World: r3wp

Try searching for "beer" alone ... I found out how to do suggested 
links

featured links

Looks like you can have a great deal of fun setting up the search 
engine parameters

If you can get to them from the root, then they are fair game, unless


.....they have a rel=nofollow......We have that on a few simply because 
they duplicate content (eg viewing a script, viewing a script in 
color, downloading a script


....Mailing list -- best to index either the individual posts (http://www.rebol.org/ml-display-message.r) 
or the complete threads (http://www.rebol.org/ml-display-thread.r) 
but not both.


....you may get a __lot__ of duplication when spidering the AltME 
archive as every post has a URL, but we display in batches of 50.....So 
perhaps only spider URLs like
http://www.rebol.org/aga-display-posts.r?post=r3wp291xNNNN
when NNNN is 1, 51, 101, 151, etc....


....I think You already have indexed the ML as on REBOL.,net and 
Carl's latest 300 AltME messages, eg
   http://mail.rebol.net/cgi-bin/mail-list.r?msg=45305
   http://host4.altme.com/altweb/rebol3/chat771.html

It would be better _not_ to have index those; it just creates duplicates 
once you have indexed the equivalents on REBOL.org (especially as 
the AltMe last 300 goes out of date so quickly).

Tell me what is unclear there!

[oops -- that was meant for Graham, privately]

I wondered if that were the case

I've already indexed the mailing list on rebol.net so I guess I should 
avoid ml-display-thread.r and display-message.r

I think the REBOL.org archive is a better place to send people than 
then REBOL.net one for the ML (it has threads not just messages in 
id sequence).
So perhaps index REBOL.org's then drop the REBOL.net one?

not sure how to drop something ...

Must be possible --- whe pages or sites are removed, or if they later 
post robots.txt exclusions.

interesting, I don't see it documented but it does use site:

site:rebol.org sunanda

Interestingly this search engine has a REST interface so you can 
wrap your own custom search around it.  95k pages and still going 
...

has anyone did anything about openId with rebol ?

Any pointers of how to incorporate PayPal payment method into one's 
website?

paypal has an example. It's pretty easy.

http://www.diigo.com


Looks like an interesting way to add annotations to rebol.com etc 
and share them with others without actually altering the pages

If you add a public annotation, then any diigo enabled browser will 
see them.

umm.. perhaps you have to do something ..
how about this ?  http://diigo.com/0ca2d

looks pretty nice, but the question (as for so many other web apps) 
is - will it be there tomorrow if I put all my stuff on it?

I think this fear of having to redo a bunch of effort is keeping 
quite a few sites from really breaking through - you only feel confident 
about these kind of features from already established sites

that's why my view is that you should always own your stuff - the 
cloud should only be used as a temporary cache, or when you need 
grid computing  - just some infrastructure that can be easily substituted 
with another

actually for grid computing, I meant when you need huge power - otherwise, 
a grid of personal computing devices could constitute a powerful 
enough grid

I guess if one wanted this sort of security ... one would want your 
private data copied from their S3 store to your own somehow.

they've been running for 4 years now .. so I guess that's a little 
reassurance

@Gabriele - I'd like to hear your comments on what this article says 
(in sec 2.3) about allowing some basic expressions in the view, wrt 
your templating engine?


http://www.simple-is-better.org/template/#why-if-for-calculations-etc-are-necessary-in-the-view

In my approach, the mapping between the data and the "view" is defined 
by a dialect. I guess, one might want to make at least part of that 
mapping part of the view; this does not change the fact that the 
template file can then be simple HTML.

Formatting

 to me does not seem a templating issue though. It's about localization 
 and customization (different users will want different formats for 
 dates, and almost everything else). So, this is a completely separate 
 axis. As Carl often says... the problem is multidimensional, reducing 
 it just to "model" and "view" is not really going to work.

(for example, when he says that "the calculation of a subtotal belogs 
to the view", i disagree that the actual code to compute the sum 
belongs to the view. the view should allow for declaratively defining 
such kind of things. page subtotals, for example, was a feature i 
intended to have in PDF Maker 2 tables - since you don't know where 
the table is going to be split at, the engine has to be able to do 
the work for you.)

has anyone here worked on a firefox plugin before?

any tips for wget tool allowing to go via form authentication? We 
need to make off-line copy of one site. The site works only from 
one PC at a time from our organisation, it seems it uses cookies 
or so ....

use the cookie-enabled http protocols on rebol.org (by Cyphre IIRC)

what for? :-) we have wget like tool for rebol?

building the wget is trivial in rebol... but you still need the http 
calls to support cookies.  though we can probably probe the headers 
for cookie returns in newer R2 versions.

Petr, I think you need to take a look into curl stuff for this.

MaxV spots an interesting product for website generation. Written 
in REBOL:
   http://synapse-ehr.com/forums/showthread.php?139-Rebol-CMS

From Dario Giacomelli
Registered on the R3 chat as Dario

wow, interesting

http://it.linkedin.com/in/ingdariogiacomelli

anyone game to test it out?

Many of the concepts are a lot like how I started the first phase 
of my CMS three years ago, but the implementation is a bit different

Is anyone here familiar with the current state of WYSIWYG web designers? 
Are there any decent current free ones? I'm only familiar with Kompozer 
(the new Nvu) and have heard good things about WebMatrix. This is 
for a person familiar with Frontpage, but I'm trying to disuade them 
from using it (for all our sakes), and don't want to encourage piracy 
of Dreamweaver or Expression Web.

I haven't used Nvu or Kompozer for years, and only use code editors 
for my web work (or don't do web front end stuff at all).

There's been a lull in KompoZer development, but it's reasonable 
now

It still has a lot of quirks, though

Are there any decent alternatives to it?

Sadly, no

Would using any modern word processor then saving as HTML be any 
worse than Frontpage? ;) I wouldn't do a complex site this way, but 
for a couple of static pages it may suffice.

Just don't use Frontpage. If you end up having to edit the pages 
later by hand, it's faster simply to build them by hand.

If Frontpage is the alternative, WebMatrix looks a lot better. However, 
it's focused on developers and complex backends. If the goal is simple 
web pages for a common user, KompoZer is probably better, and more 
in the ballpark of Frontpage

I just remembered that there's a successor to KompoZer:

http://bluegriffon.org

It seems to be usable by now, but I have no experience with it