Should the Extra repo have better Moderation and Management

I am not sure of the current process of approving Extra and if any ever get denied when submitted. I am somewhat starting to feel the Android Market (Google Play) effect kicking in with the Extras on the MODX website. I find that some of the Extras allowed in, have no real value to them and are just making finding useful Extras a bit harder, expecially with the lack of details as to what some of these Extras do.

I do believe that some Extras are only submitted by the developers due to their not being a way to setup your own provider to pass these Extras through. I can see this happening when a developer want to be able to update multiple Client's website with the latest version of their extra without having to upload it via FTP to each client manually.

There are also Extras that are out of date and are no longer being maintained by the developer that should be removed from the Extras repo.

I won't be naming the Extras I find to be clutter for one reason or another. I would like to hear others view on what I have stated here and also a official view from the MODX team.

I agree that there are completely pointless Extras but shouldn't the relevancy be decided by the community? Personally I look mostly at download count and when it was last updated. I usually ignore the thumbs up/down ratings because hardly anyone ever uses them.
I wonder if allowing comments on addons would be beneficial.

I completely agree that the moderation could be better. Personally I find it kind of annoying to browse the extras only to find a bunch of them with obscure names and no description, leaving you wondering what the extra is even for, why anyone would want it, and even more so how do you even use them? Having to browse the source code only to find out what the extra does is not something that should be necessary.

The point of developers submitting extras for their own benefit doesn't sound far fetched when you realise how many extras lack proper description. Often times I've found extras using the latest entry, in what should be the change log, as the description. This isn't particularly helpful to 3rd parties.

Checking that the extra provides proper description and that the correct fields are used for their respective data shouldn't take long.

Can only agree, this has been one of my biggest fears that someday the Extras get WP effect and 95% of them are
#1 useless
#2 Do lot of things but not really well
#3 Do lot of things but most of the time only help 1 out of 100

Maybe this would be the opportunity to call out for community, there are lot of good extras that would need more fine tuning, new way of thinking the good idea and finalize them. Most of the extra developers have contributed the code as open source to GitHub so the sources are easily accessible.

So I propose new movement !!! how bout each of us who see a good extra and notice that it could do lots of other stuff actually contributed. Fork the reporsity, make your own branch by MODX guidelines http://rtfm.modx.com/display/community/Community+Contributor%27s+Guide. Submit a new pull request with clear description what it does etc... Though this might cause some more work for original developer, but most people do them for love of MODX and community. At this point it is obviously the main developers responsibility to discard or do what ever wants with the pull request. If 3 months for example has passed since the pull request, create own package and submit it to MODX extras with added note that the "original seems like it is not developed actively anymore and this new package comes with this and that etc... bug fixes or new functions"

Also original package creators could do bit and really pull out the reporsities or mark them with "Development discontinued" and give hint that people could actually fork and continue the good work.

Better documentation is very important. With any form of development, developers often do a terrible job describing what they created. It's too hard to write from a perspective of knowing nothing about it when you know everything about it. Perhaps we should also reach out and describe what our addon does to other developers and have them help us better describe them.

About a year ago there was a thread opened where the same question was asked, I recommended that everyone that submits packages should also add how to use it, and they should also be version controlled so that if a package was created for 2.1 you can't install it on 2.2 until it has been updated to say it's compatible, unfortunately I was shot down in flames and basically told that people would stop submitting them ( still something I don't believe ) but I decided to leave that thread to save upsetting people.

In short my views are that no package should be accepted without documentation and testing ( yes us the community would be happy to do that ), and every package should be version controlled.

Just my 2c and I probably wont look at this thread again as I'll probably get shot down again....

I can not say it better then Paulp:

In short my views are that no package should be accepted without documentation and testing ( yes us the community would be happy to do that ), and every package should be version controlled.

You know I actually agree with sufficient documentation being required, even though I probably even have a few addons that aren't up to snuff. Part of me wants to say 'beta' versions of addons should be added regardless...but I guess that's what GitHub is for. As MODX grows, we definitely don't want to find ourselves in a Wordpress plugin situation.
Maybe when MODX rejects submissions based on lack of documentation, the community should step in and help document the addon. We all want well documented, useful addons in the repo.

I think it would be great to have a 1-10 rating system, where users could give an overall rating and separate ratings for documentation, features, and ease of use.

I have mixed feeling about reviews. They can be very helpful, but if the first one or two reviewers lack competence at MODX and trash the extra, everyone thinks the package you spent a month developing is a piece of crap and others don't bother to check it out.

I do agree that projects that are not adequately described or documented should be rejected, with feedback to the developer and the suggestion to resubmit. The problem is, who will spend the time to evaluate them, and will doing this lower the number of good extras submitted.

BTW, I think the new version of MyComponent may help some, because it's going to create a readme.txt, file and empty tutorial and lexicon files as well as providing a place for a package description -- and it has a place to enter your GitHub username and the name of the GitHub repository for the extra (and will automatically put GitHub links in the readme.txt file). I'm hoping that will serve as a strong suggestion to people to document and internationalize their extras, and to put them on GitHub.

On updating and improving existing, substandard components -- I often have the urge to do this, but don't want to step on the toes of the original developer and usually don't have time anyway because I've got a backlog of 10-20 new extras of my own to work on.

It's also very difficult to weed out less useful extras from the repository, because an extra that I consider useless might be just what someone else is looking for.


------------------------------------------------------------------------------------------
PLEASE, PLEASE specify the version of MODX you are using.
MODX info for everyone: http://bobsguides.com/modx.html

Extras are automatically given a forum topic in the Addons -> General Repository Items Support. There used to be a link to this automatically added to the extra's page. This might be a good place for those who are so inclined to add documentation, propose fixes and otherwise add their own contributions to the support of the extra.

I find, though, that many of the links that are available on the extra's page are broken, either 404 or pointing to totally irrelevant posts.