Home » Eclipse Projects » EGL Development Tools » What's next???(Can't find any future planning materals...)
|Re: What's next??? [message #1015208 is a reply to message #1014981]
||Wed, 27 February 2013 18:22
| Dan Darnell
Registered: November 2011
Thank you. I am pleased to see that that IBM has responded and acknowledged that there is in fact a re-assessment taking place where IBM's involvement is concerned.|
I'll note a few things in response:
1. Why wasn't this re-assessment made public until now? Reading between the lines, I think there is concern about the lack of involvement in the project outside of IBM but you can't lament the lack of involvement in the project outside of IBM if you always treat the project like it is an internal IBM project. Want partners in this effort? Act like an entity that is willing to be a partner. (I would love to talk more about this and how the project team could both communicate better about the project and involve more non-IBM'ers...if there is any interest in such a discussion.)
2. There are a few of us who have developed things that we hope to include in the project but the 0.8.2 release was a ground-up redevelopment of so much infrastructure that (at least in my case) I wanted to wait until it dropped before I invested more time in preparing my code for submission to the project.
3. I know of few people or even companies that have the skills to contribute to the work that has taken place on the EDT project so far. The (IBM) project team has been putting together core infrastructure and (I think) bringing a lot of code over from internal projects like RBD. In my opinion, the project is just now at a point where meaningful but ancillary contributions like mine seem possible and relevant. But if IBM steps back from the project now you can be certain that nobody will step forward. IBM stepping back before 1.0 says quite clearly to people like me: "Dead project. Run away."
4. For those of us in the IBM i community, IBM walking away from EDT before 1.0 would be a huge blow. EGL in RBD was pushed hard as the "way forward" for us. Every "way forward" direction from IBM (let me count the ways...VisualAge RPG, HATS, WebFacing, PHP, "learn Java or you will be flipping burgers," and on, and on) has proven to be short-lived (maybe with the exception of PHP). Unfortunately, when IBM gives advice, we suckers tend to take it to heart and make serious investments before the plug is unceremoniously pulled. It has long been apparent that EDT represents the logical evolution and way forward for EGL. We're on board. We are investing. But we've been down this road with IBM before and consistently come out feeling misled, betrayed, confused, and angry. Are you really going to pull the rug out from under us again?
5. I have expressed more than once that I wish the project team would work harder to get the word out about EDT. My articles for iPro Developer and IBM Systems Magazine along with Joe Pluta's articles for IBM Systems Magazine are the only mentions of EDT that I can find online. (Oh, and the Slashdot article I wrote that announced the first release of the tool and the work I started and several of us continued on Rosetta Code.) You can't fail to tell people about EDT and then wonder why nobody knows about it. Contrast this (as I have many times) with an effort such as Opa -- many, many mentions in trade publications and an active ongoing effort to get the word out about its existence. If you want an active, thriving project with many contributors, make an effort to tell people you exist.
As for my take on the direction IBM takes (not that you are asking):
- Deliver 1.0 and then assess where IBM should stand going forward. I guarantee you that by the time you get to 1.0 you will have contributions from outside IBM. You know that there are people like me out here who have already invested in the project by writing some significant code, creating extensions (one example being for HTML 5 features), submitting bug reports, participating in design discussions (one example being the IBM i extensions).
- Start (helping us) talk about EDT in the outside world.
|Re: What's next??? [message #1015506 is a reply to message #1015208]
||Thu, 28 February 2013 22:40
| Richard Moulton
Registered: August 2011
Location: Devon, UK
On the whole I share similar views to those already made by Dan and others on this thread, and the equivalent thread on the RBD forum.|
I have contributed, in some small way, to the project where I am able; through using/testing, bug reports, responding to forum posts, and sharing code samples. So, my contributions are from a 'user of the langauage/tool' perspective. I will continue to contribute, where I can, but I don't have the skills to contribute code to the core language, generators or IDE.
I glanced at the proposal for EDT (http://www.eclipse.org/proposals/egl/) and looking at the Initial Contribution section I would say that IBM are there, and then some. So, where do we go from here? Will anyone pick up the EDT baton and commit to maintaining the core code? I have no idea, though I would assume that along the road IBM have been trying to get other parties involved with that express goal.
I don't know what EDT 1.0 would look like compared to EDT 0.8.2 but unless someone does commit to maintaining and extending that core code would there be any purpose in continuing?
Will IBM continue to maintain the core code? I don't know the answer to this one but I'm also not sure why IBM started the project and where they saw it going. How are they measuring the success of the project? I'm assuming it was to get EGL 'out there' and create a larger community of developers using the language. If that is the case how successful has that been? How many people are/have been using EDT? Could the community be doing more to get the EGL message 'out there'? If so, how?
What would it take for someone to commit to maintaining the core code? Are there goods reasons why organisations have not committed to maintaining the core code? If so, could the community help to address those issues?
From my own perspective, my initial interest in EGL was to gain an understanding of web technologies and being an IBM i server RPG developer, EGL seemed a good fit, and of course the prospect of only needing to learn one language across a multi-tiered application environment was a big selling point for me. I've invested a significant amount of my time in EGL over the years and therefore want to see it succeed. When I was looking for a web application development language I chose EGL over .net - I'm an IBM i server guy, it just felt right. Is that a good enough reason for the majority of developers?
What did I find difficult and could therefore be a detractor for new developers?
1) The biggest hurdle for me was with the documentation and online resources, if there were a larger community of users then there would generally be a larger pool of resources to call upon. I regularly use google to help me solve my RPG problems but there just aren't (currently) the volume of EGL developers to give you that pool. I find howtos, walk throughs and tutorials immensely useful.
2) The second is with the richness of the RUI widgets. The existing widgets are good but they just don't give the language/toolset the breadth. I know, I can create my own widgets, and I do, but if you're just starting out on a pilot project and are already grappling with the language and tooling, trying to then create new widgets, or extend the existing widgets, is a big ask.
Don't get me wrong I love EDT, it truly is a joy to use. For example, I'm in the process of putting together a new mobile app, which I'll share once I've beaten out the last few gremlins, and I made use of the HTML5 LocalStorage code originally developed by Chris Laffra and then adapted by Dan. Implementing this was a breeze, I truly couldn't believe how easy it was to get this working. The EGL team have done a superb job with the product. I always felt there were areas within RBD/EGLCE that didn't seem quite right, they always felt a bit a weird, a bit non-intuitive, they left you thinking "oh ... so that's how you do that!" and EDT has addressed a lot of those areas. Also, the support I've received on the EDT forum is first class, both from IBM'ers and other EGL developers.
If I was holding the purse strings where would I invest the money?
1) Enrich the RUI widgets, make them world class.
2) Get the EGL message out there. How? I would showcase the product at every IBM event there is. I would include (in the showcase) a real-world problem that organisations are facing today and demonstrate how EGL is the solution to that problem. If the EGL message can't be sold to existing IBM customers and partners then what are the chances of selling it to non-IBM customers? IBM has a very loyal customer base.
I know; to achieve those two points will take a significant investment.
Current Time: Wed Nov 26 16:40:08 GMT 2014
Powered by FUDForum
. Page generated in 0.94942 seconds