I wish retake old good practice

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
34 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Simon Michael
+1


Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Simon Michael
In reply to this post by Andreas.Raab
+1.

On 3/10/10 8:58 AM, Andreas Raab wrote:

> I think that's exactly right. Perhaps we should formulate the tasks for
> a release more clearly. Here's an attempt at doing this:
>
> Release Tasks:
> * Drive the process to decide which packages to include besides the core
> packages. This needs to be a community process where the community
> decides what should be in Squeak and what shouldn't. The release team's
> task is to implement the results of this discussion.
>
> * Drive the process about what the out-of-the-box look should be. Which
> projects should be loaded by default, what they should contain, how the
> preferences should be set, etc.
>
> * Ensure that important bugs are closed for the new release, solicit
> feedback on the missing issues, decide what are considered release blockers.
>
> * Ensure all tests are green.
>
> * Ensure proper packaging of the result on all platforms, verify
> installation procedures, find a variety of testers for the result.
>
> * Ship it.
>
> Am I missing something?

Yes, I think these things also should happen in some shape or form:

* Update web presence and docs. This includes ensuring that
   latest packages are properly linked and downloadable;
   all "official" web pages are updated to reflect current status;
   release notes are available; #squeak topic is updated.
   This should be live before announcements go out.

* Announce the release. This includes preparing one or more official
   release announcements (eg full and short version), with
   sufficient community preview and refinement; identifying suitable places
   to announce; posting the announcement to those places at the right time.

* Release followup. This includes watching replies to the announcement,
   and release-related queries on irc and squeak lists, and when appropriate:
   following up with public replies, forwarding problems to the appropriate place,
   and generally helping ensure the release is a successful one, up to
   and including making follow-up releases to fix critical issues.


"Releasing is always four times as much work as you think"



Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Ian Trudel-2
In reply to this post by Bert Freudenberg
2010/3/10 Bert Freudenberg <[hidden email]>:
> Please do *not* take release discussion off this list. It was not "good practice". Instead, it was one of the reasons that the larger community had no idea what was going on. Everyone here should be interested in the next release so it makes no sense to take it off squeak-dev.


I agree with you, Bert. It's important to keep some buzz and
excitement but more importantly involvement of the community in any
such discussion. Having too much low traffic or too specific mailing
lists will only fragment the community's attention. Sometimes going
smaller is making things bigger.

Ian.
--
http://mecenia.blogspot.com/

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Randal L. Schwartz
>>>>> "Ian" == Ian Trudel <[hidden email]> writes:

Ian> I agree with you, Bert. It's important to keep some buzz and excitement
Ian> but more importantly involvement of the community in any such
Ian> discussion. Having too much low traffic or too specific mailing lists
Ian> will only fragment the community's attention. Sometimes going smaller is
Ian> making things bigger.

In particular, I think what was missing was a call to action on the part of
*developers* on squeak-dev, about how to use Keith's tools.  The reasons for
this boggled me when I heard them -- apparently, Keith was forbidden to post
on the developer list because of his work arrangement.  I'm still a bit
flabberghasted at that... the guy creating the tools couldn't talk about tools
in the one place that the developers who wanted to use the tools would
consistently be.  Certainly not a recipe for communication or leadership.

There's certainly a need for a smaller group to talk about the final steps, so
a specific, short-term release mailing list makes sense.

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<[hidden email]> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

keith1y
>>>>>> "Ian" == Ian Trudel <[hidden email]> writes:
>
> Ian> I agree with you, Bert. It's important to keep some buzz and  
> excitement
> Ian> but more importantly involvement of the community in any such
> Ian> discussion. Having too much low traffic or too specific mailing  
> lists
> Ian> will only fragment the community's attention. Sometimes going  
> smaller is
> Ian> making things bigger.
>
> In particular, I think what was missing was a call to action on the  
> part of
> *developers* on squeak-dev, about how to use Keith's tools.

Those that were interested were invited to have a tour of the bob  
server via vnc and a look around in February I think it was, and to  
try builds for themselves.

The one person who tried it out liked what they saw and is the only  
person qualified to comment. That person is Ken Brown, and as Ken says  
if he can do it anyone can.

>  The reasons for
> this boggled me when I heard them -- apparently, Keith was forbidden  
> to post

Not forbidden just unwise.

> on the developer list because of his work arrangement.  I'm still a  
> bit
> flabberghasted at that... the guy creating the tools couldn't talk  
> about tools
> in the one place that the developers who wanted to use the tools would
> consistently be.  Certainly not a recipe for communication or  
> leadership.

Except that we had already decided that the release list was the place  
for that conversation, and this arrangement had worked for the two  
previous releases. We also had irc,  a wiki, and mantis, and allegedly  
a board liaison person.

The squeak-dev conversation that I saw was predictably driven by a  
group of about 5 newbies whom I had never seen before, and as far as I  
know have no community contributions to squeak before or since, and no  
knowledge of any historical context.

These guys beat up on my 3.11 consolidating maintainance development,  
(needed as 3.x reaches the end of its life according to the board)  
when really they should have been complaining that the brand new  
flashy promised Squeak5.0 was nowhere to be seen.

In my book the only people qualified to make significant direction  
changes as far as a release goes are those who have an active  
interest, and those who are making an active contribution, and those  
who have put their time and effort into actually volunteering to do  
stuff. Those people would be on the release list. The rest is noise,  
that would only confuse and confound.

Keith


Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Igor Stasenko
On 11 March 2010 04:54, keith <[hidden email]> wrote:

>>>>>>> "Ian" == Ian Trudel <[hidden email]> writes:
>>
>> Ian> I agree with you, Bert. It's important to keep some buzz and
>> excitement
>> Ian> but more importantly involvement of the community in any such
>> Ian> discussion. Having too much low traffic or too specific mailing lists
>> Ian> will only fragment the community's attention. Sometimes going smaller
>> is
>> Ian> making things bigger.
>>
>> In particular, I think what was missing was a call to action on the part
>> of
>> *developers* on squeak-dev, about how to use Keith's tools.
>
> Those that were interested were invited to have a tour of the bob server via
> vnc and a look around in February I think it was, and to try builds for
> themselves.
>
> The one person who tried it out liked what they saw and is the only person
> qualified to comment. That person is Ken Brown, and as Ken says if he can do
> it anyone can.
>
>>  The reasons for
>> this boggled me when I heard them -- apparently, Keith was forbidden to
>> post
>
> Not forbidden just unwise.
>
>> on the developer list because of his work arrangement.  I'm still a bit
>> flabberghasted at that... the guy creating the tools couldn't talk about
>> tools
>> in the one place that the developers who wanted to use the tools would
>> consistently be.  Certainly not a recipe for communication or leadership.
>
> Except that we had already decided that the release list was the place for
> that conversation, and this arrangement had worked for the two previous
> releases. We also had irc,  a wiki, and mantis, and allegedly a board
> liaison person.
>
> The squeak-dev conversation that I saw was predictably driven by a group of
> about 5 newbies whom I had never seen before, and as far as I know have no
> community contributions to squeak before or since, and no knowledge of any
> historical context.
>
Oh, good.. looks like we finally found who didn't allowed you to deliver: noobs!
Thanks for sharing this conclusion with us.

> These guys beat up on my 3.11 consolidating maintainance development,
> (needed as 3.x reaches the end of its life according to the board) when
> really they should have been complaining that the brand new flashy promised
> Squeak5.0 was nowhere to be seen.
>
> In my book the only people qualified to make significant direction changes
> as far as a release goes are those who have an active interest, and those
> who are making an active contribution, and those who have put their time and
> effort into actually volunteering to do stuff. Those people would be on the
> release list. The rest is noise, that would only confuse and confound.
>

So, Keith. What is the purpose of squeak-dev mailing list to your thinking?
A place for noob flame wars?
I'm interested in answer (oh .. looks like we using this list to conversate)...

> Keith
>


--
Best regards,
Igor Stasenko AKA sig.

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Igor Stasenko
Keith's references to noobs reminded me the song (guess who) ;)

Hey noob, don't make it bad
Take a bad soft and make it better
Remember to let it into your heart
Then you can start to make it better

Hey noob, don't be afraid
You were made to go out and get it
The minute you let it under your skin
Then you begin to make it better
....

--
Best regards,
Igor Stasenko AKA sig.

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Michael Haupt-3
Igor!

Am 11.03.2010 um 06:11 schrieb Igor Stasenko <[hidden email]>:
> Hey noob, don't make it bad
> Take a bad soft and make it better
> Remember to let it into your heart
> Then you can start to make it better
>
> Hey noob, don't be afraid
> You were made to go out and get it
> The minute you let it under your skin
> Then you begin to make it better

Thank you, this made my day before it has even started. The  
Beatles ... :-)

Proclaim this the OSS anthem. :-)

Best,

Michael

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Andreas.Raab
In reply to this post by Edgar De Cleene
On 3/10/2010 9:54 AM, Edgar J. De Cleene wrote:
> On 3/10/10 2:58 PM, "Andreas Raab"<[hidden email]>  wrote:
>> Am I missing something?
> What about complete new ways of work ?

I'm not quite sure what you mean here but the release process isn't
usually the place to innovate. See for example the articles here:

    http://en.wikipedia.org/wiki/Release_management
    http://en.wikibooks.org/wiki/Release_Management

This is a bit industry-parlance heavy but it does relate to open source
projects quite directly.

Cheers,
   - Andreas

> Because just complete SqueakCore-9371-alpha, in the way to go Experiments
> folder.
> And I wish this become 4.2 when all you saw in the videos on youtube is
> polished enough to all CoreDevelopers, not to all in squeak dev or to Board
> will.
> For that I working hard.
> For that on each Monday we have reports and the next Monday, two reports.
> One for FunSqueak , or the process for have the most big image with the most
> packages which could load and work .
> Another for SqueakCore.
>
> Edgar
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Edgar De Cleene



On 3/11/10 4:15 AM, "Andreas Raab" <[hidden email]> wrote:

> I'm not quite sure what you mean here but the release process isn't
> usually the place to innovate. See for example the articles here:


Well , if I can't do in Release new ways and only business as usual ....
And what the trunk is ? A new way you start.
Too bad "my expertise", which is not to much do not have more exposure.
I like SL3 become SqueakCore, but if you do not want it, stop me until
someone else comes with the same is veeeery unfair.

All now talk about modularization. JustTalk.
I very old for JustTalk, need SmallTalk now and not in two years.

But if the only thing I could do is YOUR list , well I do.
And  continue with Mendieta at my own.
Can't use Experiments for it?
And let bold Squeakers have now 4.2 ?

Edgar



Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Bert Freudenberg
On 11.03.2010, at 07:40, Edgar J. De Cleene wrote:

>
>
>
>
> On 3/11/10 4:15 AM, "Andreas Raab" <[hidden email]> wrote:
>
>> I'm not quite sure what you mean here but the release process isn't
>> usually the place to innovate. See for example the articles here:
>
>
> Well , if I can't do in Release new ways and only business as usual ....
> And what the trunk is ? A new way you start.
> Too bad "my expertise", which is not to much do not have more exposure.
> I like SL3 become SqueakCore, but if you do not want it, stop me until
> someone else comes with the same is veeeery unfair.
>
> All now talk about modularization. JustTalk.
> I very old for JustTalk, need SmallTalk now and not in two years.
>
> But if the only thing I could do is YOUR list , well I do.
> And  continue with Mendieta at my own.
> Can't use Experiments for it?
> And let bold Squeakers have now 4.2 ?

Well, IMHO you have to wear two hats at the same time.

For new stuff that should get into a release eventually, wear your core-developer hat. Discuss on squeak-dev, get it into trunk.

For actually making the release, wear your release-manager hat. Package what's in trunk. No more.

And btw, nobody is saying the trunk process is perfect. We need to refine it, and possibly switch to another model some day if we figure out an even better way to gather contributions. But for now it's what we have, and to get wide support for your ideas IMHO it would be best for you to stick to it.

- Bert -


Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Edgar De Cleene



On 3/11/10 8:42 AM, "Bert Freudenberg" <[hidden email]> wrote:

> On 11.03.2010, at 07:40, Edgar J. De Cleene wrote:
>>
>>
>>
>>
>> On 3/11/10 4:15 AM, "Andreas Raab" <[hidden email]> wrote:
>>
>>> I'm not quite sure what you mean here but the release process isn't
>>> usually the place to innovate. See for example the articles here:
>>
>>
>> Well , if I can't do in Release new ways and only business as usual ....
>> And what the trunk is ? A new way you start.
>> Too bad "my expertise", which is not to much do not have more exposure.
>> I like SL3 become SqueakCore, but if you do not want it, stop me until
>> someone else comes with the same is veeeery unfair.
>>
>> All now talk about modularization. JustTalk.
>> I very old for JustTalk, need SmallTalk now and not in two years.
>>
>> But if the only thing I could do is YOUR list , well I do.
>> And  continue with Mendieta at my own.
>> Can't use Experiments for it?
>> And let bold Squeakers have now 4.2 ?
>
> Well, IMHO you have to wear two hats at the same time.
>
> For new stuff that should get into a release eventually, wear your
> core-developer hat. Discuss on squeak-dev, get it into trunk.
>
> For actually making the release, wear your release-manager hat. Package what's
> in trunk. No more.
>
> And btw, nobody is saying the trunk process is perfect. We need to refine it,
> and possibly switch to another model some day if we figure out an even better
> way to gather contributions. But for now it's what we have, and to get wide
> support for your ideas IMHO it would be best for you to stick to it.
>
> - Bert -

Fine , you win.
End of discussion.

Arriba Mendieta .
El Sur tambien existe.

Edgar



Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

keith1y
In reply to this post by Igor Stasenko
>
> So, Keith. What is the purpose of squeak-dev mailing list to your  
> thinking?
> A place for noob flame wars?

Since squeak is a development tool, squeak-dev is a place for users of  
squeak. i.e. Developers that "use" squeak to "develop" things. Hence  
the name squeak-dev.

> I'm interested in answer (oh .. looks like we using this list to  
> conversate)...


The release list, surprisingly is for those who want to develop or  
contribute to developing a release of squeak.

I am stuck here because with "cuis", Juan has chosen the honourable  
path of being different, but not using the difference as an excuse for  
splitting the community.

I want to develop stuff on top of smalltalk, I don't like low level  
coding much, but after 4 years looking at the same cruft and lack of  
modularity in squeak it is tiresome. There still isn't a process that  
enables fixing any of the cruft.

When it comes to the release I have always had one and only one  
question that has bee top of my list since August 2006

What is the timeline for replacing FileDirectory, and how are you  
going to achieve it, across all forks and all 700+ packages, even one  
fork would be a start.

If you cant answer that question then you aren't doing anything I am  
interested in. Closures I can do without, fancy graphics I can ignore,  
multiheaded beasts are not useful to me, fancy tools I don't need I am  
happy with the PackagePaneBrowser, but every time I have to use  
FileDirectory I feel the brain-cells dying.

regards

Keith

 

Reply | Threaded
Open this post in threaded view
|

Re: I wish retake old good practice

Igor Stasenko
On 11 March 2010 13:32, keith <[hidden email]> wrote:

>>
>> So, Keith. What is the purpose of squeak-dev mailing list to your
>> thinking?
>> A place for noob flame wars?
>
> Since squeak is a development tool, squeak-dev is a place for users of
> squeak. i.e. Developers that "use" squeak to "develop" things. Hence the
> name squeak-dev.
>
>> I'm interested in answer (oh .. looks like we using this list to
>> conversate)...
>
>
> The release list, surprisingly is for those who want to develop or
> contribute to developing a release of squeak.
>
> I am stuck here because with "cuis", Juan has chosen the honourable path of
> being different, but not using the difference as an excuse for splitting the
> community.
>
> I want to develop stuff on top of smalltalk, I don't like low level coding
> much, but after 4 years looking at the same cruft and lack of modularity in
> squeak it is tiresome. There still isn't a process that enables fixing any
> of the cruft.
>
> When it comes to the release I have always had one and only one question
> that has bee top of my list since August 2006
>
> What is the timeline for replacing FileDirectory, and how are you going to
> achieve it, across all forks and all 700+ packages, even one fork would be a
> start.
>
> If you cant answer that question then you aren't doing anything I am
> interested in. Closures I can do without, fancy graphics I can ignore,
> multiheaded beasts are not useful to me, fancy tools I don't need I am happy
> with the PackagePaneBrowser, but every time I have to use FileDirectory I
> feel the brain-cells dying.
>

So what is the problem? Write it and use it. Give it away. If somebody
will use it also, its cool,
if not - not your problem.

> regards
>
> Keith
>
>
>



--
Best regards,
Igor Stasenko AKA sig.

12