[isf-wifidog] User profiles in WifiDog

Benoit Grégoire bock at step.polymtl.ca
Jeu 11 Mai 18:57:00 EDT 2006


Originally I was hoping to build a profile system with a core set of generic, 
semantically interoperable data elements (including links to services), that 
could be extended with network specified, non interoperable elements.  I was 
hoping to start with the core set, which could have been implemented quickly 
before solving the much more complicated and delicate problem of supporting 
totally generic profile elements (and being able to actually do stuff with 
them).  

I'm now convinced that agreeing on such a core set of elements will be 
polluted with everyone trying to push it's grocery list, and can't be done in 
the current social context.  So the entire generic system will have to go in 
at once, and there won't be any core elements.

Yep, that means it will be very difficult to EVER exchange profile information 
with other services.  It also means a lot more code for such simple things as 
setting an avatar.  Don't blame me, it would appear that the only other 
alternative is for me to receive nothing but complaints once it is done have 
people consider that it doesn't solve their problem. (And unlike some other 
stuff in the past, profiles are things I am doing completely on my free 
time).

I wish this energy could have been spent discussing and documenting the 
database schema I planned to use to solve the general problem of profiles (or 
implementing it).  But that was not to be.  

Instead I spent the entire afternoon defending my professional judgement, my 
UI and web development experience (Somehow people imagine that all these 
popular services (flickr, skype, gmail) are somehow fundamentally new and 
that I don't understand them!  Anyone remember what I DO for a living? (no, 
websites isn't the answer, try consulting for officially standardizing the 
exchange,  gathering and management of semantic learning object data.  That's 
an order of magnitude more complicated than the very conceptually simple 
services that are all the rage right now).  So here it is the hopefully very 
slightly productive rant:   

<rant>
I just had a major skirmish with someone at ISF over profiles that deeply 
saddened me. 

Sometimes I simply can't understand the reasoning of some of the "web 2.0" 
fans, specifically their willingness to tie themselves irrevocably to 
commercial services and/or and/or fads and/or create non-interoperable data 
structures to gain a little convenience for very specific use cases.

How did the argument start?  I dared claim that having a flickr specific 
element in the core profile set made absolutely no sense.  OMG how dare I say 
that relying on a trendy commercial photo service is a bad idea and not 
future proof?  That we should instead have some photo collection element, 
with handlers for (at least potentially) different services.  Isn't that 
blasphemous?  After all, to paraphrase a famous misquote:  A flickr account 
ought to be enough for anybody.  

I was promptly accused of trying to make decisions for everyone, and that I 
should ask people at ISF what they think before implementing or not 
implementing just about every important feature.  I was astonished!  I was 
expected to ask everyone at ISF for their opinion for every field I did or 
did not put in in profiles, but at the same time it would be ok for ISF to 
impose flickr as an image service to every group using wifidog!  Surely I'm 
not the only one seeing the irony of being accused of tyranny for trying to 
impose letting each group (and future groups) make their own choices!  Even 
if most (or all) the groups agreed, that would still be incredibly 
shortsighted.

Suppose for example that at the beginning of wifidog development (the public 
CVS went online 2004-03-08), we decided to somehow push and support a single 
webmail service to make life life easier for our users.  That service would 
most definitely not be the beloved Gmail:  it didn't even exist yet! (the 
beta was announced on 2004-04-01).

Our community wireless movement claims to be inspired by the open-source/free 
software movement, that the semantic web is the future, that open standards 
and network neutrality are important issues and free speech is paramount.  I 
though that in such a movement I wouldn't even have to explain this!

But off course, for all it's faults, slavery is extremely convenient is it 
not?
</rant>

P.S.:  I have a confession to make:  I always go way out of my way to build 
consensus.  But in the absence of it, I make decisions that I believe in my 
experience to be the best for the greater good, taking consideration problems 
that I think groups will face in the future even if they don't know it yet 
and dismissing people's opinion when they complain but don't propose a viable 
alternative.  And yes, everything else being equal, I give considerably more 
weight to the opinions of those that actively contribute code and ressources 
to the wifidog project.  When I am right, people call it leadership; when I 
am not, they call it egocentrism.  

Luckily, this is a meritocracy.

-- 
Benoit Grégoire, http://benoitg.coeus.ca/
-------------- section suivante --------------
Une pièce jointe non texte a été nettoyée...
Nom: non disponible
Type: application/pgp-signature
Taille: 191 octets
Desc: non disponible
Url: http://listes.ilesansfil.org/pipermail/wifidog/attachments/20060511/98ed35a1/attachment.pgp


More information about the WiFiDog mailing list