connotea-code-devel — Discussion and devlopment of the core Connotea code and plug-ins.

RE: [Connotea-code-devel] Customising look and feel

[Lund, B. <B....@na...>]

Hi Sambit,

The majority of the HTML code is generated by the components themselves.
These are places in the page layout by code such as:

${component:SearchForm}

in the page templates.

However, there is a generic Inc component that you can use to include simpl=
e
common elements in pages.  For example, the following code:

${component:Inc(filename=3Dwelcome)}

Would include the content from a file named welcome.inc in the html
directory.

Ta,
Ben



-----Original Message-----
From: con...@li...
[mailto:con...@li...]On Behalf Of Sambit
Patnaik
Sent: 10 January 2006 23:24
To: con...@li...
Subject: [Connotea-code-devel] Customising look and feel




  We were trying to understand the way we customize the 'look and feel' of
the Connotea bookmarks so that it looks uniform with the rest of the
website.

  The way we understood it is that the in the Connotea code, the Components
module has the various perl files (for ex header.pm, footer.pm) which
construct the html and output to the browser. So the code in the Header.pm,
Footer.pm,Login.pm etc can be modified to have a customized header, footer
or login box.

 Is the above understanding correct? Or is there any alternate way in which
we can create common templates such as header and footer which can then be
just invoked in the .tpl file.

I would really appreciate if someone could answer this question.

Thanks

-Sambit
=20





=20

***************************************************************************=
*****  =20
DISCLAIMER: This e-mail is confidential and should not be used by anyone wh=
o is
not the original intended recipient. If you have received this e-mail in er=
ror
please inform the sender and delete it from your mailbox or any other stora=
ge
mechanism. Neither Macmillan Publishers Limited nor any of its agents accept
liability for any statements made which are clearly the sender's own and not
expressly made on behalf of Macmillan Publishers Limited or one of its agen=
ts.
Please note that neither Macmillan Publishers Limited nor any of its agents
accept any responsibility for viruses that may be contained in this e-mail =
or
its attachments and it is your responsibility to scan the e-mail and=20
attachments (if any). No contracts may be concluded on behalf of Macmillan=20
Publishers Limited or its agents by means of e-mail communication. Macmilla=
n=20
Publishers Limited Registered in England and Wales with registered number 7=
85998=20
Registered Office Brunel Road, Houndmills, Basingstoke RG21 6XS  =20
***************************************************************************=
*****

[Connotea-code-devel] Customising look and feel

[Sambit P. <sa...@ua...>]

  We were trying to understand the way we customize the 'look and feel'
of the Connotea bookmarks so that it looks uniform with the rest of the
website.=20

  The way we understood it is that the in the Connotea code, the
Components module has the various perl files (for ex header.pm,
footer.pm) which construct the html and output to the browser. So the
code in the Header.pm, Footer.pm,Login.pm etc can be modified to have a
customized header, footer or login box.=20

 Is the above understanding correct? Or is there any alternate way in
which we can create common templates such as header and footer which can
then be just invoked in the .tpl file.

I would really appreciate if someone could answer this question.

Thanks

-Sambit=20
 =20
=20
=20



 =20

[Connotea-code-devel] OPML file generator for social bookmarking RSS feed bundles

[Tony H. <ton...@gm...>]

Hi -

Just a quick note to say that I have posted an OPML generator for bundling
user/tag combination RSS feeds from delicious, connotea and simpy (so
far...)

http://blogs.open.ac.uk/Maths/ajh59/005502.html
tony

Re: [Connotea-code-devel] remote user

[Martin F. <ma...@ne...>]

Sambit Patnaik wrote:
> Hi,
> 
> I was exploring the source code for Connotea and I wonder if there is a 
> way by which the REMOTE_USER var (Apache's) can be used by Connotea to 
> manege the authentication.
> 
> Your answers will be helpful!

You could probably write code to use that, but:
- you probably will have to call $r->user on the Apache request object 
instead, as I'm not sure that REMOTE_USER will be set under mod_perl
- you will of course need some way to authenticate that the user is who 
they say they are
- that authentication needs to be pretty snappy as this occurs on every 
request

The code to rewrite is in Bibliotech::AuthCookie.

The current code has two authentication methods.

Upon user login, a cookie is sent that contains the user ID, username, 
Unix login time, and an MD5 hash based on this data plus a secret key. 
Bibliotech::AuthCookie validates the cookie data by simply verifying the 
MD5 hash, so there is no database access. This is the primary method.

Because users may want to fetch user-personalized data (e.g. include you 
private bookmarks) from applications that cannot show a login page (e.g. 
RSS reader) and/or do not store cookies properly, we have an 
EXPERIMENTAL secondary system which is also coded in 
Bibliotech::AuthCookie; if the URL path information begins with /auth/ 
then the browser is challenged for credentials and once it supplies the 
username and password these are checked against the database. We are not 
fond of this method, and thus may remove it in future versions, but it 
does work.

Martin

[Connotea-code-devel] remote user

[Sambit P. <sa...@ua...>]

Hi,

I was exploring the source code for Connotea and I wonder if there is a
way by which the REMOTE_USER var (Apache's) can be used by Connotea to
manege the authentication.

Your answers will be helpful!

Thanks

-Sambit

[Connotea-code-devel] Distributed Social Bookmarking

[Tony H. <ton...@gm...>]

I just posted a couple of doodles about the possibility of social
bookmarking interoperability, with an outline case scenario ( Distributed
Social Bookmarking <http://blogs.open.ac.uk/Maths/ajh59/005336.html>) and a
preliminary comparison of google and delicious ( Social Bookmarking
Interoperability <http://blogs.open.ac.uk/Maths/ajh59/005332.html>, also
commented on here <http://www.downes.ca/cgi-bin/page.cgi?post=3D32659>).

Would there be any interest/use in setting up an informal working group
looking at interoperability in social bookmarking systems?

ALso - has anyone had a play with IBM's
Dogear<http://www.acmqueue.com/modules.php?name=3DContent&pa=3Dshowpage&pid=
=3D344&page=3D1>
?

tony

[Connotea-code-devel] Connotea under Windows

[C.B.Rasrendra <C.B...@ru...>]

Hello folk,

is there people do Connotea installation under Windows 
machine  ? Ofcourse  with Apache, mysql, dan perl 
installed to that machine.
I only have windows machine and , probably, i will try to 
port Connotea runing under my windows machine.

Regards,

CB Rasrendra

Re: [Connotea-code-devel] Tags and RSS: a proposal

[Fil <fi...@re...>]

> We actually put the tags in a Dublin Core subject tag in RSS as well.

Yes I know, and so do I generally. However <dc:subject> is not satisfying
for "tags", in my view, in the sense that it lacks a URL.

> I believe you're talking about the content:encoded portion, which is a 
> copy of the XHTML that appears on the normal web page. That's really for 
> the convenience of people looking at the feed in a newsreader that has 
> an embedded HTML viewer; we didn't actually expect it to be parsed for data.

Well that's the beauty of microformats: instead of giving content for the
machines, that humans have a hard time reading, it reverses the question and
tries to make human-readable content (aka XHTML) also readable by machines.

In the long run it could mean that we don't have to duplicate content (HTML,
RSS) for the two audiences (humans, machines). In the short run, it just
gives clickable links in syndicated content, which proves very practical,
and really helps enrich our usage of feeds.

If you want to check out what one can do with this, just check our SPIP's
"portal" at http://sedna.spip.org/sedna/

Some sources (like, eg del.icio.us/spip) do send their "tags" and we make
them clickable, which is very interesting from a users' experience point of
view.

However, del.icio.us, just like connotea and flickr, currently have a
specific parsing code inside SPIP's aggregator. I wish this could just be
resolved generically instead of specifically for each social tagging
service.

(And we also use the rel="category" attribute inside <a ...> links in order
to retrieve categories with their full URLs.)

-- Fil

Re: [Connotea-code-devel] Tags and RSS: a proposal

[Martin F. <ma...@ne...>]

Hi Fil,

We actually put the tags in a Dublin Core subject tag in RSS as well.

So for any item:
stated in an XPath-ish way:
all the /item/subject paths will give you the tags
or stated in an RDF-ish way:
any triples matching <rss:item> dc:subject ? will give you the tags.

I believe you're talking about the content:encoded portion, which is a 
copy of the XHTML that appears on the normal web page. That's really for 
the convenience of people looking at the feed in a newsreader that has 
an embedded HTML viewer; we didn't actually expect it to be parsed for data.

But nonetheless your proposal is worth discussion. We'll take a look at 
the links you cite.

Thanks,
Martin Flack


Fil wrote:
> in order to make tags+RSS really work, it should work both ways: you should
> be able to fetch not only "all news related to a tag", but also "the tags
> related to the news you retrieve".
> 
> After checking a handful of social-tagging websites and software, I've been
> thinking that we need a common way of sending tags along within RSS, and
> that the best candidate for this would-be common format is the relTag
> microformat:
> 
> Check out, for example, Le Monde diplomatique's RSS file at
>         www.monde-diplomatique.fr/recents.xml
> 
> There you have, in the content, the tags, expressed as (when unescaped):
>         <a rel="tag" href="address....">tag</a>
> 
> This is a "relTag" compatible way of expressing tags
>         http://microformats.org/wiki/reltag
> 
> Connotea, for one, exports tags as (i.e. in www.connotea.org/rss/tag/AvianFlu )
>         <a href="http://www.connotea.org/tag/flu" title="flu" class="postedtag">AvianFlu</a>
> 
> So, basically, we're almost there: with an extra rel="tag" attribute, it
> would be standardized.
> 
> 
> In fact, I didn't just think about it; I have already implemented the thing
> in the aggregator part of SPIP (www.spip.net), and it's working very nicely.
> 
> (On a side note, I have also implemented the detection of connotea's current
> scheme, but as it's not standardized it's almost guaranteed to fail one day
> or the other, and, more to the point, we should avoid a multiplication of
> formats that means essentially the same thing).
> 
> I'd really appreciate it if Connotea was to support this idea -- which means
> just, technically, adding the rel="tag" attribute in the RSS schema (in
> other pages too, why not, but this is another story).
> 
> What do you think about this?
> 
> -- Fil

[Connotea-code-devel] Tags and RSS: a proposal

[Fil <fi...@re...>]

Good evening,

in order to make tags+RSS really work, it should work both ways: you should
be able to fetch not only "all news related to a tag", but also "the tags
related to the news you retrieve".

After checking a handful of social-tagging websites and software, I've been
thinking that we need a common way of sending tags along within RSS, and
that the best candidate for this would-be common format is the relTag
microformat:

Check out, for example, Le Monde diplomatique's RSS file at
        www.monde-diplomatique.fr/recents.xml

There you have, in the content, the tags, expressed as (when unescaped):
        <a rel="tag" href="address....">tag</a>

This is a "relTag" compatible way of expressing tags
        http://microformats.org/wiki/reltag

Connotea, for one, exports tags as (i.e. in www.connotea.org/rss/tag/AvianFlu )
        <a href="http://www.connotea.org/tag/flu" title="flu" class="postedtag">AvianFlu</a>

So, basically, we're almost there: with an extra rel="tag" attribute, it
would be standardized.


In fact, I didn't just think about it; I have already implemented the thing
in the aggregator part of SPIP (www.spip.net), and it's working very nicely.

(On a side note, I have also implemented the detection of connotea's current
scheme, but as it's not standardized it's almost guaranteed to fail one day
or the other, and, more to the point, we should avoid a multiplication of
formats that means essentially the same thing).

I'd really appreciate it if Connotea was to support this idea -- which means
just, technically, adding the rel="tag" attribute in the RSS schema (in
other pages too, why not, but this is another story).

What do you think about this?

-- Fil

Re: [Connotea-code-devel] PLoS citation plug-in

[Martin F. <ma...@ne...>]

Bartosz,

This looks really great, thank you!

I will try and get it running internally here and get back to you with=20
some feedback.

Martin

Bartosz Tele=C5=84czuk wrote:
> I have developed a citation plugin for PLoS journals. Although it is
> possible to bookmark PLoS articles with DOI, it is not comfortable and
> doesn't parse authors' names. Instead I use RIS export.
>=20
> The plugin is based on NPG plugin with small modifications. You can
> download it from:
>=20
> http://www.tvk.wroc.pl/~tvk111a/PLoS.pm
>=20
> I tested it with my local connotea server and it performed well (there
> are still some minor issues with article pages).
>=20
> Could you try the plugin and give me your opinion about it? Is it
> possible to put it on official Connotea?
>=20
> I warn you that it is my first Perl script, so it is highly
> experimental. However, I hope it will be useful.
>=20
> Best regards,
>=20
> Bartosz
>=20
> ----------------------------------------------------------------------
> INTERIA.PL | Kliknij po wiecej >>> http://link.interia.pl/f18c1
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by the JBoss Inc.
> Get Certified Today * Register for a JBoss Training Course
> Free Certification Exam for All Training Attendees Through End of 2005
> Visit http://www.jboss.com/services/certification for more information
> _______________________________________________
> Connotea-code-devel mailing list
> Con...@li...
> https://lists.sourceforge.net/lists/listinfo/connotea-code-devel

[Connotea-code-devel] PLoS citation plug-in

[<muc...@po...>]

I have developed a citation plugin for PLoS journals. Although it is
possible to bookmark PLoS articles with DOI, it is not comfortable and
doesn't parse authors' names. Instead I use RIS export.

The plugin is based on NPG plugin with small modifications. You can
download it from:

http://www.tvk.wroc.pl/~tvk111a/PLoS.pm

I tested it with my local connotea server and it performed well (there
are still some minor issues with article pages).

Could you try the plugin and give me your opinion about it? Is it
possible to put it on official Connotea?

I warn you that it is my first Perl script, so it is highly
experimental. However, I hope it will be useful.

Best regards,

Bartosz

----------------------------------------------------------------------
INTERIA.PL | Kliknij po wiecej >>> http://link.interia.pl/f18c1

Re: [Connotea-code-devel] Citation plug-in

[Martin F. <ma...@ne...>]

Bartosz Tele=C5=84czuk wrote:
> I am considering contributing to Connotea by writing a citation plug-in.
> The website I want to enable exports bibliographic information only
> by SOAP protocol. Is it possible to use it in Biblotech?
>=20
> I think it needs overriding get sub in CitationSource class and using
> Perl SOAP library such as SOAP::Lite. How should I do it?
>=20
> I am a Perl beginner and will appreciate any help.

Hi Bartosz,

Thank you for your interest in Connotea!

I would recommend that as a first step you write a script that can take=20
a URI as input, download the data over SOAP that you need from the=20
service, and then print out fields like title, authors, etc. Then we can=20
help you turn that script into a Connotea Code citation module by=20
rearranging the code to fit a Bibliotech::CitationSource::* format.

That's how I usually start them.

Let us know how you go.

Cheers,
Martin Flack

[Connotea-code-devel] Citation plug-in

[<muc...@po...>]

I am considering contributing to Connotea by writing a citation plug-in.
The website I want to enable exports bibliographic information only
by SOAP protocol. Is it possible to use it in Biblotech?

I think it needs overriding get sub in CitationSource class and using
Perl SOAP library such as SOAP::Lite. How should I do it?

I am a Perl beginner and will appreciate any help.

Regards,

Bartosz Telenczuk

----------------------------------------------------------------------
INTERIA.PL | Kliknij po wiecej >>> http://link.interia.pl/f18c1

[Connotea-code-devel] Connotea Web API Development

[Ben L. <b....@na...>]

Hi All,

We're beginning the process of developing a Web API for Connotea -- that=20
is to say, we're defining a way of remotely programming against the=20
Connotea database.

We've put together a basic outline proposal of how it should work, which=20
I've copied here.  I wanted to share this with you all as a starting=20
point for discussion -- so please feel free to give your feedback or=20
suggestions for changes here.  Martin (who'll be doing the programming)=20
is on this list, so he'll be basing the final design on your feedback.

------------------------------------------
Connotea Web API -- Proposed Specification
------------------------------------------

=3D=3D=3D General rules =3D=3D=3D

  * The API will use basic HTTP requests to perform all operations,=20
using a simple RDF/XML syntax as the data format

  * All API URIs will begin /data

  * The rest of the URI will follow the existing link syntax for=20
Connotea as closely as possible.
   * For example, to retrieve your Connotea library:
   * /data/user/yourusername

  * The following HTTP verbs are used:
   * GET - to retrieve information about a resource (bookmarks, tags,=20
users, etc) or list of resources
   * POST - to create a new resource
   * PUT - to replace/edit an existing resource
   * DELETE - to remove a resource
   * OPTIONS - to view all options that are available for a particular=20
resource

  * For example:
   * to delete your copy of Bookmark xyz, you would send a DELETE=20
request to /data/user/yourusername/uri/xyz.
   * to edit your copy of Bookmark xyz, you would send a PUT request to=20
/data/user/yourusername/uri/xyz, including the RDF (or key=3Dvalue URL=20
encoded data with appropriate Content-type header, see below)=20
representation of the new resource in the body (including URI, tags,=20
description, etc.).  This would overwrite the existing information.

  * As an alternative to PUT and DELETE, you can instead send a POST=20
request to  /data/edit and /data/delete to achieve the same things.

  * To create a new resource, send a POST request to the /data/add=20
service, including the RDF/XML (or key=3Dvalue, with appropriate=20
Content-type header, see below) representation of the new resource in=20
the body

  * It will support both cookie-based authentication and HTTP authenticatio=
n
   * All requests will require a Connotea account


=3D=3D=3D Resources =3D=3D=3D

  * The Connotea Web API will offer access to the following resources:
   * User's Bookmarks [/posts or just /]
   * Bookmark URLs      [/bookmarks]
   * Tags           [/tags]
   * Users          [/users]
   * Tag Notes      [/tagnotes]
   * Comments       [/comments]

  * The API supports two different data constructs to represent each of=20
these resource types:
   * RDF/XML
   * key=3Dvalue (HTML form-style) lists
   * Each must have the correct Content-type header associated with it

=3D=3D=3D Errors and Status Codes =3D=3D=3D

  * Errors are signaled by:
   * appropriate HTTP status codes
   * Content of the body returned

  * 200 OK - for all successful GETs, edits and deletes

  * 201 Created - all went well in creating a resource (POST to /add)
   * the Location: header will give the URI of the newly cerated=20
resource (e.g. /user/username/uri/hash)
   * as will the body.
   * appropriate /edit and /delete URIs also given in the body

  * 401 Permission denied - if there are no authorisation details or=20
they are are incorrect

  * 409 Conflict - if there's a conflict between the operation requested=20
and the data submitted, e.g.:
   * Adding something you already have
   * Adding a disallowed URI type
   * Using an invalid tag
   * The body of the response gives the appropriate reason

  * 404 Not found - for PUTs and GETs on resources that don't exist

  * 400 Bad request - for malformed queries

  * 500 Internal server error - for everything else...


=3D=3D Example Operations =3D=3D

  * Return a list of Bookmark XML chunks:

{{{

GET /data/uri/[hash]

}}}

or

{{{

GET /data?uri=3D[uri]

}}}

  =3D>

{{{

<rdf:RDF
  xmlns:rdf=3D"..."
  xmlns=3D"http://www.connotea.org/schema#"
 >
<rdf:Description=20
rdf:about=3D"http://www.connotea.org/data/user/username1/uri/[hash]">
.=2E..  content here -- see below
</rdf:Description>

<rdf:Description=20
rdf:about=3D"http://www.connotea.org/data/user/username2/uri/[hash]">
.=2E..
</rdf:Description>

.=2E..

</rdf:RDF>

}}}

   * Each chunk contains:
    * Connotea user_bookmark identifier (e.g.=20
http://www.connotea.org/user/ben/uri/22681b7ac687784b8263d99aeb089107)
    * URI
    * creator
    * tags
    * bibliographic information
    * /edit and /delete links

  * To Create an new bookmark

{{{

POST /data/add
uri=3D[uri]
.=2E..

}}}

or

{{{

POST /data/add

<rdf:RDF>

.=2E..standard RDF/XML construct goes here -- see below

</rdf:RDF

}}}


  * To get a list of all tags used on a particular URI

{{{

GET /data/tags/uri/[hash]

}}}

or

{{{

GET /data/tags?uri=3D[uri]

}}}

  =3D>

{{{

<rdf:RDF
  xmlns:rdf=3D"..."
  xmlns=3D"http://www.connotea.org/schema#"
 >
<rdf:Description rdf:about=3D"http://www.connotea.org/data/uri/[hash]">
   <dc:subject>tag1</dc:subject>
   <dc:subject>tag2</dc:subject>
   ...
</rdf:Description>
</rdf:RDF>

}}}

=3D=3D Data Formats =3D=3D=3D

=3D=3D=3D RDF/XML Serialization =3D=3D=3D

{{{

<rdf:RDF
  xmlns:rdf=3D"..."
  ...
  xmlns=3D"http://www.connotea.org/schema#"
 >
   <rdf:Description rdf:about=3D"[Connotea data resource path]">
     <dc:title>User edited title</dc:title>
     <dc:creator>ben</dc:creator>
     <dc:date>2005-10-16T14:11:07Z</dc:date>
     <dc:subject>tag1</dc:subject>
     <dc:subject>tag2</dc:subject>
<!--
or  <tags>Tag string to be parsed by Connotea</tags>
//-->

     <editservice rdf:resource=3D"http://www.connotea.org/edit?uri=3D[hash]=
" />
     <deleteservice=20
rdf:resource=3D"http://www.connotea.org/delete?uri=3D[hash]" />

     <uri>
       <dcterms:URI rdf:about=3D"[bookmark URL]">
       <dc:title>Title of webpage or citation</dc:title>
       ... citation details would go here, as in the current RSS feed forma=
t
       </dcterms:URI>
     </uri>
     <annotate:reference=20
rdf:resource=3D"http://www.connotea.org/data/comments/uri/ff0f778a7e5af9fd6=
88a78d0d4e7f33b"=20
/>
   </rdf:Description>
</rdf:RDF>

}}}

=3D=3D=3D Key=3DValue List =3D=3D=3D

As with the current HTML /add form

If the application has already separated the tags:

{{{

uri=3D[uri]
tag=3D[tag1]
tag=3D[tag2]
tag=3D...
title=3D[title]
description=3D[description]
comment=3D[comment]

}}}

or, if the application wants Connotea to separate the tags:

{{{

uri=3D[uri]
tags=3D[tag string to be parsed by Connotea]
title=3D[title]
description=3D[description]
comment=3D[comment]

}}}

-----------------------------------------------------------



As I said, this is meant to be just a starting point for discussion, so=20
please follow up if you have any comments or alternative ideas.

Thanks very much,
Ben

-- Ben Lund
-- New Technology
-- Nature Publishing Group
---

***************************************************************************=
*****  =20
DISCLAIMER: This e-mail is confidential and should not be used by anyone wh=
o is
not the original intended recipient. If you have received this e-mail in er=
ror
please inform the sender and delete it from your mailbox or any other stora=
ge
mechanism. Neither Macmillan Publishers Limited nor any of its agents accept
liability for any statements made which are clearly the sender's own and not
expressly made on behalf of Macmillan Publishers Limited or one of its agen=
ts.
Please note that neither Macmillan Publishers Limited nor any of its agents
accept any responsibility for viruses that may be contained in this e-mail =
or
its attachments and it is your responsibility to scan the e-mail and=20
attachments (if any). No contracts may be concluded on behalf of Macmillan=20
Publishers Limited or its agents by means of e-mail communication. Macmilla=
n=20
Publishers Limited Registered in England and Wales with registered number 7=
85998=20
Registered Office Brunel Road, Houndmills, Basingstoke RG21 6XS  =20
***************************************************************************=
*****

RE: [Connotea-code-devel] Problems with 'ListOfRecent' and Firefox import

[Mary R. <mro...@st...>]

Hi Peter - I had some of the same questions regarding use of the =
templates and components --=20
If you can find the posting from Martin Flack dated July 15, subject =
"Re:Questions" -- that message contained quite a bit of useful =
information (from my point of view).  =20

-----Original Message-----
From: con...@li... =
[mailto:con...@li...] On Behalf Of =
Peter Weiland
Sent: Friday, October 21, 2005 8:12 AM
To: con...@li...
Subject: [Connotea-code-devel] Problems with 'ListOfRecent' and Firefox =
import

Hello,

first of all, thank you for releasing this great piece of software as
open source.

Now, after getting the application up and running, I have noticed some=20
problems:

1) The 'Recent Activity' link shows a blank page. There is an error=20
message in the source code of the page:

<!-- No "ListOfRecent": DBD::mysql::st execute failed: You have an error =

in your SQL syntax near '(DISTINCT=20
CONCAT(g.gang_id),':/:',IFNULL(g.name, '+NULL'),':/:',IFNULL(g.owner, '=20
at line 1 [for Statement "SELECT ub.user_bookmark_id, ub.user, ub.bo...


2) The import of Firefox bookmarks fails. The server displays the=20
following error message:

ERROR
The requested URL could not be retrieved

While trying to retrieve the URL: http://136.199.85.25/bibliotech/upload

The following error was encountered:

     * Write Error

The system returned:

     (32) Broken pipe

An error condition occurred while writing to the network. Please retry=20
your request.


I'm using the following setup:

RedHat Enterprise Linux 3
Perl 5.8.0
Apache/2.0.46
MySQL 4.1.14
mod_perl-1.99_09-10.ent


Apart from these problems it would be quite helpful to have some kind of =

example site in the distribution that shows how to use the different=20
templates and components (not to mention a guide for developing=20
additional import filters :)

Thanks in advance
Peter

--=20
ZPID - Zentrum f=FCr Psychologische Information und Dokumentation
Peter Weiland MA                  Fon: +49(0)651-201-2601
Informationstechnologie           Fax: +49(0)651-201-2604
Universit=E4t Trier                 E-Mail: we...@zp...
54286 Trier                       http://www.zpid.de/



-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, =
discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Connotea-code-devel mailing list
Con...@li...
https://lists.sourceforge.net/lists/listinfo/connotea-code-devel

[Connotea-code-devel] Problems with 'ListOfRecent' and Firefox import

[Peter W. <we...@zp...>]

Hello,

first of all, thank you for releasing this great piece of software as
open source.

Now, after getting the application up and running, I have noticed some=20
problems:

1) The 'Recent Activity' link shows a blank page. There is an error=20
message in the source code of the page:

<!-- No "ListOfRecent": DBD::mysql::st execute failed: You have an error=20
in your SQL syntax near '(DISTINCT=20
CONCAT(g.gang_id),':/:',IFNULL(g.name, '+NULL'),':/:',IFNULL(g.owner, '=20
at line 1 [for Statement "SELECT ub.user_bookmark_id, ub.user, ub.bo...


2) The import of Firefox bookmarks fails. The server displays the=20
following error message:

ERROR
The requested URL could not be retrieved

While trying to retrieve the URL: http://136.199.85.25/bibliotech/upload

The following error was encountered:

     * Write Error

The system returned:

     (32) Broken pipe

An error condition occurred while writing to the network. Please retry=20
your request.


I'm using the following setup:

RedHat Enterprise Linux 3
Perl 5.8.0
Apache/2.0.46
MySQL 4.1.14
mod_perl-1.99_09-10.ent


Apart from these problems it would be quite helpful to have some kind of=20
example site in the distribution that shows how to use the different=20
templates and components (not to mention a guide for developing=20
additional import filters :)

Thanks in advance
Peter

--=20
ZPID - Zentrum f=FCr Psychologische Information und Dokumentation
Peter Weiland MA                  Fon: +49(0)651-201-2601
Informationstechnologie           Fax: +49(0)651-201-2604
Universit=E4t Trier                 E-Mail: we...@zp...
54286 Trier                       http://www.zpid.de/

[Connotea-code-devel] Integrating with a CMS

[Ajay G. <aj_...@ya...>]

Hello,

I discovered Connotea a few days ago and am going gaga
over it. This is 
something I would love to integrate into the intranet
website that I run 
for my research group (http://virus.chem.ucla.edu).

The site is built on a CMS called Geeklog
(http://geeklog.net). Could 
someone help me with strategies to get connotea to
pull user information 
from the Geeklog user database and use the CMS
authentication instead of 
its own? If there is general documentation on
integration to CMSes out 
there some pointers would be much appreciated.

Cheers!

Ajay



		
__________________________________ 
Start your day with Yahoo! - Make it your home page! 
http://www.yahoo.com/r/hs

[Connotea-code-devel] Re-populate citation table after db upgrade...?

[Mary R. <mro...@st...>]

Hello - do you have a script to re-populate the old citation data into
the new citation table, or could you provide some more information about
the retroactive citation utility/feature that's mentioned in
schema_alter.sql?

=20

Congratulations to all the  Nature developers on your award, by the
way..

Re: [Connotea-code-devel] Connotea Code v1.5 Released

[Alf E. <al...@hu...>]

On 12 Sep 2005, at 16:14, Ben Lund wrote:

> Hi All,
>
> We've just released the latest version of Connotea Code.  This  
> brings the functionality of the open source release up to date with  
> that available on http://www.connotea.org/.
>
> http://sourceforge.net/projects/connotea
>
> New features in version 1.5:
>
>     * RIS Export
>     * Private Bookmarks
>     * User Groups
>     * RIS and Firefox Bookmarks Import
>     * DOI Bookmarking
>     * OpenURL Links

For the cases where either the user isn't logged in, doesn't have a  
Connotea account or hasn't registered a preferred OpenURL resolver,  
would you consider adding
a) an invisible COinS <http://ocoins.info/>
and perhaps b) a visible link to OCLC's OpenURL Gateway <http:// 
worldcatlibraries.org/registry/gateway>
as defaults?

Apologies if something like this is already present - I had a quick  
look but couldn't see anything.

alf.

[Connotea-code-devel] Connotea Code v1.5 Released

[Ben L. <b....@na...>]

Hi All,

We've just released the latest version of Connotea Code.  This brings=20
the functionality of the open source release up to date with that=20
available on http://www.connotea.org/.

http://sourceforge.net/projects/connotea

New features in version 1.5:

     * RIS Export
     * Private Bookmarks
     * User Groups
     * RIS and Firefox Bookmarks Import
     * DOI Bookmarking
     * OpenURL Links
     * Tag Notes
     * Geotagging


The DOI bookmarking features of Connotea rely on our membership of=20
CrossRef.  We have included the DOI plug-in with this release, but it is=20
not enabled in the default configuration, since it won't function unless=20
you are at an institution that is a member of CrossRef.  Feel free to=20
contact me if you have any questions about this.

Thank very much,
Ben

-- Ben Lund
-- New Technology
-- Nature Publishing Group
---


***************************************************************************=
*****  =20
DISCLAIMER: This e-mail is confidential and should not be used by anyone wh=
o is
not the original intended recipient. If you have received this e-mail in er=
ror
please inform the sender and delete it from your mailbox or any other stora=
ge
mechanism. Neither Macmillan Publishers Limited nor any of its agents accept
liability for any statements made which are clearly the sender's own and not
expressly made on behalf of Macmillan Publishers Limited or one of its agen=
ts.
Please note that neither Macmillan Publishers Limited nor any of its agents
accept any responsibility for viruses that may be contained in this e-mail =
or
its attachments and it is your responsibility to scan the e-mail and=20
attachments (if any). No contracts may be concluded on behalf of Macmillan=20
Publishers Limited or its agents by means of e-mail communication. Macmilla=
n=20
Publishers Limited Registered in England and Wales with registered number 7=
85998=20
Registered Office Brunel Road, Houndmills, Basingstoke RG21 6XS  =20
***************************************************************************=
*****

[Connotea-code-devel] Re: [Connotea-discuss] Moderating submissions

[Ben L. <b....@na...>]

Hi Tony,

[I'm moving this to the Connotea-code-devel list so that other=20
developers are aware of your questions.]

Thanks very much for this, and please accept my apologies for not=20
replying sooner.  I've added some comments below:

Tony Hirst wrote:
> Hi-
>=20
> We're about to start running a couple of  online study groups for kids,=20
> and one of the restrictions we're working under is that content is=20
> pre-moderated (so kids post a message, a moderator checks it's okay,=20
> then releases it to the rest of the forum).
>=20
> I've been thinking about providing a social bookmarking facility to=20
> support this, and in the worst case reckon there will be the following=20
> contraints:
>=20
> 0) accounts can only be released by an admin furnisshed with a list of=20
> legitimate users, and can't be requested...

I'm fairly sure this could be accomodated with the current Connotea=20
Code.  You'd have to remove the /register option and have create an=20
interface to allow an administrator to create accounts.

> 1) each study group will have a closed/private membership (no-one=20
> outside the study group will be able to see either the group or the=20
> bookmarks generated by group members);

This is possible with the groups and privacy options already present,=20
although you'd have to modify the code slightly to make these the default.

> 2) an individual may be a member of more than one study group;

Yes, that's fine.

> 3) all bookmarkeds/posts will have to be cleared by a moderator before=20
> the other memebers of the group can see them

This is not possible right now, and so would require quite a lot of new=20
code to be written.

> [4) a weaker use of the environment - as a personal rather than social=20
> bookmarking tool - would mean that each person is in only their own=20
> personal group and no-one else (excpet the moderator) will be able to=20
> look at the bookmarks)]

This is much easier to accomplish, but, as you say, would throw out many=20
of the benefits of social bookmarking.

> 5) in a study group, there may be smaller project groups (eg 2 or 3=20
> individuals); their bookmarks may be: a) only released to each other=20
> after moderation; b) opened to the rest of the complete study group=20
> after a certain period of time/on a particular date).

There's no support for nested groups within Connotea, but since users=20
can be members of more than one group, you might be able to set up a=20
series of groups that would meet this requirement.

> To what extent can connotea cope with the above constraints at the=20
> current time? How much effort  (apologies - I haven;t had a chance to=20
> look at the code yet) is likely to be required to add a moderation=20
> check? Would a workaround would be that all posts are embargoed/kept=20
> private until a moderator releases them?

This workaround might work, although would still require a reasonable=20
amount of work in modifying the permissions system and consructing an=20
interface to let the moderator edit other people's bookmarks.


Ta,
Ben

-- Ben Lund
-- New Technology
-- Nature Publishing Group

***************************************************************************=
*****  =20
DISCLAIMER: This e-mail is confidential and should not be used by anyone wh=
o is
not the original intended recipient. If you have received this e-mail in er=
ror
please inform the sender and delete it from your mailbox or any other stora=
ge
mechanism. Neither Macmillan Publishers Limited nor any of its agents accept
liability for any statements made which are clearly the sender's own and not
expressly made on behalf of Macmillan Publishers Limited or one of its agen=
ts.
Please note that neither Macmillan Publishers Limited nor any of its agents
accept any responsibility for viruses that may be contained in this e-mail =
or
its attachments and it is your responsibility to scan the e-mail and=20
attachments (if any). No contracts may be concluded on behalf of Macmillan=20
Publishers Limited or its agents by means of e-mail communication. Macmilla=
n=20
Publishers Limited Registered in England and Wales with registered number 7=
85998=20
Registered Office Brunel Road, Houndmills, Basingstoke RG21 6XS  =20
***************************************************************************=
*****

[Connotea-code-devel] RE: Questions

[Mary R. <mro...@st...>]

Martin -  the explanation helps quite a lot, thanks! =20

-----Original Message-----
From: Martin Flack [mailto:ma...@ne...]=20
Sent: Friday, July 15, 2005 9:05 PM
To: con...@li...
Cc: t.h...@na...; b....@na...
Subject: Re: Questions

Hi Mary,

We're happy to have you using Connotea Code, and we agree that the
documentation is thin at this stage in the game.

Ben Lund is deciding when the next update to the open source release
will occur and what will go in it, so perhaps next week he will address
your questions about new features.

I'm going to offer a few technical points to perhaps answer several
peoples' questions about setting up Connotea Code. In that spirit I'm
sending this to the connotea-code-devel list.

Yes, the "recent" template is for all practical purposes the default. I
confess some ambiguity in our design that "recent" should probably be a
sorting method, not a "page" type, but that's how it is now, because the
social bookmarking sites we based the design on all emphasized recent
entries.

The incoming URI is parsed as:
/[output]/[page]/user/[user]/tag/[tag]/date/[date]/uri/[uri|hash]

(The brackets are just there to show where variable data goes, there are
not actual brackets on the URI.)

So "/user/martin" uses the default output, default page, and one
qualifier which is user.

- an output type:
  - text
  - html (default)
  - rss
  - ris
  - data
- a page:
  - home (default for no query qualifiers, see below)
  - recent (default when some query qualifiers are included)
  - library
  - bookmarks
  - users
  - tags
  - popular
  - register
  - verify
  - loginpopup
  - login
  - logout
  - search
  - commentspopup
  - comments
  - addpopup
  - addcommentpopup
  - addcomment
  - add
  - editpopup
  - edit
  - upload
  - remove
  - retag
- query qualifiers
  - user
  - tag
  - date
  - uri

The qualifier is introduced by a keyword and is then followed by one or
more values. Slashes between values mean OR and pluses mean AND.
So /user/martin/ben means martin OR ben whereas /user/martin+ben means
martin AND ben. This affects what user_bookmarks are found.

Without any qualifiers, the page is "home". As soon as you add a
qualifier like /user/uuu the page is defined as "recent".

The parsed URI with the /user/uuu and /tag/ttt parts are what create the
"current query". From this query come the main user_bookmarks. A
user_bookmark is our most important data object - it defines one user's
posting of one bookmark.

Components fall into three types:
- output that serves a function, like the 'add' form, where a query is
not expected, and qualifiers would be ignored
- output that is separate from the query, like the header or footer, and
qualifiers would be ignored
- output that reacts to the query, like the main list of user_bookmarks
or a list of tags that is limited by the current query

In a template, a component call looks like one of these:
${component:xxx}
${component:xxx(y=3Dz...)}
${component:xxx:ppp}
${component:xxx:ppp(y=3Dz)}

The xxx is the name:
  AddCommentForm
  AddForm
  EditForm
  Comments
  Footer
  Header
  Inc
  Infobar
  ListOfBookmarks
  ListOfUsers
  ListOfActiveUsers
  ListOfTags
  ListOfActiveTags
  ListOfRecent
  ListOfPopular
  LoginBox
  LoginForm
  LogoutForm
  RegisterForm
  RemoveForm
  RetagForm
  SearchForm
  Toolbox
  UploadForm
  VerifyForm
 =20
The ppp is an optional "part" (applies to the ListOf* components):
  cabinet
  query
  linked
  related

This will generate a list of entities related in some way to the
user_bookmarks found via the current query, and can play a big part of
the exploratory nature of the interface.
e.g. on a /user/uuu page (a "library" page as we call it):
- ListOfTags:cabinet shows all that user's tags
- ListOfTags:query shows tags used in the query (i.e. the actual tags
that appear on the URI, if any), that are not cabinet tags
- ListOfTags:linked shows tags linked on the main user_bookmarks on the
current page, that are not cabinet or query tags
- ListOfTags:related shows tags linked one step removed, that are not
cabinet, query, or linked tags

The y=3Dz bits are options that can be set. An option without an equals
sign is the same as =3D1 (i.e. setting it to true). The most important
thing is to mark your main component "(main,verbose)" so you get full
output, as most components can give off different output in verbose vs
non-verbose mode and some even have slightly different behavior when
used with the main option. But remember it is the query that is deciding
what user_bookmarks are to be displayed; the component just reacts.

Options are:
  main - this is the main component on this page
  verbose - be verbose (recommended to be always used for "main")
  noheading - suppress the normal heading that comes with the data
  id - set a CSS id that may be used by the component
  element_class - set a CSS class that may be used by the component
  filename - use with Inc to select file to display*

*Inc can take these types of filenames:
//some/file - an absolute filesystem path, (leading slash is taken off)
/some/file - a filesystem path relative to the document root, plus .inc
file - a file in the document root, plus .inc (e.g. test =3D test.inc)

Inc can also take its filename from the URI. If you create a file called
guide.inc and visit /guide it should be displayed using inc.tpl at
wherever ${component:Inc(main,verbose)} is located.

You are correct that in each template there should be exactly one "main"
component. In a file called recent_user.tpl which will be served for
library pages, this should probably be the main component call:
<div id=3D"main">${component:ListOfRecent(main,verbose)}</div>

You'd probably also want somewhere laid out on the same page:
<div id=3D"cabinet">${component:ListOfTags:cabinet}</div>
<div id=3D"linked">${component:ListOfTags:linked}</div>
<div id=3D"related">${component:ListOfUsers:related}</div>

Template files should be in the document root and end in a .tpl
extension. The system selects which template file to use based on the
"base name" which is the page requested (so "recent" is quite important
as it is the default when qualifiers are present) and the qualifiers in
the query URI. So here are some example mappings:
/user/martin --> recent_user.tpl
/user/martin/tag/rss --> recent_user_tag.tpl
/tag/rss --> recent_tag.tpl
/popular/tag/rss --> popular_tag.tpl
/date/2005-01-01 --> recent_date.tpl

The mappings will downgrade, so if there is no recent_user_tag.tpl then
recent_user.tpl will be used. We actually have better code to pick the
template checked into our dev branch but it does the same thing.

default.tpl a last resort.

The template files also support variable replacement in these alternate
forms:
$variable
${variable}

They are:
${bookmarklet_add}
${bookmarklet_add_popup}
${bookmarklet_comments}
${bookmarklet_comments_popup}
${bookmarklet_addcomment}
${bookmarklet_addcomment_popup}
${bookmarklets}
${sitename}
${siteemail}
${docroot}
${location}  <-- very useful for prefixing href's and src's
${link}
${title}
${heading}
${description}
${error}
User record items of the logged-in user:
${user_id}
${username}
${password}
${active}
${firstname}
${lastname}
${email}
${verifycode}
${last_deletion}
${created}
${updated}

It also supports Perl-like string case changers:
\u
\U...\E
\l
\L...\E

e.g. "Welcome to \u$sitename!" will display your site name with a
capital letter.

Hope this helps a bit. :-)


--=20
Martin Flack
Web Development Director
NeoReality, Inc.
http://www.neoreality.com/

[Connotea-code-devel] Re: Questions

[Martin F. <ma...@ne...>]

Hi Mary,

We're happy to have you using Connotea Code, and we agree that the
documentation is thin at this stage in the game.

Ben Lund is deciding when the next update to the open source release
will occur and what will go in it, so perhaps next week he will address
your questions about new features.

I'm going to offer a few technical points to perhaps answer several
peoples' questions about setting up Connotea Code. In that spirit I'm
sending this to the connotea-code-devel list.

Yes, the "recent" template is for all practical purposes the default. I
confess some ambiguity in our design that "recent" should probably be a
sorting method, not a "page" type, but that's how it is now, because the
social bookmarking sites we based the design on all emphasized recent
entries.

The incoming URI is parsed as:
/[output]/[page]/user/[user]/tag/[tag]/date/[date]/uri/[uri|hash]

(The brackets are just there to show where variable data goes, there are
not actual brackets on the URI.)

So "/user/martin" uses the default output, default page, and one
qualifier which is user.

- an output type:
  - text
  - html (default)
  - rss
  - ris
  - data
- a page:
  - home (default for no query qualifiers, see below)
  - recent (default when some query qualifiers are included)
  - library
  - bookmarks
  - users
  - tags
  - popular
  - register
  - verify
  - loginpopup
  - login
  - logout
  - search
  - commentspopup
  - comments
  - addpopup
  - addcommentpopup
  - addcomment
  - add
  - editpopup
  - edit
  - upload
  - remove
  - retag
- query qualifiers
  - user
  - tag
  - date
  - uri

The qualifier is introduced by a keyword and is then followed by one or
more values. Slashes between values mean OR and pluses mean AND.
So /user/martin/ben means martin OR ben whereas /user/martin+ben means
martin AND ben. This affects what user_bookmarks are found.

Without any qualifiers, the page is "home". As soon as you add a
qualifier like /user/uuu the page is defined as "recent".

The parsed URI with the /user/uuu and /tag/ttt parts are what create the
"current query". From this query come the main user_bookmarks. A
user_bookmark is our most important data object - it defines one user's
posting of one bookmark.

Components fall into three types:
- output that serves a function, like the 'add' form, where a query is
not expected, and qualifiers would be ignored
- output that is separate from the query, like the header or footer, and
qualifiers would be ignored
- output that reacts to the query, like the main list of user_bookmarks
or a list of tags that is limited by the current query

In a template, a component call looks like one of these:
${component:xxx}
${component:xxx(y=z...)}
${component:xxx:ppp}
${component:xxx:ppp(y=z)}

The xxx is the name:
  AddCommentForm
  AddForm
  EditForm
  Comments
  Footer
  Header
  Inc
  Infobar
  ListOfBookmarks
  ListOfUsers
  ListOfActiveUsers
  ListOfTags
  ListOfActiveTags
  ListOfRecent
  ListOfPopular
  LoginBox
  LoginForm
  LogoutForm
  RegisterForm
  RemoveForm
  RetagForm
  SearchForm
  Toolbox
  UploadForm
  VerifyForm
  
The ppp is an optional "part" (applies to the ListOf* components):
  cabinet
  query
  linked
  related

This will generate a list of entities related in some way to the
user_bookmarks found via the current query, and can play a big part of
the exploratory nature of the interface.
e.g. on a /user/uuu page (a "library" page as we call it):
- ListOfTags:cabinet shows all that user's tags
- ListOfTags:query shows tags used in the query (i.e. the actual tags
that appear on the URI, if any), that are not cabinet tags
- ListOfTags:linked shows tags linked on the main user_bookmarks on the
current page, that are not cabinet or query tags
- ListOfTags:related shows tags linked one step removed, that are not
cabinet, query, or linked tags

The y=z bits are options that can be set. An option without an equals
sign is the same as =1 (i.e. setting it to true). The most important
thing is to mark your main component "(main,verbose)" so you get full
output, as most components can give off different output in verbose vs
non-verbose mode and some even have slightly different behavior when
used with the main option. But remember it is the query that is deciding
what user_bookmarks are to be displayed; the component just reacts.

Options are:
  main - this is the main component on this page
  verbose - be verbose (recommended to be always used for "main")
  noheading - suppress the normal heading that comes with the data
  id - set a CSS id that may be used by the component
  element_class - set a CSS class that may be used by the component
  filename - use with Inc to select file to display*

*Inc can take these types of filenames:
//some/file - an absolute filesystem path, (leading slash is taken off)
/some/file - a filesystem path relative to the document root, plus .inc
file - a file in the document root, plus .inc (e.g. test = test.inc)

Inc can also take its filename from the URI. If you create a file called
guide.inc and visit /guide it should be displayed using inc.tpl at
wherever ${component:Inc(main,verbose)} is located.

You are correct that in each template there should be exactly one "main"
component. In a file called recent_user.tpl which will be served for
library pages, this should probably be the main component call:
<div id="main">${component:ListOfRecent(main,verbose)}</div>

You'd probably also want somewhere laid out on the same page:
<div id="cabinet">${component:ListOfTags:cabinet}</div>
<div id="linked">${component:ListOfTags:linked}</div>
<div id="related">${component:ListOfUsers:related}</div>

Template files should be in the document root and end in a .tpl
extension. The system selects which template file to use based on the
"base name" which is the page requested (so "recent" is quite important
as it is the default when qualifiers are present) and the qualifiers in
the query URI. So here are some example mappings:
/user/martin --> recent_user.tpl
/user/martin/tag/rss --> recent_user_tag.tpl
/tag/rss --> recent_tag.tpl
/popular/tag/rss --> popular_tag.tpl
/date/2005-01-01 --> recent_date.tpl

The mappings will downgrade, so if there is no recent_user_tag.tpl then
recent_user.tpl will be used. We actually have better code to pick the
template checked into our dev branch but it does the same thing.

default.tpl a last resort.

The template files also support variable replacement in these alternate
forms:
$variable
${variable}

They are:
${bookmarklet_add}
${bookmarklet_add_popup}
${bookmarklet_comments}
${bookmarklet_comments_popup}
${bookmarklet_addcomment}
${bookmarklet_addcomment_popup}
${bookmarklets}
${sitename}
${siteemail}
${docroot}
${location}  <-- very useful for prefixing href's and src's
${link}
${title}
${heading}
${description}
${error}
User record items of the logged-in user:
${user_id}
${username}
${password}
${active}
${firstname}
${lastname}
${email}
${verifycode}
${last_deletion}
${created}
${updated}

It also supports Perl-like string case changers:
\u
\U...\E
\l
\L...\E

e.g. "Welcome to \u$sitename!" will display your site name with a
capital letter.

Hope this helps a bit. :-)


-- 
Martin Flack
Web Development Director
NeoReality, Inc.
http://www.neoreality.com/

RE: [Connotea-code-devel] Using components in pages - more information?

[Mary R. <mro...@st...>]

Can you give me any more information on this?  Please...?

=20

It's hard to know what list can be used where, and what the effect will
be.  Does a page always have to contain a "main" component that is a
list of bookmarks , and the user and tag lists are relative to that?
Perhaps there's some documentation that I've missed?

=20

Also - why does a user or tag query use the "recent" template?   It's
using my "popular", "home" , and "recent" templates, but not my "user"
template (it uses "recent", rather than "default".

=20

I can tell that there was some overall philosophy to how all this
functions, but it would be good to know what it is - it's very
experimental / hit-or-miss right now, at least for me.

=20

-----Original Message-----
From: con...@li...
[mailto:con...@li...] On Behalf Of
Mary Rosasco
Sent: Tuesday, July 05, 2005 3:15 PM
To: con...@li...
Subject: [Connotea-code-devel] Using components in pages - more
information?

=20

To the developers - Could you expand on the instructions for adding
components to a page?  I am getting mixed results - "ListOfActiveTags"
is working for me, but I can't get a "recent" list (other than the main
component on the "recent" page).  I haven't had any success with the
ComponentName:subsection syntax you describe in the documentation.

=20

Could you summarize which options can be used with which list
components, and any caveats related to their use?  For instance, I am
assuming that "related" in your ListOfTags example means related to the
list that is the main component on the page.

=20

Thanks...

=20

=20

=20