chuq@sun.uucp (Chuq Von Rospach) (11/13/85)
[If you like this bugeater line, please send $5.00 to...]
The furor over net.sources.mac has me thinking. I was one of the people involved
in getting net.sources.mac started, and I've gotten a LOT of useful stuff in it
over the last year. At the same time I'm not satisfied with the kind of postings
I've seen in the last few months. In trying to figure out why I think I've found
two problems with net.sources.mac that we need to look at and resolve:
Lack of Source
There are historical reasons why there isn't much source. Early on we were all
trying to make the machines work and there simply wasn't a good development
environment for the Mac. The only way to allow people to use things was to pass
a binary. Now, there is the compatibility problem. There simply isn't a
development environment dominant enough that "everybody" will have it (unlike,
for example, Unix [C] or Apple II [Basic]). Mac C doesn't talk to megamax or
sumacc, and there is pascal, lisp, forth, and various assemblers.
I've come to the conclusion that the compatibility issue is a red herring. We
post binaries because we've always posted binaries, and "compatibility" is a
good excuse. What I think we really need is solid examples of how to make the
achine roll over. You can read Inside Mac for weeks and learn more from SKEL
in a few hours when it comes to the practicality of working with the machine.
Binaries don't teach you how to program the beast and, more importantly, don't
allow you to change the program to do what you want. Resources notwithstanding,
you can't hack a binary. You also can't learn from it.
It is time to bite the bullet and start posting sources again. I may not be able
to compile it, but if it is something I really want, I'll either convert it (and
learn in the process) or buy the system that CAN compile it. Binaries are handy.
Sources are useful, and it is time to make sources.mac useful as the teaching
tool it should be so we can all learn how to make better tools.
The Shareware Problem
In trying to decide what to do about the "shareware crisis" on the net, I went
looking through my archives. I found a LOT of "shareware" programs in it. The
number of "shareware" programs I have that I paid for turns out to be nil. In
trying to reconcile this position, I did some research into what "shareware"
should be."Shareware" (or "freeware") was a term coined by the late Andrew
Flugelman for programs he wrote for the IBM-PC called PC-TALK and PC-WRITE. He
wanted to stay away from the normal distribution channels -- 40% for the
distributor, advertising, and the normal administrative overhead involved in
getting code to market.
His basic philosophy was that shareware was commercial quality code. If you look
at what his view of shareware is versus what is being posted as shareware, there
is a BIG discrepancy. He distributed a word-processor and a telcom program. We
are posting things like Binhex, disk initializers, and other toys. Is the stuff
being posted as shareware something that could have been sold in a store?
With very few exceptions, no. Most of the stuff being posted as shareware would
have been written anyway, and in the old days given away at user group meetings.
Now, chasing the almighty dollar, people who might have shared their discoveries
about the machine are locking them away in undecipherable binaries in hope of
earning a few dollars. I can think of only two things I've seen pop by that I
felt deserved the "shareware" label. One is "Red Ryder" and the other is the
"Dungeons of Doom" game. Both of these are non-trivial programs with a lot of
work and finishing put into them, and both of them could easily stand on their
own in the commercial market.
What Do We Do?
The people posting "shareware" should seriously consider whether or not their
program deserves to be shareware. I think we all stand to gain a LOT more from
examining and discussing each others code than we could ever gain from passing
around binaries. The prime question would have to be "Is this something I'd want
to sell in a computer store?" and if you don't think the code is quite robust
enough or the documentation quite clear enough, the answer is no. If you aren't
sure, post the sources as shareware and let us pay you for the algorithm. We
need to pressure the off-net people to reconsider their positions, too --
shareware gives a few people some short term gains, but gives the Mac community
long term losses because we simply aren't sharing our knowledge with each other.
If I show you what I know, and you show me what you know, we both get better at
programming the machine, which is what it is all about.
Isn't it?
Anyway, Shareware has its place on the net, and it always will. But we should be
a lot more careful defining what is and isn't shareware and start posting the
sources that we will need to make everyone a better Mac programmer. It's time to
stop posting binaries, folks -- we simply don't need them anymore.
--
:From the Crystal Caves of Avalon: Chuq Von Rospach
sun!chuq@decwrl.DEC.COM {hplabs,ihnp4,nsc,pyramid}!sun!chuq
Our time is past -- it is a time for men, not of magic.ems@amdahl.UUCP (ems) (11/16/85)
> ... It's time to > stop posting binaries, folks -- we simply don't need them anymore. > -- > :From the Crystal Caves of Avalon: Chuq Von Rospach What about the new MacOwners who don't have all the binaries yet? When I first got my mac, net.sources.mac was indespensable in making it a usable machine. What about those who don't have a compiler yet? Those who don't want to lay out $N, where N is large, for M, where M is large, different environments/compilers? -- E. Michael Smith ...!{hplabs,ihnp4,amd,nsc}!amdahl!ems 'If you can dream it, you can do it' Walt Disney This is the obligatory disclaimer of everything. (Including but not limited to: typos, spelling, diction, logic, and nuclear war)
chuq@sun.uucp (Chuq Von Rospach) (11/18/85)
> > ... It's time to > > stop posting binaries, folks -- we simply don't need them anymore. > > What about the new MacOwners who don't have all the binaries yet? When > I first got my mac, net.sources.mac was indespensable in making it > a usable machine. What about those who don't have a compiler yet? > Those who don't want to lay out $N, where N is large, for M, > where M is large, different environments/compilers? Perhaps we want to ship both source and binary, if the binary isn't huge. The new Macowner is a small enough part fo teh group that we can handle them as we go (look at how few requests for downloading software we're getting on the net). If we always aim at catering to the lowest base of knowledge, we'll never teach each other to be experts. Here is a great case in point: Jeff Shulman at topaz posts three messages of Diatom in net.sources.mac. To quote, "I wrote it to learn how to make the Mac do all the wonderful things it does". What he posted was a binary, a doc file, and a configuration file. Great. He learned how to do lots of wonderful things, and I have a program binary that prints pretty pictures. If I had a source, I could learn how to print pretty picutres myself, but he didn't post the source, so I can't. He gains, but nobody else can share in it without decompiling his program. I don't see how this helps out the new user. Or even the not so new user... chuq -- :From the Crystal Caves of Avalon: Chuq Von Rospach sun!chuq@decwrl.DEC.COM {hplabs,ihnp4,nsc,pyramid}!sun!chuq Our time is past -- it is a time for men, not of magic.
tim@ISM780B.UUCP (11/21/85)
> d) [this is working in a very different direction] > the definite compiler review showing that we should all use I doubt that this is possible. For example, consider Aztec and Megamax. Aztec is commonly considered to be the best C compiler for the Mac ( at least until Green Hills is available ). Megamax is commonly considered to be very good. But consider the price. Aztec is around twice the price of Megamax. You can go ahead and decide that we all should use Aztec, and I will fart in your general direction, 'cause I don't want to spend that much money. Or we can decide that Megamax is to be prefered. Then those who have the money and want the vi like editor and the shell will rightly fart in your general direction. It gets even more complex if we bring in Consulair, Desmet, Hippo, and Softworks. Maybe we should all use Neon or Smalltalk, since those are the only two Mac languages that have only one implementation. Tim Smith ihnp4!cithep!tim ima!ism780!tim
chuq@sun.uucp (Chuq Von Rospach) (11/21/85)
> Here is a great case in point: Jeff Shulman at topaz posts three messages > of Diatom in net.sources.mac. To quote, "I wrote it to learn how to make > the Mac do all the wonderful things it does". What he posted was a binary, > a doc file, and a configuration file. Great. He learned how to do lots of > wonderful things, and I have a program binary that prints pretty pictures. > If I had a source, I could learn how to print pretty picutres myself, but > he didn't post the source, so I can't. He gains, but nobody else can share > in it without decompiling his program. I don't see how this helps out the > new user. Or even the not so new user... One thing I didn't make clear, and I apologize to Jeff for that, was that Jeff didn't write diatom, but was making something off the Delphi system available. Personally, I think the work he is doing with net.*.mac is wonderful. Unfortunately, it just complicates the problem. Since we DO get stuff off of Compuserve and off of Delphi, we don't always have access to the source. In the case of a number of things (like uw, for example), the binary is useful enough to warrant posting without source. In other cases (and I count Diatom among them) there just doesn't seem to be a practical use. If I had the source, I could use it as a learning tool, but what else is Diatom good for, and how useful is it without source? So the question is: Do we want the binaries anyway, if we can't get the source, or can we come up with a reasonable standard of when something ought to be posted and when it shouldn't? If nothing else, I'd like to find a way to prevent people like Jeff from getting flamed by people like me every time they post something from off then net that we don't like. If we can give them an idea of what we don't want, maybe we can cut some of the volume in net.sources.mac without loss of the useful stuff. -- :From catacombs of Castle Tarot: Chuq Von Rospach sun!chuq@decwrl.DEC.COM {hplabs,ihnp4,nsc,pyramid}!sun!chuq Let us now take the sacre oath. As of now, he is no longer an elephant!
tim@ISM780B.UUCP (11/21/85)
/* Written 11:15 am Nov 21, 1985 by me in net.micro.mac */ Maybe we should all use Neon or Smalltalk, since those are the only two Mac languages that have only one implementation. /* End of text from net.micro.mac */ Yes, I know there are others.
pangrle@uiucdcsp.CS.UIUC.EDU (11/27/85)
>/* Written 11:15 am Nov 21, 1985 by tim in net.micro.mac */ >.... Or we can decide that Megamax is to be prefered. Then those who >have the money and want the vi like editor and the shell will rightly >fart in your general direction. > >It gets even more complex if we bring in Consulair, Desmet, Hippo, and >Softworks. Maybe we should all use Neon or Smalltalk, since those are >the only two Mac languages that have only one implementation. > >/* End of text from net.micro.mac */ >/* Written 12:52 pm Nov 21, 1985 by tim in net.micro.mac */ > >Yes, I know there are others. >/* End of text from net.micro.mac */ Too late. They probably have already farted in your general direction. Barry Pangrle "... and my favorite word is: 'you never know'." Joaquin Adujar