World: r3wp

Lucid Hydra 200 - HW allowing load balancing between GPUs, even if 
there are 2 or more cards from different vendors in one system?

http://www.anandtech.com/video/showdoc.aspx?i=3646

hmm, even OSes as Haiku get Qt4 - the power of open-source?

100 cores,  55 w

http://www.tilera.com/products/TILE-Gx.php

about above cpu... this figure is impressive, if true:  • Up to 200 
Tbps of on-chip mesh interconnect

with 100 cores running very small apps in parralel... I can finally 
see something built to run elixir  :-)

yes, was thinking the same. It could also be a elegant sales offer 
to intel :-))

that is what I want to port rebol on.  I am sure Carl would like 
rebol on that  :-)

yes, or erlang. hopefully task! will arrive soon...

Well, I haven't found an instruction set description anywhere. So 
chances are high that it's Vaporware.

because they use "partners" the way I see it is that you have to 
sign an agreement in order to receive the dev SDK/API.

I have developed such things 10 years ago with runtime reconfigurable 
deadlock free communication network (we called it worm-routing). 
And getting C compiled down to such a thing is not easy because every 
CPU needs a good access to memory. Either local (than how to exchange 
data?), global (how to do locking) etc.

Beside a bunch of other problems coming up. But maybe they have really 
made it in which case we should see a bunch of announcements in a 
short time.

Up to 200 Tbps of on-chip mesh interconnect    that sounds like enough 
to make it pretty fluid to me  :-)

hehe :-) http://www.slashgear.com/nvidia-ion-le-directx-10-hack-suggests-purposefully-crippled-features-2962210/

Could this be a good news?


Type designers and Web designers have reached a consensus on a format 
specification for embedding fonts on the Web. Mozilla is already 
including support for the font format in Firefox 3.6, and wide adoption 
could come sooner than many expected.


http://arstechnica.com/web/news/2009/11/web-open-font-format-backed-by-mozilla-type-foundries.ars?utm_source=rss&utm_medium=rss&utm_campaign=rss

If it gets added to HTML5, that would be great. And some year IE 
might support it :)

Sooner than you think.
 One of the two designer groups is from MS

Microsoft and some new UI concept ideas - http://www.istartedsomething.com/20091106/microsoft-college-tour-09/

http://openofficemouse.com/

You have got to be kidding...

:-D

Get one .. no, get two!

the OOMouse, the first multi-button application mouse
Those who forget the 1990s are doomed to repeat them.

Its this very thinking that makes OO less than a desirable application(s)...

Everytime I've used it it seems bloated, bogged down and slow..

new Google Go language - http://www.techcrunch.com/2009/11/10/google-go-language/

intended to be a systems programming language

Another new Mobile OS - this time from Samsung. The OS is called 
Bada - http://www.osnews.com/story/22476/Is_There_Room_for_a_New_Mobile_OS_

A lot of negative talk about too meny mobile OS's already but the 
phone market is a lot different than PC Market...with contracts every 
one or two years most people trade in their phone at most every 2 
years and each time re-evaluate the field (some of anyway). This 
gives new guys a chance to enter the fray. If you got somethng unique 
you may gain share quickly.

It seems Win Mobile is doomed. MS woke up too late. Sony and HTC 
completly moved off of Win Mobile. Nokia probably too. And now MS 
has mostly lost Samsung too ...

More on Bada here - http://www.linuxfordevices.com/c/a/News/Samsung-Bada/
, http://www.bada.com

HTC is still on Win Mobile, but it doesn't matter: They write their 
own UI and just port it from OS to OS. Right now HTC has Win and 
Android phones (the vast majority being Win), but they could drop 
both easily if they want.

Pretty cool optical technology coming next year:
http://techresearch.intel.com/articles/None/1813.htm

http://blog.chromium.org/2009/11/2x-faster-web.html


Google experiment with a new protocol to speed webserver transfers 
up about 2x from HTTP.

hmm, where does google lead us? Thru averaging web technologies to 
advancements? so screw the standards, let's go the google way.

Hehe, ppl seem to smash them, having enough of the We-are-gods-work-for-us-as-we-have-nothing-innovative-ourselves-we-just-ripp-others-work 
...

http://www.osnews.com/comments/22486

I don't get that statement. This is a network protocol, which they 
came up with (aka: innovation) and they are providing as a proposed 
open standard with an open source reference implementation (read: 
giving, not taking). Network protocols require both ends of the connection 
to talk the same protocol, so if they want others to talk to their 
servers, they need for the others to understand and use the protocol. 
Which they are providing for free.


This is an example of best practices, and of Google being a good 
corporate citizen (which they aren't always). The protocol is even 
pro-privacy (take that, China and Iran!). It even seems easy enough 
to implement that we could build it into the HTTP scheme of R3. So, 
where is the downside of this, exactly?

I see commenters claim that this is a ripoff of Opera Unite (it isn't 
in any way anything like Opera Unite), and others claim that Google 
is trying to get others to do their work (misunderstanding the concept 
of open standards).

The downsize is in google, being "always in beta", dictating "standards", 
just from the multibillion position

Sorry, "dictating" standards? There is no dictation here. It's a 
proposal, a Request For Comment (though less formal than an RFC yet).

if it would come from any other company, it would get ignored. The 
same goes for Chrome - it is in no way unique, just a rip-off of 
others, plus minus few things done differently (tasks per tab).

So you can't see it? There is no concept in google, just slow domination. 
They are either dumb enough, not having some top level gurus/designer, 
not having complete idea, or they are way too much clever - throwsing 
various things at us, slowly leading to their total domination.

If people start getting pissed off at Google for actually having 
and using the money to fund research *which they are giving away*, 
then we are doomed. The protocol looks good so far. If it sucks, 
it should get ignored (see SOAP). If it doesn't suck, it should be 
adopted. There is no reason to give a crap about "domination" because 
Google isn't trying to control network protocols, just to improve 
them for all.


It makes sense to complain about their domination in search and advertising, 
and their kowtowing to local tyrants at times. But this is not one 
of those cases. They are giving the protocol away for free. They 
aren't tying it to a platform like MS. It is even encrypted end-to-end, 
so the tyrant governments can't easily read it. They even are providing 
an open-source reference model, *and* asking for advice on implementation 
strategies.


There is no down side for us here. The only upside for them is not 
exclusively for them: Anyone who implements a protocol like this 
would gain the same benefit. For that matter, there is no way for 
them to gain from this over anyone else in the only ways which they 
do dominate: search and advertising, or even online apps. If they 
were closing this protocol then maybe they could gain over others, 
but they are opening it so it is only gain for all.

I read the same OSNews coments that you did, and these people need 
to learn to read the article before commenting.

yes, you can see it in reactions. I have much deeper respect to proprietary 
guys like IBM or MS lately. Their technologies give me total picture 
of what I can use in our company. Well designed stuff. Those things 
might be complex, but well engineered (WebSphere). I will always 
be one refusing the servility. I have the same problem with Apple 
(Jobs). There is no problem with their products, but the problem 
is with the attitude and it starts to show. Even if Jobs introduces 
new icon on the desktop, he would get fanatical following. I can 
see the same wave of google fanatics emerging. The so called "google 
culture" is ... hyped.

They can't even patent this protocol since they have already released 
the description of how this works *and* reference code.

In comparison to MS or IBM I can see no top designers in google, 
having actually a vision, a complete one. They throw things here 
or there, they can do whatever (almost unlimited resources), and 
you can bet, that they lead us to lock-in ....

I don't give a crap about Google culture. If the protocol is good 
(and it looks good so far) I'll write the R3 support for it.

The lock in is in mentality. All the cloud crap, not having the date 
at my location everything on server. Welcome matrix :-)

An open protocol doesn't have to be used with Google servers.