We need your help!!! Where is the Community ?

[raphael]

Quote:So the general advice is dont use ISPCP Omega as is at the moment because its unstable and won't be upgradable. Then we have people who dont want to build new servers using VHCS as a control panel.

I know it's not easy; I'm also in this situation but all I can do is try to do the things ASAP, but without collaboration I can't go any faster. RatS wants me to test all my changes before committing, but, does he even realise that that means to recreate a vm, get the new sources, install, and test? well, I don't have enough time for that, I barely have time now.

Also, there are things that really disappoint me from other people. An example? well, I introduced the new way to handle the translations but even when all that needs to be done to update the main catalogue is run one script and then an other one to update the translations... nobody does it. I have to do it (and after that people complain because I made some other changes to the translations; how does people want me to see those changes between thousands of code indention changes? updating the main catalogue would have been more than enough).

I joined the project trying to make some changes that were really required, instead of only making them on our systems. I even contacted VHCS RELOADED and VHCS OMEGA teams so they could work together, but it didn't really happen.

If people can't handle the way I do the things, that really need to be done, then I'll better step apart and work on my own (and probably end up rewriting everything from scratch)

[raphael]

Quote:We do not need any new features now and we do not need any enhancements. We can do this afterwards, this is the point I don't agree with Raphael's opinion.

but I think we really do, why? because if right now I migrate all our customers to ispCP Omega I would still have to make many other changes in order to get it into shape. Like?
* amavis/maia support
* awstats support
* multiple nameservers support
(those are only the main ones right now, but there are some other structure changes that also need to be done)

[joximu]

It's getting interesting now...

well - we have some important points:
a) there should really be some sort of stable release now otherwise more people are using something other (see post 88 from rsmithgs).
There are peoplo needing a new panel - they should not start with VHCS now - should they?
b) the structure is not in a final state for further enhancements.

It's not easy - both are important.
I only see something like this as a way out:
- trying to release a stable version ASAP (1.0 or 0.9 - I don't care)
- continuing the work on the base/structure for the next step (1.1 or 1.0)

We have Raphael who lift's the whole application on some higher level but as long as he is doing this many other people are stuck with their work.

We should really discuss this!

Peace!
Joximu

[joximu]

raphael Wrote:

Quote:We do not need any new features now and we do not need any enhancements. We can do this afterwards, this is the point I don't agree with Raphael's opinion.

but I think we really do, why? because if right now I migrate all our customers to ispCP Omega I would still have to make many other changes in order to get it into shape. Like?
* amavis/maia support
* awstats support
* multiple nameservers support
(those are only the main ones right now, but there are some other structure changes that also need to be done)

You're right: a webhosting without Spam-Management is a nonsense today. I poersonally can live some months without awstats but this may depend...

So, Raphael, could you point out what sort of basic changes have to be done. I think if more of the other devs know what's planned (what you are planning) we could help in a better way. Just know it's a "one man show". You don't give much attention to the release plans - you do what's good for your own server. Ok - this is not bad, but if we know that, we can better help you - or am I wrong?

J

[raphael]

Quote:you do what's good for your own server

because many people will also use them

I've also talked to Clemens (a friend of mine) who is the one who takes care of the multiple nameservers support on our service to see if he can think about a design that is easy to implement in ispCP.

Quote:but as long as he is doing this many other people are stuck with their work.

because when people only want to fix things instead of adding must haves and fixing those too

Quote:could you point out what sort of basic changes have to be done

first of all: finish awstats support (RatS is said to be taking care of it), amavis/maia support, multiple nameservers support
other things:
* libapache2-mod-fcgi support
* cronjobs for customers (the engine side is almost done, GUI side needs to be finished, some database changes are required)
* backups in a per domain basis switch
* same as above but for apache logs
* ispCP data backups (mainly only the ispcp database)
* fix all the javascript-displayed messages so they show useful information (e.g. the domain it is asking you to confirm you want to delete). For this I've found this and this other lib that would provide the same syntax of the translations system currently being used by the PHP code. But the first one should not be used until a license is specified (I already contacted the author via mail some weeks ago but sill no response), and the second one is GPL which would conflict with the rest of the MPL js code
* add option to change, at least, the MX records of a domain

[rsmithgs]

Gawd whats been started!

I think many of us adopted VHCS originally in good faith to find issues that the dev team were unwilling to communicate. Obviously this is a situation non of us want to repeat!

I agree with Joximu, there probably needs to be a discussion, probably not with the likes of me as I'm unimportant but obviously between the important members of the team.

Regarding testing I feel both Rats and Raphael are right. Changes do need testing but I feel that Raphael's time is better spent doing development, therefore would it be worth getting some official testers on board and having offical testing procedures?

I'm of the believe that when any project or business gains a certain size there needs to be clearly defined roles and procedures, otherwise people start to loose sight of the end goal.

What is the end goal, I believe it is to produce a high quality control panel product that has real community support behind it. Obviously a huge goal and needs to be broken down a bit.

I love what this product is developing into and would hate for it to fall apart now. There are a lot of people out there (including me) who cant code but want to help in any way. If it's equipment testing or cash I'm sure the community will provide.

So Structure, Procedure, Communication and using the resources of the community could help ICPCP Omega become a stable product quicker without the sacrifices.

Though I could be completely wrong and have a very loose understanding of the situation.

[BeNe]

First at all - this is a very good situation and should have already in former times happened.
What i can see: We have to discuss more here !!! This is very important.

Let us made a plan what is really to be done for a RC3.
Raphael posted them - thanks!
Which Developers can help and want to help active ?
We have a Dev Corner in the Forum - time to use it.

A Testing-Team sound good.
But the testing must be very near to the development, so that the Dev´s can fix them ASAP.

Greez BeNe

[raphael]

Quote:I'm of the believe that when any project or business gains a certain size there needs to be clearly defined roles and procedures, otherwise people start to loose sight of the end goal.

RatS is supposed to be the project leader, but I haven't seen the leadership I would expect to see on a project; the first thing to be done at project management/leadership level is to set some goals, plan the release rules, the QA rules, the will break changes rules and some other stuff. There's not too much of this neither on the public forums nor the dev ones

[raphael]

Quote:But the testing must be very near to the development, so that the Dev´s can fix them ASAP.

your demo has been very useful during my development , since I don't have enough time to setup everything; and since the official test server doesn't let me connect to it via SSH

[rsmithgs]

Would I be correct in assuming everyone thinks there should be some communication?

If so surely this needs to be sorted first, communication method and procedure?