Hi Al, I'll take a stab...
Al Rogers wrote:
While I like what I've experienced of Plum, I'm a bit concerned about it's "proprietary" nature and what appears to be the lack of a robust user community (except for this somewhat invisible list).
As far as proprietary is concerned, you are not allowed to removed the copyright notices from the custom tags and components that are included as part as the framework. I don't really see a problem with this, because OSS licenses usually require the same. I have heavily modified some of both the components and tags, and I just added a "portions modified by Jeff Fleitz" comment to the bottom of their header, and annotate my changes so that they are easier to see when I am trying to merge changes from updates.
As far as the robust user community, you have a point, although I am unsure of the cause. Traffic been light lately, and I like to think that maybe it is due to the fact that once you get a project under your belt and get under the hood a little bit, you really don't need that much help until you delve into something really complex, where you are going to have to make modifications to code. The other issue is the one of Windows as a dev platform. Plum uses a .NET-based code generator, so all those folks (and they are many) who use Macs to develop CF applications don't really give it a second thought. You don't really need the Plum 'IDE' to develop Plum apps; I only use the IDE marginally anymore, as I have setup form templates and setup code hinting/completion in my editors so that I can pretty much build stuff with an editor manually now. But it hasn't been packaged that way, so the community is smaller.
I'm also concerned about using Plum on the job I'm attempting to do. I'd appreciate any advice anyone can offer about the wisdom of forging ahead with Plum.
...
I'm beginning I think that I could develop (or at least prototype) my yet-another-new application in Plum, and then go back and retrofit the existing applications into the Plum method. However we already have close to 100K users in my existing user table, with several related tables that are used by the various sub-applications. For instance, we use a STATE and COUNTRY lookup table to translate CA into California and US into United States, for example.
There are a couple custom tags for this now, that populate pull downs. I have modified the User stuff to reflect this kind of enhancement as well. Not tough to do.
Also, some of our applications demand more contact information than others (we work with teachers and students). Teachers who want to participate in any of our projects must include both school and home contact information before they can register for the project (otherwise, we require minimal information: school name, state, country and zip). Hence, we have separate tables for school addresses and home addresses.
You can extend the default tables and components that handle them. I have done this almost every Plum project I have worked on.
Furthermore, the email address of EVERY registrant must be validated via email delivery and their reply, after which their record must THEN be reviewed and approved by one of our moderators before they become eligible to participate in one of our projects. Thus, at the minimum, I believe I will have to modify the login/NewUserForm.cfm in order to retrofit the authentication to my existing data tables. The first question is, does this sound feasible in Plum?
Probably, but I would encourage you to look seriously at Plum's security model and try to see if you can't bring your requirements into the Plum framework instead of the other way around. A lot of Plum's provided functionality is built around those Plum tables (PlumUser). Trying to decouple all that logic with those tables is not going to be fun. I heavily modify that table along with the security.cfc component to handle these login customizations.
--------------------------------------------------------' Question 2: While PLUM will be useful for my administrative management of the web site, much of our data requirements for our applications that teachers and students interact with simply don't lend themselves to simple form tables. For instance, in some instances we need to do live-interactive drill-down queries that present form choices based on the previous input so as not to overwhelm the client with too many choices presented all at once. Thus, I'll want to include a sophisticated form and display generator within the Plum application framework for several of our client applications. Will Plum let me do this?
Well, Plum has drill down/up functionality built into it's lists already. You can essentially build lists with built-in search forms that have drill down/up capabilties to navigate between Parent Child forms, pretty much unlimited levels. Plum also has master-detail forms that allow you to drill down/up from edit forms, view forms, etc. I have some of these built that go down 6 levels. You can use custom queries/views if you need more than a single table form to work with. Several of the tags that deal with lookups allow you to specify additional query filtering via a whereClauseExtension attribute. They are very functional
--------------------------------------------------------- Question 3: We also want to include frame generators, DHTML dropdown navigation menus, and eventually integrate open source applications such as forums, blogs, etc. Will permit and/or support these kinds of additions?
Don't see why not. DHTML menus, etc are easy. Using frames or iframes is easy. It's just code so you should be able to integrate anything. Making it all appear cohesive may take some work. Adam and David have a forums package in beta that I have played with. Most of that kind of stuff Plum would be great at building anyway. The one issue we have is with wysiwyg editors for the CMS. Right now you have the choice of htmlArea, which is obsolete, or ActivEdit, which costs money. I have downloaded copies of FCKEditor and TinyMCE, but have not had the time to integrate them (custom tag) yet. We also need to integrate with OSS databases like MySQL and PostG
--------------------------------------------------------- Question 4 (final question): One of the things that lends credence to Fusebox is the vigorous user support that exists in an open source framework. Plum seems to have a fairly limited following at this time, and it's not clear to me at this point how much commitment the Churvis brothers have to the ongoing support and development. I know it's very reasonable, even to purchase for all that it does, but I'm not clear about ongoing support and development.
That's fair. I don't know either. They are working on a huge project (using Plum) and have been incommunicado for the past couple months, and will probably not participate until towards the end of this year. It would definately help if they showed up more. I have been involved with them for several years now, and you get those periods of inactivity, due to the way they have chosen to market the product. Their training courses and development projects are their bread and butter and I dont' see that changing. Plum is a 'gift to the community', and they don't want to be a ISV selling a product that will lock out a lot of users due to pricing. It's their call.
It is that way with pretty much all OSS anyway. I have been lurking the Fusebox forums for several years, and am playing with CFEclipse, Ruby on Rails, etc. All of the folk that contribute to these have a job to do and the other stuff gets done in their spare time. Most of the time, it's not quick enough for those not in the development community. You do have more community participation in those environments, though. Adam and David have encouraged participation, and we definately need to grow this area. I am in the process of getting into the 'blogsphere' in the near future so that I can share some of my thoughts/findings on Plum and provide a better public record of what is going on.
The question is: What am I missing?
You should really build something with it to give it a fair shake, even if it is just a simple app that you can use. It is very powerful, and builds apps quickly. Plum is really an agile technology, which is why I like it. It is great for iterative development. But I am an independent and don't work in a corporate environment. Most of those folk are using some form of MVC, because of all the benefits that model provides to teams.
I am on my 12th app with Plum, and it fits like a glove for me. I still play around with other stuff (I have several Fusebox books on my shelf) but my job is to get robust apps out the door as fast as I can, and Plum does a great job of that for me.
-------------------------------- I'll appreciate any insights you can provide, and any additional third-party evaluations and reviews you can steer me to (As I said, I've already read everything I can reasonably google on this subject).
The only eval I know of is the one I did for CFDJ: http://coldfusion.sys-con.com/read/48233.htm HTH, Jeff ********************************************************************** You can subscribe to and unsubscribe from lists, and you can change your subscriptions between normal and digest modes here: http://www.productivityenhancement.com/support/DiscussionListsForm.cfm **********************************************************************
