lauren@rand-unix.ARPA (03/22/85)
I strongly suggest that we AVOID the use of net.sources (or new
subgroups) for new, large software distributions. Except for
fairly small postings, net.sources is an EXTREMELY UN-economical
way to distribute software. Little programs are fine, but
big source distributions never reach large portions of the network
intact and result in endless repostings and lost money. Instead
of forcing the entire network to pay the charges for such postings,
I suggest that people consider using the more traditional methods
of sending tapes when distributions are more than a couple of
100K, or using direct mail/FTP/kermit if practical. But posting many
hundreds of thousands of bytes to Usenet just DOESN'T WORK as
a practical matter most of the time, and the amount of waste
that results (in terms of transmission time, phone costs, etc.)
is incredible. It just results in more nails in the Usenet coffin.
---
By the way, the software-ethics list is still in formation. One interesting
point that several people already pointed out to me is that
the use of notices in sources such as:
"May be freely distributed but not used for profit, or
many only be distributed if you do this and that..."
appear to have virtually no force of law. If something is marked for
open distribution, it apparently is a no-op to try put specific
restrictions on "modes" of distribution--people can sell or distribute
such things in any manner they choose. Apparently there have been
at least two recent court cases (involving sales of software that
were marked "not to be sold for profit" [or some such]) where the
software WAS sold and the courts ruled in favor of the sellers.
The reasons for the rulings are interesting, but I won't go into
them here at this time. I don't necessarily agree with those decisions,
nor do I suggest that they should guide anyone's actions (since I
have not looked up the court references myself and so have not
verified the information that was sent to me on this topic) but
I just thought I'd pass the basic info (as I received it) along.
--Lauren--bill@persci.UUCP (03/26/85)
> I strongly suggest that we AVOID the use of net.sources (or new > subgroups) for new, large software distributions. Except for > fairly small postings, net.sources is an EXTREMELY UN-economical > way to distribute software. Little programs are fine, but > big source distributions never reach large portions of the network > intact and result in endless repostings and lost money. Instead > of forcing the entire network to pay the charges for such postings, > I suggest that people consider using the more traditional methods > of sending tapes when distributions are more than a couple of > 100K, or using direct mail/FTP/kermit if practical. But posting many > hundreds of thousands of bytes to Usenet just DOESN'T WORK as > a practical matter most of the time, and the amount of waste > that results (in terms of transmission time, phone costs, etc.) > is incredible. It just results in more nails in the Usenet coffin. > --Lauren-- I quite disagree. It's all fine and dandy for those with ARPA access to smugly dictate that item x may be had by logging into some system and transferring with 'ftp', but there are many of us out there who DON'T have ARPA access. Frankly, one of the justifications my (for profit!) employer has for supporting the net is the useful (and over 100K!) utilities that are posted on it. If the only things passed over the net were the type of blather I see in most newsgroups, we wouldn't bother! Also, a lot of that software reaches companies that aren't even on the usenet. My former employer is using LOADCONTROL, SCAME, and VC, all acquired from other companies on the net. Software available 'by just sending a tape' won't get distributed anywhere near as widely. I am sure Lauren would not be so quick to condemn these postings if he didn't have ARPA access. It's ALWAYS easier to deny the next guy something when you don't have to give up anything at all!
joe@petsd.UUCP (Joe Orost) (03/27/85)
Lauren is partially right. net.sources should not be used to distribute
large, important, and useful programs.
Instead, they should be posted to "mod.sources" via "genrad!sources". This
way, the moderator can limit the amount of source being transferred per day,
and can check that the "shar" packages are appropriately built.
My "compress" program was distributed via mod.sources, and I didn't see
any requests for re-transmissions.
Limiting access to ARPA sites, or via tape/snail mail severely limits the
distribution, and causes much more work to be performed by the author after
donating the freeware.
regards,
joe
--
Full-Name: Joseph M. Orost
UUCP: ..!{decvax,ucbvax,ihnp4}!vax135!petsd!joe
ARPA: vax135!petsd!joe@BERKELEY
US Mail: MS 313; Perkin-Elmer; 106 Apple St; Tinton Falls, NJ 07724
Phone: (201) 870-5844
Location: 40 19'49" N / 74 04'37" Wdwight@timeinc.UUCP (Dwight Ernest) (03/27/85)
The most practical and sensical method for the distribution of large sources to those who want it, without using postings to net.sources, would seem to me to be that used by Oklahoma State--post an L.sys uucp file entry to the net, with an abstract of the software that's being made available, and let those sites who wish to do so assume the cost of a direct uucp link to the offering site. If necessary, for security reasons, the offering site could permit, through uux, the running of small shell procedure related to the uucp logon that would simply uucp all of the related source code to the calling system that's using the restricted logon. The vehemence of the previous poster who said that those without ARPA access are just left "out in the cold" without any ability to get this public domain source code that's being offered is simply NOT JUSTIFIED when you realize that this pathway is available. As someone who deals with limited budgets, I agree whoeheartedly with Lauren's opinion that the posting of large source code packages via general net posting is just not practical, since you are indeed posting all of the systems on the net to assume the cost of this large posting and its associated communications time. Let's just all realize that there are very practical alternatives available through the use of the same kind of imaginative thinking that got Usenet started in the first place, and use these alternatives whenever possible.
roy@phri.UUCP (Roy Smith) (03/27/85)
> > I strongly suggest that we AVOID the use of net.sources (or new > > subgroups) for new, large software distributions. [...] net.sources > > is an EXTREMELY UN-economical way to distribute software. > > I quite disagree. It's all fine and dandy for those with ARPA access to > smugly dictate that item x may be had by [...] transferring with 'ftp'. > [...] I am sure Lauren would not be so quick to condemn these postings > if he didn't have ARPA access. Oh, come on. I don't have ARPA access either; this has not been a serious detriment to my acquiring software. Sometimes I ask people to email the source to me, sometimes I send a tape, or occasionally people publish public uucp info so I can do the uucp version of anonymous ftp and get it myself (the best way, in my mind). In my opinion, a short description of the software (a la the toolchest) and instructions for getting the distribution is preferable to posting the whole source. -- cmcl2!rocky2!cubsvax -\ vax135!timeinc -> !phri!roy (Roy Smith, System Administrator) allegra -/ The opinions expressed herein do not necessarily reflect the views of the Public Health Research Institute.
bill@persci.UUCP (03/27/85)
> Lauren is partially right. net.sources should not be used to distribute > large, important, and useful programs. > Instead, they should be posted to "mod.sources" via "genrad!sources". This > way, the moderator can limit the amount of source being transferred per day, > and can check that the "shar" packages are appropriately built. > Limiting access to ARPA sites, or via tape/snail mail severely limits the > distribution, and causes much more work to be performed by the author after > donating the freeware. > regards,joe Thank you, Joe. You are quite right. I had forgotten about "mod.sources" (is that because I haven't seen anything on it for a while?). I didn't really intend to come down on Lauren as heavily as my previous posting sounded, but I do have a genuine concern that the system may become divided into the 'haves' and 'have nots', and messages like the one I read appeared to advocate that division. It's not quite so bad with the 'have' vs. 'sort-of-have' system we've got now, where much of the stuff still gets to us 'lesser' sites. By the way, I hope people don't mind my compressing their (> included) letters. I prefer to keep my postings short (< 1 screen). -- Bill Swan (..uw-beaver!tikal!persci!bill) Live in Seattle! 100000000000 slugs couldn't be wrong!
dave@circadia.UUCP (David Messer) (03/29/85)
> > > I strongly suggest that we AVOID the use of net.sources (or new > > > subgroups) for new, large software distributions. [...] net.sources > > > is an EXTREMELY UN-economical way to distribute software. > > > > I quite disagree. It's all fine and dandy for those with ARPA access to > > smugly dictate that item x may be had by [...] transferring with 'ftp'. > > [...] I am sure Lauren would not be so quick to condemn these postings > > if he didn't have ARPA access. > > Oh, come on. I don't have ARPA access either; this has not been > a serious detriment to my acquiring software. Sometimes I ask people to > email the source to me, sometimes I send a tape, or occasionally people > publish public uucp info so I can do the uucp version of anonymous ftp > and get it myself (the best way, in my mind). > > In my opinion, a short description of the software (a la the > toolchest) and instructions for getting the distribution is preferable > to posting the whole source. Although Lauren's suggestion has some merit, it is only true if only a small percentage of people are interested in the posting. The total cost to the USENET community would be greater if everyone had to call long-distance to get a copy of a program. Another objection is that I am not sure if postings to net.sources should be discouraged. I find that net.sources it the single most useful group in the network, and unless there was some other method of distributing sources, I think that I would miss out on some very nice programs. Also, because it is easy to post a program to net.sources, I think more people will take the trouble to do it. I know that if I had to set up some special way to distribute the few programs that I have posted, (i.e. if I had to actually THINK about how to do it), I don't think that I would have bothered. In any event, Lauren's objection seems to be based largely on the fact that distribution of ANYTHING on USENET is unreliable. My respose to that is, rather than putting a band-aid on the problem, fix it so that postings WILL get through. I am not an expert on they way netnews is distributed, but the times that I have personally seem postings get lost seems to always be related to the receiving system running out of disk space. It seems to me that it would be possible to confirm that an article has been correctly received, and keep trying every day until it does get through. (Up to some time- limit of course). Is this over simplified? I do agree that specialty software distributions should not be posted to the whole world. Those who wish to post any large article should certainly consider the cost to others when they send it out. -- Dave Messer ...ihnp4!stolaf!umn-cs!circadia!dave
david@ukma.UUCP (David Herron, NPR Lover) (04/01/85)
>From: bill@persci.UUCP >Message-ID: <130@persci.UUCP> > [Quote of --Lauren--'s article on use of net.sources] >I quite disagree. It's all fine and dandy for those with ARPA access to >smugly dictate that item x may be had by logging into some system and >transferring with 'ftp', but there are many of us out there who DON'T >have ARPA access. Frankly, one of the justifications my (for profit!) >employer has for supporting the net is the useful (and over 100K!) utilities >that are posted on it. If the only things passed over the net were the type >of blather I see in most newsgroups, we wouldn't bother! >Also, a lot of that software reaches companies that aren't even on the >usenet. My former employer is using LOADCONTROL, SCAME, and VC, all acquired >from other companies on the net. Software available 'by just sending a tape' >won't get distributed anywhere near as widely. >I am sure Lauren would not be so quick to condemn these postings if he didn't >have ARPA access. It's ALWAYS easier to deny the next guy something when you >don't have to give up anything at all! But we have our own version of ftp. It is called uucp. Y'know? that thing that delivers the news to us? It transfers files don't it? I agree with --Lauren--. For large postings, they should be available via "anonymous uucp" much like the OKstate kermit setup. (And thank you OKstate for doing it too!) The only problem I see is distributing it to overseas sites. Something will have to be worked out. Like secondary repositories in europe and asia or some such. But thats a detail. Any other comments? -- --- David Herron --- ARPA-> ukma!david<@ANL-MCS> or david%ukma.uucp@anl-mcs.arpa --- Or even anlams!ukma!david@ucbvax.arpa --- UUCP-> {ucbvax,unmvax,boulder,oddjob}!anlams!ukma!david --- cbosgd!ukma!david "The home of poly-unsaturated thinking".