[geeklog-devel] WSRP, GL2 and other stuff

[Tony Bibbs]

Vincent, my comments are below.  I cc'd the list as I think my repsonses 
will shed some light for others.

--Tony

Vincent Furia wrote:

> Tony,
>
> There are several of us 'non-core' (no cvs access) developers waiting 
> in the wings, wondering when you're going to start farming out work 
> for geeklog.  I can only really speak for myself, but I'm pretty 
> frustrated at the lack of communications we get. 


The communication structure for GL2 doesn't really exists because, to be 
frank, it has been a one person effort thus far.  I would love to add 
people to the mix and if you are volunteering to be engaged and 
accountable for certain pieces that is great.  So, for the record, I 
will assume you are interested unless I hear otherwise.

>
> A couple of months ago I gave you a bunch of input on the MVCnPHP 
> library, that appear to have just been ignored.  I don't mind being 
> ignored if I can see that my code is (a) being incorporated in CVS or 
> (b) the problems I found were being addressed in another way.  
> Otherwise a response back would be nice indicating what you think of 
> the changes (even a "don't bother me with this" would be better than 
> being ignored).  This was especially frustrating because you even 
> asked people on the devel mailing list to review your MVCnPHP code. 


Yes, you have.  I haven't touched the MVC code since my last message to 
the list.  You aren't being ignored and I can understand how my silence 
can give you that impression so my apologies.  Your input I took as 
someone who simply gave an honest look at the code (and I do appreciate 
it) but not as a developer.  Had your intents been more clear I would 
have responded differently I'm sure.

>
>
> The lack of communication is hurtful in other ways.  About two weeks 
> ago a developer on the IRC channel was wondering what happened to 
> you.  It seems he had been trying to contact you to help with GL2 for 
> about a month and hadn't been able to get a response.  He was getting 
> so frustrated he considered spawning his own GL2 (obviously under a 
> different name).  I think this is a pretty common feeling with many 
> people who are looking forward to GL2. 

Who is this 'developer'?  To be honest, I have received no real input or 
interest for helping with GL2.  A quick search of in mailbox confirms 
this.  For someone to be to a point to spawn his own GL2 is a bit 
premature considering nobody has been contact with me.  If you can dig 
this person up, let me know and I can follow-up because, again, there is 
more than enough work to go around.

>
>
> In this case, I had recently spoken to you on IRC, so I mentioned to 
> this developer that you were planning on putting a story up on GL2 to 
> get input on the plugin API for GL2 ("the next big thing to do").  
> That was more than a month and a half ago.  I haven't seen the 
> developer's handle on IRC since. 

Well, that was about the point I started getting frustrated.  I have 
literally thousands of lines of code and I decided that before I started 
on the API I should get some feedback on what has already been done.  
Much of the code is still alpha or beta quality and it didn't make sense 
to start something new until some polish was given to things that have 
been done.  The only portions of code that have had any feedback to date 
is the MVC code and, in part, the Translation library.  That still 
leaves a lot of code to be reviewed (authentication and authorization in 
particular).

>
>
> Right now I'm working on a generic configuration module that uses 
> mysql to store config variables and caches them in a php file 
> (hopefully no more hassle with config.php...).  My intent is for this 
> code to be used for GL2, and possible GL1.3.x and other php projects.  
> I'd really prefer to be working on some core GL2 code, but it's 
> impossible to know what to work on when things just "appear" in CVS, 
> and then attempts to contribute are ignored.
>
> To make all this constructive, I'd like to recommend better 
> communication.  Maybe a weekly meeting in IRC that goes through status 
> reports (it could be really short) and a GL2 TODO list maintained on 
> the website and update regularly as task are accomplished and new work 
> appears. 

Actually, I just messaged Dirk that I plan on install the latest gForge 
code (http://gforge.org) and if it is usable I will put GL2 in there and 
work with Dirk to get 1.3.x in there as well.  That shoud serve as a 
suitable project website to facilitate better communication.  IMHO, IRC 
or IM is not worth much as I prefer archives of stuff so email should be 
the focus.  Should we create a geeklog2-devel list? That seems to be of 
value as a lot of mail goes through geeklog-devel that isn't relevant to 
GL2.  Just a thought.

>
>
> This sounds as bad as your post did.  I'm sorry, but this has building 
> for a while and then your post, "a cry for help", struck a pretty raw 
> nerve.   I think I'll send this to you exclusively, and let you mail 
> it to the list or keep it to yourself as you see fit (and I apologize 
> for the two email's, I'm not sure which box your using or even if they 
> are the same box).

No, don't worry, the result is all I am after.  As long as this spurs 
the conversation we need to get things going I will be happy.  So 
consider yourself 'in'.  Track me down in IRC today and I can set you up 
in CVS so you have direct access to the code and then we can talk about 
divying up the tasks.

Please be sure to send me the handle or email of that 'other developer' 
you eluded to.

>
> Thanks for all the hard work though Tony.  No one can fault your 
> dedication or commitment to GL2.
>
> -Vinny
>
> Tony Bibbs wrote:
>
>> Holy bloat batman.
>>
>> Don't get me wrong, I see and understand the value in this.  However, 
>> my hunch is that simply implementing the 1.0 spec (if it is even out) 
>> would take months (at my current rate) and you haven't even begun to 
>> address the core GL2 features.
>>
>> Furthermore, I have no feel for any progress being made on a PHP 
>> Portlet API.  I think having one would be nice but a google search on 
>> "PHP Portlet API" brings up an old geeklog-devtalk post first which 
>> leads me to believe we are the only ones even thinking out load about 
>> this.  I think we could recruit the right people from various PHP CMS 
>> systems to start building the API but do we really want to spend our 
>> time doing this?  I would love to but at the current rate GL2 will be 
>> lucky to be in beta by this winter given the general lack of 
>> additional support I have.
>>
>> Which brings me to my next point.  I am growing increasing frustrated 
>> with the general disinterest in GL2.  I have architected what I 
>> believe to be a good foundation thus far but have received little to 
>> no input and hardly any programming support.  Doing no one any 
>> discredit, I understand why it is important to keep 1.3.x moving and 
>> I don't want to stop that.  However, long ago we had ear marked a few 
>> new features in 1.3.x and had talked about freezing that codebase 
>> except for security and bug fixes.  Now, I admit there wasn't broad 
>> consensus then but I guess I had always hoped for this.  In an effort 
>> to correct this, I have been back in contact with Jason and he is 
>> still onboard albeit it in a limited capacity.
>>
>> I guess my point is I feel we are at a point where we should shit or 
>> get off the pot.  Perhaps I woke up on the wrong side of bed and I 
>> admit today the glass is half empty so I apologize if this sounds 
>> harsh.   With all this said, is a 1.3.x code freeze feasible?  If 
>> not, I will need help building a general roadmap for GL2 and help 
>> recuiting new help because, to be honest, GL2 has enough code that I 
>> need extra eyes to validate or invalidate what I already have and I 
>> will need consistent help moving forward with it.
>> I think we all agree GL 1.3.x is quickly outgrowing it's 
>> capabilities, I think the user community is interested in the 
>> prospect of GL2 so unless I am way off base I think the need is 
>> there.  I just don't feel the commitment is there and it is to the 
>> point that I have begun to lose interest in it and I hate that.  I 
>> think if we could all get engaged with GL2 to some degree that would 
>> provide sufficient motivation.  My point is something needs to 
>> change.  I have ideas of what those things are but they all require 
>> more fresh bodies...a rare commodity in building OSS.
>>
>> Again, none of this is meant to be personal.  It is just my current 
>> perspective and a cry out for help.
>>
>> --Tony
>>
>> Blaine Lang wrote:
>>
>>
>>> There are few evolving standards with Portals but WSRP (Web Services 
>>> for Remote Portals) looks to be a key standard that we will want to 
>>> support.
>>>  
>>> This presentation explains it well: 
>>> http://xml.coverpages.org/wsrp-overview200206.pdf
>>>  
>>>  
>>>  
>>>  
>>
>>
>>
>>
>>
>> _______________________________________________
>> geeklog-devel mailing list
>> geeklog-devel at lists.geeklog.net
>> http://lists.geeklog.net/listinfo/geeklog-devel
>>