The Blog of Maxim Porges

Archive for July, 2005

  • Java/Spring/ColdFusion White Paper Update

    How time flies.

    Last weekend was spent in South Florida with the girlfriend’s family, which more or less completely destroyed the possibility of any work on the white paper getting done. I did manage to squeeze out the outline for the Java/Spring/CF white paper, so with any luck I’ll crank it out this weekend (maybe without 100% tested code samples) and then I can clean it up after posting it if necessary. I’m more interested in time-to-market than total accuracy at this point.

    Otherwise, the schedule has been the usual mix of completely hectic/exhilarating. I’m heading up the technical side of our Oracle team’s first J2EE-based application, which I more or less put the brakes on until we decide upon (a) a development/architecture tool, (b) an SDLC/requirements gathering process, and (c) a front end platform. It’s pretty interesting, since we’re taking a group of primarily Oracle Forms developers, and matching them up with the Java/OO experience of myself and one of our sharpest developers, with the intent being that we build the first fully layered OO system to be deployed by the Oracle team. While the Oracle team is highly proficient and has been developing Forms/PL-SQL apps for many years, this is their first forage in to OO. It’s a sure sign of the level of intestinal fortitude of our Director (my boss) and the openess of the team to investigate the implementation of new technologies.

    As far as the SDLC goes, we’ve been using FLiP very successfully on the Web Team at Westgate Resorts going on three years now, but we’re the only software development team in the department using FLiP full force. The waterfall-based methodology used by the Oracle team is not FLiP. It works, but I’d like to tweak it a bit before we try to do anything OO with it; maybe a bit more forward engineering, and a slightly different tack on gathering business requirements.

    From the tool standpoint, we’ve got demos from Rational and Oracle next week to look at OO modeling solutions, which I’m looking forward to. Between something like Rational’s Software Architect software and Adalon, we’ll have a pretty formidable tool set on the Web Team. To date, there is very little from a modeling standpoint being used by our Oracle teams, and they have expressed an interest in changing that situation.

    I’m interested to see how the product demos go, especially when we do some short term trials. I’m not sure what it is about modeling tools, but they always seem to be slightly under cooked from my experience (read: buggy). I won’t complain, though – they’re certainly better than writing out your system documentation/forward engineering by hand.

    The Oracle tool option we’re considering is JDeveloper, with a potential front end option of JSF/ADF. I’m not too concerned about which front end option we go with, but I must say that after 5 years of web development, I’d rather stay away from JSF. I’m automatically wary of anything that requires a browser to render HTML and execute JavaScript as a platform for enterprise system development; there’s just too many areas for “interpretation” at the client level. As an alternative, I’m extremely interested in Macromedia Flex, for a number of reasons that I won’t list in their entirety – but essentially it comes down to (a) consistent performance and development techniques across all platforms/browsers (including backward compatibility), and (b) fantastic flexibility in user interface design. I’m looking forward to our front end debate to see how these points stack up, especially when we see JDeveloper in action – Oracle tends to build tools that enable rapid development, which is always a plus. Then again, when I explained our goals and needs to our Oracle rep, she told me that Rational would probably be a better option; and how often do you get that kind of candor from a sales rep! Nonetheless, my mind remains open until I see the tools in action.

    Of course, I’ll update you all on the outcome as the project progresses.

    2005.07.14 / 2 responses / Category: Uncategorized

  • Brief Post on An Architect’s View, Adalon 3.0 Broken on Tiger

    I just noticed that Sean Corfield posted briefly on his web site about my FLiP presentation at CFUNITED. Thanks Sean!

    Sean also emailed me earlier today letting me know that Adalon 3.0 does not work on Mac OS X Tiger (a.k.a. OS X 10.4). Not having Tiger on my laptop, I haven’t run in to this, but I’ll be investigating the cause and posting the issue/resolution (if found) once I have it.

    2005.07.05 / 2 responses / Category: Uncategorized

  • FLiP Presentation, Max On Tape, Agile Development

    Just a quick update to let you all know that I am still alive. On my return to work, I was buried under a metric assload of email and other such niceties, which of course dominated my day.

    I supplied a link to the resources for my CFUNITED FLiP presentation on my previous post (look for the link at the end of the post CFUNITED Day 2). (I would have linked directly to the resources again here, but I’m trying not to scatter resource links on too many posts until I finish migrating my blog to its final host on a hybrid Fusebox/Blogger site.)

    Sean Tierney posted an audio recording of me and a bunch of other CF-nerds going off about interfaces and OO techniques in CFCs. It turns out from Sean Corfield’s comment on the post that he and I disagree somewhat about this topic, which I expect will result in some interesting discourse as I post my white papers in the coming weeks. I’m going to work on the first part of the CFC/Java series this weekend, assuming I can squeeze it in around a trip to South Florida that will be happening simultaneously.

    I had an interesting discussion today with Michael about changing our architecture process to being more agile. Our architecture process is currently based upon procedural Fusebox 3.0 development, even though we’re using a quite different toolset now: Java on the back end and Fusebox 4.1 MVC on the front end. Now that we’re doing pure OO on the back end, we should be able to break our architecture process up a bit more and get some more throughput. In tune with my usual eagerness for constant process improvement and experimentation, we’ll be experimenting with this over the next few projects. I’ll let you all know what transpires.

    2005.07.05 / 4 responses / Category: Uncategorized