Questions about COODBMS vs MOODBMS, 9 day fast
The 9 day fast is over and we are all tired. This is the tenth day, victory day, and we had a big fire puja for the Goddess Durga's blessing. So it is a good time to start the preliminaries for MOODBMS.
My first thought is that notifications should be part of MOODBMS, as the COODBMS client is dumb but the MOODBMS clients will need to have a local cache. On the other hand, the MOODBMS help is not open ended. So we need to start a commands set in COODBMS and have help implemented on the server side. This way an application can define its own help. (I am a big believer that clients should be application-independent and back-end driven.)
posted by Bill la Forge @ 9:27 PM
6 comments
6 Comments:
Your project caught my attention, but the more I tried to understand what unique value proposition your software concepts will supposedly bring to the table, the less I could discern any such thing. I do not intent to be negative here, I just want to let you know that as a software developer of many years, I really can't help thinking that this is just "yet another" one-person's-view of how software should be done, and that it will suffer the same fate as most of these projects: near zero adoption.
Why? Because you have not presented any compelling argument in layman's terms for how this is going to improve life in software development as we know it. I see strings of common phrases about how objects and databases and all this stuff is just not working well together (I agree), but I really don't *quickly* see how your approach is going to help me.
I can't help thinking how much great software (great in theory at least) goes down the drain because people can't spend as much time putting forth compelling examples and textual discussion to back their project as they do writing code. Code is the problem - and it is what you are trying to "fix", with more *code*. Consider this a sales job too... convince me and others BEFORE building all this.
A common problem software people have is that, in their *own* mind, the solution they envision makes total sense. But, invariably, they can't present it to others (that don't read minds) in a way that is clear. I'm left feeling overwhelmed after reviewing your various projects (COWDB, etc) and wondering if I should ever look at this in more depth. I'm all for technology and s/w advancement, but you have not sold me on this.
You are not negative, you are correct. There have been many such attempts. And while writing a framework or an API may be difficult, it is easy compared to getting adoption.
In this case I am not trying to implement a better way of doing things, but trying to support a new way of thinking about things. I do not contend that I am doing anything especially novel in the particulars, only that I am attempting to enable a new approach to software.
There have been other, much smaller, programs written to Norm's specifications (proprietary) and these have proven to be quite interesting. The implementations are more "natural" and "intuitive" than is usually the case.
The problem is that it is quite difficult explaining the differences that arise from a different world view. I've sweated over the value proposition and made only small progress.
Part of the problem is that once you begin to think in Rolonic terms, it is all so intrinsically obvious that explaining it is really tough. As Norm recently sayed (partly in frustration methinks), "Wholeness, partness, structure, stream. What else is there?"
So I'm left with building something which, once complete, will have obvious utility and which addresses some of the big issues in the industry. I think we all know enough that you will not find any particular details that are especially novel. Only that the work is organized and balanced in terms of a new perspective, and a powerful one.
So I feel a bit like the stranger in a strange land (a 60's SF by RAH). I can't argue a strong case for what I am doing. Only I can try to demonstrate the effectiveness of the view and hope that others will see its value.
Bill,
Thanks for the feedback. I'll keep an eye on your project. I do think your pursuit may be with merit, and I'll be hoping you succeed with your objectives. And, I will hopefully at some point reach that "aha!" moment where it just makes sense and all. Rolonic - rock on! :)
I am currently pinning my hopes on MOODBMS, as that has the best chance of being able to demonstrate the value of a Rolonic approach. With applications being both implemented and modeled using Rolonic stuctures, any advantages to be had should be compounded to the point where it overshadows the quality (good or bad) of the implementation.
Mike, thanks for you kind words, but please give me another 3 to 4 months. I'll be diving into an application of Rolonics that is new to me, so I may spend some good portion of the effort going around in circles from time to time.
I also believe that "clients should be application-independent and back-end driven". And that brings me to this: what's to stop the back-end from being application-independent? Is this a silly question?
Well, we've got to put the business logic somewhere, ideally as a script in a Descriptor Entry stored in the database.
Actually I want to create a typing system. Rolons would have one or more Roles (i.e. types) and we can then attach business logic to the Role definition.
Post a Comment
<< Home