Topic: Why I am not involved in the improvements of this software

[beast]

This text was written for another thread but I have created a new thread for this post because I think it need more attention

In the past I really testet the beta releases etc. Was a bit involved!

The problem come with the bug tracker system that every developer seam to point at (and I understand why from a developer stand point). Every time somebody ask a question or have a problem they are directed to the bug system. The problem is that, in my view, the system is made to be used by the developers and most "ordinary" people is repelled by this system. It does not look nice to use or understand in any way. There is also a problem with the split between the forum that everybody understand and the bug tracker. What is the difference, when shall we use the forum and when shall we use the bug tracker etc. You also need a specific access name and password for each system.

The end result is that the developers do not get so many errors to solve, but the error are more documented and easy to list - but does it make a better system in the end? I do not think so! It is also the reason why people ask for the status of version 8 - nothing about it anywhere else but in the bug tracker really (everything else is outdated). May also be the reason why we still wait for version 8. Nobody like to get involved with the current system.

The solution is to make a better interface for a bug system and hopefully integrate it with a simple forum in some way. I know this is quite a project and out of scope for the SME-server people, but there may be a better system out there somewhere.

[Mat78]

I think that's true.

The forums is easy to be used and everybody use it.

The actual bug tracker is really hard to be used. An easier way of use and a good gui can help people reporting bugs and potential bugs so that developers can solve it.

[Mjohnson]

I hate these kinds of posts.  This issue has been rehashed so many times. 

First, I have used SME for several years.  I am not a developer and have little talent necessary to advance the project along.  Tried to get more involved a couple of years ago, but frankly it is a tight group without a lot of horizontal thinking and my outreach didn't amount to much.  All good people, just not a good fit at the time.

More to the point, the bug tracker is the way this distro moves ahead.  You can rant and rave, but that is the way it is.  This is a developer driven project and any attempt to introduce an executive or marketing function has been unsuccessful.  As unfriendly as the Bug Tracker appears at first, it is a very effective problem solving technique.  I have never been left hanging when I have used it.  What's more, while I have never made any tight personal friends out of the development team, they have also never responded in a belittling or arrogant fashion. 

Prior to the release of the SME 7 version, a lot of the problems were being dealt with via the forums.  The last days of the SME 6 version were really messy.  All sorts of contribs, how-to's and hacks were being disseminated via the forum which resulted in an unworkable system to help those in trouble and particularly to the real coders.  Personally I had modded my SME installation to the point where an upgrade to Version 7 had to be done via a clean install.  It was ugly and on a production machine, just plain stupid.

I salute the current system in that it has maintained a large degree of control and accountability for the users and those regulars who patrol the forums dispensing advice and assistance.  I realize the mantra of post it in the bugtracker gets old after a while, but it really is meant as the first step in a solution.  By the way, when you post your problem in the bug tracker, you get email responses to your post.  It is outstanding.


MJ

[beast]

I hate these kinds of posts.  This issue has been rehashed so many times. 

Maybe something is wrong then 

[steever]

No, I think the bugtracker system is needed and it is obviously working well for the devs!  What I'd like to see is a little more communication from the devs when things are solved/changed. 

For example, the wiki lists ldap authentication as one of its top 5 projects http://wiki.contribs.org/Top_5, but there's no information anywhere on the wiki about how or if it works.  Just a pointer to the bugtracker which presents the user with a lot of scored out lines.  Somewhere in the middle of it there's some stuff about getting ldap auth to work but does it?  And how?  I'd like to see a little documentation come out of the devs into the wiki which is the easiest to understand interface to what's possible with the SME Server.

Likewise, try to get a handle on what's actually in SME 8 results in the same thing - directions to look at the bugtracker.  Why not a couple of lines in the wiki describing the improvements in plain language? 

Before I get flamed - which is becoming increasingly common on contribs whenever someone offers some feedback or advice- I'd like to restate how much in awe I am of the development team.  What a great distro.

Steve

[pfloor]

Steever,

No flaming, just some facts for you to chew on.

As of today there are exactly 3 people with the distinguished "Developer" label next to their names and one of those is absent due to other commitments.  That's right THREE (really only 2).  There is are 3 additional that belong to "Quality and Testing" and a (very) small handfull of others that help develop the software on a constant basis.

There are 122 members on the "Documentation" team.

We know what the 2 developers and their helpers are doing, they are doing their jobs and writing code.

Where are all the documentation people?  They need to scour the forums and bugs and update the wiki.

There are almost 3200 members here and a tiny group that actually do all the work.  The developers are busy and don't have time to update the documentation so folks start complaining about the wiki.  When those working on the code stop and take time to update the wiki and then other folks start complaining about why SME8 is taking so long, etc.

It's a vicious cycle and there is nothing anyone can do about it except get busy on the wiki or start threads like this one, it's everyones individual choice but past experience tells me that this thread in itself won't accomplish anything.

Beast,

If you don't like the system here, please build a new one.  The developers like it the way it is and are not going to change it.  Many have asked but no one else has put in the effort and the developers just don't have the time.  This project is open and you are welcome to offer what you can come up with.  If the developers like it, they will use it, if they don't then they will not.  It's really as simple as that.  You can't ask them to do everything, it just isn't possible or practical.

By the way, it is totally impossible to use a forum as a bug tracker, a forum just isn't designed for that purpose and it won't work.  You have to use some sort of bug tracking system and I have tried most of them and they are all as unfriendly as bugzilla.  They are complex, developer oriented programs designed to help developers develop and track software.  Bug trackers are not meant to be easy or pretty.  If they were, they would be nothing more than another forum.

Bugzilla is the way it is because it has to be to accomplish what it does in an orderly fashion, nothing more, nothing less.

Also, until SME7, the bug tracker for SME (formally Mitel) was not an open bug tracker and that's why you never saw "report a bug to the bug tracker" before 7.x, you couldn't as it was only accessable (AFAIK) to Mitel employees.  When Mitel had the OS, they would scour the forums and look for bugs and report them to the internal bug tracker themeselves but Mitel had many more developers and paid them to do it.  That doesn't exist anymore and the developers now work for free and can't do it all.  They ask for the communities help.

The developers depend on the community to report bugs just as the community depend on the developers to keep their servers secure and running...Another vicious cycle that will most likely never end

Lots of folks have great ideas but that's all they are, ideas.  Until someone has time to apply the ideas, they are still just ideas and will always remain ideas.

The fine people here toiling over this great piece of work (I consider it art) just don't have the time to do everything.  The software is the most important part to them and they do their part for absolutely free.  The rest (the wiki, forums, etc.) is second in line to them and they don't have any time to work on anything else so someone else is going to have to step up here.  Instead of "Why I am not involved in the improvements of this software" topics, I would like to see this for once...

"I have developed a completely new look and feel for contribs.org.  It involves A, B, and C and uses X, Y and Z.  Can the developer in charge please contact me so we can convert the existing website to my new and improved version."

Now that would definitely get their attention.

[Paul Howard]

The forums is easy to be used and everybody use it.

The actual bug tracker is really hard to be used.

Forums are good for users seeking help from other users. The bugtracker is good for developers in tracking and solving issues. There is always going to be divide as the requirements are different and there has to be a middle ground. I have to say, it has to come from users in reporting bugs since we are the ones who are going to find them one way or another.

Initially the Bug Tracker looks like the mother in law on a bad hair day without her HRT but it is a matter of use rather than apathy. I have grasped the nettle and posted a bug report, yes I did feel like I was stumbling around the first time but I'm sure will be easier next time or at least with a little practice since like most software it has to be learned.

It is the job of the developers to investigate, confirm or dismiss bugs rather than users, something which we easily forget but we do need to feed them. It is a matter (for me) in changing my psychology for the usage of the forum and bugtracker and using them together rather than seeing them as two separate entities.

From my own perspective, I would post in both about potential problems (after searching) ensuring I referenced the bug report in the forum post to get the best of both worlds ie any help the community as well as ensuring any potential bug comes to the attention of the developers in the bugtracker.



SME server is a good distro and a personal thanks to all those who are involved with the project and contribute to it. Without SME server I really would be stuffed as there is nothing out there for free for a non familiar linux admin in a small business environment, which does everything that its needs to.

[beast]

I feel that I have to react to some of the comments made in this thread. First of all I am educated as a developer myself - but not really in use any more. I have tried to post a bug in the bugtracker so I more or less know how the system works. I like SME and use it on a regular basis.

My current complain right now is the way people react when you try to tell them about a problem. In my view people react hostile when somebody try to improve things by telling about something that does not work. The first thing that happen is that they tell you that the current system is the only way that the system will work. Next they tell me that I can just get involved if I like to change things (my main complain is that the current system stops me from getting involved).

I do not say that the bugtracker is not needed in any way and we can just rely on the forum, but the current bug tracker is made for the developers and of cause they are more happy about the situation now. I just try to step aside and take a look from the outside. And with this view on things I see a problem. The current way we handle software development stops a lot of people from getting involved - this was the only issue that I tried to raise. I do not have a ready solution for this, but maybe by starting this thread I can start something?

The only other bugtracker software I know of is Mantis that I use internally in my company and for what my opinion is worth - I think it is better from an ease of use standpoint.

/Beast

NB: I give all the credit I can to the 2 or 3 developers that are active 

[janet]

beast & others

My comments are by no means a personal attack, so please do not take them that way.
I see that some people so easily place their own values, interpretations and wrongly conceived ideas on other peoples words, and immediately go on the attack, instead of just reading the text as unemotionally as possible, and answering in a similar fashion.

Are you (& others) really serious when you suggest/imply that bugzilla is hard to use, and that hardness therefore stops you (& others) from posting bugs or being involved ????

If you (or anyone else who uses sme & the forums) can install sme server & use the forum software, then I think you are all capable of using bugzilla.

How hard is it ?

Register
Create new post
select category
Say what you have to say
Commit the entry

The steps are not really that hard or daunting.
I think that any issues that users may have are really just psychological
eg ooh I never used that before and don't know what to do, it looks too complicated !

The only answer I can give to that sort of attitude, is to give it a go and try using it. After your first or second post to bugzilla, I'm sure you will no longer consider it hard to use, it's quite easy really.

The only thing stopping people from getting involved, is their willingness to get involved.
Blaming the bugzilla as difficult to use, or saying that some people display an unpleasant attitude towards newcomers or those asking for change, is also just an excuse.

The only way that something will happen is to get in and "do it".
I don't mean to just blaze away at your own version of things, but to interact with the existing developer community, find out what needs doing, see if people will support those changes, and then start developing code or provide financial or other support. In this case the code has been "sanctioned" to suggest a phrase, by developers.

There is no real use creating something that developers are not interested in implementing.
The current developers are experts on sme, and know it "inside out".
There are often many reasons that a newcomer or even an experienced person who is unfamiliar with sme, will not realise about the pros and cons of introducing some new wiz bang feature. Perhaps more importantly would be a limited understanding of the work involved to bring a seemingly simple suggestions to fruition (codewise), ie the task is often more complicated than it would initially seem, as is usually the case with much software development.

To quote what I believe are two examples (and I am not being critical or judgemental about either of them, just obvservational), one from a poster arne, wanting to implement firewall code in a non sme standard way. He initially seemed adamant that he could do what he wanted to do and why shouldn't his work be included in sme base code or otherwise be a good contrib. He did not get support from the developers generally, and one main reason was that he did not give them well developed code, that was written in the standard way that sme does things. It seems from reading posts that arne initially had a poor understanding of how sme server worked internally, and he had difficulty seeing why his ideas were not valid (in the sme sense).
I'm pleased to say that arne now seems to understand sme much better and has even praised the way it works (feature & codewise).

Another feature request that has had a significant amount of code already developed, is LDAP authentication. The coder is asking for it to be included in sme8 release, but a key developer is asking for a new feature bug to be created to discuss the pros and cons of actually implementing that feature and how widspread the need for it is. Here we have a case of an external developer "ready to go" with some code, but the main key developers have not really been all that much involved in the work done to date.

Until the key developers support the new feature and are happy with the rationale for the introduction of that feature, and then get involved in reviewing the associated code, then the feature will never get into the base code (no matter how well advanced the code may already be). Another important element is the time effort required to fully and completely carry out the integration into the base code, and any future maintenance requirements for that code ie upgrade issues etc. There are many things to consider (beyond just the desirability) when adding a major new feature.

My comments are not made as a coder, as I am not one. I do consider myself an experienced user, and have used sme for over 8 years.

[arne]

I can see I am mentioned, or my firewall project was mentioned.

What I needed to do my project was one single line of technical information.

I got some pages of flaming but between all those negative feedbacks I found the single little line of information that I needed to do my firewall project.

After obtaining the sentence, I made a list of the firewalling requests I have seen in this forum, during the years, and actually solved all of them, I think, plus those I could imagine myself.

(2, 3, 4 networks adapters, dmz, wireless client, wireless access point etc, etc.)

One basic problem with the SME server design is that it was done some years ago and based on that generation of Linux kernels that existed at that time. It looks for me that some of this basic seup around this is still in use, also today.

Doing all kind of firewall variants is actually next to nothing, when you do it more on direct basis on the Linux technology that exist today.

If such a project should be done as if the technology were the old one, such a project, that normally should be “an easy one”, would be so difficult that it would be almost impossible to do.

When the requirement for how to make a project is “and the only allowed method is that one that can not work” then it is not a question off “the good will”, but that the requirements is set in such a way that the project will not be possible to do.

Because of this “SME server conservatism” I think you will not see, such general firewall solutions on the SME server like those on for instance Smoothwall for quite a long time.

With some new thinking, I think such a project would not be to difficult to do at all. (But not necessarly with all the functionality of Smoothwall.)

All big and long lasting projects like the SME server project will always need some new thinking and innovation from time to time.

For my own part, its not a problem at all. During all the hard discussions, I got my single line of information, so now I can do all the firewall design I want or need and there is no restrictions or problems at all, and everything works quite like it should, always.

Personally I think there could be some more room for general discussions related to security and other related stuff.

Nothing in this world is perfect. (But 8.0 is still on its way, and somebody is doing it.)

I think that the SME server is a very nice project with a lot of clever people.

As long as bad words like firewall, dmz, wireless or security is not mentioned it is also a very friendly forum. So why mention the bad words ?

Back to the header of this tread: “Why I am not involved in the improvements of this software.” – Well, I would like to, but the problem is that the only things I know anything about, is stuff that contains bad words.

I have got a lot of flames because of my firewall variants, but I did never answer it that way. Collecting information is some time quite like washing for gold. There is a lot of sand, but in between the sand there is the valuable small grain of gold that can do your project.

I do not think different about the SME server project than I did before, but I think that all the firewall problems I have had ever, is now solved, and that there is no reason to discuss this theme all over again. (I can do all the "on SME server variants" and the main method I actually use today is via Virtualization. This gives a multiple and more than 100 % solution for all problems.)

[cactus]

One basic problem with the SME server design is that it was done some years ago and based on that generation of Linux kernels that existed at that time. It looks for me that some of this basic seup around this is still in use, also today.

The analogy would not change for multiple cards that much AFAIK, however it was not in the scope and goal of the project as SME Server is geared to be an alternative for products like Microsoft Small Business Server.

Doing all kind of firewall variants is actually next to nothing, when you do it more on direct basis on the Linux technology that exist today.

And that was explicitly not in the scope of this project.

Because of this “SME server conservatism” I think you will not see, such general firewall solutions on the SME server like those on for instance Smoothwall for quite a long time.

It is not all conservatism, it is a design goal. To develop and keep a system stable you need to keep it simple and rule out some of the multitude of possibilities.

As long as bad words like firewall, dmz, wireless or security is not mentioned it is also a very friendly forum. So why mention the bad words ?

Those are not bad words, but the extensiveness you desire is outside the scope of SME Server.

To go short, you should stop comparing Smoothwall or any other extensive firewall with SME Server. SME Server has a firewall but it's not the main goal of SME Server, which it is for Smoothwall. It's like apples and bananas

[cactus]

The solution is to make a better interface for a bug system and hopefully integrate it with a simple forum in some way.

Or document it, like is done here: http://wiki.contribs.org/Bugzilla_Help. If you think things are missing, add them, if you still have questions ask them. Saying you can not help, without stating where you get stuck does not provide us clues to help you.
The latter is one of the points where both forums and bug tracker are stated to be not working for them. This mechanism is called externalization in psychology/philosophy, blaming your environment for your lack of knowledge or problems, where you could have been (part of) your own solution.

[electroman00]

Or document it, like is done here: http://wiki.contribs.org/Bugzilla_Help. If you think things are missing, add them, if you still have questions ask them. Saying you can not help, without stating where you get stuck does not provide us clues to help you.
The latter is one of the points where both forums and bug tracker are stated to be not working for them. This mechanism is called externalization in psychology/philosophy, blaming your environment for your lack of knowledge or problems, where you could have been (part of) your own solution.

Cactus my friend, go back and re-read the original post, then post what you perceive as the main primary point in as few words as possible, let's say 5 words or less.

Keep in mind the topic title.

[e[nt]e]

Cactus my friend, go back and re-read the original post, then post what you perceive as the main primary point in as few words as possible, let's say 5 words or less.

Keep in mind the topic title.

I actually don't wanted to post to this topic, although some things came to my mind when I first read the original post but that was posted by mary later on.
But know I don't think that you need to rebuke Cactus for simply posting his opinion on, mind you, the first post.
Now again my post could be seen as rude and offtopic but I ask you not to do so, at least not here in this thread.

[cactus]

Cactus my friend, go back and re-read the original post, then post what you perceive as the main primary point in as few words as possible, let's say 5 words or less.

At the risk of being flamed or considered harsh. My reaction was to the original post, so if the contents does not suit you, fine. If you can not get the message from my relatively short post, fine as well, but do not blame me or my writing!

It would have helped if you would have stated what part of this message needs more explaining or why and where it goes off-topic so we could have discussed that instead of lecturing me (only):

Or document it, like is done here: http://wiki.contribs.org/Bugzilla_Help. If you think things are missing, add them, if you still have questions ask them. Saying you can not help, without stating where you get stuck does not provide us clues to help you.
The latter is one of the points where both forums and bug tracker are stated to be not working for them. This mechanism is called externalization in psychology/philosophy, blaming your environment for your lack of knowledge or problems, where you could have been (part of) your own solution.

I was not willing to be offensive, and choose a more elaborate way of writing what I perceive, in order to not be battered as being cruel, sarcastic or anything like that.
But if you really want it in five words (for the first part): defining problems will solve them and five more (for the second part): fix yourself before blaming others.

Keep in mind the topic title.

Taking the risk of raising a religious discussion here: Do you mean I have no right in posting here as I do contribute to this distribution in many ways I can, like reorting issues, fixing issues, documenting stuff, providing help in the forums and writing contribs? (I dislike the subject from the start as it is not very constructive IMHO at all, but I have no power to change that).

One final word of advice (in five words again): Think in solutions not problems!