Re: The parallel projects (was Re: Apache 2.4 and php 5.6 test branch)

[Michael Tremer <michael.tremer@ipfire.org>]

[-- Attachment #1: Type: text/plain, Size: 7740 bytes --]

Hi,

On Wed, 2016-01-20 at 08:50 +0100, ummeegge wrote:
> Hi all,
> 
> Am 19.01.2016 um 12:41 schrieb Michael Tremer <
> michael.tremer(a)ipfire.org>:
> > 
> > When starting with something we often don't know where we will end
> > up.
> > That's quite normal. However going to the very end won't work
> > either.
> 
> That´s true but i´am not sure what you mean with "going to the very 
> end" ? Regarding to the mailproxy/server project it is beneath an 
> also surely commercial intend also a kind of fun for the involved 
> people there to build up some features they want to have but it 
> contains also a little knowledge (may for other interested people) of 
> how it is possible to build your own stuff in a Linux from Scratch 
> environment by examples and realize your own wishes but also concrete 
> steps how it could be in general possible to deliver bugs or to 
> handle a little more with Git and i know about the will of the 
> involved people to cooperate with the IPFire project and deliver 
> their work too but there are sadly some obstacles sometimes with 
> language (english in this case) and possibly also the how to 
> integrate themselves into the development process.

I agree with that it is always hard to get in touch with the right
people and that many people want to be really sure they did everything
right. They are too afraid to make mistakes. That is on them I suppose.
I cannot repeat myself enough to share your work even if it is not
perfect, yet, and even if you have many questions. Doing mistakes is
normal and talking about them avoids some.

The language barrier is a second thing. However I think that everyone
can do this. One just has to start and then learn to become better with
reading and writing English.

Marcel got in touch on the forums about this matter where he sent me a
private message. In that one he signals that he has no interest what so
ever in this mailing list and no interest what so ever in posting
patches on here. *That* is a completely different thing which I cannot
accept at all.

> I´am not sure how much + work you have had causing this project but 
> basically this circle (3-5 people) tried to help themselves at all 
> and not to stick to the core developers and their time, also a bigger 
> part of the work which was done in there concerns also the core part 
> of IPFire.

I haven't had any coding work with this at all. I haven't looked at the
git branches because at a first glance this is just a mess and it is
quite pointless to find out what is going on without putting too much
time info this.

Nobody of the people who are involved in this mail proxy/server project
have been in touch with me except you.

Other people have been trying to extract some parts from the Git branch
with help of the original developers and make them ready to be merged
into the main distribution which completely failed.

> May it is the wrong part to discuss the sense of Addons like Postfix,
> Amavisd or even Spamassassin on IPFire since they where and are
> available, some people are interested in the usage of it and the
> project starts initially with an idea to implement another security
> layer like the mailproxy. This aren´t a core part of IPFire since you
> need to install this Addons first, it can be seen a little similar to
> the Copfilter project. Nevertheless i do not see updates for Apache,
> PHP, Perl and a lot of other basic (common) components as a beside
> work or 3rd party stuff please correct me if i´am wrong.
> 
> The whole Email flood you deliver may cause also of this kind of
> projects which are going on in the forum but how should we deal with
> this ? The forum is a good possibility to find a wider range of
> people with various systems, i think it is also a voice from and for
> the community beneath small talk and support. It is hard to see
> before what will happens with this or that topic or how much
> questions will be addressed to the wrong people. Could it be a
> solution to redirect them just to the appropriate topics if you have
> nothing to do with this code ?

I do not see why it is not possible to have a conversation on a mailing
list just in the same way you are having that on the forums.

If someone doesn't agree with the emails being sent to him, just
unsubscribe. Nobody is forced to stay on this list. Nobody is even
forced to read all the emails. But if you want to have a word and if
you want a peer-reviewed distribution, there must be a place where that
will happen.

> > We can't ship that all in just one update. There are many reasons
> > for
> > that.
> > 
> > 1) The updates will get too big to ship.
> > 
> > 2) The updates will be too big to get tested. Not many people are
> > engaging in this any way.
> > 
> > And we actually don't need to have this so stressful.
> > 
> > Some things of that like PHP can be updated individually or with
> > fewer dependencies. So we start with that and go on from there.
> 
> For sure. I tried to deliver this many time also in other forum
> sections, in that specific i tried to deliver ideas of how we can
> split all that into bite-sized morsels, mentioned also many times the
> usage of the mailinglist and have called for the usage of it and i
> don´t know why it is such a big hurdle even a lot of stuff are
> already available. Although some help also in there is definitely
> needed cause we stuck there also with some unsolvable problems.
> PHP and Apache was also a splitted working part of that and i
> currently do not know what the state or problem is in there ?

The current state is that Marcel deleted the branch where he worked on
that. It seems that all the effort that was put into that was for
nothing.

The reason for that might have been a misunderstanding, but I do not
know to whom he was talking (not to me) and what was said, because that
didn't happen in public on this list, but was in private.

> > Does anyone know if they are even all subscribed?
> 
> Not sure about that, have seen Dirk so far, Daniel has delivered the
> mail correspondence with Marcel in here, Thomas wrote also in here
> some time ago i think and Matthias helped in the forums section also
> out and is shurley also here present ;-).
> 
> So let me mentioned in the end. Like Dirk said it already, it was and
> is not our intend to make things worst with our Forum projects for
> anybody, the opposite was the case and i feel a little sorry that all
> the achieved work produces frustration cause it was also meant to
> save time for the core developers by delivering tested updates also
> for core system components where it sadly seems to struggles at the
> end by finding a solution how to deliver them or simply to speak with
> another.

I didn't assume that anyone is trying to make things worse
deliberately.

> For myself, i have also some other open projects in the forum (e.g.
> Portspoof, Pmacct, Ossec) which may also never comes to an end simply
> because it is nice to research for new functionalities also with
> other people but also to bring on new ideas and cast an eye on new
> stuff. I can understand the hassle which you have Michael, so just
> say it and i will close them and stop this kind of work if it makes
> no sense at all but only stress.

I didn't consider them to be parallel projects that fall into this
category which is supposed to be merged right away. This is a test and
you guys play around with that. If that is supposed to be merged is a
different thing since there needs to be a group of people who are
interested in using this and willing to maintain the package.

An update of essential core components is something different though.

Best,
-Michael

> 
> So far from here.
> 
> Greetings,
> 
> Erik
> 
> 
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]