose that in an environment like Android, REBOL apps could have 
equal opportunity to HTML5 apps - so the best tech should win.

actually, I find the AS noticeably better.  the steps are pretty 
much invisible even on almost horizontal/vertical lines.

I'm sure the font rendering would be much more stable if this AA 
were used.

Yeah, I find significant difference in the almost vert/horizontal 
lines too, but I guess it's subjective - lots of people tend not 
to notice all kinds of artifacts in graphics, some of which drive 
me crazy.

very thin lines also keep much more of their color , instead of blending, 
which is why I think this AA would make a big difference in font 
rendering.

Maxim: yes the steps aren't visible too much...it looked too blury 
to me but it can be because of the LCD I'm using now.

Subjectivity: yeah, just check some forums about image/gfx filtering..lot 
of flame wars :)

we don't notice detail as much in image drawing, but in graphics 
and slow moving pictures, AA is a killer

in GLASS I had to take AA off in some text sizes for the font I am 
using.   some letters looked bold and some didn't.  On some letters 
with round corders (like S, g, D) the AA changes the width of the 
line so much its annoying to read.

font rendering: this is separate chapter imo..it would be great if 
this algorithm looks as good on text as on the images drawings.

corders=corners

The issue with AGG font rendering is that we are not using the 'clear 
type' like techinque yet. It would look much better even with current 
AA. I believe it will be part of R3 once HostKit is avaliable.

isn't cleartype patented and illegal to use in USA unless we have 
a license from MS?

on the other hand, if the hostkit allows us to use system font rendering 
routines, then all the font issues will be solved.

dunno, but what McSeem did is not exact cleartype clone iirc.

AdrianS: I don't think you can convert the Web people to Rebol easily 
just by replacing the JS in a browser. Better to build some 'cool' 
product based on Rebol with modular interface so people can play 
with it and build own 'gadgets' using dialects etc. I believe there 
are still some 'oportunities' to do it on misc embedded systems but 
first you need HostKit source and fully abstracted Rebol dll versions 
compiled for such systems.

the main feature of REBOL today when comparing to web, today, is 
the  "one language for everything" aspect of it.

people are also impressed when I build a parse dialect in ten lines 
in 5 minutes and show them that its actually spitting new structured 
data out of other data.

Building a new visual dialect from scratch in twenty lines also blows 
them away.

Let me refine what I said earlier - the idea would be to use some 
dialect that would create all the necessary bits using REBOL (JS, 
CSS, DOM elements) to be rendered by the embedded WebKit renderer. 
If, at the same time, people could run 'standard' HTML5 code in this 
same container (with integration to the hosting REBOL runtime), the 
whole would still be perceived as one integrated environment with 
HTML5 possibly being simplified into REBOL over time.

just having a webkit plugin will allow us a lot of new possibilities.

Max, could Liquid/GLASS code be translated to HTML5 without huge 
amounts of pain?

I don't mean the actual framework code, but the code that defines 
the UI and interaction for a specific application.

don't know.  I think it would require me to port liquid to JS in 
the least.

Remark is my solution for web stuff..  its basically what you describe 
above.   One engine to manage all aspects of a web application.  
Using rebol dialects compiled in real-time, you can build any data. 
 but the Dialects can be embeded within any other file, using a twist 
on the html friendly <tag> notation.

I haven't pinned the "web application" portion of it, but all the 
file building & serving aspects can already be used with remark.

building a layout dialect for Remark which is compatible to GLASS/VID 
3 is pretty easy.  its my next main Remark project.

(AFTER I've released all my stuff ;-)

when we'll have a webkit REBOL plugin, then I (and hopefull others) 
will be able to do a real-time web dev software using the remark 
framework.

I wonder what's involved in hosting WebKit - pulling down the source 
now

Looks like it's been discussed over the years, but I don't see that 
anything significant has been done, by googling.

when I went to the webkit site is was severely under-documented.

it might actually be easier to look at the chrome source code and 
starting from there!

yeah - the embedding API docs are not there at all

Adobe Flex 4 ActionScript 3 wraps WebKit, right?

a good list here:

http://trac.webkit.org/wiki/Applicationsusing WebKit

Hosting Webkit wouldn't help here: The whole point to HTML5 etc. 
acceptance is that people don't have to install another program - 
they can just use their existing web browser. Hosting Webkit would 
only help us if we want to display existing web browser code; it 
wouldn't be necessary for generating code to run in Webkit, because 
the copy of Webkit that people would be displaying your GUI in would 
usually be a separate program, often on a separate computer. And 
HTML/JS/CSS is just text - we can generate text already.

plug-in still might be the way to go. Google is introducing native 
extensions too ...

Native Client would be an interesting way to get REBOL into a browser, 
particularly once they get the portable version up and running :)