From owner-freebsd-chat Tue Sep 14 7:44:47 1999 Delivered-To: freebsd-chat@freebsd.org Received: from lariat.lariat.org (lariat.lariat.org [206.100.185.2]) by hub.freebsd.org (Postfix) with ESMTP id F3F96150E8 for ; Tue, 14 Sep 1999 07:44:43 -0700 (PDT) (envelope-from brett@lariat.org) Received: from mustang (IDENT:ppp0.lariat.org@lariat.lariat.org [206.100.185.2]) by lariat.lariat.org (8.9.3/8.9.3) with ESMTP id IAA03935; Tue, 14 Sep 1999 08:44:33 -0600 (MDT) Message-Id: <4.2.0.58.19990914080305.047718d0@localhost> X-Sender: brett@localhost X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Tue, 14 Sep 1999 08:44:16 -0600 To: "Jordan K. Hubbard" , Jonathan Lemon From: Brett Glass Subject: Re: FreeBSD Distributions: Leveling the playing field Cc: jkh@zippy.cdrom.com, chat@FreeBSD.ORG In-Reply-To: <80892.937289098@localhost> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org At 11:04 PM 9/13/99 -0700, Jordan K. Hubbard wrote: >It depends on how "separate" they are, and quantifying that is pretty >difficult when it's really such a matter of "feel" for what's >reasonable and what's not. I think we'll have to simply keep doing >what we're doing, which is to look at each thing calling itself >"FreeBSD" or using the word FreeBSD somewhere in its title and see if >it looks appropriate. In that case, there's a problem in that the standards are not cut and dried and objective. This creates a big risk for anyone who might consider producing a product based on FreeBSD. Would a product be denied use of the name because it was "too good" and therefore likely to divert sales from Walnut Creek? Worse still, case-by-case review would require the creator of a product to explicitly state his product plans to an employee of Walnut Creek CD-ROM, which (as you yourself say) might consider itself to be a direct competitor. Furthermore, the permission would likely have to be granted by you -- an employee of that competitor who (as you yourself said in an earlier message) would likely be working on those competitive products. THE developer of Walnut Creek's products would know every time someone was considering a new release, or maybe even a new feature. I hope that you can understand why this is cause for concern! >Things which simply call themselves FreeBSD in the software sense >(e.g. having a CD labelled "FreeBSD distribution" or even "Turbo >FreeBSD" :) have to conform to the basic structure of the official >release, from the installation bits In the source code for the current FreeBSD installer, there's a remark from one "JKH" saying, "This utility is a prototype which lasted several years past its expiration date and is greatly in need of death." Therefore, it does not seem to me that it is a good idea to discourage add-ons such as a better installer, so long as those improvements are clearly labeled as such and the original is included. >to the layout of sources and other >components that would otherwise create great confusion were they to >migrate or mutate substantially from the official version. Again, this could cause problems. Suppose a product shipped with an enhanced source browser or an improved development environment? If so, it makes sense to include the sources in a format that these additions could use. >Companion products, like "The FreeBSD coloring book" or the "FreeBSD >games pack" which contain just "reasonably related" material for >FreeBSD can also use the trademark as long as it's in good taste and >contains some reasonable tie-in to the core product. I think the >FreeBSD coloring book would be kind of cool, for example, whereas I >think the "FreeBSD pop-up porno calendar" (or "Chucky the 13th") >really wouldn't be. :). > >Some existing examples: > >"The Complete FreeBSD" is a book by Greg Lehey about FreeBSD. As >such, it qualifies handily to use the trademark. Were the book >actually about beekeeping, it would not. > >"The FreeBSD toolkit" is a 6-CD set containing all the extra FreeBSD >packages and ports distfiles (+ other FreeBSD related things) that we >can possibly cram onto them. It doesn't claim to be a distribution, >and it's not, it's just an add-on pack specifically useful to FreeBSD >people and it also thus qualifies to use the trademark in this fashion. I, personally, consider this to be an enhanced distribution, since it contains a full "snapshot" plus extras. >Some hypothetical examples which would NOT work: > >"FreeBSD Ultra LEET" - A 10 CD FreeBSD distribution with a new >re-vamped installer and a special bonus port to the Atari ST. This >would definitely not qualify because of the different installer (and, >as much as I may dislike my own installer, I at least know what people >are talking about when they report "an installer bug in FreeBSD") and >the Atari ST port would also be rather less than official since the >project has no such thing in its CVS repository. Again, it's odd that you would deny use of the trademark due to the replacement of an installer which is "in need of death" with one that makes FreeBSD easier to use. >"Really Small FreeBSD" - A binary-only distribution for embedded >systems work, A distribution for embedded systems work would not succeed if it were binary-only. Embedded systems developers INSIST on having source. I wouldn't mind, though, if the source were on online rather than on the disk, so long as I could always get to it and recompile from it if I chose. >also containing proprietary network boot-ROM images and >special drivers for various flash products. This would be a fine >product, I am sure, but again it'd be just too different from >"standard out-of-box expectations" that any user might reasonably have >when hearing the FreeBSD name. I would think that appending something appropriate to the name, such as "BlobWare Embedded FreeBSD," should be sufficient to indicate that the product was different. I, personally, could probably use a product like the above -- but only if source was either on the disk or online. > Something with a lot of value-add for >a specific use and no source tree would not meet those expectations, >even though the product itself might be really great for that specific >use. It doesn't seem to me that the FULL source tree should be on disk in every case. Kernel sources are necessary for a kernel rebuild, but the rest is so easily obtained from the Net, and so seldom used even by technically oriented users, that requiring it to be on the disk is a waste of space for many. I myself would be quite happy if the product were able to bring in the full tree from the Net upon request, but I certainly don't need it for most installs. I'd rather see the space on the first disk taken up by utilities I need; the sources could go on a later disk or even just be made available for automatic download (as they already are, in fact). Currently, I have to switch, awkwardly, between CDs to get all of the utilities I prefer to install on a new FreeBSD system. I'd rather have more of them on one CD. > I'd recommend they call it something like "AtomAntOS" and just >put "Based on FreeBSD" in their sales literature somewhere if they >wanted to give us a nice acknowledgement and maybe catch a little of >the open source buzz, however peripherally, at the same time. I think that the identification provided by a footnote such as "Based on FreeBSD" would be too weak to prevent the world at large (and especially Linux fanatics) from calling it a fork. Which would harm the reputation of all of the BSDs, most especially the one they would say it was forked from. This would hurt the project. --Brett Glass To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message