[Axiom-developer] Re: next release

[Gabriel Dos Reis]

On Tue, 27 Feb 2007 address@hidden wrote:

| Gaby,
|
| > This is true in general, including your changes to Gold we don't see
| > until the Big Day. I'm trying to single you out. We all have local
| > experimental trees that would do more harms if published than kept local.
|
| I published my whole research tree (in git, since svn was broken)
| in November when we had this same discussion but got no response.

It appeared to me to be an "experiment".  I did not exactly what I
should "answer".

| All of my axiom-related work is now kept in that tree.
|
|
|
|
| >...[snip]... Please note also
| > that because Gold and Silver have gone, until very recently, into very
| > uncertain status ...[snip]...
|
| The plan was to publish a new Gold at the end of November.  But in the
| beginning of November we had the big discussion about me being "a
| single point of failure" so I stepped out of the way and told everyone
| that any changes they wanted in Gold they would have to do
| themselves.  Everyone had permission to do that. Since that time no-one
| has changed anything. It became clear that nobody was going to make
| the effort to create a new release.

My understanding was that decision was the outcome of
miscommunication, therefore I was not going to do any release or do any
commit to Silver or Gold (I don't know exactly which trees are those)
_until that was resolved_.  Clearly, I'm not going to take anybody's
place if that someone thinks he is being "kicked out".  Sorry to put
it that way, but that is how your message sounded.


[ For example, I needed an "official" version for my last semester
  class project, but in the end I distributed an official version
  of build-improvements so that students can work.  I would have
  preferred it was official. ]

|  I've decided to step back into the role of releasing systems and am
| trying to catch up to the outstanding changes.

I very much welcome your stepping back into the role of releasing the
system.

| I tried maintaining silver but I was the only one making changes.

Going back to last semester discussion, you're the only one doing the
commit to Silver, because it appeared to me that:
  (1) the system was set up in a way that you're the only one to
      effectively commit to silver, e.g. any commit to SF SVN
      reposiotyr is effectively "lost" from your perspective.

  (2) you wanted to take over silver, which I was originally maintaining.

| Plus the additional copies of Axiom were causing rising server costs.
| If it depends on my effort silver is dead. I only have enough time
| available to do Gold and do my own axiom-related work. I certainly
| don't have time to maintain silver. Since advocacy is volunteering
| you're welcome to create and maintain a silver version that is a
| test version for gold changes.

So, we are again back to what things were in April-September 2006?
This is an important question I need a clear answer for.

As I already pointed out, we should encourage people to commit
themselves their patches to Silver.  That reduces the load on the
maintainer.

-- Gaby