directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Karasulu <>
Subject Re: Txn heads up
Date Sat, 31 Dec 2011 09:04:20 GMT
On Friday, December 30, 2011, Emmanuel Lécharny <>
> On 12/30/11 8:25 AM, Alex Karasulu wrote:
>> I'm trying to balance two different concerns with the activity I am
>> in the TXN branch. First I know we need to move forward on this matter
>> second I know Selcuk will be trying to work although with less time on
>> while dealing with a relocation and we're altering much of it: the moving
>> target problem. Then again having more people capable of working in this
>> region is a big plus. Let's just make sure that the changes are really
>> necessary instead of just moving code around. That way we can make it
>> easier for Selcuk to get back into the code when he has time.
> I don't know what make you feel that we are just 'moving code around'.
This is the first important commit done on this branch, and I carefully
balanced the pros and cons of doing it. The reason why I pushed this big
chunk of code is that I wasn't able to move forward without making a
difference from mandatory changes (those I committed) and some changes I
may want to revert. In any case, having pushed one unique commit is the
guarantee we can rollback it completely in one shoot.

Cool. Just nervous since this is so important for us. Not suggesting what u
are doing is needless. Sometimes we do however make some formatting
changes. I did that too while looking in this branch. It gives me a comfort
level to have it in our format. Although tees changes are not a big deal.

> Now, everything in this commit has been discussed on the mailing list.
There is no surprise, and nothing critical has been changed to the
transaction layer (yet).

Thats great - please disregard my concerns here - they're seeming now not
to be driven by reality but by fear. I myself need to give this more time
but as what seems to be the usual I get less and less time :(.

> There is one single issue that need to be addressed, the encapsulation of
transactions. This is not easy to address.

I would like to help on this

> Obviously, it would be way easier if Selcuk had time to dedicate on this,
but as he said, he is currently relocating, and is not available. He even
asked on the mailing list for someone to relay him, which is what I'm
doing. I guarantee you I'm extremely careful, and I'm not trying to go too

Cool again I apologize for the unfounded concerns.

> Last, not least, as you said, having more than one committer being able
to work on the code is not only important, it's vital. I have no idea if
Selcuk's employer will allow him at all to continue to work on an OSS
project (yes, this is a burden, and no, it's not about Selcuk, it's a
problem for each one of us. In France, for instance, you have to ask your
employer if you want to work on an OSS project, even on your own time). It
would be very, very sad if it were the case, because Selcuk's code is
really good and bring us to a level we would not have been able to reach
without a lot of efforts.

Yeah we need more peeps in this area and u are doing a great job.

>> I know that none of us would make changes just to make a change. But here
>> if you can just give some extra effort and consideration before making an
>> alteration so we can keep it as familiar to Selcuk as possible when he
>> back. Of course on the other hand if we have to advance and the change is
>> critical then we must change it.
> I think I have given all the extras efforts, plus a bit more, trust me on
that. I *wish* I was not the only one spending time on this part of the
code (Selcuk is helping as much as he can), it could be much better and
faster. I appreciate that you care and share such a concern, though.

I'm perhaps more panic oriented having less time to be involved - this my

> I'll be MIA for the next two days, so I wish an happy new year to all of
you guys !

Have a great new year!


> --
> Regards,
> Cordialement,
> Emmanuel Lécharny

Best Regards,
-- Alex

View raw message