Why does my customization (mod/theme) take so long to be reviewed?

Started by Joshua Dickerson, March 02, 2007, 02:16:59 AM

Previous topic - Next topic

Hambil

Quote from: Rudolf on March 27, 2007, 06:48:22 AM
My point is that the selection for mod "reviewers"  and/or "approvers" has to be pretty much like the selection of team members - man-handled. There's no automatic way to decide. 
I disagree - see my above post.

www.catnine.net/smf

Rudolf

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has opted-in via a checkbox in their profile
Good so you already have the approval of the people, you don't need to ask them.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of approved mods (3 sounds like a good talking point)
BAD. See my previous post.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of mod downloads (1000 for a talking point)
BAD.
You are implying that the success of a mod depends on how good a person is at coding and not about the modification it offers.
Take Chris here on this thread. He's working on a Calendar mod, right? He can revamp completely the calendar and write something just perfectly gorgeous from the coding point of view, with no bugs whatsoever, and perfectly secure. But I will never download it because it offers a feature I don't need and want.
So it means that the system will penalize the mod authors based on the usefullness of their mod.
Addedum: There could be a person who releases 3 mods that are downloaded 30000 times, but all three are full of bugs. Would that person be suitable to review other mods?
I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Rudolf

Quote from: Hambil on March 27, 2007, 06:50:41 AM
Quote from: Rudolf on March 27, 2007, 06:48:22 AM
My point is that the selection for mod "reviewers"  and/or "approvers" has to be pretty much like the selection of team members - man-handled. There's no automatic way to decide. 
I disagree - see my above post.
I disagree - see my above post.
:D :D
I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Hambil

Quote from: Rudolf on March 27, 2007, 06:54:36 AM
Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has opted-in via a checkbox in their profile
Good so you already have the approval of the people, you don't need to ask them.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of approved mods (3 sounds like a good talking point)
BAD. See my previous post.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of mod downloads (1000 for a talking point)
BAD.
This implies that the success of a mod depends on how good a person is at coding and not about the modification it offers.
Take Chris here on this thread. He's working on a Calendar mod, right? He can revamp completely the calendar and write something just perfectly gorgeous from the coding point of view, with no bugs whatsoever, and perfectly secure. But I will never download it because it offers a feature I don't need and want.
So it means that the system will penalize the mod authors based on the usefullness of their mod.
If you are determine to see it that way, then you would be correct. However, I feel your logic is flawed.

First, all mods should be considered 'well coded' by the very nature of the system you have created. If a lot of 'not well coded' mods are being approved then all that does is show even more cracks in your current system. If almost exclusively 'well coded' mods are being approved, then your system works (which you keep claiming it does), and 3 approved mods should be a strong indicator that a member knows the guidelines.

Second, can someone create three small mods that get 1000 downloads and get into the beta-group faster or easier than another person who creates 1 large mod that doesn't get a lot of installs? Sure, but I think looking at the actual stats for mod authors will show this to be a rather rare thing. And, if it does happen, worse case scenario is some mod authors get into the beta-group faster or easier than others, so what? It doesn't invalidate the proposed system in anyway.  

www.catnine.net/smf

Sarke

All they need is to understand the situation, their ability to code is not the biggest concern.  Even an average joe can install a mod and see if it gives errors or not, and as long as they understand the risks that is ok. 

I don't think these beta testers should have any power though, all they should have is access to a sub-board where they can download and give feeback to other mod authors.  That's all that is needed, a sub-board with one topic per unapproved mod.  People in this group can download the unapproved mods, test them and give feedback if they wish.  There is no need to support these mods because they're not supposed to be used in a live enviroment, and those who do use it like that do so at their own risk.

I would say as little as one approved mod is needed, and if that is a one-liner then so be it.  They've set themselves apart from the general community by being able to put some code together and that should weed out most of the people who don't understands the risk of a beta.  A big red sign should take care of the rest.


To put this into perspective.  By paying MONEY people are given access to betas of SMF, is that not correct?  How does money qualify those people?  Are they somehow smarter?

My MODs          Please don't PM me for support, post in the appropriate topic.

Hambil

Quote
To put this into perspective.  By paying MONEY people are given access to betas of SMF, is that not correct?  How does money qualify those people?  Are they somehow smarter?
Really? I did not know this.  :o

Quote
With your Charter Membership you are given access to a closed section of our Community forums, as well as a spiffy badge under your name showing that you're a Charter Member. Here you're able to discuss the latest betas, mods which haven't been publicly released yet, as well as provide feedback about SMF.

So, now some things are starting to click into place. We're asking them to give away what they are already charging for. No wonder they don't want to. This changes some things IMHO. I thought there was NO avenue for getting a look at unapproved mods. If all I have to do is pay $50 a year, well, maybe it's not so much an issue for me anymore.

The question is, how can they make use of the paid forums to improve the speed at which mods are approved. I mean, all the protests about skill levels have already just gone right out the window because a total noob can get access to unreleased mods and provide feedback, for $50 a year.  

www.catnine.net/smf

Dannii

AFAIK Charter Members don't have access to unreleased mods. Maybe they should. If they all started beta testing, would that be a big enough group to speed up the process?
"Never imagine yourself not to be otherwise than what it might appear to others that what you were or might have been was not otherwise than what you had been would have appeared to them to be otherwise."

Rudolf

Quote from: Sarke on March 27, 2007, 07:19:11 AM
To put this into perspective.  By paying MONEY people are given access to betas of SMF, is that not correct?  How does money qualify those people?  Are they somehow smarter?
You are wrong. Look I am a charter member. We have access to betas before you. But the difference is minimal. Most of the time what you can access and what I can acces is the same. Access to betas a couple of weeks before releasing to public is just a small plus to what the Charter members get.
I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Rudolf

Quote from: Hambil
Quote from: Rudolf on March 27, 2007, 06:54:36 AM
Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has opted-in via a checkbox in their profile
Good so you already have the approval of the people, you don't need to ask them.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of approved mods (3 sounds like a good talking point)
BAD. See my previous post.

Quote from: Hambil on March 27, 2007, 06:44:04 AM
Member has x number of mod downloads (1000 for a talking point)
BAD.
This implies that the success of a mod depends on how good a person is at coding and not about the modification it offers.
Take Chris here on this thread. He's working on a Calendar mod, right? He can revamp completely the calendar and write something just perfectly gorgeous from the coding point of view, with no bugs whatsoever, and perfectly secure. But I will never download it because it offers a feature I don't need and want.
So it means that the system will penalize the mod authors based on the usefullness of their mod.
If you are determine to see it that way, then you would be correct. However, I feel your logic is flawed.

First, all mods should be considered 'well coded' by the very nature of the system you have created. If a lot of 'not well coded' mods are being approved then all that does is show even more cracks in your current system. If almost exclusively 'well coded' mods are being approved, then your system works (which you keep claiming it does), and 3 approved mods should be a strong indicator that a member knows the guidelines.

Second, can someone create three small mods that get 1000 downloads and get into the beta-group faster or easier than another person who creates 1 large mod that doesn't get a lot of installs? Sure, but I think looking at the actual stats for mod authors will show this to be a rather rare thing. And, if it does happen, worse case scenario is some mod authors get into the beta-group faster or easier than others, so what? It doesn't invalidate the proposed system in anyway. 
Look, you are right about the number of mods approved. They do show in some way how good a person is, or how active is.
But in reality it doesn't works like that, and it's not only rare thing. I can write three mods that change one line each in the code - change not add- and just wait until 1000 people download it, and I will get the right to approve other mods. This is very close to reality, and you couldn't even blaim the cracks in the system.
I can write 3 mods and just don't come back anymore for a year. In the meantime people kept downloading my mods, and I will reach the limit to enter the "mod approval team". Is that right towards a person who is actively and supporting developing mods over the year, but can't get the number of downloads because maybe he implemented a thing that not many need? Maybe only 700 people need it, so he still wouldn't have any chance to be considered even after years and years of contribution. See the fundamental flaw in your system? This is reality.
One other thing you should consider is that the code base of the software is changing, and you can bet that the next big release of SMF (not the small upgrades to the 1.1 line) will have their code and logic changed big times. There are mod creators who don't update anymore their mods, they lost interest or they don't have the time to do so... yet they would be eligible and part of the "team approval team". Is this right towards a person who actively follows the development of the software, keeps himself up to date with the news, but can't achieve a certain number of downloads or a certain number of mods released?

You just can't automatize the process, there must be some human evaluation... because whatever criterias you put in the program every one of them will have a major flaw. There will be cases when the system wouldn't take the "right" decision. That's why you have to use people in this process. At least you have who to blaim if things go wrong. If a software fails to chose the right people, then you'll blame the developers, if the developers make decisions based on their "gut feeling" you will blame the developers. In any case they will get the blame, so why would they bother creating a system that doesn't solves their problem?


First of all, lets make one thing clear. Maybe I explained myself clear.
My main issue is with the number of downloads, but you can take any fixed criteria and apply it.
The criteria of number of downloads is just bad, because it doesn't measures in any way how well a person knows to code and how well he/she knows the guidelines. It just measures how many people are using his/her mod. In this case you can have two members that have similar level of knowledge, but one can have approve/review mods and the other not. This is simple discrimination based on how good = useful their mods are. Which has nothing to do with how good they are at coding. It's like saying: you are white so you can approve mods or you are over 18 so you can approve mods, you are female so you can't approve mods.

The entrance of any group whatsoever should be based on merit, in this case how well they know to code, to understand the software and to manipulate, and not on any other criteria (sex, color, religion, age, total number of downloads <-- I repeat: these have nothing to do with the professionality of a coder)

As a last note to you Hambil: there were times when I saw mods written (and approved) by customization-team members that were -IMO- kind of a disaster. They worked, but the way they did their thing was just awful. OK? It is not a very rare case. No one is on the same level, and in an open source project there are a lot of noobs hanging around learning. (I was one once... and 80% of the times I still feel a noob)
So what you suggest would lead soon to disaster.
I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Hambil

It can't lead to disaster if it's a moderated process, which is one of the things I proposed.

Look, any system like this is about checks and balances. You can take any one criteria and say "that can be abused", but the idea is that the sum is more than the parts. By combining number of mods, number of downloads, and hell add in a few other criteria if needed like "member is active in the last month" and members hacks are xKB in size when totaled. You can come up with all kinds of criteria. Each added criteria works together with the others to balance the system -  no single point of failure. You could even allow a mod author to qualify by reaching 3 of 4 criteria. You can configure this however you desire, and it's really not hard to do.

And, I reiterate, because the mod must be approved by a certified SMF team member before it hits the public, very little harm can be done.

Honestly, the resistance to ideas like this is just mind-boggling. It has elitism written all over it. 

www.catnine.net/smf

Chris Curran

Quote from: groundup on March 26, 2007, 09:59:22 PM
Quote from: Chris Curran on March 26, 2007, 08:17:18 AM
It's very frustrating to have to wade through *snip* the poorly documented coding guidelines
SMF Coding Guidelines - http://www.simplemachines.org/community/index.php?topic=159824

groundup,

Yes, you have that link NOW, but it wasn't always there with that content and the info in it keeps changing. Didn't you guys have a meeting at the front of this project and nail down your guidelines then? Lets take a current example. My mod uses GPL style function parameter declarations. This is the style that's used in apache, gnome and many, many other projects. It's a very old and well established coding style. An example:

functionName($var1,
                      $var2,
                      $var3)
{

... code ....
}

There are many benefits to this style, but that conversation is beyond the scope of this post. As you know, the SMF coding style does not cover this topic. I get a message yesterday from an unnamed SMF member that I need to change that code because "it looks weird". WTF!?!?! You have GOT to be kidding me! The youth, lack of experience and lack of formal education in the field of computer science is starting to show at the SMF team level. Frankly, I'm not going to change the code. I don't much care if the mod is approved or not at this point. It's classic, valid code and I'm not jumping through yet another hoop because some kid doesn't know what he's looking at.

cheers,
Chris

Chris Curran

Quote from: Rudolf on March 26, 2007, 05:51:00 PM
Dudes,
I gave up a long time ago.
Whatever you do do it for the people who use SMF, not for the people who make it. They don't need your work, and you certainly don't need their help.
So what if it takes two months for a mod to be approved? Big deal. If you really want to share your mods find alternative ways. You are mod creators, I bet you can figure out something.

Well said Rudolf. You make a good point and thanks for pointing it out. Since I house my home PC's/servers in a 42u rack, I think I'll start my own mod server and not be bothered with the kids running this show.... Forgive me for not reading the rest of this thread - it's just not worth my time.....

cheers,
Chris

Rudolf

Quote from: Hambil on March 27, 2007, 09:22:01 AM
And, I reiterate, because the mod must be approved by a certified SMF team member before it hits the public, very little harm can be done.

In this case it would be enough a separate boards where people can kind of submit they beta mods to the whole community. Team members could follow the threads in it to see the status of the mod based on beta testers. Check out this, the mod author posted the mod in that board on my suggestion (I suppose) and he has an thread where peolpe can download his mod and test it. It is something I did with most of my mods.
So everyone can use it and give feedback. The customization team could check once in a while to see if there are obvious complaints about the mod and/or to interact with the mod author - and the best of all everyone could see that work is done and things are moving.

Is this a good enough solution for you?
I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Hambil

Quote from: Chris Curran on March 27, 2007, 09:40:03 AM
Quote from: Rudolf on March 26, 2007, 05:51:00 PM
Dudes,
I gave up a long time ago.
Whatever you do do it for the people who use SMF, not for the people who make it. They don't need your work, and you certainly don't need their help.
So what if it takes two months for a mod to be approved? Big deal. If you really want to share your mods find alternative ways. You are mod creators, I bet you can figure out something.

Well said Rudolf. You make a good point and thanks for pointing it out. Since I house my home PC's/servers in a 42u rack, I think I'll start my own mod server and not be bothered with the kids running this show.... Forgive me for not reading the rest of this thread - it's just not worth my time.....

cheers,
Chris

Chris,

There are a few realities to deal with here (for some of us at least). One of those realities is that this SMF site has a huge amount of traffic. If you have a desire to create professional mods at some point, you'll need to leverage this site just like you need to leverage vbulletin.org for vb hacks.

It can get frustrating, and I understand your desire to give up, and should you do so I wouldn't fault you. But, it would be sad to see experienced developers turn away. Yes, there are some "kids" who haven't been living in front of a computer since 1980 involved in running the show here - that's not uncommon with open source. But, they are valuable. They bring young and energy and a passion I couldn't reclaim even if I could still live on Mountain Dew and Hot Pockets.

You and I (and others I'm sure) have age and wisdom and experience to offer, and I hope they find ways to hear our voices. I'm going to keep talking one way or another :D

www.catnine.net/smf

Hambil

Quote from: Rudolf on March 27, 2007, 09:44:39 AM
Quote from: Hambil on March 27, 2007, 09:22:01 AM
And, I reiterate, because the mod must be approved by a certified SMF team member before it hits the public, very little harm can be done.

In this case it would be enough a separate boards where people can kind of submit they beta mods to the whole community. Team members could follow the threads in it to see the status of the mod based on beta testers. Check out this, the mod author posted the mod in that board on my suggestion (I suppose) and he has an thread where peolpe can download his mod and test it. It is something I did with most of my mods.
So everyone can use it and give feedback. The customization team could check once in a while to see if there are obvious complaints about the mod and/or to interact with the mod author - and the best of all everyone could see that work is done and things are moving.

Is this a good enough solution for you?
I am not able to check out that link. But, your idea certainly sounds like a move in the right direction :)

www.catnine.net/smf

Rudolf

I will update all my mods in the next few weeks. Thanks for your patience.

SVG-Collapse (you need an SVG compliant browser)

Dannii

I've had beta testing threads for both of my two major mods, I think its a great idea. Maybe it should be formalised, with a special board?
"Never imagine yourself not to be otherwise than what it might appear to others that what you were or might have been was not otherwise than what you had been would have appeared to them to be otherwise."

karlbenson

Quote from: eldʌkaː on March 27, 2007, 10:05:41 AM
I've had beta testing threads for both of my two major mods, I think its a great idea. Maybe it should be formalised, with a special board?
Seconded.

I've got a large mod in developed for ad revenue sharing. Lots of ideas, lots of bug fixes.
A special place for us to gather would help improve each others mods (and hopefully mean less to fix/do come moderating time)

Sarke

Quote from: eldʌkaː on March 27, 2007, 10:05:41 AM
I've had beta testing threads for both of my two major mods, I think its a great idea. Maybe it should be formalised, with a special board?

See that's all I've been asking about.  People can already post a new topic with a link to download of their unapproved mod, so why the big fuss about all the potential problems of making it a sub-board?

My MODs          Please don't PM me for support, post in the appropriate topic.

karlbenson

a child board would be better as the betas won't get lost along with 'how can I do this' queries and such in the coding section.

Just as you have people who tend to hang around the importers/converters section, you would probably have people that would do the same around a beta mod board.

Advertisement: