[squeak-dev] What Constitutes a Complete and Final Release?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
18 messages Options
Reply | Threaded
Open this post in threaded view
|

[squeak-dev] What Constitutes a Complete and Final Release?

Ken Causey-3
I don't believe the 3.10 release is in a state that we should consider
complete and final.  My complaint does not concern anything about the
image contents, there are perhaps concerns there as well but I do not
intend to address them here.  What I'm concerned with is the appearance
to someone who is looking for the latest release of Squeak.

My ideal example of a release is 3.8:

http://ftp.squeak.org/3.8/

You find here both basic and full images clearly labelled.  There is a
gzipped sources file.  You also find platform support directories.
There is also an ISO file for a CD which I don't really remember
anything about.

By the way I've had some complaints about gzipped source files and the
lack of support for ungzipping on all platforms.  Perhaps we should
rethink that?

I could also wish for README, LICENSE, and maybe NEWS files.

The 3.9 release

http://ftp.squeak.org/3.9/

is not too bad.  They changed the filename convention to the worse in my
opinion.  There is a new sources file (gzipped).  There are also
platform support files although RiscOS is missing (Tim wasn't interested
any longer?).  There is no full image and the image there is not clearly
labelled as Basic.

The 3.10 release

http://ftp.squeak.org/3.10/

is not even close.  Firstly this is the development directory and really
there probably should be a new cleaned up directory with only the
release and related files, not the interim releases.  The release file
is named correctly (I brought this up with Edgar seperately).  There is
no platform support, there is no sources file, there is no full image.
There are certainly not my wished for text files.

My comments here of course represent my idea of a release.  I think it's
something we should discuss and try to form a clear policy on so that
future release teams have a target to aim for.

Ken




signature.asc (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Klaus D. Witzel
May I add to Ken's remarks, What Constitutes a Complete and Final Release?

1 - a release team which does not run away after the release
2 - a release team which does not run away before the release

It seems to me that the board has a problem here, with both 3.9 and 3.10 :(

/Klaus

On Sat, 05 Apr 2008 00:16:16 +0200, Ken Causey wrote:

> I don't believe the 3.10 release is in a state that we should consider
> complete and final.  My complaint does not concern anything about the
> image contents, there are perhaps concerns there as well but I do not
> intend to address them here.  What I'm concerned with is the appearance
> to someone who is looking for the latest release of Squeak.
...
> My comments here of course represent my idea of a release.  I think it's
> something we should discuss and try to form a clear policy on so that
> future release teams have a target to aim for.
>
> Ken
>



Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: What Constitutes a Complete and Final Release?

timrowledge

On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>
> It seems to me that the board has a problem here, with both 3.9 and  
> 3.10 :(


Well, no. It's not the board that has any sort of problem; the board  
is a bunch of people that meet every couple of weeks to discuss an  
agenda that is primarily organisation based (ie getting to a state  
where we can join the SFLC foudation thingy). The board is not a team  
that can be asked or expected to solve all *our* problems. *You lot*  
get to do the actual work. Until and unless you provide funds to pay a  
team....

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
"Virtual Memory" means never knowing where your next byte is coming  
from.





Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Klaus D. Witzel
On Sat, 05 Apr 2008 06:00:21 +0200, tim Rowledge wrote:

> On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>
>> It seems to me that the board has a problem here, with both 3.9 and  
>> 3.10 :(
>
> Well, no. It's not the board that has any sort of problem; the board is  
> a bunch of people that meet every couple of weeks to discuss an agenda  
> that is primarily organisation based (ie getting to a state where we can  
> join the SFLC foudation thingy). The board is not a team that can be  
> asked or expected to solve all *our* problems.

Someone must be there to be asked, from whom is to be expected that  
problems are addressed (if not solved). And at the moment that's the board.

> *You lot* get to do the actual work. Until and unless you provide funds  
> to pay a team....

How do you mean that, perhaps in the sense of "give money and receive a  
guarantee that something will be done somehow"?

> tim
> --
> tim Rowledge; [hidden email]; http://www.rowledge.org/tim
> "Virtual Memory" means never knowing where your next byte is coming from.
>


Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Andreas.Raab
In reply to this post by timrowledge
tim Rowledge wrote:
>
> On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>
>> It seems to me that the board has a problem here, with both 3.9 and
>> 3.10 :(
>
>
> Well, no. It's not the board that has any sort of problem;

Well, yes. It is the board's problem if it assumes the authority to
decide on the release team (which it has). Half-finished results of a
release team *are* the board's problem. It is not the boards task to fix
the ftp directory on the server; it's the boards task to fix the process
that led to selecting a team that doesn't get that job done. If the
board isn't capable of doing that, then it has a *big* problem in my
understanding.

Cheers,
   - Andreas

  the board is

> a bunch of people that meet every couple of weeks to discuss an agenda
> that is primarily organisation based (ie getting to a state where we can
> join the SFLC foudation thingy). The board is not a team that can be
> asked or expected to solve all *our* problems. *You lot* get to do the
> actual work. Until and unless you provide funds to pay a team....
>
> tim
> --
> tim Rowledge; [hidden email]; http://www.rowledge.org/tim
> "Virtual Memory" means never knowing where your next byte is coming from.
>
>
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: What Constitutes a Complete and Final Release?

timrowledge
In reply to this post by Klaus D. Witzel

On 4-Apr-08, at 9:18 PM, Klaus D. Witzel wrote:
>
>> *You lot* get to do the actual work. Until and unless you provide  
>> funds to pay a team....
>
> How do you mean that, perhaps in the sense of "give money and  
> receive a guarantee that something will be done somehow"?

Unless someone works out a way to have a paid team to do all the work  
people want done then *you lot* - all the users of Squeak - have to  
communally do the work.

On 4-Apr-08, at 10:18 PM, Andreas Raab wrote:

>
> tim Rowledge wrote:
>> On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>>
>>> It seems to me that the board has a problem here, with both 3.9  
>>> and 3.10 :(
>> Well, no. It's not the board that has any sort of problem;
>
> Well, yes. It is the board's problem if it assumes the authority to  
> decide on the release team (which it has). Half-finished results of  
> a release team *are* the board's problem. It is not the boards task  
> to fix the ftp directory on the server; it's the boards task to fix  
> the process that led to selecting a team that doesn't get that job  
> done. If the board isn't capable of doing that, then it has a *big*  
> problem in my understanding.

Sadly, the board can't conscript people - not in this universe anyway.  
Unless we magically find that way to have a paid team  etc etc. then  
*you lot* have to step forward in sufficient numbers to do the work.  
The last two releases have been shouldered by a tiny number of people  
who got rewarded with complaints for their efforts.

If you want it done, volunteer to help. If you want to complain... go  
away.

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
C++ is history repeated as tragedy. Java is history repeated as farce.



Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Dan Ingalls
In reply to this post by Andreas.Raab
>>On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>>
>>>It seems to me that the board has a problem here, with both 3.9 and 3.10 :(
>
>tim Rowledge wrote:
>>
>>Well, no. It's not the board that has any sort of problem;

Andreas replied...

>Well, yes. It is the board's problem if it assumes the authority to decide on the release team (which it has). Half-finished results of a release team *are* the board's problem. It is not the boards task to fix the ftp directory on the server; it's the boards task to fix the process that led to selecting a team that doesn't get that job done. If the board isn't capable of doing that, then it has a *big* problem in my understanding.

I have to agree with Andreas here.  Any board worth its salt should consider itself to have a pretty wide scope, and it should feel pain whenever something is not working right.  It may not be able to fix the problem directly, but it ought to be trying to understand it, and working with doers to fix it, to work around it, or to chart another better path.

I'd like to have a pithy solution here, but I need to catch up on some of the context, and I happen to be in the middle of a work-for-pay crunch.  But maybe if we all take more of an "I am 100% responsible" approach to our problems we'll start to get some more clarity about what is needed.

        - Dan
---------------------------------------
> the board is
>>a bunch of people that meet every couple of weeks to discuss an agenda that is primarily organisation based (ie getting to a state where we can join the SFLC foudation thingy). The board is not a team that can be asked or expected to solve all *our* problems. *You lot* get to do the actual work. Until and unless you provide funds to pay a team....
>>
>>tim
>>--
>>tim Rowledge; [hidden email]; http://www.rowledge.org/tim
>>"Virtual Memory" means never knowing where your next byte is coming from.


Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: What Constitutes a Complete and Final Release?

Randal L. Schwartz
>>>>> "Dan" == Dan Ingalls <[hidden email]> writes:

Dan> I have to agree with Andreas here.  Any board worth its salt should
Dan> consider itself to have a pretty wide scope, and it should feel pain
Dan> whenever something is not working right.  It may not be able to fix the
Dan> problem directly, but it ought to be trying to understand it, and working
Dan> with doers to fix it, to work around it, or to chart another better path.

I too agree here.  I think it's the job of the board to be accountable for the
overall vision of Squeak.  I'd rather the board get out of the way when things
are already working, but if it's clear that things aren't working and some
decisions need to be made, I'll be personally motivated to make sure that
those shortcomings are on the board's agenda.

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<[hidden email]> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] What Constitutes a Complete and Final Release?

Edgar J. De Cleene
In reply to this post by Ken Causey-3



El 4/4/08 7:16 PM, "Ken Causey" <[hidden email]> escribió:

> The 3.10 release
>
> http://ftp.squeak.org/3.10/
>
> is not even close.  Firstly this is the development directory and really
> there probably should be a new cleaned up directory with only the
> release and related files, not the interim releases.  The release file
> is named correctly (I brought this up with Edgar seperately).  There is
> no platform support, there is no sources file, there is no full image.
> There are certainly not my wished for text files.
>
> My comments here of course represent my idea of a release.  I think it's
> something we should discuss and try to form a clear policy on so that
> future release teams have a target to aim for.
>
> Ken

Well, surprise me we have a list "Discussion about development of Squeak
3.10" <[hidden email]>, where you, Klaus, Andreas, (not
remember if Tim) was and never said nothing about this.
Blaming me for Ralph unknown reasons is unfear.
I work , but I don't was the man in charge.
Send several unanswered  mails to release list, private and here.
I wish move on, send tentative goals and tentative image for 3.11.
With Matthew, Craig, Pavel and some more people we still meeting and
discussing.

Remember you I was who said to Ralph we should go to 3.8 practices , I
always said this is the way to go.

> Unless we magically find that way to have a paid team  etc etc.
said Tim

Not bad.
You must know in Argentina we have talented unpaid students.
With the same money ESUG trough away for each (and still I wish see the
results), you can have a six person team here.
And I could do for the price to maintain a good reliable mirror here.

> I'd like to have a pithy solution here, but I need to catch up on some of the
> context, and I happen to be in the middle of a work-for-pay crunch.  But maybe
> if we all take more of an "I am 100% responsible" approach to our problems
> we'll start to get some more clarity about what is needed.
>
>     - Dan

You are the same guy who start all don't you ?
So I take every word you said about how and start to work.
If you wish can send a list of troubles addressed by Ken and solutions to
each.

To all, read the "Discussion about development of Squeak 3.10"
<[hidden email]> post , please.

Waiting directions....

Edgar



Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Klaus D. Witzel
In reply to this post by timrowledge
On Sat, 05 Apr 2008 07:50:56 +0200, tim Rowledge wrote:

> On 4-Apr-08, at 9:18 PM, Klaus D. Witzel wrote:
>>
>>> *You lot* get to do the actual work. Until and unless you provide  
>>> funds to pay a team....
>>
>> How do you mean that, perhaps in the sense of "give money and receive a  
>> guarantee that something will be done somehow"?
>
> Unless someone works out a way to have a paid team to do all the work  
> people want done

Tim, are these just your words, or more a proposal to potential donators.  
And if so, what does the board say to this?

> then *you lot* - all the users of Squeak - have to communally do the  
> work.
>
> On 4-Apr-08, at 10:18 PM, Andreas Raab wrote:
>>
>> tim Rowledge wrote:
>>> On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>>>
>>>> It seems to me that the board has a problem here, with both 3.9 and  
>>>> 3.10 :(
>>> Well, no. It's not the board that has any sort of problem;
>>
>> Well, yes. It is the board's problem if it assumes the authority to  
>> decide on the release team (which it has). Half-finished results of a  
>> release team *are* the board's problem. It is not the boards task to  
>> fix the ftp directory on the server; it's the boards task to fix the  
>> process that led to selecting a team that doesn't get that job done. If  
>> the board isn't capable of doing that, then it has a *big* problem in  
>> my understanding.
>
> Sadly, the board can't conscript people - not in this universe anyway.  
> Unless we magically find that way to have a paid team  etc etc. then  
> *you lot* have to step forward in sufficient numbers to do the work. The  
> last two releases have been shouldered by a tiny number of people who  
> got rewarded with complaints for their efforts.
>
> If you want it done, volunteer to help. If you want to complain... go  
> away.
>
> tim
> --
> tim Rowledge; [hidden email]; http://www.rowledge.org/tim
> C++ is history repeated as tragedy. Java is history repeated as farce.
>
>
>
>



Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: What Constitutes a Complete and Final Release?

garduino
In reply to this post by Andreas.Raab
Full agree.

After read all the thread, some points come to my mind:

1. The release (aka the Squeak itself) is our main asset, then the
Board must participate with decisions if the things don't go ok (As
Edgar commented, seems very clear that in 3.10 we lossed the Team
Leader).

2. Personally think that is a bad thing (or at least homy thing) to
talk about money when problems of this nature appear. This is an open
source community, we aren't here by money, and talk about money simply
to elude responsabilities is unfair. If a guy don't have time to the
work he promised to do, must step out and find a replacement, but
claim by money isn't fair I think.

Just my 0.02.




2008/4/5, Andreas Raab <[hidden email]>:

> tim Rowledge wrote:
>
> >
> > On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
> >
> > >
> > > It seems to me that the board has a problem here, with both 3.9 and 3.10
> :(
> > >
> >
> >
> > Well, no. It's not the board that has any sort of problem;
> >
>
>  Well, yes. It is the board's problem if it assumes the authority to decide
> on the release team (which it has). Half-finished results of a release team
> *are* the board's problem. It is not the boards task to fix the ftp
> directory on the server; it's the boards task to fix the process that led to
> selecting a team that doesn't get that job done. If the board isn't capable
> of doing that, then it has a *big* problem in my understanding.
>
>  Cheers,
>   - Andreas
>
>
>   the board is
>
> > a bunch of people that meet every couple of weeks to discuss an agenda
> that is primarily organisation based (ie getting to a state where we can
> join the SFLC foudation thingy). The board is not a team that can be asked
> or expected to solve all *our* problems. *You lot* get to do the actual
> work. Until and unless you provide funds to pay a team....
> >
> > tim
> > --
> > tim Rowledge; [hidden email]; http://www.rowledge.org/tim
> > "Virtual Memory" means never knowing where your next byte is coming from.
> >
> >
> >
> >
> >
> >
> >
>
>
>


--
Germán S. Arduino
http://www.arsol.biz
http://www.arsol.net

Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] What Constitutes a Complete and Final Release?

Ken Causey-3
In reply to this post by Edgar J. De Cleene
On Sat, 2008-04-05 at 05:34 -0300, Edgar J. De Cleene wrote:
> To all, read the "Discussion about development of Squeak 3.10"
> <[hidden email]> post , please.

Edgar,

I just want to make it clear I was not intending my email as an attack
on you or the 3.10 (or even 3.9) release teams.  I was trying to start a
discussion to form a community consensus on what would consitute a
release file/directory organization which I then expected we would bring
to you and ask you to conform to (and offer assistance in doing so).  At
this point I think it is best to continue to discuss this with the
community as a whole and later come to you with a clear definition of
what we believe is desirable, so I would like this discussion to
continue on squeak-dev for now.  Of course you and all the other release
team members should feel free to participate.

Ken



signature.asc (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Andreas.Raab
In reply to this post by Edgar J. De Cleene
Edgar J. De Cleene wrote:
> Well, surprise me we have a list "Discussion about development of Squeak
> 3.10" <[hidden email]>, where you, Klaus, Andreas, (not
> remember if Tim) was and never said nothing about this.
> Blaming me for Ralph unknown reasons is unfear.

Indeed. And I never intended to. My criticism wasn't about any release
in particular, it was aimed at Tim's comment effectively saying that
problems with the release aren't the board's problems.

Chers,
   - Andreas

> I work , but I don't was the man in charge.
> Send several unanswered  mails to release list, private and here.
> I wish move on, send tentative goals and tentative image for 3.11.
> With Matthew, Craig, Pavel and some more people we still meeting and
> discussing.
>
> Remember you I was who said to Ralph we should go to 3.8 practices , I
> always said this is the way to go.
>
>> Unless we magically find that way to have a paid team  etc etc.
> said Tim
>
> Not bad.
> You must know in Argentina we have talented unpaid students.
> With the same money ESUG trough away for each (and still I wish see the
> results), you can have a six person team here.
> And I could do for the price to maintain a good reliable mirror here.
>
>> I'd like to have a pithy solution here, but I need to catch up on some of the
>> context, and I happen to be in the middle of a work-for-pay crunch.  But maybe
>> if we all take more of an "I am 100% responsible" approach to our problems
>> we'll start to get some more clarity about what is needed.
>>
>>     - Dan
>
> You are the same guy who start all don't you ?
> So I take every word you said about how and start to work.
> If you wish can send a list of troubles addressed by Ken and solutions to
> each.
>
> To all, read the "Discussion about development of Squeak 3.10"
> <[hidden email]> post , please.
>
> Waiting directions....
>
> Edgar
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] What Constitutes a Complete and Final Release?

Ken Causey-3
In reply to this post by Ken Causey-3
Oh and one detail I forgot:  The naming convention for the entire
package (zip file) containing the release should also be followed by the
contained .image and .changes files.  This is not the case in 3.10.

Ken

On Fri, 2008-04-04 at 17:16 -0500, Ken Causey wrote:

> I don't believe the 3.10 release is in a state that we should consider
> complete and final.  My complaint does not concern anything about the
> image contents, there are perhaps concerns there as well but I do not
> intend to address them here.  What I'm concerned with is the appearance
> to someone who is looking for the latest release of Squeak.
>
> My ideal example of a release is 3.8:
>
> http://ftp.squeak.org/3.8/
>
> You find here both basic and full images clearly labelled.  There is a
> gzipped sources file.  You also find platform support directories.
> There is also an ISO file for a CD which I don't really remember
> anything about.
>
> By the way I've had some complaints about gzipped source files and the
> lack of support for ungzipping on all platforms.  Perhaps we should
> rethink that?
>
> I could also wish for README, LICENSE, and maybe NEWS files.
>
> The 3.9 release
>
> http://ftp.squeak.org/3.9/
>
> is not too bad.  They changed the filename convention to the worse in my
> opinion.  There is a new sources file (gzipped).  There are also
> platform support files although RiscOS is missing (Tim wasn't interested
> any longer?).  There is no full image and the image there is not clearly
> labelled as Basic.
>
> The 3.10 release
>
> http://ftp.squeak.org/3.10/
>
> is not even close.  Firstly this is the development directory and really
> there probably should be a new cleaned up directory with only the
> release and related files, not the interim releases.  The release file
> is named correctly (I brought this up with Edgar seperately).  There is
> no platform support, there is no sources file, there is no full image.
> There are certainly not my wished for text files.
>
> My comments here of course represent my idea of a release.  I think it's
> something we should discuss and try to form a clear policy on so that
> future release teams have a target to aim for.
>
> Ken
>



signature.asc (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Klaus D. Witzel
In reply to this post by Randal L. Schwartz
On Sat, 05 Apr 2008 09:28:54 +0200, Randal L. Schwartz wrote:

>>>>>> "Dan" == Dan Ingalls <[hidden email]> writes:
>
> Dan> I have to agree with Andreas here.  Any board worth its salt should
> Dan> consider itself to have a pretty wide scope, and it should feel pain
> Dan> whenever something is not working right.  It may not be able to fix  
> the
> Dan> problem directly, but it ought to be trying to understand it, and  
> working
> Dan> with doers to fix it, to work around it, or to chart another better  
> path.
>
> I too agree here.  I think it's the job of the board to be accountable  
> for the
> overall vision of Squeak.

These are strong words, Andreas, Dan and Randal and I'm sure that  
everybody here likes to hear them (every now and then) and see some action  
into that direction.

/Klaus

>  I'd rather the board get out of the way when things
> are already working, but if it's clear that things aren't working and  
> some
> decisions need to be made, I'll be personally motivated to make sure that
> those shortcomings are on the board's agenda.
>



Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] What Constitutes a Complete and Final Release?

stephane ducasse
In reply to this post by Ken Causey-3
Hi ken

Once there is a precise description of what should be done, we can do  
it.

Stef


On Apr 5, 2008, at 4:31 PM, Ken Causey wrote:

> On Sat, 2008-04-05 at 05:34 -0300, Edgar J. De Cleene wrote:
>> To all, read the "Discussion about development of Squeak 3.10"
>> <[hidden email]> post , please.
>
> Edgar,
>
> I just want to make it clear I was not intending my email as an attack
> on you or the 3.10 (or even 3.9) release teams.  I was trying to  
> start a
> discussion to form a community consensus on what would consitute a
> release file/directory organization which I then expected we would  
> bring
> to you and ask you to conform to (and offer assistance in doing  
> so).  At
> this point I think it is best to continue to discuss this with the
> community as a whole and later come to you with a clear definition of
> what we believe is desirable, so I would like this discussion to
> continue on squeak-dev for now.  Of course you and all the other  
> release
> team members should feel free to participate.
>
> Ken
>


Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: What Constitutes a Complete and Final Release?

Hilaire Fernandes-4
In reply to this post by timrowledge
tim Rowledge a écrit :

>
> On 4-Apr-08, at 8:49 PM, Klaus D. Witzel wrote:
>>
>> It seems to me that the board has a problem here, with both 3.9 and
>> 3.10 :(
>
>
> Well, no. It's not the board that has any sort of problem; the board is
> a bunch of people that meet every couple of weeks to discuss an agenda
> that is primarily organisation based (ie getting to a state where we can
> join the SFLC foudation thingy). The board is not a team that can be
> asked or expected to solve all *our* problems. *You lot* get to do the
> actual work. Until and unless you provide funds to pay a team....


I found this statement pretty irresponsible :(
And reading this statement one may think what is the Sqf good for
(beside the never ending registration in SFLC...) and outsiders may even
fell that Squeak community is not a good place to go (no clear roadmap).

Hilaire



Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: What Constitutes a Complete and Final Release?

Michael Davies-2
In reply to this post by Randal L. Schwartz
On Sat, Apr 5, 2008 at 8:28 AM, Randal L. Schwartz
<[hidden email]> wrote:
...
>  I think it's the job of the board to be accountable for the
>  overall vision of Squeak.  I'd rather the board get out of the way when things
>  are already working, but if it's clear that things aren't working and some
>  decisions need to be made, I'll be personally motivated to make sure that
>  those shortcomings are on the board's agenda.

It strikes me that Edgar and co have got us 95% of the way to a
release that everyone would be happy with, and that the problem is
understanding who has to do what to complete that final 5%. The board
could help here by clarifying the release process. This could be a
straightforward task, which then puts the onus on the community to do
the legwork:

1) Agree and publish (on squeak.org) a set of criteria for releases to
be considered complete (formats, naming, pre-prepared bundles for key
OSes). Ken's email looks a good starting point. It may also be worth
considering whether specific criteria should be added to each future
release (eg target date, feature list).

2) Agree and publish (as above) a simple release process (e.g. release
manager announces to Board and squeak-dev list that they believe that
the image is ready for release. If no-one objects, the release manager
prepares all required files according to the agreed criteria and loads
them to a staging area, and announces to Board and squeak-dev list
that the packages are ready.  All then have the opportunity to check
that it meets the criteria,  and if no-one objects, the Board
announces the release. This authorises the release manager to copy the
files to their official home, the web team to update the squeak.org
links and the news team to publicise).

3)  For future releases ensure that there is a clearly defined release
manager who is aware of, and agrees with, the criteria and process.
The board have no further direct involvement with the release under
normal circumstances.

4) Once the criteria in (1) have been established, the Board should
consider if the work done for 3.10 meets them, and if not, ask for a
volunteer to co-ordinate the work required to do so.

I hate to add to the workload of people who already contribute their
free time, but given the importance of the Squeak.org releases to the
community, I suggest that this should be considered the main agenda
item for the next board meeting. Certainly getting (1) and (4) sorted
will help us put this release to bed quickly, and (2) and (3) will
help shape the discussions about the next release.

Based on the nature of the discussion so far, I will update the news
item on the Weekly Squeak to clarify that work is still under way to
produce final and fully packaged release files.

Cheers,
Michael