MWOP Has Resigned

Showing 1-33 of 33 messages
MWOP Has Resigned pmjones 2/27/13 10:49 AM
Hi all,

I for one am sad to see him go but I cannot blame him.

    <http://www.mwop.net/blog/2013-02-27-resigned-from-php-fig.html>

The final straw was what I will call the "mob" crying for tabs to be added in PSR-2.



-- pmj

Re: MWOP Has Resigned Andreas Möller 2/27/13 10:51 AM

> I for one am sad to see him go but I cannot blame him.
>
>    <http://www.mwop.net/blog/2013-02-27-resigned-from-php-fig.html>
>
> The final straw was what I will call the "mob" crying for tabs to be added in PSR-2.

:-(
Re: MWOP Has Resigned Mike van Riel 2/27/13 10:56 AM
It is a very sad thing to see him resign, I can understand his frustration (I stopped reading halfway through the tabs vs. spaces threads due to the level of hostility).



--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Re: MWOP Has Resigned Guilherme Blanco 2/27/13 12:08 PM
That reminds me when we decided to open the mailing list for everyone... some people foresee all the listed topics MWOP highlighted as "It will happen sooner or later". =\
--
Guilherme Blanco
MSN: guilher...@hotmail.com
GTalk: guilhermeblanco
Toronto - ON/Canada
Re: MWOP Has Resigned Timothy Boronczyk 2/27/13 12:14 PM
... And this is why we can't have nice things.  Sorry to see Matthew leave. :(

-Tim

Timothy Boronczyk, ZCE
tboro...@gmail.com
Re: MWOP Has Resigned drgomesp 2/27/13 12:20 PM
Sad T_T


Daniel Ribeiro Gomes Pereira
iPhone: +55 (48) 9111-0931


2013/2/27 Timothy Boronczyk <tboro...@gmail.com>

Re: MWOP Has Resigned Lukas Kahwe Smith 2/27/13 12:21 PM

On Feb 27, 2013, at 21:14 , Timothy Boronczyk <tboro...@gmail.com> wrote:

> ... And this is why we can't have nice things.  Sorry to see Matthew leave. :(
>
> -Tim
>
> Timothy Boronczyk, ZCE
> tboro...@gmail.com
>
>
> On Wed, Feb 27, 2013 at 3:08 PM, guilher...@gmail.com <guilher...@gmail.com> wrote:
> That reminds me when we decided to open the mailing list for everyone... some people foresee all the listed topics MWOP highlighted as "It will happen sooner or later". =\
>
>
> On Wed, Feb 27, 2013 at 1:56 PM, Mike van Riel <draco...@gmail.com> wrote:
> It is a very sad thing to see him resign, I can understand his frustration (I stopped reading halfway through the tabs vs. spaces threads due to the level of hostility).

its sad indeed .. and i hope someone else from ZF will step up to take his place.
all the fussing about tabs vs. spaces might have been the final straw but as his parting post clarifies its not like he sees much left for this group, since he seems to disagree with the concept of adding PSRs with interfaces (see also http://www.mwop.net/blog/2012-12-20-on-shared-interfaces.html).

regards,
Lukas
Re: MWOP Has Resigned Andreas Möller 2/27/13 12:48 PM

> its sad indeed .. and i hope someone else from ZF will step up to take his place.
> all the fussing about tabs vs. spaces might have been the final straw but as his parting post clarifies its not like he sees much left for this group, since he seems to disagree with the concept of adding PSRs with interfaces (see also http://www.mwop.net/blog/2012-12-20-on-shared-interfaces.html).

Now, let's move on and continue debating tabs over spaces!


Not.


Best,

Andreas
Re: MWOP Has Resigned Andrew Eddie 2/27/13 5:07 PM
On Thursday, 28 February 2013 04:49:38 UTC+10, pmjones wrote:
The final straw was what I will call the "mob" crying for tabs to be added in PSR-2.

I'll take that as a compliment. As a member of "the mob" (nice one), I'd just like to say after reading his blog, there has obviously been a lot of unwritten understanding of how things should be working. The fact that our current bylaws don't reflect what was intended to be done is a fundamental problem.

Regards,
Andrew Eddie
Re: MWOP Has Resigned pmjones 2/27/13 5:27 PM

On Feb 27, 2013, at 7:07 PM, Andrew Eddie wrote:

> On Thursday, 28 February 2013 04:49:38 UTC+10, pmjones wrote:
> The final straw was what I will call the "mob" crying for tabs to be added in PSR-2.
>
> I'll take that as a compliment. As a member of "the mob" (nice one),

The voting members here are excluded from that appellation.  I apologize.


-- pmj
Re: MWOP Has Resigned Sebastian Krebs 2/27/13 5:30 PM



2013/2/28 Andrew Eddie <mamb...@gmail.com>

On Thursday, 28 February 2013 04:49:38 UTC+10, pmjones wrote:
The final straw was what I will call the "mob" crying for tabs to be added in PSR-2.

I'll take that as a compliment. As a member of "the mob" (nice one), I'd just like to say after reading his blog, there has obviously been a lot of unwritten understanding of how things should be working. The fact that our current bylaws don't reflect what was intended to be done is a fundamental problem.

You can create as many bylaws as you like, but it doesn't change anything, when many ("the mob") simply ignore them (probably because they don't know them, because they didn't read them, because they aren't members ;)). Tabs for example: You can deny the proposal, but wait half a year and the next one will come and even if you deny it once more by pointing to the bylaws, the discussions will come up (again).
There are statements in PSR-2, that I don't completely agree too, but do I need a special bylaw to keep me quiet? I am not a voting member, but the I (the mob ;)) can create much noise, if I really like.

BTW: I like the "mob"-metapher :D

Regards :),
Sebastian

 

Regards,
Andrew Eddie

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/php-fig/-/c3YfMCN9MvgJ.

For more options, visit https://groups.google.com/groups/opt_out.
 
 



--
github.com/KingCrunch
Re: MWOP Has Resigned Guilherme Blanco 2/27/13 5:35 PM
Can we create a ML that only has the approved members?
We could still take the opinions from everyone on this one during idea/feedback collection.
That would solve these issues...

Cheers,
Guilherme Blanco
MSN: guilher...@hotmail.com
GTalk: guilhermeblanco
Toronto - ON/Canada
Re: MWOP Has Resigned Justin Hileman 2/27/13 5:38 PM

On Wed, Feb 27, 2013 at 5:35 PM, guilher...@gmail.com <guilher...@gmail.com> wrote:
Can we create a ML that only has the approved members?
We could still take the opinions from everyone on this one during idea/feedback collection.
That would solve these issues...

I doubt it. Because the same misunderstanding would exist, but instead of a group of "dictators" making up rules for the rest of the PHP community to follow, it would be a group of dictators *making up rules in secret* for the rest of the PHP community to follow.

Re: MWOP Has Resigned Guilherme Blanco 2/27/13 5:41 PM
So can we ban people that constantly message the ML without useful suggestions?
What would be approach to handle this situation?


--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 



--
Guilherme Blanco
MSN: guilher...@hotmail.com
GTalk: guilhermeblanco
Toronto - ON/Canada
Re: MWOP Has Resigned Andrew Eddie 2/27/13 8:16 PM
On 28 February 2013 11:30, Sebastian Krebs <kreb...@gmail.com> wrote:
You can create as many bylaws as you like, but it doesn't change anything, when many ("the mob") simply ignore them (probably because they don't know them, because they didn't read them, because they aren't members ;)). 

There are always going to be people that make life difficult - short of becoming a hermit, that's life. Maybe I'm just used to dealing with them on a larger scale than most people. But if there's something in writing, at least I can give them a link to chew on - what they do with it then, I don't care about. This is simply a lesson in being slack on i-dotting and t-crossing.

Regards,
Andrew Eddie
Re: MWOP Has Resigned Andrew Eddie 2/27/13 8:18 PM
On 28 February 2013 11:38, justin <jus...@justinhileman.info> wrote:
PHP community to follow, it would be a group of dictators *making up rules in secret* for the rest of the PHP community to follow.

I'm having that "hulk" moment again. Keep it open.

Regards,
Andrew Eddie 
Re: MWOP Has Resigned Paul Scott 2/27/13 8:55 PM
On 28/02/2013 03:38, justin wrote:
> I doubt it. Because the same misunderstanding would exist, but instead
> of a group of "dictators" making up rules for the rest of the PHP
> community to follow, it would be a group of dictators *making up rules
> in secret* for the rest of the PHP community to follow.
>
>

This is a very bad idea. Secrecy is never a way to approach an open topic.

With that being said, I agree mostly with Matthew's post, except for the
interfaces bit. I too, have mostly distanced myself from discussions
recently, as they are cyclic, redundant (for the most part) and I simply
do not have the bandwidth to read all of it. I end up picking up
snippets from the names that I recognise and respect, which sometimes is
not enough to make an informed decision.

I too had/have high hopes for this group. This, too, has been a dream of
mine ever since the inception of my framework and PHP4 for that matter,
so I really would like to see it work...

With that being said, I think most of us have stuck through the growing
pains. We need to now refocus and carry on into the next level. There
are issues, but nothing that an entire group of intelligent people can't
figure out!

-- Paul
Re: MWOP Has Resigned Sebastian Krebs 2/27/13 11:56 PM



2013/2/28 Andrew Eddie <mamb...@gmail.com>

On 28 February 2013 11:30, Sebastian Krebs <kreb...@gmail.com> wrote:
You can create as many bylaws as you like, but it doesn't change anything, when many ("the mob") simply ignore them (probably because they don't know them, because they didn't read them, because they aren't members ;)). 

There are always going to be people that make life difficult - short of becoming a hermit, that's life. Maybe I'm just used to dealing with them on a larger scale than most people. But if there's something in writing, at least I can give them a link to chew on - what they do with it then, I don't care about.

Like they do :) 

Of course there are always trolls, but a group cannot work, when the background noise starts to paralyzing useful discussion.
 
This is simply a lesson in being slack on i-dotting and t-crossing. 

Regards,
Andrew Eddie

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 



--
github.com/KingCrunch
Re: MWOP Has Resigned Lukas Kahwe Smith 2/27/13 11:57 PM

On Feb 28, 2013, at 5:55 , Paul Scott <psco...@gmail.com> wrote:

> On 28/02/2013 03:38, justin wrote:
>> I doubt it. Because the same misunderstanding would exist, but instead of a group of "dictators" making up rules for the rest of the PHP community to follow, it would be a group of dictators *making up rules in secret* for the rest of the PHP community to follow.
>>
>>
>
> This is a very bad idea. Secrecy is never a way to approach an open topic.

Yeah, I dont think that this really solves anything. What could maybe work is to encourage people to create separate mailing list for specific topics (cache etc) with the assumption that this way people more genuinely interested in the topic will subscribe and discuss.

But in the grand scheme of things we are in the same boat as php-src .. or maybe even worse, since contrary to them we are even more dependent on consensus and our barrier to entry is even lower (ie. most PHP devs can come up with a PSR, the number of people that can code is much smaller).

That being said, we also need to get better at documenting why the PSRs we passed are the way they are. I noticed this especially with the Logger PSR, which within hours got lots of negative feedback about things that we explicitly had to do differently because as we were discussing it realized that the "naive" solution didnt work.

> With that being said, I agree mostly with Matthew's post, except for the interfaces bit. I too, have mostly distanced myself from discussions recently, as they are cyclic, redundant (for the most part) and I simply do not have the bandwidth to read all of it. I end up picking up snippets from the names that I recognise and respect, which sometimes is not enough to make an informed decision.
>
> I too had/have high hopes for this group. This, too, has been a dream of mine ever since the inception of my framework and PHP4 for that matter, so I really would like to see it work...
>
> With that being said, I think most of us have stuck through the growing pains. We need to now refocus and carry on into the next level. There are issues, but nothing that an entire group of intelligent people can't figure out!

I think the process will continue to be painful, we will improve our workflow, we will learn how to avoid some issues, but even if within this group we manage to get back to a less aggressive tone and a more constructive POV, there will be enough people outside that will do their best to ensure that every PSR will be known for how many kittens it kills.

regards,
Lukas
Re: MWOP Has Resigned Drak 2/28/13 1:23 AM
A members only list does not have to be private. It can be on world read, just no post rights for non-members. And, we could use it for voting without any spurious people voting/commenting.

Regards,

Drak



--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Re: MWOP Has Resigned Paul Dragoonis 2/28/13 3:32 AM



On Thu, Feb 28, 2013 at 7:57 AM, Lukas Smith <sm...@pooteeweet.org> wrote:

On Feb 28, 2013, at 5:55 , Paul Scott <psco...@gmail.com> wrote:

> On 28/02/2013 03:38, justin wrote:
>> I doubt it. Because the same misunderstanding would exist, but instead of a group of "dictators" making up rules for the rest of the PHP community to follow, it would be a group of dictators *making up rules in secret* for the rest of the PHP community to follow.
>>
>>
>
> This is a very bad idea. Secrecy is never a way to approach an open topic.

Yeah, I dont think that this really solves anything. What could maybe work is to encourage people to create separate mailing list for specific topics (cache etc) with the assumption that this way people more genuinely interested in the topic will subscribe and discuss.

Don't push this idea please Lukas.
 

But in the grand scheme of things we are in the same boat as php-src .. or maybe even worse, since contrary to them we are even more dependent on consensus and our barrier to entry is even lower (ie. most PHP devs can come up with a PSR, the number of people that can code is much smaller).

That being said, we also need to get better at documenting why the PSRs we passed are the way they are. I noticed this especially with the Logger PSR, which within hours got lots of negative feedback about things that we explicitly had to do differently because as we were discussing it realized that the "naive" solution didnt work.

As part of our new workflow we should have an "evidence" discussion or a section dedicated to links to integral discussion threads and also any surveys or supplementary document references.
 

> With that being said, I agree mostly with Matthew's post, except for the interfaces bit. I too, have mostly distanced myself from discussions recently, as they are cyclic, redundant (for the most part) and I simply do not have the bandwidth to read all of it. I end up picking up snippets from the names that I recognise and respect, which sometimes is not enough to make an informed decision.
>
> I too had/have high hopes for this group. This, too, has been a dream of mine ever since the inception of my framework and PHP4 for that matter, so I really would like to see it work...
>
> With that being said, I think most of us have stuck through the growing pains. We need to now refocus and carry on into the next level. There are issues, but nothing that an entire group of intelligent people can't figure out!

I think the process will continue to be painful, we will improve our workflow, we will learn how to avoid some issues, but even if within this group we manage to get back to a less aggressive tone and a more constructive POV, there will be enough people outside that will do their best to ensure that every PSR will be known for how many kittens it kills.

regards,
Lukas

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Re: MWOP Has Resigned Drak 2/28/13 8:48 AM
I am not surprised that someone left... frankly, as soon as I saw the sudden vote thread, my heart sank and I thought to myself, "I don't think I can be bothered with this crap any more". You can expect more people to bail if we don't have some kind of decorum here. Even the kinds of discussions around a sort of "compromise/appease the tabs religion with a PSR-4" makes me wonder if this is just a matter of time before mobs erode our willpower and overthrow everything, followed by overthrowing that.  The fact people might be thinking to make such a change to appease the mob is already a clear disaster in waiting, and a sign of things to come.

Additionally, the attempt to reduce noise by creating the CS group is flawed to me because I know there are not so many subscribed by choice, certainly not all the voting members - and who blames them? who wants the extra noise? Then I fear discussions and proposals might be made there and suddenly we'd see another vote effectively broadsiding many voting members.  Aside,  I think we really do need a voting members list (on public read access) so that we can vote there and a final closing arguments time before a vote goes out, so the vote is expected and cannot be used as a weapon. I also think it's time people started using their filters to ignore conversations they are not interested in rather than making knee jerk reactions (like the CS list was all about). Yeah, I'll be flamed by some for that, but there is truth in it.

So anyway, I feel the frustration too. I know others are not being vocal about their frustration - and others have just tuned out entirely. If we don't tread carefully here, there will be more casualties and this group could lose it's credibility very quickly.

Regards,

Drak



--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Re: MWOP Has Resigned Robert Hafner 2/28/13 10:25 AM

As a "non-member" who has worked on the caching proposal for over a year now, I find this offensive. Just because I'm not part of the ruling clique doesn't mean my voice shouldn't be heard.

The problem with this group is simple- there's no actual leadership. There's no one who steps up and says "no, this was already voted on, I'm closing this conversation". There's no enforcement of the bylaws. That alone will help the problem.

Another way to deal with this is to vote on *having* discussions. It seemed fairly obvious to me that there weren't goign to be any changes with regards to tabs versus spaces, so maybe instead of arguing about it for dozens of emails someone should have just called a vote to see if there was even interest in the discussion. If it turns out that a majority of the voting members weren't going to ratify any changes regardless, then the conversation could have been shut down as pointless. 

Robert
Re: MWOP Has Resigned Amy Stephen 2/28/13 10:37 AM
There probably is some danger to things falling apart if others also get discouraged. For that reason, it might be better to be a little overly positive when posting. Pretend it until you feel it. =)

Drak - a lot of your comments about the membership, noise, identifying membership, voting, were helpful for workflow. Appreciate that.

We'll get some sidewalks in, signs that say "Enter here", doors shut when an issue is closed, membership voting area with public viewing,  things will straighten out. Crowd proof the place.

Good to see discussion on process, too. If it's not written in the bylaws - it's only your assumption. No sense in being frustrated with a member who could have just as easily assumed the other.

Fake your optimism for awhile, until there is a little more structure and improvements to the rules for process in place. Then, it'll be as good as new.
Re: MWOP Has Resigned Sebastian Krebs 2/28/13 1:44 PM



2013/2/28 Robert Hafner <ted...@tedivm.com>


As a "non-member" who has worked on the caching proposal for over a year now, I find this offensive. Just because I'm not part of the ruling clique doesn't mean my voice shouldn't be heard.

The problem with this group is simple- there's no actual leadership. There's no one who steps up and says "no, this was already voted on, I'm closing this conversation". There's no enforcement of the bylaws. That alone will help the problem.

Regarding spaces-vs-tabs episode 13: I remember one or two, who mentioned, that there was already a vote about it andthe topic is not under discussion anymore...



--
github.com/KingCrunch
Re: MWOP Has Resigned Drak 2/28/13 1:45 PM
On 28 February 2013 18:25, Robert Hafner <ted...@tedivm.com> wrote:
The problem with this group is simple- there's no actual leadership. There's no one who steps up and says "no, this was already voted on, I'm closing this conversation". There's no enforcement of the bylaws. That alone will help the problem.

That's really interesting. Would you mind suggesting some text for this? I would be interested to see options here.

Regards,

Drak
Re: MWOP Has Resigned Lukas Kahwe Smith 2/28/13 1:48 PM
Robert seems to be talking about the benevolent dictator, however imho this model only works if there is a natural indisputable leader, which is in most cases I have seen a founder. I do not think that such a person is identifiable for FIG. at most we could have someone evolve over time that becomes the ethical/moral compass but this will take a lot of time and lucky coincidences.

regards,
Lukas Kahwe Smith
sm...@pooteeweet.org



Re: MWOP Has Resigned Drak 2/28/13 1:53 PM
Robert, could you put some thoughts into the PR that's running at the moment: https://github.com/php-fig/fig-standards/pull/93/, or even start something new (PR or gist) if you have a specific idea.

Regards,

Drak


On 28 February 2013 18:25, Robert Hafner <ted...@tedivm.com> wrote:

As a "non-member" who has worked on the caching proposal for over a year now, I find this offensive. Just because I'm not part of the ruling clique doesn't mean my voice shouldn't be heard.

The problem with this group is simple- there's no actual leadership. There's no one who steps up and says "no, this was already voted on, I'm closing this conversation". There's no enforcement of the bylaws. That alone will help the problem.

Another way to deal with this is to vote on *having* discussions. It seemed fairly obvious to me that there weren't goign to be any changes with regards to tabs versus spaces, so maybe instead of arguing about it for dozens of emails someone should have just called a vote to see if there was even interest in the discussion. If it turns out that a majority of the voting members weren't going to ratify any changes regardless, then the conversation could have been shut down as pointless. 

Robert



On Feb 28, 2013, at 1:23 AM, Drak <dr...@zikula.org> wrote:

A members only list does not have to be private. It can be on world read, just no post rights for non-members. And, we could use it for voting without any spurious people voting/commenting.

Regards,

Drak


On 28 February 2013 07:57, Lukas Smith <sm...@pooteeweet.org> wrote:

On Feb 28, 2013, at 5:55 , Paul Scott <psco...@gmail.com> wrote:

> On 28/02/2013 03:38, justin wrote:
>> I doubt it. Because the same misunderstanding would exist, but instead of a group of "dictators" making up rules for the rest of the PHP community to follow, it would be a group of dictators *making up rules in secret* for the rest of the PHP community to follow.
>>
>>
>
> This is a very bad idea. Secrecy is never a way to approach an open topic.

Yeah, I dont think that this really solves anything. What could maybe work is to encourage people to create separate mailing list for specific topics (cache etc) with the assumption that this way people more genuinely interested in the topic will subscribe and discuss.

But in the grand scheme of things we are in the same boat as php-src .. or maybe even worse, since contrary to them we are even more dependent on consensus and our barrier to entry is even lower (ie. most PHP devs can come up with a PSR, the number of people that can code is much smaller).

That being said, we also need to get better at documenting why the PSRs we passed are the way they are. I noticed this especially with the Logger PSR, which within hours got lots of negative feedback about things that we explicitly had to do differently because as we were discussing it realized that the "naive" solution didnt work.

> With that being said, I agree mostly with Matthew's post, except for the interfaces bit. I too, have mostly distanced myself from discussions recently, as they are cyclic, redundant (for the most part) and I simply do not have the bandwidth to read all of it. I end up picking up snippets from the names that I recognise and respect, which sometimes is not enough to make an informed decision.
>
> I too had/have high hopes for this group. This, too, has been a dream of mine ever since the inception of my framework and PHP4 for that matter, so I really would like to see it work...
>
> With that being said, I think most of us have stuck through the growing pains. We need to now refocus and carry on into the next level. There are issues, but nothing that an entire group of intelligent people can't figure out!

I think the process will continue to be painful, we will improve our workflow, we will learn how to avoid some issues, but even if within this group we manage to get back to a less aggressive tone and a more constructive POV, there will be enough people outside that will do their best to ensure that every PSR will be known for how many kittens it kills.

regards,
Lukas

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.




--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Re: MWOP Has Resigned Amy Stephen 2/28/13 2:13 PM
On Thu, Feb 28, 2013 at 12:25 PM, Robert Hafner <ted...@tedivm.com> wrote:


A members only list does not have to be private. It can be on world read, just no post rights for non-members. And, we could use it for voting without any spurious people voting/commenting.

Regards,

Drak


Robert - Having a public read, member-only posting area for discussions and voting would be helpful. It supports the membership role and responsibilities and it helps the public not accidentally get in the way, something most people would be horrified to do.

Now, it might be helpful to add and in a working group lead(s) during the discussion phase for their PSR. They could then be available for questions. That maybe was your point? Then, those folks could exit for the vote. Keeps the historical record, clear, too.
RE: MWOP Has Resigned Michael Cullum 2/28/13 3:32 PM
Perhaps a system where voting members nominate themselves, then the 'chair'
would be voted for by members, they stay in the position for 1 year (or 8-10
months), then another is chosen (and a member cannot be a chair for two
terms in a row)? That would make it fair and it would stop it becoming too
much of a dictatorship but still provides a leader that can ensure rules are
enforced, make sure things are merged when accepted etc. Should anything
untoward happen (although I'm sure it wouldn't), all that would be needed
would be a vote of no confidence.

Thanks,
Michael Cullum
Re: MWOP Has Resigned Paul Dragoonis 2/28/13 4:54 PM



Spot on
 

regards,
Lukas Kahwe Smith
sm...@pooteeweet.org



--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



Re: MWOP Has Resigned Robert Hafner 2/28/13 10:55 PM

This is actually a discussion I tried to start almost a year ago ( https://groups.google.com/group/php-fig/browse_thread/thread/211edd4e46855572/c3069182bca0d071 ), and one that I think is going to be more important as time goes on. I'm not speaking of a benevolent dictator at all, although I can see where you'd get that. I think a better idea would be something of a steering committee, with a more meta set of roles. 

Right now there are a lot of answered questions about this group. Off the top of my head-

1. What is the actual mission?

2. How is it decided what's currently being worked on?

3. Who qualifies to be a member?

4. How are bylaws made or modified?

5. What should the process for proposing, amending and ratifying a PSR be?

6. What if a flaw is found in a ratified proposal? Should that be different for minor (typos) versus functional ones?

7. How should "amendments", or superseding standards, be dealt with?

8. What rules of civility should be in place, and how should they be enforced?

9. Who maintains the website? The repository? How are they picked?


Again, those are just off the top of my head, I'm sure there are many others.


The thing is this group just kind of sprang out of no where (at least as far as most people can tell), and so most people don't know the answers to these questions. For things like software projects this works okay- typically the person who started it has a good idea where it should go, and it can easily be forked. For a standards group that's a completely different story though.

This is why I think a steering committee is a good idea. The steering committee's immediately focus should be on coming up with a comprehensive set of bylaws that answer those questions. After that the steering committee should enforce them, handle things like calling votes, keeping tasks on focus, and probably doing management things like website changes, creating github repositories, and approving pull requests.

I think with a little organization this group can be great, but I can tell you that as it stands how it's extremely frustrating. I tried, really hard, to get people engaged on the caching proposals. What ended up happening was that I had to answer the same questions, over and over, had no clear way of seeing if people besides the handful emailing actually wanted the proposal to move forward, nor any consensus on which direction forward meant. On top of that every couple of months a new proposal pops up and we start from square one. The lack of process, and the inherent issues with doing this on a mailing list, make it seem impossible to actually make new standards. This is a fixable problem, but it has to be addressed.

Robert
Re: MWOP Has Resigned John Mertic 3/1/13 5:37 AM
I think you are spot on. A board of directors, helping us recognize the trends going on in the industry and emerging patterns, would do wonders for helping us build direction. 

The inherent bias of this group is our current focus I'm our day jobs; building software. With the day to day realities of that role, it's no surprise that topics like indents vs spaces keep bubbling up. As valuable as they are in a standards ( which I'd argue is diminishing with the rise of IDEs that take care of recognizing tokens for you ), they don't have as much impact on interoperability, which judging from the title is the focus of the group. It's a question of tactical vs strategic, and I feel the community as a whole benefits from the group leaning their expertise on the side of strategic and less tactical.

I think was you are alluding to Robert is the need for a GOSPA exercise for the group. I'd be happy to help lead such an exercise if the group feels there is merit.

Thanks!

John Mertic
SugarCRM


On Friday, March 1, 2013, Robert Hafner wrote:

This is actually a discussion I tried to start almost a year ago ( https://groups.google.com/group/php-fig/browse_thread/thread/211edd4e46855572/c3069182bca0d071 ), and one that I think is going to be more important as time goes on. I'm not speaking of a benevolent dictator at all, although I can see where you'd get that. I think a better idea would be something of a steering committee, with a more meta set of roles. 

Right now there are a lot of answered questions about this group. Off the top of my head-

1. What is the actual mission?

2. How is it decided what's currently being worked on?

3. Who qualifies to be a member?

4. How are bylaws made or modified?

5. What should the process for proposing, amending and ratifying a PSR be?

6. What if a flaw is found in a ratified proposal? Should that be different for minor (typos) versus functional ones?

7. How should "amendments", or superseding standards, be dealt with?

8. What rules of civility should be in place, and how should they be enforced?

9. Who maintains the website? The repository? How are they picked?


Again, those are just off the top of my head, I'm sure there are many others.


The thing is this group just kind of sprang out of no where (at least as far as most people can tell), and so most people don't know the answers to these questions. For things like software projects this works okay- typically the person who started it has a good idea where it should go, and it can easily be forked. For a standards group that's a completely different story though.

This is why I think a steering committee is a good idea. The steering committee's immediately focus should be on coming up with a comprehensive set of bylaws that answer those questions. After that the steering committee should enforce them, handle things like calling votes, keeping tasks on focus, and probably doing management things like website changes, creating github repositories, and approving pull requests.

I think with a little organization this group can be great, but I can tell you that as it stands how it's extremely frustrating. I tried, really hard, to get people engaged on the caching proposals. What ended up happening was that I had to answer the same questions, over and over, had no clear way of seeing if people besides the handful emailing actually wanted the proposal to move forward, nor any consensus on which direction forward meant. On top of that every couple of months a new proposal pops up and we start from square one. The lack of process, and the inherent issues with doing this on a mailing list, make it seem impossible to actually make new standards. This is a fixable problem, but it has to be addressed.

Robert




On Feb 28, 2013, at 4:54 PM, Paul Dragoonis wrote:




On Thu, Feb 28, 2013 at 9:48 PM, Lukas Kahwe Smith <smith@pooteeweet.org> wrote:

On Feb 28, 2013, at 22:45 , Drak <drak@zikula.org> wrote:


> On 28 February 2013 18:25, Robert Hafner <tedivm@tedivm.com> wrote:
> The problem with this group is simple- there's no actual leadership. There's no one who steps up and says "no, this was already voted on, I'm closing this conversation". There's no enforcement of the bylaws. That alone will help the problem.
>
> That's really interesting. Would you mind suggesting some text for this? I would be interested to see options here.

Robert seems to be talking about the benevolent dictator, however imho this model only works if there is a natural indisputable leader, which is in most cases I have seen a founder. I do not think that such a person is identifiable for FIG. at most we could have someone evolve over time that becomes the ethical/moral compass but this will take a lot of time and lucky coincidences.

Spot on
 

regards,
Lukas Kahwe Smith
smith@pooteeweet.org



--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+unsubscribe@googlegroups.com.
To post to this group, send email to php-fig@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.




--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+unsubscribe@googlegroups.com.
To post to this group, send email to php-fig@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+unsubscribe@googlegroups.com.
To post to this group, send email to php-fig@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.
 
 


--
John Mertic
jme...@gmail.com | http://jmertic.wordpress.com
More topics »