World: r3wp

include JavaScript, VB, Ruby, Python

Reichart: It's smaller, faster and IMO more streamlined. Really good. 
And it's price is very fair.

I can't wait for word processors to no longer exist.

They waste my time.

Teh formality of it all.

Reichart, seen the Etoilé desktop? Early concepts of it shows how 
apps are banished and everything is made up of smaller bits which 
you put together to an "app". You do it on the fly.

so if you want to write a document, bring up a piece of "paper" which 
uses a type setting service. if you want to print, you call a printing 
service. if you want to spell check, you call a spell checking service. 
if you want dynamic content in your document, you can call services 
which can respond to various input with an output, tied dynamically 
to the paper, such as the current date, or a customer database entry

whether it will be like this in the final version, I don't know. 
it seems to be application oriented right now.

reichart: http://www.stud.fit.vutbr.cz/~xcapmi00/etoile/

More accurately: http://www.stud.fit.vutbr.cz/~xcapmi00/etoile/task_a.html

Reichart;  We (a dev team) duked it out way back with Word for DOS. 
 It was a

complete waste of our time.  We handed management a text file with 
some fairly

complex technical information and a "beautiful" word doc, full of 
near gibberish.

Management picked the gibberish doc...it looked better, to pass up 
the line.  We

giggled, then informed him of the insider joke, and spent the day 
wrestling with

Word to make the real tech spec "look good".  Sex sells.  When we 
wanted a

faster network, the document started with "Your pipe is very small" 
 No manager
wanted a small pipe! Very effective.

I'm glad I don't have to deal with this kind of management... Brian, 
I've read stories about how network equipment purchases were based 
on how many blinking lights there were on the front panel and how 
an admin created a fake light panel to get his manager off his back, 
because the manager complained that the equipment "wasn't doing anything".


I know it can't get this simple, but management should never be a 
position you could get hired directly into... it should be a position 
one can only advance to through plain skill.

Well, to be fair.  I wouldn't really want techs running a large corporation. 
 Skill sets

are skill sets and techs are good at techie and (most) bosses are 
good at money
(and requisitioning bigger pipes).

Etoilé .... interesting...

Your pipes are very small
 LOL, and smart...

I've been following the Silverlight and DLR developments a lot this 
week. It seems to me that this would be a good way to get REBOL in 
the browser. You could market a REBOL based on the DLR as a /Services 
integration library. Rebol Universal Services Transport, a way to 
bind all of those Iron languages to light-as-air REBOL/Services :)

I don't know something about etoile seems like its not really changing 
the actual workflow of use.  I still sense a "software" in the GUI... 
but I agree its much more pervasive.

I would like it to completely ban the use of actual apps and just 
rely on services to do everything. And then on top of that, make 
the whole damn thing scriptable. It would be a hell of a bold move, 
but I think it would work.

They are talking about banning the concept of files, and rely fully 
on persistent stores, but there is still not a solution on how to 
do that.

why services... services are softwares.

I think you will like elixir.   :-)

its close to etoile... but it also loose the false desktop metaphor.

when I say close, I mean in many underlying motivations.

What's the difference between a service and an app? 

PickOS used a DB as it's file system.

and its fully scriptable  :-)  antidote is actually used BY the GUI 
instead of coding many of the things internally... for example, all 
hotkeys are actually within an external file with character and command 
scripts which applied when that key is pressend and no focus is detected.

Services are small. They do one single thing and they do that one 
thing very well. OSX has them and they've been there for ages, but 
the system only relies on them for manipulating things in apps, not 
to construct ad hoc apps themselves. How often have you not wanted 
a cool feature from program X in program Y and vice versa? This would 
do the trick.

elixis is much more granular than etoile.

The service versus app distinction is a big gray area IMO. No great 
answers here, but if you don't provide "preconfigured service bundles" 
a.k.a. applications, how does Grandma use them?

its much simpler.

when you see the relationships.

in etoile they still consider a "desktop" to be a viable and intuitive 
interface... when in fact it isn't

I'm totally behind the concept of modular scriptable software though 
Henrik.

Gregg, compare it to how you use your real life items, like a paper, 
a pencil, eraser, etc. Grandma does not want to know that she has 
to open Word or some <weird open source name app> to write a document. 
She wants a piece of paper. A service will give her a piece of paper 
as a view port. On the technical side, you don't load a bajillion 
features into memory that you don't need, only a viewport and a text 
renderer.

hearing and seeing alan kay in the last few days has only solidified 
many of my ideas.  One capital sentence he repeats:


Adults have too many context and concepts, to be able to think simply 
and understand the most basic ideas.   kids have a "fresh" take on 
things... and they are much better at chosing simple things.

elixir, for example will seem like a bizare work environment for 
some, I guess, but its sooo simple, it needs no real learning... 
actually, the only thing people will have to learn is the panels 
which people will add to interface the internals... but at least 
we will be able to SEE the relationships and associations they have 
with the "innards"

so once you "understand" the concept of an self managing atom of 
information...

you understand EVERYTHING.

Maxim, yes, it's because we go to school. When I went to public school 
I liked electronics and wanted to work with it. I found it fun and 
could even put together little circuits that did fun stuff. When 
I became an engineer, the fun went away and everything became immensely 
complex, so what I had learned as a kid, I lost.

are you comming at the devcon henrik? I don't rememebr?

nope

The goal of elixir, is to get many people writing VERY small procedures 
and tasks.

darn... would really have liked to go over elixir with you in person.

 A service will give her a piece of paper as a view port.

 -- But what features does the service provide, and when does it become 
 an application? i.e. how do you save something, find something you 
 wrote before, add spell checking, print something, etc. These are 
 things that can be answered in different ways, and I think we'll 
 see a lot more big changes in software in the next 10 years.

the ubiquity is the key.

a pencil which actually stores data on a sheet.

a real sheat.

electronic paper already exists and is being sold by sony as a small 
book reader... no back lit.  0 consumption until you edit the page.

things like that and gestural workflow.

Gregg, you don't save anything. Like the piece of paper, the information 
is persistent the moment you write on it. How they want you to access 
documents, I'm not sure they are done working that out. I suppose 
that services that are tied together exist in contexts, so that putting 
text writing, printing and spell checking together makes sense, while 
bitmap painting and spell checking does not.

anyhow... I've got to go... but I am started to be exited for devcon... 
things are starting to "work"

I remember actually that Amiga Inc. were some of the first to start 
talking about persistent storage in desktop environments. Now everyone 
else but Amiga Inc. are doing it. :-)