gnu@hoptoad.uucp (John Gilmore) (05/13/88)
jay@splut.UUCP (Jay Maynard) wrote: > People who argue for source-only distributions don't believe in letting > an author do as he likes with HIS work. This person who argues for source-only distributions has no trouble with letting an author do as he likes with his work -- as long as he doesn't spend other peoples' money doing it. My argument against binaries runs along one line: Binaries are very hard to maintain and very hard to port. What good is the last 30 years of progress in software engineering if we never get the sources? Maintenance is always the largest cost of any piece of software. Binaries are just throwaways; change CPU or OS and toss 'em. Don't tell me an IBM PC binary good today will be good forever...I know better. How many binary programs from PC-DOS 1.0 (before the hierarchical file system) still run unchanged? There is 8088 code that doesn't run on the 386 -- Minix is one (the 386 broke the LOCK prefix on some instructions). I don't think the net should be a medium for "free" (==> other people pay for it, not you) distribution of binary software. There are plenty of IBM PC user groups, BBS's, Fidos, and companies that charge $3/floppy for PD software, doing that. It should be a medium for the continuing evolution of software (and programmers). Binaries don't evolve; when they stop working, they go extinct. Many of the people on the net are on the leading edge of software development. Their binaries won't teach us anything, but their sources will. -- John Gilmore {sun,pacbell,uunet,pyramid,ihnp4}!hoptoad!gnu gnu@toad.com "Use the Source, Luke...."
fyl@ssc.UUCP (Phil Hughes) (05/16/88)
In article <4577@hoptoad.uucp>, gnu@hoptoad.uucp (John Gilmore) writes: > jay@splut.UUCP (Jay Maynard) wrote: > Binaries are just throwaways; change CPU or OS > and toss 'em. Don't tell me an IBM PC binary good today will be good > forever...I know better. > ... > I don't think the net should be a medium for "free" (==> other people > pay for it, not you) distribution of binary software. There are plenty > of IBM PC user groups, BBS's, Fidos, and companies that charge > $3/floppy for PD software, doing that. It should be a medium for the > continuing evolution of software (and programmers). Although I have nothing against free software, net distribution of binaries is not free. As a usenet node we pay for that distribution every day. As far as I know there are no MS-DOS or Apple users down stream from us but I hesitate to chuck all the binary groups as I expect there would be someone who cared somewhere. In the mean time, we use an Atari ST to generate graphics for the classes we teach and to do flyers and ads. The public domain software we have for the Atari came from the old $3/disk vendors. I would rather send $3 to someone for the disk than deal with watching for the binary I want, extracting it, transfering it, unpacking it and attempting to find out what system screwed up on the transfer so the binary is useless. -- Phil uunet!pilchuck!ssc!fyl
dhesi@bsu-cs.UUCP (Rahul Dhesi) (05/17/88)
In article <1220@ssc.UUCP> fyl@ssc.UUCP (Phil Hughes) writes: >...net distribution of >binaries is not free. As a usenet node we pay for that distribution >every day. >...I hesitate to chuck all the binary groups as I expect >there would be someone who cared somewhere. >I would rather send $3 to someone >for the disk than deal with watching for the binary I want, extracting >it... Phil Hughes prefers no binaries on the net. As I understand it, his site chooses to accept and transmit binaries, because he believes others value them, yet he prefers that no other site accept and transmit binaries, because his site doesn't value them. I'm getting tied up in knots analyzing this reasoning. -- Rahul Dhesi UUCP: <backbones>!{iuvax,pur-ee,uunet}!bsu-cs!dhesi
chuq@plaid.Sun.COM (Chuq Von Rospach) (05/18/88)
>Phil Hughes prefers no binaries on the net. As I understand it, his >site chooses to accept and transmit binaries, because he believes >others value them, yet he prefers that no other site accept and >transmit binaries, because his site doesn't value them. >I'm getting tied up in knots analyzing this reasoning. Hmm. It seems perfectly clear to me: o Phil does't believe binaries believe on the net. o The net still has legal binary groups o Phil has not yet chosen to deviate his site from the "official" list of legal groups. o Phil believes that the net, as a whole, should make a decision to remove binaries from the net, rather than having sites drop the distribution in an ad hoc manner. How much clearer can it get? For the record, I'm with Phil. Binaries have outlived their usefulness. Chuq Von Rospach chuq@sun.COM Delphi: CHUQ Robert A. Heinlein: 1907-1988. He will never truly die as long as we read his words and speak his name. Rest in Peace.
manes@dasys1.UUCP (Steve Manes) (05/18/88)
In article <4577@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >What good is the last 30 years of progress in software engineering if >we never get the sources? It's always seemed a bit curious to me that folks who balk the loudest at PD or Shareware distribution of binaries just nod in resignation when they're required to pay upwards of $50,000 to peek at the source code for a program or operating system they paid >money< for. As a Shareware author (who doesn't release code to my most substantial offering) I find something in this attitude to resent. If a developer creates a program that others find useful enough to download, who's being harmed? If you paid $350 for dBASE-2 in 1982 and you've got a '386 today, you've still got an 8088 translation of a CP/M program and there's nothing you can do to change that. At the moment, Shareware authors represent about the only progress in certain software areas, notably communications and file archiving tools, so I think it's counterproductive to restrict access to these programs, and hamstring the competition it promotes, to the Outside World. >Binaries don't >evolve; when they stop working, they go extinct. Many of the people on What do you mean by "stop working"? Do you mean "stop evolving"? Again, you'll find Shareware authors far more responsive to bug fixes and timely enhancements than the commercial developers. This all depends on what subculture of binary download you're referring to. If it's just some quick 'n nasty toss-off, I agree: publish the source. If, on the other hand, it's a program that the author has spent a great deal of time writing, chances are that he wants to protect his source... for a number of good reasons. Magpie (my Shareware BBS offering) took three years to write. I won't release the source because (a) I want to provide support and I don't need munged source code to make the job difficult, (b) I don't want to encounter what other source-published BBS programs have encountered, namely its getting dicked with by some junior terrorist and re-uploaded as the genuine article, complete with a backdoor and media incinerator, and (c) I have long-term plans already in the works for Magpie and don't want to get myself painted into a corner with a lot of third-party enhancements that may well be incompatible with standard files that have been hacked to provide Feecher X or "fix" Feecher Y. I agree with you: source code is a great study and a way to improve your own coding chops. But there are a LOT of people out there who don't know how to program, don't own a compiler and don't care to do any more than USE the product. For them, binaries are the only way to go. -- +----- + Steve Manes + decvax!philabs!cmcl2!hombre!magpie!manes Magpie BBS: 212-420-0527 + SmartMail: manes@magpie.MASA.COM
jpn@teddy.UUCP (John P. Nelson) (05/18/88)
>For the record, I'm with Phil. Binaries have outlived their usefulness.
I just had to add my $.02
I certainly prefer sources over binaries. Given the nature of USENET,
with a diversity of computer types, and where EVERY site must
redistribute EVERY article, it makes more sense to post SOURCE, so that
the program is usable by as wide an audience as possible.
On the other hand, there are a LOT of PC's (all types, not just IBM)
out there. Often those who write programs for PCs are not willing
to distribute source (silly, perhaps, but that's how it is). Also,
often SOURCE for such programs is steeped in machine dependencies:
the program wouldn't be much use to anyone who didn't have the right
type of machine, anyway.
It can be argued that USENET is primarily made up of machines that are
NOT PCs, and users of such require SOURCE. While this may be true to
sume extent, there are also a LOT of users who use both a USENET machine
and a personal PC: These people find binaries quite useful (I do myself!)
So what is the answer? There ARE other forums for PC binaries (BBS
systems around the world, pay services like compuserve and the SOURCE,
etc.) On the other hand, a significant fraction of users of USENET are
benefitted by binaries. Who makes that kind of decision? Do we turn
off rec.arts.startrek just because only 1% of USENET readers benefit?
(no flames please :-) How about talk.jokes? The binary groups have a
large readership, as large as most of the other popular groups, and
larger than many of the "fringe" groups!
--
john nelson
UUCP: {decvax,mit-eddie}!genrad!teddy!jpn
ARPA (sort of): talcott.harvard.edu!panda!teddy!jpnnather@ut-sally.UUCP (Ed Nather) (05/19/88)
In article <53618@sun.uucp>, chuq@plaid.Sun.COM (Chuq Von Rospach) writes: > > For the record, I'm with Phil. Binaries have outlived their usefulness. > Neat. Let's get rid of comp.binaries.ibm.pc and send all our source code to comp.sources.ibm.pc, OK? ...what? That newsgroup doesn't exist? Well, then, perhaps we should not be too hasty ... -- Ed Nather Astronomy Dept, U of Texas @ Austin {allegra,ihnp4}!{noao,ut-sally}!utastro!nather nather@astro.AS.UTEXAS.EDU
fyl@ssc.UUCP (Phil Hughes) (05/19/88)
In article <3083@bsu-cs.UUCP>, dhesi@bsu-cs.UUCP (Rahul Dhesi) writes: > In article <1220@ssc.UUCP> fyl@ssc.UUCP (Phil Hughes) writes: > >...net distribution of > >binaries is not free. As a usenet node we pay for that distribution > >every day. > >...I hesitate to chuck all the binary groups as I expect > >there would be someone who cared somewhere. > Phil Hughes prefers no binaries on the net. As I understand it, his > site chooses to accept and transmit binaries, because he believes > others value them, yet he prefers that no other site accept and > transmit binaries, because his site doesn't value them. > > I'm getting tied up in knots analyzing this reasoning. Ok, let me try again. Fact: it costs a lot of money to send traffic over the net. It may not be out of your pocket but someone pays. Fact: Binary programs don't convey timely information. You may learn something from playing a game or solve a problem by running a utility or whatever but you don't learn something by reading it or whatever. Fact: The world is full of BBSs and companies that distribute binaries of public domain programs for cost or there abouts. Fact: It takes time to snatch a binary off the net, get it unpacked and written to appropriate media for use. Now, my point, again, is that the overall it would cost less and take less time to buy a disk full of binaries for $3. My comment about "someone would complain ..." reflects my laziness. I need to ask those we feed if they care but I haven't as yet. -- Phil uunet!pilchuck!ssc!fyl
fyl@ssc.UUCP (Phil Hughes) (05/19/88)
Here is another proposal: how about a group where you post info about available binaries. Availability info would include a blurb about what it does and a list of ways to get it: Acme BBS, send $3 to Sleazy Software Company (no, that isn't what SSC stands for) or whatever. We could just try this and see binaries on the net would just die. -- Phil uunet!pilchuck!ssc!fyl
bobmon@iuvax.cs.indiana.edu (RAMontante) (05/20/88)
In article <4782@teddy.UUCP> jpn@teddy.UUCP (John P. Nelson) writes: > >I just had to add my $.02 Ditto. Machines that run MSDOS do not generally come with any compiler as a standard component. (Okay, IBM used to include BASIC. What a lifesaver.) I was in a workstation discussion recently, and (I'm no expert, but) at least one person thought it worth mentioning that one of the systems _did_ package a 'cc' with their UN*X-like OS. My point is that, as brands proliferate, the likelihood of having a standard C compiler available goes down. If you have a machine that didn't come with a compiler for a standard language (i.e., C), then source code is not much help. The appropriate binary may be a lifesaver. [Personally, I now prefer source. Thank you, Philippe Kahn, for marketing a more-or-less functional C compiler at a price I could afford.] -bob montante
tneff@atpal.UUCP (Tom Neff) (05/21/88)
You know, the more I listen to this...
I can definitely see Chuq's (and others') point. Binaries have the
following problems: they are (often) HUGE and they are of only limited
net-wide interest. Now comp.cog-eng is also of limited interest, but
it is not so HUGE so the net doesn't mind.
If we could eliminate the HUGE aspect of binaries, then, we could solve
some of the problem. So how about a size limit of 64k or less? This
would allow the posting of useful stuff in response to user inquiries,
e.g.,
> How do I enlarge my PC keyboard buffer? I'm desperate...
OK, here is a uuencode of KBDBIG.COM which will give you 128 bytes...
=====> cut here <=====
etc.
which I do feel is a legitimate function for usenet; but discourage
treating usenet like one supergiant BBS where everyone's 300K demos
are available for download.
This point probably needs to be made forcefully: PC users (at home or
work) with access to Usenet nearly always have access to regional
Bulletin Boards (BBS) as well, not to mention commercial services
like CompuServe and GEnie. For only a few dollars in access time,
you can get anything you want from these places, and you can put your
own software in front of tens of thousands of PC owners. DO IT! Don't
impose on the hospitality of the backbone and thousands of other sites
trying to pervert usenet into Zoonet. (No offense, Rahul - ARCnet is
already a trademark. <grin>)
It may be that after all voices have been heard on this issue, binaries
will seem appropriate for the alt.* hierarchy or even a new bin.*
hierarchy as Jamie Watson suggests. For selected sites it may make
real sense to integrate substantial PC binaries flow into the uucp
network. But not for the net at large.
Comments invited, here or in the 'box.
--
Tom Neff UUCP: ...uunet!pwcmrd!skipnyc!atpal!tneff
"None of your toys CIS: 76556,2536 MCI: TNEFF
will function..." GEnie: TOMNEFF BIX: are you kidding?roskos@csed-1.IDA.ORG (Eric Roskos) (05/21/88)
>Also, often SOURCE for such programs is steeped in machine dependencies: >the program wouldn't be much use to anyone who didn't have the right >type of machine, anyway. This is an important point. I personally prefer sources too, but of the PC sources which are posted (and, in fairness to the posters, it should be mentioned that a good many of the comp.binaries.ibm.pc postings include sources in the PKARC files, along with the binaries), I find *very few* which I can compile; and I use the MSC 5.1 compiler, which is about as close to ANSI and System V compatible as they come. Too many PC and Macintosh compilers don't follow the library conventions, even when it looks like they do, so the sources are not portable. This is especially true of compilers like Turbo C, which seem to have various structures in their include files whose content and function you have to try to guess when porting the code. C source code is just not that portable: it's easy to port most Unix source to the PC (less so to the Macintosh), but often very hard to go the other way. However, there is a second, I think equally important, issue involved. Some people have claimed, "you should use BBS's to distribute binaries." But I trust Usenet binary postings much more than any BBS simply because I trust the technical competence of the people on the Usenet a lot more than I do most BBS operators I have encountered. This is probably less true than formerly, since it appears that now people are interfacing BBS systems to the Usenet, but the Usenet still has a much higher proportion of people whose profession is programming, and who can spot malicious programs more quickly than someone who is primarily a user. As I mentioned earlier, it seems pointless to keep arguing over this issue; if a site doesn't like a particular newsgroup, they don't have to receive it. I know if we ever have to do this here, we will discontinue the voluminous "soc" and "talk" groups long before the binary groups. But the point is, it seems as if people who make such an issue out of the binary groups are concerned with more than their own site's costs; they seem to be out to deprive other sites of the groups too. This doesn't seem very considerate of other sites who do find the groups useful. -- Eric Roskos, IDA (...daitc!csed-1!roskos, or csed-1!roskos@DAITC.ARPA, or Roskos@DOCKMASTER.ARPA) "To tell you the truth, I can't tell what most of these buttons are for." "And, it talks like my car. I don't like that." -- Passengers heard commenting on an "ergonomically designed" elevator.
sl@van-bc.UUCP (pri=-10 Stuart Lynne) (05/21/88)
In article <4491@dasys1.UUCP> manes@dasys1.UUCP (Steve Manes) writes: >In article <4577@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes: >>What good is the last 30 years of progress in software engineering if >>we never get the sources? > >It's always seemed a bit curious to me that folks who balk the loudest at >At the moment, Shareware authors represent about the only progress in >certain software areas, notably communications and file archiving tools, so >>Binaries don't >>evolve; when they stop working, they go extinct. Many of the people on I think it's time to just switch the binaries over to a separate distribution. There is obviously both a great need for them, both as freebies and shareware, but it is equally obvious that a large number sites just don't want to carry them because a) they are binary not source; b) they tend to be very large and c) they tend to be available elsewhere. Now we went through this type of discussion last year and the end result was the alt groups. So how about: alt.pc.binaries alt.shareware As per usual the stuff would be carried by uunet and be available fairly easily to most sites *IF* they want it. I would venture to say that some other noisy groups might be appropriate for moving to alt as well. We have to do something to help cut back the acceleration in the rate of growth. Personally I don't subscribe to *all* of the alt groups but do maintain a link with uunet to get the ones I do want (or are wanted by someone I feed). I probably would subscribe to alt.pc.binaries. -- {ihnp4!alberta!ubc-vision,uunet}!van-bc!Stuart.Lynne Vancouver,BC,604-937-7532
elg@killer.UUCP (Eric Green) (05/22/88)
in article <166@atpal.UUCP>, tneff@atpal.UUCP (Tom Neff) says: > treating usenet like one supergiant BBS where everyone's 300K demos > are available for download. > This point probably needs to be made forcefully: PC users (at home or > work) with access to Usenet nearly always have access to regional > Bulletin Boards (BBS) as well, not to mention commercial services > like CompuServe and GEnie. For only a few dollars in access time, > you can get anything you want from these places, and you can put your > own software in front of tens of thousands of PC owners. DO IT! Don't > impose on the hospitality of the backbone and thousands of other sites > trying to pervert usenet into Zoonet. (No offense, Rahul - ARCnet is > already a trademark. <grin>) AMEN!!! Sites which post 300K demos of commercial programs should be immediately evicted from the net for abuse of net.priviliges. The net is NOT for the marketing benefit of its participants... the net's primary function is INFORMATION, in particular, the kind of information and discussion needed to keep up-to-date and commercially viable (beancounter fodder :-). I've learned a lot from sources -- how to handle termcap, the difference between Sys V & BSD tty drivers (:-(), how to open console ports under AmigaDOS.... but, not a darned thing from binaries. There ARE some valid places where binaries are appropriate. Generally, for nifty LITTLE programming utilities, such as, for example, a "grep" utility for whatever brand of personal computer that you have. But, only with full sources -- you may not have Brand X compiler, but sources are still useful for other purposes (for example, if you write a little public domain text editor, you might want to rip the screen handling out of MicroEmacs and the pattern-matching out of PD-grep). -- Eric Lee Green {cuae2,ihnp4}!killer!elg Snail Mail P.O. Box 92191 Lafayette, LA 70509 "Is a dream a lie that don't come true, or is it something worse?"
jpederse@encad.Wichita.NCR.COM (John Pedersen) (05/26/88)
In article <1227@ssc.UUCP> fyl@ssc.UUCP (Phil Hughes) writes: |> |>Here is another proposal: how about a group where you post info about |>available binaries. Availability info would include a blurb about what it |>does and a list of ways to get it: Acme BBS, send $3 to Sleazy Software |>Phil uunet!pilchuck!ssc!fyl Boy that really opens up some doors, maybe that group could be called commercial.advertising Whats the difference between someone "advertising" a program for a few bucks and someone "offering" a program for a few hundred bucks. Or for that matter someone "selling" software, hardware, cars, houses, ,.......... -- John.Pedersen@Wichita.NCR.COM NCR Engineering & Manufacturing EMC Engineering Wichita KS 316-636-8837
karl@ddsw1.UUCP (Karl Denninger) (05/28/88)
In article <438@encad.Wichita.NCR.COM> jpederse@encad.UUCP (John Pedersen) writes: >In article <1227@ssc.UUCP> fyl@ssc.UUCP (Phil Hughes) writes: >|> >|>Here is another proposal: how about a group where you post info about >|>available binaries. Availability info would include a blurb about what it >|>does and a list of ways to get it: Acme BBS, send $3 to Sleazy Software >|>Phil uunet!pilchuck!ssc!fyl > >Boy that really opens up some doors, >maybe that group could be called commercial.advertising > >Whats the difference between someone "advertising" a program for a few bucks >and someone "offering" a program for a few hundred bucks. Or for that matter >someone "selling" software, hardware, cars, houses, ,.......... Which is exactly what the 'biz' distribution is for. It's been posted before, so here's it again: BIZ is a distribution (top-level name) for commercial messages. 'ddsw1' was the founding point for the distribution. Although it has not received much use, 'biz' could easily handle a lot of traffic -- we (and I suspect many of the other 'biz' sites) have plenty of extra capacity and a willingness to feed anyone and everyone who can poll us. The structure is set up such that a site receiving the 'biz' distribution can easily select which groups are desired; those not wanted simply are not sent or forwarded. There is a top-level break for software, hardware, and computer-related services, as well as provisions for eventual expansion to cover more than just the computer field. EMAIL or voice phone for a connection to the 'biz-net'! ---- Karl Denninger | Data: +1 312 566-8912 Macro Computer Solutions, Inc. | Voice: +1 312 566-8910 ...ihnp4!ddsw1!karl | "Quality solutions for work or play"
fyl@ssc.UUCP (Phil Hughes) (05/31/88)
In article <438@encad.Wichita.NCR.COM>, jpederse@encad.Wichita.NCR.COM (John Pedersen) writes: > In article <1227@ssc.UUCP> fyl@ssc.UUCP (Phil Hughes) writes: > |> > |>Here is another proposal: how about a group where you post info about > |>available binaries. Availability info would include a blurb about what it > |>does and a list of ways to get it: Acme BBS, send $3 to Sleazy Software > Boy that really opens up some doors, > maybe that group could be called commercial.advertising > > Whats the difference between someone "advertising" a program for a few bucks > and someone "offering" a program for a few hundred bucks. Or for that matter > someone "selling" software, hardware, cars, houses, ,.......... Actually, not much. People advertise houses and cars in misc.forsale and all kinks of new products are 'advertised' in comp.newprod. Convention seems to dictate that this is a problem if other than the individual owner places the ad. In other words, if I am a realtor, most people would be offended (including myself) if I posted real estate ads. It is hard to imagine that posting 5 lines of information about where a program is available for $3 (or even $3,000) could be considered more commercial than posting a binary demo version of a program. The binary would certainly be much larger and would probably contain information on where to purchase the 'real' version. The intent of posting the demo binary is to get you to buy the real one. We sell UNIX software (as well as other stuff) and use the newprod group to announce the availability of new products. We hope everyone will rush out and buy it so we can retire and move to Tahiti or whatever (which hasn't happened yet) but we also do it so people will know that a new product is available. I have been on the other end, looking for a specific piece of software, and see this type of posting as a service. Anyway, USENET is here as a service to the users. This binary distribution discussion started because people felt that the benefit of sending binaries on the net may not be worth the cost. Telling me that something exists takes a lot less bandwitdh than sending it. If the announcement also tells me where I can get it and what it costs, I can now decide if I an willing to pay the price. -- Phil uunet!pilchuck!ssc!fyl
heiby@mcdchg.UUCP (Ron Heiby) (06/01/88)
Phil Hughes (fyl@ssc.UUCP) writes: > all kinks of new products are 'advertised' in comp.newprod. Gee! I thought the stuff I approved for comp.newprod was all pretty tame! Some people get off on wierd stuff, huh? :-) P.S. My wife *does* say that I like my computers more than her. I don't believe it, though! -- Ron Heiby, heiby@mcdchg.UUCP Moderator: comp.newprod & comp.unix "Failure is one of the basic Freedoms!" The Doctor (in Robots of Death)
peter@ficc.UUCP (Peter da Silva) (06/02/88)
Here's my vote, for what it's worth: Sources only. And sources should be distributed in a generally readable format (yes, comp.sources.mac, this means you). -- -- Peter da Silva, Ferranti International Controls Corporation. -- Phone: 713-274-5180. Remote UUCP: uunet!nuchat!sugar!peter.