<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"><html><head><meta content="text/html;charset=UTF-8" http-equiv="Content-Type"></head><body >I agree with Matt.
<br>We have just started to work on a API-basis. The APIs will be
<br>difficult to use initially no doubt. I expect the APIs to be APIs like by 5.2 only as it is a huge change in mindset.
<br>Moreover, whatever APIs we have now, have to be properly documented and made much more usable.
<br>Matt has pointed out quite a few lacunaes and I do assure you that we do intend to fix them.
<br>Guidance and direction on the implementation toward achieving the same are welcome.
<br>
<br>To all System Integrators, yes, I do understand your pain. But, please understand getting rid of legacy code is much easier said than done.
<br>As of now, we will have to live with what we have.
<br>Post 5.0, we will make appropriate changes.
<br>
<br>Richie<br><br><br><br><br>---- Matthew Brichacek<mmbrich@fosslabs.com> wrote ---- <br><br><blockquote style='border-left: 2px solid #0000FF; padding: 6px;'><html>
<xbody>
> [snip]<br>> ><br>> > Our current installation is a heavily-modified (and I do mean HEAVILY)<br>> > modified 4.3.2 installation. To be honest, I painted myself into a bit<br>> > of a corner with the way I implemented our code changes, such that<br>> > upgrading to newer versions is a formidable challenge.<br>> <br>> this is a mistake many people do.<br>> you have choosed to fork the project instead of trying to work with<br>> the community.<br>> <br>> and all this people end up in the same corner as you... :-)<br>> <br>While I would normally agree 100% with Sergio on this point, I think we<br>need to take the history of this project into consideration and maybe<br>look at the bigger picture.<br><br>There was a point in this project where you could come along, see a<br>project with a fairly vibrant community forming, an interesting product<br>being developed and a complete lack of direction and management. This<br>type of environment is ripe for internal forks and the fact that the<br>project didn't fork into a new OSS project is nothing short of a miracle<br>(seriously!).<br><br>I think Dennis is just the first one to cry out for help publicly, there<br>are probably many more out there like him (I know of 6). The system<br>integrators had a choice to make... Stand behind the OSS project in all<br>it's glory and be at the mercy of a project with no direction or<br>leadership, or fork and be in charge of their own destiny. While in<br>most cases I would say you're loony for privately forking a strong OSS<br>project, in those cases I would say it was a justifiable move (i'm<br>biased of course).<br><br>Ok, so there have to be a _ton_ of cool features and ideas floating<br>around in all these private forks and I think the project managers<br>should make a full effort to get these integrators back into the<br>community and help them get their features in 5.x. Why not put them in<br>4.x? Here are the options as far as I see them for people who have<br>forked from 4.x:<br><br>1) Keep running with your own 4.x fork even after the community drops<br>4.x in favor of 5.x. Good luck, have fun, and let us know how that<br>works out :).<br><br>2) Submit patches for all your features into 4.x, get them integrated,<br>then submit them for 5.x (or hope some wonderful soul did it for you).<br>If you have to submit them yourself after you've done 4.x, chances are<br>you will miss the final release of 5.x and will be in a stable period<br>that may be harder to get your features into.<br><br>3) If not already done, stabilize your 4.x fork and get busy forward<br>porting everything to 5.x. You'll need to create your own upgrade path<br>for your company/customers but IMHO you're chances of success are better<br>in this case. Some of your features may not be accepted, but maybe the<br>underlying framework that you need to enable those features can be, or<br>if you're lucky, the API will do what you need (don't hold your breath).<br><br>It really comes down to options 2 and 3, and who wants to double all<br>their work when you know damn well that 4.x is going to be abandoned now<br>that 5.x shows the promise it does. And I mean that with absolutely no<br>disrespect for the people working hard on maintaining 4.x for the<br>community. Will 4.x be maintained forever? I highly doubt it, once 5.x<br>stabilizes and upgrade paths are figured out we'll certainly want those<br>developer resources moved to 5.x right?<br><br>About code comments:<br>I'm in no position to preach about code comments, but I am getting<br>better at it (code comments that is :). What I think we could use _way_<br>more than code comments is a sane API with at least _some_<br>standardization and logic to it. But I don't have the time to do it or<br>fix the millions of things that will break, so I'll just shut up now.<br><br>Matt<br><br><br><br>_______________________________________________<br>Get started with creating presentations online - <a href="http://zohoshow.com?vt">http://zohoshow.com?vt</a> <br>
</xbody>
</html></blockquote></body></html>