peepagg-developers — People Aggregator Development

FASD project: Online survey launched

[<ben...@id...>]

Dear Open Source developer

I am doing a research project on "Fun and Software Development" in which I kindly invite you to participate.
You will find the online survey under http://fasd.ethz.ch/qsf/. The questionnaire consists of 53 questions and you will need about 15 minutes to complete it.

With the FASD project (Fun and Software Development) we want to define the motivational significance of fun when software developers decide to engage in Open Source projects. What is special about our research project is that a similar survey is planned with software developers in commercial firms. This procedure allows the immediate comparison between the involved individuals and the conditions of production of these two development models. Thus we hope to obtain substantial new insights to the phenomenon of Open Source Development.


With many thanks for your participation,
Benno Luthiger


PS:
The results of the survey will be published under http://www.isu.unizh.ch/fuehrung/blprojects/FASD/.
We have set up the mailing list fa...@we... for this study. Please see http://fasd.ethz.ch/qsf/mailinglist_en.html for registration to this mailing list.

_______________________________________________________________________

Benno Luthiger
Swiss Federal Institute of Technology Zurich
8092 Zurich

Mail: benno.luthiger(at)id.ethz.ch
_______________________________________________________________________

I hope you all agree with me

[Marc C. <ma...@br...>]

http://blogs.it/0100198/2004/05/16.html#a2757
 
:-)

Re: Drupal Puzzle Pieces

[Anthony B. <abo...@ga...>]

Personally, I can't wait for "Milestone 5 - Collect Underpants". Might I
suggest a "Revenge of the Nerds" style raid. o_O

----- Original Message -----
From: "Jonas M Luster" <jl...@jl...>
To: <pee...@li...>
Cc: "Dries Buytaert" <dr...@bu...>; "Julian Bond"
<jul...@vo...>; "Grant K Rauscher" <gr...@2c...>; "Nick
Chalko" <ni...@ch...>; "Anca Mosoiu" <an...@an...>;
<crs...@ui...>; <wo...@ac...>; "Joel De Gan" <jo...@ta...>;
<an...@an...>; "adrian rossouw" <ad...@ob...>;
"Anthony Borders" <abo...@ga...>; "Danny Ayers"
<dan...@vi...>; "Brian Dear" <br...@ev...>; <wa...@wa...>;
"David L. Sifry" <da...@si...>; "Mitch Ratcliffe"
<go...@ra...>; "Neil Drumm" <dr...@de...>; "'Phillip
Pearson'" <pp...@my...>; "Eric Sigler" <es...@2n...>; "Zack Rosen"
<za...@sk...>; "Marc Barrot" <ma...@pr...>; "rich
seidner" <se...@mi...>; <Ca...@Pe...>
Sent: Sunday, May 09, 2004 5:15 PM
Subject: Re: Drupal Puzzle Pieces


> Hopefully we’re all in sync and know what we’re gonna do – right?

I guess what we're missing is a set of tangible milestones and short
term design goals. Let me try to outline some of the things I see...

We have a core:

[ core ]

this core holds basic user information, as well as a set of pointers to
instances:

[ core ]
|   \   \   \   \
[ instances ]

each instance represents a "persona" of the account in question. The
account can hold an infinite amount of personae, each of which is
linked to either an internal or an external service, which in turn
holds data, such as weblog postings, news, images, your bank statement,
or the list of girlfriends to call next week.

What is needed here:

a) A core
b) Instantiations
and
c) communication protocols between external and internal sources

there are a number of ways, how this could be glued together - FOAF,
some proprietary system, or something else. I assume you want FOAF.

so here:

                   [ core ]
                  / <----- FOAF glue
[ instance ]
                 \   <----- infoexchange protocol
      [ external or internal source ]

Taking this to tangible milestones:

a) we need a core. The core must be able to transparently address any
number of instances, all of which should be agnostic towards the data
they receive and send. In Drupal talk, we need to expand user.module
and profile.module to include those changes.

b) we need means to communicate and extract info from FOAF data. Again,
this might be handled by Drupal, and requires something like the
drupal.module.

c) We need APIs between the external and internal services. This can be
done by creating a new node-type and attaching the service to it.
node-types then are tied to personae and all personae are tied to one
account.

Milestone 1: Make Drupal multi-personae capable
Milestone 2: Ensure Drupal's ability to authenticate against and
retrieve data from FOAF
Milestone 3: Create node-types and tie them to personae
Milestone 4: Create "commons" module for registered users
Milestone 5: Collect underpants
Milestone 6: ...
Milestone 7: Profit

jonas

RE: Drupal Puzzle Pieces

[Marc C. <ma...@br...>]

Hey dude

Crazy busy getting ready to split for lala for E3.

Only two comments on this:

- I don't think we have to worry about keeping things internally
represented as FOAF.  FOAF is all about the import/export.  Whatever
efficient way you wanna store things internally - is fine by me.  I know
Drupal has lots of built-in structs.

- You can decide what best order to do things in.  We're (BBM) obviously
interested in being able to point to something as 'PeepAgg' - but you
can best decide how that unfolds.  As you can see from my drawing, DLA
is a separate module and brand - as well.  So 'eventually' we'll want to
have a module (or collection of modules) which can be called the
PeepAgg.  I'm just not up on the politics and reality of extending
modules versus starting from scratch.  How does that work?

=====

http://blogs.it/0100198/DrupalArchRoadmap.jpg

So assuming:
	- Joel gets the pDNS going
	- Chris gets FOAFnet functionality into the drupal.profile
	- you can take it one step further - by turning it into multiple
persona and implementing the goals you map out.
	- so what can we get walkah to do?

- Marc


P.S.  The current live server for 1UP.com is www.ogamo.com  Feel free to
come on by.


-----Original Message-----
From: pee...@li...
[mailto:pee...@li...] On Behalf Of
Jonas M Luster
Sent: Sunday, May 09, 2004 5:15 PM
To: pee...@li...
Cc: Dries Buytaert; Julian Bond; Grant K Rauscher; Nick Chalko; Anca
Mosoiu; crs...@ui...; wo...@ac...; Joel De Gan;
an...@an...; adrian rossouw; Anthony Borders; Danny
Ayers; Brian Dear; wa...@wa...; David L. Sifry; Mitch Ratcliffe;
Neil Drumm; 'Phillip Pearson'; Eric Sigler; Zack Rosen; Marc Barrot;
rich seidner; Ca...@Pe...
Subject: Re: Drupal Puzzle Pieces

> Hopefully we're all in sync and know what we're gonna do - right?

I guess what we're missing is a set of tangible milestones and short 
term design goals. Let me try to outline some of the things I see...

We have a core:

[ core ]

this core holds basic user information, as well as a set of pointers to 
instances:

[ core ]
|   \   \   \   \
[ instances ]

each instance represents a "persona" of the account in question. The 
account can hold an infinite amount of personae, each of which is 
linked to either an internal or an external service, which in turn 
holds data, such as weblog postings, news, images, your bank statement, 
or the list of girlfriends to call next week.

What is needed here:

a) A core
b) Instantiations
and
c) communication protocols between external and internal sources

there are a number of ways, how this could be glued together - FOAF, 
some proprietary system, or something else. I assume you want FOAF.

so here:

                   [ core ]
                  / <----- FOAF glue
[ instance ]
                 \   <----- infoexchange protocol
      [ external or internal source ]

Taking this to tangible milestones:

a) we need a core. The core must be able to transparently address any 
number of instances, all of which should be agnostic towards the data 
they receive and send. In Drupal talk, we need to expand user.module 
and profile.module to include those changes.

b) we need means to communicate and extract info from FOAF data. Again, 
this might be handled by Drupal, and requires something like the 
drupal.module.

c) We need APIs between the external and internal services. This can be 
done by creating a new node-type and attaching the service to it. 
node-types then are tied to personae and all personae are tied to one 
account.

Milestone 1: Make Drupal multi-personae capable
Milestone 2: Ensure Drupal's ability to authenticate against and 
retrieve data from FOAF
Milestone 3: Create node-types and tie them to personae
Milestone 4: Create "commons" module for registered users
Milestone 5: Collect underpants
Milestone 6: ...
Milestone 7: Profit

jonas

Re: Drupal Puzzle Pieces

[Jonas M L. <jl...@jl...>]

> Hopefully we=92re all in sync and know what we=92re gonna do =96 =
right?

I guess what we're missing is a set of tangible milestones and short=20
term design goals. Let me try to outline some of the things I see...

We have a core:

[ core ]

this core holds basic user information, as well as a set of pointers to=20=

instances:

[ core ]
|   \   \   \   \
[ instances ]

each instance represents a "persona" of the account in question. The=20
account can hold an infinite amount of personae, each of which is=20
linked to either an internal or an external service, which in turn=20
holds data, such as weblog postings, news, images, your bank statement,=20=

or the list of girlfriends to call next week.

What is needed here:

a) A core
b) Instantiations
and
c) communication protocols between external and internal sources

there are a number of ways, how this could be glued together - FOAF,=20
some proprietary system, or something else. I assume you want FOAF.

so here:

                   [ core ]
                  / <----- FOAF glue
[ instance ]
                 \   <----- infoexchange protocol
      [ external or internal source ]

Taking this to tangible milestones:

a) we need a core. The core must be able to transparently address any=20
number of instances, all of which should be agnostic towards the data=20
they receive and send. In Drupal talk, we need to expand user.module=20
and profile.module to include those changes.

b) we need means to communicate and extract info from FOAF data. Again,=20=

this might be handled by Drupal, and requires something like the=20
drupal.module.

c) We need APIs between the external and internal services. This can be=20=

done by creating a new node-type and attaching the service to it.=20
node-types then are tied to personae and all personae are tied to one=20
account.

Milestone 1: Make Drupal multi-personae capable
Milestone 2: Ensure Drupal's ability to authenticate against and=20
retrieve data from FOAF
Milestone 3: Create node-types and tie them to personae
Milestone 4: Create "commons" module for registered users
Milestone 5: Collect underpants
Milestone 6: ...
Milestone 7: Profit

jonas

Drupal Puzzle Pieces

[Marc C. <ma...@br...>]

Here  is the overall strategy masterplan puzzle pieces architecture.
 
http://blogs.it/0100198/DrupalArchRoadmap.jpg 
 
Hopefully we're all in sync and know what we're gonna do - right?
 
:-)
 
- Marc

RE: feedback?

[Marc C. <ma...@ca...>]

Perfect - after all "who was using Director in the 80's?"

-----Original Message-----
From: pee...@li...
[mailto:pee...@li...] On Behalf Of
Dries Buytaert
Sent: Thursday, April 29, 2004 2:05 AM
To: PeepAgg list
Subject: Re: feedback?

On Wed, 28 Apr 2004, Marc Canter wrote:
> http://blogs.it/0100198/draftcolorschemeandlogo.jpg

What is the target audience?

The colors in the proposed draft communicate 'excitement' (at least to
me)  
and will likely attract a young audience.  When the site is to be used
for
business purposes, you might want to consider colors that communicate
'trust', 'success' and 'quality'.

-- 
Dries Buytaert  ::  http://www.buytaert.net/


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g.

Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
_______________________________________________
Peepagg-developers mailing list
Pee...@li...
https://lists.sourceforge.net/lists/listinfo/peepagg-developers

Re: feedback?

[Dries B. <dr...@bu...>]

On Wed, 28 Apr 2004, Marc Canter wrote:
> http://blogs.it/0100198/draftcolorschemeandlogo.jpg

What is the target audience?

The colors in the proposed draft communicate 'excitement' (at least to me)  
and will likely attract a young audience.  When the site is to be used for
business purposes, you might want to consider colors that communicate
'trust', 'success' and 'quality'.

-- 
Dries Buytaert  ::  http://www.buytaert.net/

feedback?

[Marc C. <ma...@br...>]

http://blogs.it/0100198/draftcolorschemeandlogo.jpg

here's something to chew on

[Marc C. <ma...@br...>]

1Up.com - an example of a DLA (digital lifestyle aggregator)
 
The best way to understand a new technology is to have an example of it
- and the new 1Up.com <http://ntnse.com/>  is an great example of a DLA.
 
1Up is a videogame portal (with 26 front doors) combined with a social
network, journaling, gamer and game matching, contests, clubs, an arena
and many other fun filled features.  1Up brings gamers together, which
then leads to gamers forming clubs, participating in conversations,
sharing cheats and wallpaper and 'hooking up' to play together (whether
it be on-line or in front of their game consoles.)   
 
1Up.com has a rich body of reviews, previews, cheats, Top 10 lists,
downloads, news and stories at its basis, as it is published by Ziff
Media - which has 50% of the videogame magazine market.  
 
1Up.com provides every member of the system (gamer) with a rich range of
activities to choose from and their own game collection, which is then
matched to other gamer's collections.  Trade/Wish lists are matched up
as well as gamers' status - for playing games on-line.  Games can be
tracked, reviewed, rated and all sorts of end-user's submissions can be
attached to games as well.  
 
When one gamer lands on another gamer's page - they're immediately told
what matches they have together, what games they should be playing
together and what OTHER games they might like.  The social network is
immediately put into a context.  A PeoplePlace search feature enables
gamers to find other gamers, based upon what games, genres or platforms
they're associated with. 
 
Personal publishing is in it's natural setting as well, as gamers like
to talk about games, their rigs, review games, share images and videos
about games and now with 1Up.com post to their private and club journals
about games (and gamers too!)
 
1Up.com has message boards associated with each game as this is how
gamers have traditionally communicated with each other.  1up.com goes
further with message boards by combining them with the notion of Clubs
(similar to Tribe.net <http://tribe.net/> ) and providing the ability to
subscribe to these boards via RSS feeds.  Each gamer also has their own
private journal and there are also journals for ever club (creating a
sort of 'group voice'.)
 
1Up.com's built-in points mechanism rewards gamers for every action they
partake of in the system, from making friends and joining clubs, to
submitting reviews, posting images and to message boards.  These points
are then used for contests, Easter eggs, karma compiling and eventually
for redeeming products and prizes.
 
1Up.com has an image gallery which is used to manage all of the gamers
images. So if a gamer used an image for a journal post or arena
submission, they wouldn't have to upload it again to use it in as their
avatar face image, in product review or for a club journal posting.
 
1Up.com supports FOAF - so entire networks of gamers can be imported or
exported to other social networks and cell phones can utilized to post
to the system or receive updates, messages or alerts.  
 
1Up.com will eventually offer a wide range of web services available,
such as connecting into on-line gaming systems like Xbox Live and
downloading music and videos.  1Up.com will also become a road show -
which will travel from town to town offering tournaments and prizes to
attendees, which will also enable on-line friends to meet in meatspace.
 
The underlying notion driving the design and architecture of 1Up.com is
that by relying upon built-in constructs; such as points, galleries,
games, friends, reviews and feeds - incredible new kinds of end- user
experiences were possible.  1Up.com integrates many aspects of the
digital lifestyle of the gamer - into one comprehensive environment -
fine tuning certain features and functionality to the unique needs of
the gamer.
 
Not only does 1Up.com create new revenue models and opportunities, but
it also defines the notion of what a DLA can and should be.

Re: Drupal, PeopleAggregator and the People's DNS

[Joel De G. <jo...@ta...>]

Also.
check http://peoplesdns.com/=20
Eve got around to my logo :)

In addition. the filter code is non-optimized, I am going to change that
code to use bitwise shifting to escape a lot of those explode/impodes. I
have been chatting a bit with Adam (author of o'reilly's php cookbook)
who is also on the nyphp mailing list.=20
Anyway, thanks to his suggestions I should have a new version soon that
will be faster.

-joeldg

On Tue, 2004-04-27 at 12:44, Marc Canter wrote:
> As our technology spreads around the world =96 it=92s important to keep i=
n
> mind marketing these technologies.=20
>=20
> =20
>=20
> Not just getting people to use them, but also explaining what the
> technologies do, how they=92re used and (most importantly) the benefits
> they=92ll have to everyday humans.
>=20
> =20
>=20
> That=92s why we name our technologies as close as possible to what they
> do: MediaBand, ServerInTheCloset, PeopleAggregator and People=92s DNS.
>=20
> =20
>=20
> That=92s also why it=92s important that we maintain separate =91identitie=
s=92
> between these technologies =96 as developers out there may choose to mix
> and match them with the code they already have. =20
>=20
> =20
>=20
> Not everyone needs a coolio CMS, framework, plumbing platform =96 so we
> need to be able to slice out the PeepAgg code from the Drupal code and
> let folks use it with their code.  Same goes for the pDNS. At the
> install stage =96 we=92ll (eventually) let folks decide which pieces they
> want=85=85
>=20
> =20
>=20
> But first things first =96 let=92s get it all to work together!
>=20
> =20
>=20
> So just to be clear =96 these three projects are separate projects, they
> just happen to be given birth within the context of Drupal =96 and god
> dammit Drupal it is!
>=20
> =20
>=20
> -         Marc
>=20
> =20
>=20
> P.S. Joel and I have had =91some=92 discussion as to the difference
> between People=92s DNS and the PeopleAggregator.  That hasn=92t been
> entirely resolved, but let=92s leave it at this =93one could CERTAINLY
> imagine other FOAF interfaces using the pDNS =96 while they prefer to
> NOT use the PeepAgg.=94
>=20
> =20
>=20
> P.P.S. [BTW since I used the ID word above =96 that DOES bring up the
> notion of FOAF for code, certainly FOAF for groups and maybe even one
> day =96 FOAF for all things (i.e. items, products, services, etc.)  Marc
> Smith is doing interesting work like that at Microsoft.  We DO have
> FOAF for groups coming=85=85..]
--=20
joeldg - developer, Intercosmos media group.
http://lucifer.intercosmos.net

Drupal, PeopleAggregator and the People's DNS

[Marc C. <ma...@br...>]

As our technology spreads around the world - it's important to keep in
mind marketing these technologies. 
 
Not just getting people to use them, but also explaining what the
technologies do, how they're used and (most importantly) the benefits
they'll have to everyday humans.
 
That's why we name our technologies as close as possible to what they
do: MediaBand, ServerInTheCloset, PeopleAggregator and People's DNS.
 
That's also why it's important that we maintain separate 'identities'
between these technologies - as developers out there may choose to mix
and match them with the code they already have.  
 
Not everyone needs a coolio CMS, framework, plumbing platform - so we
need to be able to slice out the PeepAgg code from the Drupal code and
let folks use it with their code.  Same goes for the pDNS. At the
install stage - we'll (eventually) let folks decide which pieces they
want..
 
But first things first - let's get it all to work together!
 
So just to be clear - these three projects are separate projects, they
just happen to be given birth within the context of Drupal - and god
dammit Drupal it is!
 
-          Marc
 
P.S. Joel and I have had 'some' discussion as to the difference between
People's DNS and the PeopleAggregator.  That hasn't been entirely
resolved, but let's leave it at this "one could CERTAINLY imagine other
FOAF interfaces using the pDNS - while they prefer to NOT use the
PeepAgg."
 
P.P.S. [BTW since I used the ID word above - that DOES bring up the
notion of FOAF for code, certainly FOAF for groups and maybe even one
day - FOAF for all things (i.e. items, products, services, etc.)  Marc
Smith is doing interesting work like that at Microsoft.  We DO have FOAF
for groups coming....]

RE: New thoughts

[Marc C. <ma...@ca...>]

Coolio dude 

So how 'bout for now - work on refactoring the existing code - getting
that FOAF:profiles - PROFILES:FOAF mixup fixed and other clean up work.

:-)

maybe even get the Groups further along....

:-)

Assume that between James and Jonas - we'll get this stuff going - all
the more reason to have code to move into this new PeepAgg modules.  

:-)

Assume sometime later in the week - people will start asking for
something to plug in.

:-)

- Marc

P.S. Hopefully Eric will be getting together with Andrew this coming
week and getting the Laszlo stuff plugged in.

-----Original Message-----
From: pee...@li...
[mailto:pee...@li...] On Behalf Of
crs...@ui...
Sent: Monday, April 26, 2004 12:34 AM
To: pee...@li...
Subject: New thoughts

Alright, your local hacker here, reporting in after a hard week of 
school attempting to destroy all my life and especially free time. 

Some evolutions we're seeing over the past recent time:

Previously, I was looking at the project as something that would be 
mostly external, tying into Drupal only minimally for user accounts, 
while using Drupal as a backend to store all data. However, looking at 
the situation, it may be best to look at it the other way, to start 
with: create a new Drupal module that allows us to contain all the 
PeopleAggregator code in there.

I've described some issues and tasks associated with this module's 
development at http://devel.peopleaggregator.com/DrupalModule . This 
page lists the main issue I see, as well as the four main tasks to 
making it work. I'm no Druplicon, so I'm not offering suggestions (yet) 
I'm just working to outline what we need to do in order to get a working

module that does what we need it to do.

However, at some point, we want to be able to have PA type abilities 
included everywhere - whether it be Drupal installs, or any kind of 
other service you might use it on or with. This is more of what the 
description at http://devel.peopleaggregator.com/DrupalIntegration 
involves: how to take an external PeopleAggregator Frontend and 
integrate into your existing model: Whether it be Drupal, LiveJournal, 
or whatever else you might be needing it for.

With the arrival of a lot of people more experienced with Drupal than 
me, I'm going to step back from the main Drupal development and let the 
Gods in the field take over in that sense. I'm still going to be working

on tying things together - if you have a question, feel free to bring it

to me and I can poke at the existing code until I get an answer, or poke

whoever needs to know in order to get things done ;)

Really, I'd like to just act as a liason between you smart Drupal people

and the existing code, getting all the functioanlity we have - and how 
we have it - across to the people who need to change it up and make it 
better.

The two pages I linked up there are both what I consider relatively 
informative, but please, feel free to add to either one of them. I 
typed them both up late at night, my brain could be utterly fried ;)

Just wanted to get some goal lists going so I could figure out exactly 
where everyone is on knowing how things are going and keep everyone up 
to date.

-- Chris




-------------------------------------------------------
This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek
For a limited time only, get FREE Ground shipping on all orders of $35
or more. Hurry up and shop folks, this offer expires April 30th!
http://www.thinkgeek.com/freeshipping/?cpg=12297
_______________________________________________
Peepagg-developers mailing list
Pee...@li...
https://lists.sourceforge.net/lists/listinfo/peepagg-developers

New thoughts

[<crs...@ui...>]

Alright, your local hacker here, reporting in after a hard week of 
school attempting to destroy all my life and especially free time. 

Some evolutions we're seeing over the past recent time:

Previously, I was looking at the project as something that would be 
mostly external, tying into Drupal only minimally for user accounts, 
while using Drupal as a backend to store all data. However, looking at 
the situation, it may be best to look at it the other way, to start 
with: create a new Drupal module that allows us to contain all the 
PeopleAggregator code in there.

I've described some issues and tasks associated with this module's 
development at http://devel.peopleaggregator.com/DrupalModule . This 
page lists the main issue I see, as well as the four main tasks to 
making it work. I'm no Druplicon, so I'm not offering suggestions (yet) 
I'm just working to outline what we need to do in order to get a working 
module that does what we need it to do.

However, at some point, we want to be able to have PA type abilities 
included everywhere - whether it be Drupal installs, or any kind of 
other service you might use it on or with. This is more of what the 
description at http://devel.peopleaggregator.com/DrupalIntegration 
involves: how to take an external PeopleAggregator Frontend and 
integrate into your existing model: Whether it be Drupal, LiveJournal, 
or whatever else you might be needing it for.

With the arrival of a lot of people more experienced with Drupal than 
me, I'm going to step back from the main Drupal development and let the 
Gods in the field take over in that sense. I'm still going to be working 
on tying things together - if you have a question, feel free to bring it 
to me and I can poke at the existing code until I get an answer, or poke 
whoever needs to know in order to get things done ;)

Really, I'd like to just act as a liason between you smart Drupal people 
and the existing code, getting all the functioanlity we have - and how 
we have it - across to the people who need to change it up and make it 
better.

The two pages I linked up there are both what I consider relatively 
informative, but please, feel free to add to either one of them. I 
typed them both up late at night, my brain could be utterly fried ;)

Just wanted to get some goal lists going so I could figure out exactly 
where everyone is on knowing how things are going and keep everyone up 
to date.

-- Chris

RE: Drupal

[Marc C. <ma...@br...>]

Hey Dries,

Jonas Luster was telling me how to pronounce your name yesterday.

I lived in 'dam in 1975 and have been going back ever since.

In fact there is a certain tutorial that Dave Winer just posted - that I
created there in 1997.

Anyway - Howdy!

I was introduced to Drupal by Julian Bond almost a year ago and have
been watching your success ever since.  Congrats!

I recently met Chris Schmidt -who was Neil Drumm's room mate and we
decided that Drupal would be teh way to go - for our PeopleAggregator
project.

THEN I met Jonas Luster - who has introduced me to James Walker - and
THEN I got together with Zack Rosen and Neil - who introduced me to
Adrian Rossouw.

I has a four hour tutorial on Drupal yesterday and YES - we have many
plans.

Unfortunately I do not have the time to write it all up to for you now -
as I am with my kids - it is a wonderful day today and we are swimming
in my swimming pool.

So briefly (and I will followup later with many more details):

	- we do NOT plan on forking - so we MUST work with you and the
team!

	- we plan on adding the PeopleAggregator code into Drupal.  We
can discuss how we might maintain the PeopleAggregator 'brand' - while
at the same time provide full featured social networking to Drupal.

	- we also plan on enhancing Drupal to become a "digital
lifestyle aggregator"

	- and we plan on expanding Drupal's authentication to support
other systems (such as SourceID and the new TypeKey systems)

By full featured social networking I mean:

	- a central Commons screen

	- a public page (the About Me screen)

	- a private page (your private tools, data, setting controls,
etc.)

	- Group pages

These four kind of pages make up the PeopleAggregator architecture.

In addition we are in the midst of adding a decentralized 'distributed'
People's DNS to the system as well.

All to be handed to the Drupal Community.

:-)

I hope that is OK.

:-)

Anyway - maybe we can speak on the phone together?

We also have an IRC channel as well on freenode = #pa.

:-)

- Marc

-----Original Message-----
From: Dries Buytaert [mailto:me...@or...] 
Sent: Saturday, April 24, 2004 9:55 AM
To: Marc Canter
Subject: Drupal

Hi Mark, I read your post in the Drupal forum.  Care to elaborate a bit
about your Drupal plans?  (Btw, I went to look for you in #joiito but
you were not there.  You can - occasionally - find me in #drupal at
irc.freenode.net.)

This message was sent by Dries Buytaert to Marc Canter.

* * *

http://www.orkut.com/


To control notification emails, access your Account Settings:

http://www.orkut.com/Settings.aspx

Re: pDNS pros and cons (was Re: [rdfweb-dev] PPD.replaces.PPD)

[Ross W. R. <ro...@tu...>]

On 4/23/2004 9:26 PM Marc Canter noted that:
 > For us to be the hub of the wheel, we gotta support
> ALL formats.

Man, I wish it were easier just to pick *a* standard nowadays. Wasn't 
there a time when the word "standard" meant "that spec that everyone used"?


-- 
Regards,


	-rwr






"In the modern world the intelligence of public opinion is the one 
indispensable condition for social progress."
	- Charles W. Eliot (1834 - 1926)

RE: pDNS pros and cons (was Re: [rdfweb-dev] PPD.replaces.PPD)

[Marc C. <ma...@br...>]

Sounds good dude.

I'd recommend getting on the rdfweb list.

Since I had this 4 hour tutorial on Drupal with Jonas today, my head is
abuzz.  Totally jazzed at this point.  We didn=92t touch much on the =
PDNS
- except to think of it as our own version of Indigo (Microsoft's
universal connector kit.)=20

A hub for all kinds of authentication issue s- which turns out to be a
DRM aggregator - as well. :-)

The reality is we're faced with supporting:

	- Drupal's lite-weight authentication

	- SixApart's new TypeKey scheme - for controlling comments, but
it's really a stand alone authentication system

	- PingID's SourceID federation server

	- or other authentication schemes

And let's not forget LDAP and all those others identity systems floating
around out there.  For us to be the hub of the wheel, we gotta support
ALL formats.

:-)

- Marc

-----Original Message-----
From: Joel De Gan [mailto:jo...@ta...]=20
Sent: Friday, April 23, 2004 4:42 PM
To: dan...@vi...
Cc: Marc Canter
Subject: Re: pDNS pros and cons (was Re: [rdfweb-dev] PPD.replaces.PPD)

On Fri, 2004-04-23 at 15:50, Danny Ayers wrote:
> >
> Marvellous - sounds loosely related to Maciej's work with Bloom
filters:
>=20
> http://www.perl.com/pub/a/2004/04/08/bloom_filters.html

Bloom filters are very cool, I just spent an hour looking into them as I
am dealing with a large dataset issue right now for a domain spinner for
a work related project.=20
But what it is more similar to is probably the LOAF project
http://loaf.cantbedone.org/ except that each email will be encrypted. I
liken it more to how IM lists are set up where people need permission to
be on your list and to be able to send you messages. Except that it is
implied just by listing someone in your knows, if they list back then it
works.

> >
> Thanks.
> Please keep me informed of how you're getting on - it does sound very=20
> interesting indeed.

I am mostly staying in touch with marc, I am on the peepagg list, is
there another I should be on?

>=20
> There are a couple of people I know of that are in the late beta
stages=20
> of Semantic Web Service tools, there might be potential for sharing of

> notes/exchange of data I wonder if I could pass on your address? :
Reto=20
> Bachmann-Gm=FCr's, he's got n RDF/OWL-based RSS aggregator thing
together=20
> (http://www.gmuer.ch/), Geoff Chappell's got a search engine - better=20
> not give you an address right now, they've not gone public yet (in the

> next week,  he reckoned). You might also want to take a look at what=20
> Semaview are doing with iCal/RDF data - that might be smushable too,
and=20
> Paul Cowles seems a nice bloke.
> // Orkut eat your heart out ;-)

Interesting, it is an absolute shame that semaview is windows only it
looks interesting. Anyway, yea, pass on whatever data,in particlar any
projects you may know of that are doing smushing. PDNS is not at a point
that I would really want people poking around with things yet, I expect
a beta out shortly. At first just client demo's in PHP that can show all
the features.

Cheers

>=20
> Cheers,
> Danny.
--=20
joeldg - developer, Intercosmos media group.
http://lucifer.intercosmos.net

RE: pDNS pros and cons (was Re: [rdfweb-dev] PPD.replaces.PPD)

[Marc C. <ma...@br...>]

Coolio dudes - I'm gonna pass this long to the rest of the PeepAgg team
- by way of our mail list.

:-)

Just got back from a 4 hour Drupal tutorial with Jonas.

:-)

Time to change diapers and think about Drupal and Smarty templates.
Joel suggested not using Smarty - anybody else have an opinion on that?

- Marc

-----Original Message-----
From: Danny Ayers [mailto:dan...@vi...]=20
Sent: Friday, April 23, 2004 12:50 PM
To: jo...@ta...
Cc: Marc Canter
Subject: Re: pDNS pros and cons (was Re: [rdfweb-dev] PPD.replaces.PPD)

Joel De Gan wrote:

>Actually centralized is not the entire objective.
>It will be centralized the same way that DNS is centralized.
>Downloadable zone files and full source for the project.
> =20
>
That's wonderful!

>In the server itself which I am working on right now;
>Data will be available via HTML/RDF and via a query system using a high
>port.
>Data queries will be similar to DNS utilizing MX/A/etc.
>In additon I was looking over foaf extensions :
>http://www.perceive.net/schemas/20021119/relationship/#friendOf
>And have some great ideas for encrypting actual emails within foaf
files
>that are only viewable by the owner.. (i.e. I have marc and you in my
>foaf:knows and I can query with my password and get both your emails if
>you have me in your foaf:knows). =20
> =20
>
Marvellous - sounds loosely related to Maciej's work with Bloom filters:

http://www.perl.com/pub/a/2004/04/08/bloom_filters.html

>A lot of work here, yes.. as I do one thing, I find two more I need to
>do on this... In addition I have a fulltime job I am juggling. So, busy
>yes..  I originally thought this would be quick, but I have limited
time
>and the project is bigger that I anticipated so it is taking a bit of
>time.
> =20
>
I know the feeling well...there's just so much that could be done if=20
there was only time...

>But those are only secondary to the pdns system which obviously we will
>need to be node1 and will have to feed data out to what will
effectively
>be "registrars". Authority notes in the foaf file will state who is
>authority for you and so smushing becomes not much of an issue unless
>you want some form of an amalgamated foaf for files you have spread all
>over. But that is not a big issue as the authority note will dictate
>which file is primary etc.
> =20
>
Sounds reasonable.

>On a side note: I didn't write directdns, that is simply a product of
>the company I helped build (directnic). I did their mail systems and
>registrations (I also did a wap version for domain registrations)..
>Here is an outdated (much) resume http://tenshimedia.com/joel/ and
>current list of projects sans-pdns can be found here
>http://pagepossible.com/?go=3Dabout
> =20
>
Thanks.
Please keep me informed of how you're getting on - it does sound very=20
interesting indeed.

There are a couple of people I know of that are in the late beta stages=20
of Semantic Web Service tools, there might be potential for sharing of=20
notes/exchange of data I wonder if I could pass on your address? : Reto=20
Bachmann-Gm=FCr's, he's got n RDF/OWL-based RSS aggregator thing =
together=20
(http://www.gmuer.ch/), Geoff Chappell's got a search engine - better=20
not give you an address right now, they've not gone public yet (in the=20
next week,  he reckoned). You might also want to take a look at what=20
Semaview are doing with iCal/RDF data - that might be smushable too, and

Paul Cowles seems a nice bloke.
// Orkut eat your heart out ;-)

Cheers,
Danny.


--=20
----
Raw
http://dannyayers.com

test opml viewer page

[andrew w. <an...@an...>]

Been working on various things in peopleaggregator and I wanted to post 
a sample page here:

http://main.peopleaggregator.com:8080/lps-2.1.1/peopleaggregator/oppem/private.html

What I need help with is how to get the smarty template to look like 
this. specifically, how do I get the email address of the user?

-Andrew

dudes

[Marc C. <ma...@br...>]

'Specifcially Eric & Chris..
 
Make sure you know of and have clued in..
 
About this:
 
http://drupal.kollm.org/tmp/_contrib/drupal-contrib-phpdoc/foaf_8module-
source.html
 
http://drupal.org/node/view/6120#9177 
 
:-)
 
-          Marc
 
 

RE: hey dude

[Joel De G. <jo...@ta...>]

On Wed, 2004-04-14 at 22:40, Marc Canter wrote:
> How will PA and these others query pdns? Or more precise, how would you
> prefer? I can build a ton of different ways in, and most likely will,
> but I don't want to infringe into PA territory and what it is try to do.
> -joel

The basics here are:
yea, I can do all the lookups in the world and have a lot of this ready but I need to know how the best way
for PA to query the pdns.
I have fixed my sites on several specs for it, in a database world it needs:
1) a unique id.. any unique id, sha1 mailbox etc.. but better yet, a new form id that can handle multiple
sha1 id's to handle multiple mailboxes. I suggest property:id which can be tied to a pdns id for each id.
2) I am looking at submitting a RFC for the new foaf dns format (FOAF on a keypad = 3623) and handling 
various queries.. MX = mail, A = website, NS = authority server (tribe/livejournal/peoplesdns etc)
3) Ideas for way to integrate groups/sites/project/based_near/etc without stepping on the PA developers 
i.e. I don't want to infringe on areas that mess with PA ideas.. i.e. interactions..
4) user-A has two sub-users user-B and user-C, user-A has an email that is available to user-B and user-C so 
threfore the user-A has a new foaf property called "public-key" which allows those two users to send email to
user-A (user-B and user-C can email user-A, but user-Ba and user-Bb etc cannot, mostly to deter spammers.)
this should be a part of the dns system. 
BUT... this is just an idea and I am trying to work it all out.. input would be so very appreciated..

actually ANY ideas would be appreciated...

also note.. sorry if this is a little vauge but I am kind of off now thanks to some vodka and a long
day.. long long day...
-Joel

RE: hey dude

[Marc C. <ma...@ca...>]

How will PA and these others query pdns? Or more precise, how would you
prefer? I can build a ton of different ways in, and most likely will,
but I don't want to infringe into PA territory and what it is try to do.
-joel

>>>>>>>>>>>>> OK - free for all answer......


>>>>>>>>>>>>>> 1st of all - I'm cc:ing one James Tauber on this email -
he's expressed interest in helping as well.

>>>>>>>>>>>>>>  James is an rdf and XML nerd - and as I pondered
answering this question (about how to querry the PeoplesDNS) I suddenly
realized that the answer to that question - is the same answer that can
be used for another situation where we are faced with the battle of the
RSSs - the simple, XML based RSS 2.0 - versus the complex, rdf based RSS
1.0.

>>>>>>>>>>>>>  I'll jump to the chase and fill-in background later.....

>>>>>>>>>>>>>>>>  The problem new kinds of micro-content faces is in
getting sufficient uptake by all teh News aggregators, blogging tools
and other utils out there.  We are fortunate - within the digitalID
space - of having a format as can use and get additional specs into, in
the guise of FOAF.

>>>>>>>>>>>  It's not that simple for OpenReviews, Recipes or Events.

>>>>>>>>>>>>>>In the rest of the world - 75% of all subscriptions are in
RSS 2.0 format.  So most people expect simple XML and thus we're faced
with using namespaces to extend the format - to flow through new kinds
of micro-content.  And that fact is not such a horrible fact and since
we'll need some sort of namespace to extend teh rdf camp side of things
as well (whether it's RSS 1.0 or Atom) - then namespaces it is!

>>>>>>>>>>>>>>It doesn't take a genius (or a visionary marketing guy
like myself) to figure out that ONE schema should be established for
each of these new kinds of micro-content, and then they flow through
whichever subscription format: RSS 2.0 or RSS 1.0/Atom - and politics be
dammed.

>>>>>>>>>>>>>>>>>>So I can't help but think that the requirement of the
PeoplesDNS be - that it can take updates, pings, notification, r-t
streams, whatever teh hell someone comes up with - to notify an open,
central server about things about people.

>>>>>>>>>>>>>>>>>>>>>>Sure it would be nice ot have just one way to do
it - but (to me) that's like saying there's only kind of person, only
one way that people program, only one way to do something.

>>>>>>>>>>>>>>>>>>>>>>>>>>>> So:
		- SOAP
		- XML-RPC
		- email
		- Jabbber
		- fucking carrier pigeon if need be
		- .com

>>>>>>>>> Does that make sense at all?

RE: hey dude

[Marc C. <ma...@ca...>]

OK

Joel wrote.....

Working out some ideas for the peoplesDNS.. found RAP and have decided
that
it is the defacto rdf parser. also found a version of rdql that I might
use as it works with mysql as the one bundled with RAP does not suit my
needs..

============

Marc replies.....

>>>>>>>>>> Not up on RAP - but maybe Eric or Jonas is....

>>>>>>>>>> Check with Eric on parsers in general.  I believe he started
with an open parser and has been evolving it, fine tuning it.  Because
of the haphazard nature that many of these FOAF files take, the parser
has got to be smarter than most.

===========

back to Joel...

now I am working on some ideas for how exactly the dns will be worked
and which properites I can use for unique identifiers. Obviously the
sha1 mailbox can be used, but it is already being used in other areas,
but I need a way to tie those in. 

=============

>>>>>>>>>>>  OK - first dumb question - why can't we use sha1_sum email
as explicit unique identifier?  Wait - I know - most people have more
than one email.  So does that mean this leads us to establishing our OWN
(or shall I say) 'a unique FOAF' identifier?

====================

I have some good ideas, but also have
some other points in foaf that will need to be spec'd out very well.
I.e. query by name, by id, by email, by icq etc.. 

>>>>>>>>>>>>

============================

how to aggregate
groups of people and tie those into the above identifiers 

>>>>>>>>>>>> One thing I DO know is that FOAF has a concept of Groups.
We're instituting that class for our Groups -so we'll want to stick to
the current spec.  However (as I am sitting here today) I could just
about tell yah that no one else has done that yet.  

>>>>>>>>>>>  :-)

>>>>>>>>>>>  My dogma has preached about:
	- multiple kinds of personalities - persona - which we're using
for "kinds of accounts:
	- and we've established multiple kinds of relationships, so
that's happening as well

	- now we get to talk about multiple kinds of groups - or shall I
say multiple levels of groups.  Traditional (eGroups) logic dictates:
		- open public groups - anybody can join
		- moderated groups - you can request entrance, but must
get approval
		- private groups - which you can only be invited into -
and you must accept the offer

	- So my (our) dogma states:
		- all those kinds of groups are called groups (or
communities, or Tribes or Clubs - whatever) - they represent teh middle
circle of life

		- the inner circle is your family - these folks don't
make requests, need formal approval or even show any modicum of decorum.
They're your family - love um or hate um - you're stuck with them.
Special user interfaces will be built for these situations, where mom
and Dad get to define and control their kids interfaces, where
grandparents and distant cousins are helped out, where roomtaes and
close friends get brought in and treate dliek family.  Whatever the case
- family is teh inner circle and once they're in - there are all sorts
of new controls and rules, but not like Groups and......

		- the WWW - teh outer circle.  Basically public web
sites.  But these places have special rules and decorum as well.  And
huge upside advantages - of we can figure out a way(s) to mesh them in.

So three levels of Communities:
	- inner family
	- middle groups
	- outer public, WWW

============================

and currently
I am working on how people will call the pdns system, via webpages, via
javascript calls, via API port calls etc. Then what kind of calls.
Obviously I can borrow some ideas from true dns of MX and A records, NS
records etc.. 


>>>>>>>>>>>> I suggest pinging - on all updates.  That gives us a nice
record of what they're up to - which can be used to build interesting
activity logs.  Maybe that's not what you're asking.....

===============================

Also, updates and verification so that we know the user
updating the record owns the record. I have chosen a 256bit key for good
effect for passwords etc and will continue to use sha1 for obvious
reasons. 

>>>>>>>>>>>>>.  This verification/authentication issue can have multiple
answers..... but end-users will have to choose at least ONE!

>>>>>>>>>>>>>>  Currently two reasonable authentication answers are:
	- Drupal's
	- SourceID's

================================


More details will follow.
-Joel

On Wed, 2004-04-14 at 17:57, Marc Canter wrote:
> Here's what I have:
> 
> (504) 975-6808
> 
> a) I think I got my buddy at Tucows to be interested in tLimit.
> 
> b) I just posted an updated version of http://broadbandmechanics.com -
> which has a credits/Us page: http://www.broadbandmechanics.com/Us.htm
> 
> Just wanted to make sure:
>       Bi) it was Ok to use your name
> 	Bii) if you wanted me to link in a contact email or something
>       Biii) or plug you or any of your projects further.....
> 
> :-)
> 
> - Marc
> 
> -----Original Message-----
> From: Joel De Gan [mailto:jo...@ta...] 
> Sent: Wednesday, April 14, 2004 3:59 PM
> To: Marc Canter
> Subject: Re: hey dude
> 
> you must have the wrong number..
> today, bad day to call..
> kind of busy..
> 
> email contact is fine.
> -Joel
> 
> On Wed, 2004-04-14 at 16:02, Marc Canter wrote:
> > Just tried to call you - but got  a message from somebody named
Wendy.
> > 
> >  
> > 
> > I'm putting up some updates to our corporate site and I've listed
off
> > "who we are".
> > 
> >  
> > 
> > Haven't put Contact emails or sites up - yet. Wanted to run it by
you
> > - before I did that.
> > 
> >  
> > 
> > - Marc
-- 
joeldg - developer, Intercosmos media group.
http://lucifer.intercosmos.net

Helios is undergoing upgrades

[Eric S. <es...@2n...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This affects:
	- (www.)peopleaggregator.com
	- test.peopleaggregator.com
	- qa.peopleaggregator.com
	- devel.peopleaggregator.com
	- svn.2nw.net/pa (Source Control)

Various components are being upgraded on the machine and rearranged
behind the scenes.  (Kernel updates, DB server upgrades, etc.)  We're  
trying
to not have a solid "18 hour down" block, so occasionally the web  
server will
be returning a 404 or two.  We should be finished updating and checking
in another 4-6 hours, most of which time the web server should be up.

This was a planned maintenance window, announced 3/18/04[1].  Sorry the
reminder didn't get sent out here.  (Canter, this was also the cause of  
your
wiki woes last night.)

Ta-Ta
- -Eric

[1]  
http://sourceforge.net/mailarchive/forum.php? 
thread_id=4076907&forum_id=35366
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (Darwin)

iD8DBQFAeCiMMAsYMoETeJwRAlwFAJwO9qMCmorUJxGHWv7fzKxnW9D+ZQCfXOoK
ru0rglbdYDfqNbi9mVJMvdI=
=pTjf
-----END PGP SIGNATURE-----

Wiki down

[Marc C. <ma...@br...>]

lib/Template.php:21: Warning[2]: fopen(/web/wiki.2nw.net/): failed to
open stream: No such file or directory
lib/Template.php:22: Warning[2]: filesize(): Stat failed for
/web/wiki.2nw.net/ (errno=2 - No such file or directory)
lib/Template.php:22: Warning[2]: fread(): supplied argument is not a
valid stream resource
lib/Template.php:23: Warning[2]: fclose(): supplied argument is not a
valid stream resource
lib/Template.php:21: Warning[2]: fopen(/web/wiki.2nw.net/): failed to
open stream: No such file or directory
lib/Template.php:22: Warning[2]: filesize(): Stat failed for
/web/wiki.2nw.net/ (errno=2 - No such file or directory)
lib/Template.php:22: Warning[2]: fread(): supplied argument is not a
valid stream resource
lib/Template.php:23: Warning[2]: fclose(): supplied argument is not a
valid stream resource
lib/Theme.php:184: Notice[1024]: phpwiki.css: not found
lib/Theme.php:184: Notice[1024]: phpwiki-printer.css: not found
lib/Theme.php:184: Notice[1024]: phpwiki-topbottombars.css: not found
lib/Theme.php:184: Notice[1024]: phpwiki-modern.css: not found
lib/Theme.php:184: Notice[1024]: templates/browse.tmpl: not found
lib/Theme.php:184: Notice[1024]: templates/html.tmpl: not found