Am 23.08.2012 um 17:49 schrieb Kent A. Reed: > On 8/23/2012 3:07 AM, Michael Haberler wrote: >> As for dicing and slicing the codebase, I would think this wouldnt just be >> beneficial for the issue at hand, but also for spelling out the dependencies >> clearly, and 'selling' parts of LinuxCNC to other users, increasing use and >> community reach. For example, spinning out the HAL, RTAPI, component >> infrastructure, component compiler, and maybe the generic interface part to >> a new messaging scheme would be a boon to folks doing embedded outboard >> devices. Right now that is tightly integrated into the code base so few >> people stumble upon the fact that RTAPI/HAL is a very generic vehicle (which >> is in fact very well designed, kudos). Even I would need some time >> separating it out and make sure the recombined projects dont fall on their >> face. > > Michael: > > I believe embarking on a restructuring process is justified on technical > grounds alone and I support the idea. On the other hand, the "selling > and spinning out" part of this paragraph reflects sentiments I've seen > in several other messages on our lists and it concerns me. While there's > nothing wrong with "unlocking value," as the financial markets like to > put it, I hope the one clear goal is always to make a better LinuxCNC. > It's great if other benefits can accrue in the process but let's not let > them become the goal itself. As you say, we don't want the recombined > projects to fall on their face.* > > We're all reasonable people so nothing can go wrong? Maybe, but history > is replete with stories of good technical projects that fragmented when > they expanded their goal to be all things to all people.
Mission creep is always a problem. However, I'm a bit pressed to think starting of which mission such a creep should deviate from, as I dont see a discernible mission wrt to the future (I would understand 'being a reliable vehicle' being part of the current mission, but that is telepathy and tea sud reading). And it would be a bit pretentious to try to tack on a new 'mission' which is substantially different from what I guess is the 'non-verbal community consensus' (ha!). As for Linuxcnc3, I think it would be foolish to start such effort without a clear, stated intent, and actually a very simple, understandable one. Lets call it a mantra. As far as I am concerned, I've not condensed it into a single 'Caber Feidh' type battle cry, but among my goals would be: - simplify, simplify, simplify - take out the guru factor (can we say 'the impenetrable kitchen sink of motion'..) - regularize interfaces, and make them combinable in new ways (why cant we have layered components, say a homing comp, which USES motion to move, and which you could dump if you dont like it?) - go back to Unix roots (why is it that we cant have a readable file of motion commands, cat that on a motion device, and the machine moves? NB: we have NO way of sensible regression testing motion and motion-type vehicles due to that) - get rid of some old assumptions which turned out to be too restrictive. Fix the infrastructure first, then cut and shake up. Thats pretty much it for me, large enough goal I'd say. Simplifying is hard enough. --- sorry for the mission creep from the original thread theme, though ;) - Michael > > I'm not saying it will happen, only that it can happen. > > There's an old US television series named "Hill Street Blues" that had a > great tag line. Early episodes began with a police station's roll call. > At the end of the roll call, Sergeant Esterhaus would call out after the > departing officers "Hey, let's be careful out there." That would be my > suggestion here. > > Just my 2 cents worth. > > Regards, > Kent > > *The good news is that breaking out different parts may bring new > stakeholders and additional support, the bad news is that the different > stakeholders have to be kept focused on the goal of the original > project, else chaos ensues as they push and pull on their favorite part > to meet their own goals. The mental image that comes to my mind is the > scrum in rugby union football. > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Emc-developers mailing list > Emc-developers@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-developers ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
