On 07/27/2010 09:54 AM, Jeremy Huntwork wrote: > On Jul 27, 2010, at 9:58 AM, William Immendorf wrote: > <snipped "To Do" stuff> > 3. Either update XFCE to 4.6.2, or replace it with LXDE (and we >> should add XFCE and/or LXDE to the book, and add back HAL support >> since it's usable again.) >> > Again depends on #1. Also, it depends on what the purpose of the CD is > decided to be. I'm not 100% certain the cd should even contain X. > The purpose, IMO, is the key. If that doesn't exist, there is nothing towards which to work and nothing to prevent, or slow down, well-intentioned branching based on "this would be a nice feature." The rest--maintenance, build method, scripting and packages--are at this point if not irrelevant at least of secondary importance. I have some thoughts that will end in a recommendation. I will put them below. > >> 4. And elect a new maintainer for this package. >> > Very possible. I don't mind staying on and managing, but I can't do all the > work alone. I also don't mind handing over the reins entirely if that is what > is best for the project. > Let's not get the cart before the horse. Let's design something, get it into production, see how it works and then re-evaluate the need for a single, already present maintainer. Jeremy, I think that while resurrecting this project, if that's what happens, you need to be at the helm for, at the very least, consistency with the past.
In another post, you mentioned the difficulty of generating commit authorization. I think it needs to be difficult. There's the trust which is a completely separate issue, but a difficult authorization process may tend to weed out some whose exuberance carries them away, they actively participate but rapidly lose interest. I think, which by now I hope is obvious :) , that a purpose is the first step and that depends on what everyone thinks the end product should look like. Let me "throw out" a couple of general questions about the end product. Do we want just an up to date--whatever that means--iso image that someone can use as their host system for an LFS build? Or, do we want a cd built on an individual's machine so that they can burn it and use it for a build and use it that as a "rescue disk?" These questions lead back to something I said in another post about evaluating what users want compared to ease of maintenance--and this in large part depends on the number of people involved in the maintenance and the abilities of each. Jeremy, you responded to this concept by saying something like "that has been difficult for me to get my arms around." So, here's a recommendation: Could we, and by this I mean you, Jeremy, post on the [,A,B,C]LFS-[support,dev] lists an e-mail similar to the following: > There is a small group of people trying to assess the viability of > resurrecting the LiveCD project. We would appreciate your input to > this process by answering the following questions. > > 1. Have you ever used an LFS LiveCD. > > 2. Did you use it only for a build host? > > 3. Did you use it to build something other than LFS > > 4. Did you use it as a rescue disk? > > 5. Did you obtain the iso image by having it build on your system? > > 6. Did you obtain the image by downloading it directly? > > 7. Did you use it for anything not listed in #2-#6? > > We have posted this to many LFS mailing lists, please reply only once. > [Please reply to the LiveCD list with your answers to prevent noise on > the other lists.] Is it possible to imbed a "redirect" only to this list? If so, maybe we don't need the last sentence. I know a lot of LFS-ers monitor more than one of the lists and there are probably those who've built LFS and used the LiveCD and who monitor only one list. That's why the recommendation for the "shotgun approach." Then, if we send it, we could wait, Hmmmm, sixteen days--two week vacations plus an error margin--consider the results and generate a "purpose." Jeremy, I think you should be the author since you are the "official" maintainer. And, it goes without saying--which is a logical disconnect because I'm going to say it anyway--please change the above text in any way you see fit. It's just a "first cut." I think that a request for features is premature. We can solicit that input after we decide what the LiveCD should look like. I would be more than happy to collate the results of our data search and present the results at the end of our waiting period. I think I have pretty much exhausted my thoughts for this writing. Aren't the rest of you glad? :) Dan -- http://linuxfromscratch.org/mailman/listinfo/livecd FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
