Thread

Like to chime in here with a few thoughts. Considering that EE add-on development has exploded over the last year or so and that it probably will increase once 2.0 is out, I find the official repository quite necessary in the future. I must admit that I seldom use it, but if I do expect the add-on to work.

And I also expect, and hope, the repository won’t include all those add-ons that aren’t really necessary. One of the most difficult things about add-ons is to sift out not only those that aren’t robust enough but also those that do not add any functionality to EE. This goes especially for the commercial ones we see more and more of.

Devot:ee is great but no matter how many favs, rankings etc an add-on has, there is no knowing if it will work in the specific set-up I’m using or in the specific environment I’m developing in. Don’t know how many times I’ve installed add-ons, fought with them, only to discard them, and many of them are highly ranked in the community or at devot:ee. Rumor, enthusiasm and good will has a tendency to spiral…

Still I check out Devot:ee, I usually know which devs to trust, I keep a blind eye to the rankings, but how will it be in years’ time? How many add-ons will there be by then? The official repository can help me to narrow down my choices to vetted add-ons. But will EllisLab have time for vetting and quality checking as development speeds up? I do hope we won’t end up in that wild WordPress jungle of plug-ins that may be maintained or not, that may be compatible or not, that may, may, may…

I’m curious, does every new version of an add-on need to be re-checked before addition to the official EE add-on library? For example, the Valid URL plugin (v1.0) is in there now, but the 1.1 version is 2.0 ready. In order for Valid URL 1.1 to show in the official library, it would have to be re-submitted and reviewed again, like the iTunes App Store?

I know the 1.x add-ons will initially be separate from the 2.0 add-ons in the official library (as stated in this thread somewhere) but looking down the line at smaller incremental updates, if Valid URL 1.1 was updated to 1.1.2, for example, it would need to be checked again?

Its a great question. The short answer is no. We only reviewed major changes, but that didn’t always map to point versions. Sometimes there would be a security fix or substantial change to the code without a change in feature set.

One of the challenges will be creating a review process that is more focused on the developer submitting then the individual code pieces . That way we have a set of trusted developers and much less of a bottleneck. It also provides a difference between what is in the add-on library versus certification, which would be much, much more focused on code review.

In terms of outcome we do not want a vetting process like Apple. We’re looking for a sweet spot between no review and certification.

I should also clarify that this is a direction and not a concrete set in stone decision. Our initial goal is launching 2.0 and then focusing our attention on related items like the add-on library, the Showcase, testimonials, white papers, the Pro Net, and other website updates and changes.

We will provide an updated list of 2.0 ready add-ons as submitted by Team Beta (and we’ll link to your list as I saw you’ve got one going as well) at launch though.

I think there’s one thing missing in the interesting discussion, here. When I started developing add-ons, back in the day, I submitted them to the EE add-on library, mainly because I thought that was the way it worked. It required me to review my scripts thoroughly and actively send them to the EllisLab tech folks, who would add the add-on to the library. From that moment on, the add-on is practically out of my sight and reach.

Nowadays, I do not submit my add-ons to the library (even if submissions were open), because I want to be able to track how many people are downloading and using it, and I want to be able to quickly update the add-on if necessary (without bothering the EL dev team)—things I cannot do if the add-on is part of the official EE add-on library.

So, could it be the EE add-on library is as (seemingly) limited in numbers as it is, because some developers out there prefer to keep their add-ons on their own site? Personally, I’d start adding my add-ons again if, instead of linking to a hosted zip file, the add-on page at EE.com (like my Find and Replace plugin) would only link to the add-on page on my own site. (which would probably open another can of worms…)

I know this is said before, but I just wanted to add to the discussion the point of view of a more Expression Engine newbie like myself. I have used both the EE add-on library and Devot-ee, and both has its shares of ups and downs, but that doesn’t matter really because when you are new to ExpressionEngine you pretty much have no clue what-so-ever on what is out there, and the EE add-on library is easily the first thing you get to know about.

So of course the EE add-on library will be used a lot, especially if you are new to the whole thing.
The second thing is the upside of the whole library: It integrates with EE, so when you are inside the control panel, you can quickly find plugins or extensions or other add-ons.

Once you get to know ExpressionEngine well enough that you can go a little beyond complete newbie and over to actually making your site, you will quickly find the EE add-on library lacking, which is the current downside of it. Luckily, you pick up sites like Devot-ee or some of the private sites of add-on developers, and the forum here is also a good way of finding new add-ons.

If anything, I’d love to see either the EE add-on library have perhaps a non-certified listing of plugins (sorta like use at your own risk) while you folks at EE go through your review process, or have some API that developers could use to get their add-ons to show up in the control panel. Because right now the EE add-on library is a good place to start, but lacking a lot once you get into it, and finding updated/new add-ons can be a little hard if sites like Devot-ee didn’t exist or you didn’t know about them.

A friendly link to Devot-ee from the EE add-on library would be nice.

How about bringing in Devot-ee officially and allowing add-on members to run reporting on the number of downloads etc? It would also allow EllisLab to jump in and test the code and give it a gold star (or silver/bronze if you wanted to go down that route).

Devot-ee has a great framework and it could be built on by the team. For example some far out features could include -

- Central repository for add-ons
- Installation of add-ons from within EE 2.0 (SFTP Details required, perhaps as an FTP installer module?)
- Payment can be made based on stored card details held on secure EllisLab server or elsewhere?
- Percentage of each sale goes to EllisLab for the maintenance of the server. It could even go towards a dedicated add-on tester.

This sounds very iTunes like but I could see a greater uptake of apps if it were seen as a less disjointed effort.

*I apologise if this is all Star Trek like technology for now. It’s easier to write specs than code them!*

Obviously this would relinquish some degree of control from Ryan but let’s just assume he goes with it for now (for the sake of argument!)

How about bringing in Devot-ee officially and allowing add-on members to run reporting on the number of downloads etc?

What for?

What I would do is simply re-organize EE’s official add-on page and present it as a showcase of high quality add-ons, rather than an all-inclusive add-on library, as already suggested in this thread. Some links to external EE add-on websites could be added.

Devotee should remain Devotee.

As a long-time CI developer myself, I’m looking forward to developing some CI-powered add-ons for EE. And I’m so not being arrogant or sarcastic here, but I might even create a tutorial or two to show you EE third party developers how it’s supposed to be done.

CI >= EE

As a long-time CI developer myself, I’m looking forward to developing some CI-powered add-ons for EE. And I’m so not being arrogant or sarcastic here, but I might even create a tutorial or two to show you EE third party developers how it’s supposed to be done.

And we’ll be there to add whatever you develop to the devot:ee library. In fact, you’re invited to submit those articles to devot:ee - they’re exactly the kind of content I want on the site outside of add-on information.