Programming Paradigms

It's more than a change in the weather that has Michael scratching his head this month.


January 01, 2000
URL:http://www.drdobbs.com/programming-paradigms/184403976

Jan00: Programming Paradigms

Michael is editor-at-large for DDJ. He can be contacted at [email protected].


Jim Button: Where Is He Now?


The old saying goes, "The more things change, the more they stay the same." Now why would that be? Is it because change means abandoning the past, and abandoning the past inevitably means forgetting the past, and forgetting the past leads to repeating the past ("Those who forget the past are condemned to repeat it") and repeating the past results in things staying the same? Is that why?

Or is the old saying saying something about negative-feedback loops, systems in which feedback works as a corrective? In eras in which positive feedback loops are the rule, does the old saying itself change into something like, "The more things change, the more things change"? And is that the saying that has something to say about our collective situation in this millennial year?

Get Their Names

A reader writes to take me to task for a lapse into innumeracy in my November column. A 300-fold increase, he rightly points out, is not equivalent to a 30,000-percent increase. It is a mere 29,900- percent increase.

It annoys me to make a mistake like that because I like to think of myself as being pretty good at math. With one huge exception. When a number has a currency symbol attached, my eyes glaze over, my gears slip, and I lose all confidence that I know what I'm talking about. So perhaps I shouldn't even begin this topic. But there are some changes afoot involving businesses and markets and economies that are obviously so enormously significant that even I can grasp that there's something up. Exactly what it all means for how goods and services get bought and sold, how software gets developed, and which Internet business models make sense and which don't, I haven't the slightest idea. Then again, does anybody?

Maybe I could just list some of the signs of change: Open source is, of course, hot. That raw fact is old news, but the shamelessness with which big companies are falling all over themselves trying to prove that they're supportive of the open-source concept is remarkable. They donate money to open-source projects, invest in open-source-centric companies, and pontificate about "giving back to the community." Eric Raymond, who named the open-source phenomenon, is in high demand as a speaker. Sendmail and OpenSales have recently set up forums for discussing open-source issues and sharing code, pushing the movement forward. OpenSales is doing workshops, Sendmail has an open-source portal. And while open-source or free software doesn't mean "no-cost software," the movement does challenge traditional thinking about profits in the software business.

It needs to be said, though, that the software business is too recent a phenomenon to have any very entrenched traditions. And also that there are those who think that "software business" ought to be an oxymoron. It is at least the case that, as software invades more and more of our lives, the term will encompass too much, so that "software business" will be as meaningless as "idea business." Most ideas have nothing to do with business, but few businesses get along without any ideas. Ditto, I suspect, for software in the future.

And music? Music is a part of life, independent of producers, agents, and recording engineers. The music business is struggling with the fact that those who want to make music and those who want to hear it can connect with one another without all the middlemen, sometimes -- gasp! -- without money even changing hands. The Internet and the MP3 format have made it possible for producer and consumer to bypass all the middlemen, in music and elsewhere, but very dramatically in music. The middlemen, used to running the show, aren't taking it lying down. Mostly, they're fighting the trend, trying to make MP3 downloads illegal or technologically impossible.

Universal Music Group has, at least in part, decided to embrace the trend. It's putting up a web site where unsigned bands can upload music and their fans can download it -- apparently all for free. If you can't beat 'em, the philosophy seems to be, at least get their names for your mailing list.

The Free Market

Other companies are also deciding that there are other things in this world besides money. They're giving away expensive products to build those customer lists. Hoping, against the advice of a lot of traditional marketing gurus, that coupon clippers can be turned into real customers.

An awful lot can be had for free out there in Cyberspace (if you'll put up with the hidden cost, which usually means being forced to read ads). You can get a free personal computer, free monitor, free Internet connection, free e-mail, free web site, and free long-distance calling. You can download all the basic application programs for free: word processors, spreadsheets, databases, games, tunes. Lots of stuff that once cost big bucks, and lots more stuff that used to be shareware. Palm developers complain that applications previously sold as shareware for the Palm are being given away free by Amazon and eBay, with Palm's encouragement. They say the shareware market for Palm applications will be destroyed by these freebies. Palm says don't worry.

And the freebies just keep on coming. As one report pointed out, there are companies that will set you up with a full Internet storefront, complete with shopping cart capability and credit-card processing, presuming you can figure out something to sell at the store. Good luck on that.

The basic idea here is not as crazy as it sounds -- I guess. The Internet is where people are going to go to shop in the future, goes the thinking. They may still, in some cases, in some markets, complete their purchases offline. But they'll look first on the Net, so that's where you'll hook 'em, so right now it's worth a lot to buy their attention. One of these free marketers says we're in the midst of a shift from a retail economy to an attention economy. That seems an astute interpretation of a snapshot of the present moment, but it glosses over the end that these unusual means presumably serve. Seems like a lot of people are a little glossy about the end. If what you spend is money and what you get back is attention, eventually you have a balance-of-payments problem, I think.

In theory, of course, companies buying attention are buying a piece of a future market, and that piece will be worth a lot of money when the market becomes real. Some observers are worrying, though, that markets will be wrecked by all these freebies. How do you establish a fair market value for a computer once they've been given away like pocket calculators and AOL CD-ROMs?

Noncomputer case in point: My poor frisbee-chasing (and chewing) dog has been frisbeeless for a month because I can't bring myself to buy one for the price they charge in stores. I've just picked up too many free ones at computer shows to accept that they have any value. (My dog disagrees with me.)

Then there's this: These companies may think that they have a strategy that will pay off in long run, but are they really in control of what's happening? The Internet makes it possible, in theory, for consumers to take control of markets. And there are signs that this will not remain a mere theoretical possibility.

Mercata, Accompany, and Act Big are sites that allow customers to band together to do bulk buys. NextTag, Deja.com, Respond.com, and Epinions are offering somewhat similar services. Auction sites allow customers to offer a price for an item, directly informing vendors what the market will bear. Reverse auctions, in which many vendors bid for the business of one or a few customers, are familiar in government procurement, but could also become a tool for consumers: Suppose I offer to place an order for corn chips this Friday for every interested household in America. Bet I could negotiate a good price, and the logistics and shipping and order processing and electronic funds transfer -- those are just technical details that haven't quite been worked out yet. How long do you think it'll be before they are?

It seems plausible that the Internet is going to let consumers call the shots, create the marketplaces, dictate prices. But these are the same consumers currently being taught that the fair market value of a word processor or the computer it runs on is zero. Yikes.

Endangered Species

There are changes afoot out there in e-commerceland, and some businesses will fail to read the tea leaves right and will fail. Mostly these will be middlemen.

The Internet has been systematically eliminating the middleman in transactions. That trend is not nearly over. Many new Internet-based companies don't realize that they are themselves middlemen. Amazon.com, held up as a model of Internet business success, is a middleman, and it is not clear that Amazon's business model will survive in the long run. Amazon's model is: We dictate your choices, we set the price, we offer you random opinion in lieu of actual reviews. This model is vulnerable to attack by someone who provides an Internet-wide book search, bulk buys and auctions, and links to respected reviewers. Such a service might emerge tomorrow, and it might not even be a commercial venture. Aggregator is the name for a type of service that could challenge Amazon and the portals. Things are changing indeed.

I'm not predicting Amazon's demise; I'm just saying that the service that it currently provides may not be something that a company can charge for in a few years.

Another issue is the development of barter communities on the Internet. Currently, 15 percent of the advertising bought and sold on the Internet is done via a shadowy barter economy that plays it coy about assigning value to the swapped ads. And if no value can be assigned to a transaction, there's no tax due, right?

I suspect we're just seeing the beginning of the trend. Trading software or time seems a plausible next step in the open-source movement. I don't know how far barter could go on the Net, but pushed far enough, it could threaten governments' ability to collect revenue and undermine the concept of a single, global economy. Or could it? I say the word "economics" and my eyes glaze over.

Not Saving Face

Are we looking at the beginning of the end of the dominant computer paradigms of the past two decades?

Larry Ellison is predicting the end of the desktop, Linus Torvalds the end of commercial mega-apps. Just wishful thinking? Is it mere hype to say that Torvalds, now working with Transmeta, is involved in unseating both Intel and Microsoft? Is there some hope we could be freed of the depressing weight of x86 and Windows/DOS compatibility? Shrug off the legacy albatrosses? Microsoft looks like it's losing the battle to make Windows CE matter, but does that matter?

As Ted Nelson once said, "All the looks and feels have not been seen and felt. I think it's hilarious that they want to standardize now." The pace of change in software in the age of the Internet is so fast that it's only to be expected that yesterday's tools are not going to suffice. If you're developing applications for the Web, chances are you're using higher level tools than you would use to develop desktop applications. Or than you would have used a couple of years ago. The Internet is placing turnaround-time demands on the software-development process that yesterday's tools can't live up to. More and more, developers are turning to scripting tools like Perl and Python for web development. Yesterday's paradigms are not going to suffice, either. I thought object-oriented programming was going to last a long time. Is it really on its way out, as some suggest?

RISC is Dead

Let's lay RISC to rest. RISC was a rebellion against the Establishment. Now that it no longer embodies any of the values that drove its creation, it's time to pronounce it dead. It has been argued that CISC has moved toward RISC and vice versa, so that neither is what it was and any important difference between the two exists only in the marketing literature of semiconductor companies. That's a nice, balanced, symmetrical story, but symmetry isn't going to get us anywhere. Let's lump everything called CISC and RISC together, call it CISC (it is complex), call it the Establishment, and set the stage for the next rebellion. Let's give it a name. I suggest PUNC. Like Punk Rock. I invite you to retrofit the acronym with a meaning.

DDJ

Jan00: Jim Button: Where Is He Now?

Jim Button: Where Is He Now?

Let's not forget the past altogether. Shareware was one of the surprises of the early personal computer industry. Who would have thought that you could make a living from software paid for on the honor system? Make it publicly available, invite people to make copies of it and give them to their friends, and base your income prospects on a little notice asking people to send you a few dollars if they found the program useful: How smart is that? Very smart, as Andrew Fluegelman and Jim Button, the authors of PC-Talk and PC-File, respectively, proved. Those two programmers invented shareware, although they called it "freeware" back then. I hope to write about Fluegelman soon, but this historical note is about the mysterious Mr. Button.

Button was working for IBM in the Seattle area when the IBM PC was released. IBM employees -- including Button -- proved to be a hot market for the PC. Button, an Apple II hacker in his spare time, wanted to get his colleagues off to a good start. So he ported a database program he'd written for the Apple II over to IBM Basic and shared it with his friends and coworkers.

The friends and coworkers in turn passed the program on to their friends and coworkers, and pretty soon, Button had a good use for his database program -- keeping track of its users. He was doing a lot of bug fixes and enhancements and it was getting to be a pain notifying everyone of every change. So he tried an experiment. He included in each new copy of the program a small message encouraging people to copy and pass the program on freely, but also asking that they send him $10 -- almost immediately changed to $25 -- if they wanted to be on his mailing list to be notified of bug fixes and enhancements. And he gave the program the name PC-File. Then he and his wife went on vacation.

When they got home, brown paper bags were strewn all over the basement floor. The housesitter had to haul the checks home from the post office in grocery bags. Their son spent the summer of 1982 catching up on this avalanche of mail. Within two years, Button was making ten times his IBM salary from shareware, and quit his day job. He never had reason to regret that decision.

At its peak in the late 1980s, Buttonware had 35 employees, a broad lineup of shareware products, and a $4.5 million gross annual income. In 1992, Jim Button sold the company and retired, quite comfortably. He can now be found fly fishing in the Pacific Northwest.

Only not under that name. I called him the mysterious Mr. Button because "Jim Button" did not exist. The name was a pseudonym that this programmer chose for marketing purposes. His real name is Jim Knopf (German for button).

I'm not promising, but I'm thinking about starting a collection of "Where Are They Now?" stories on my computer history book site at http://www.fireinthevalley.com/. If I do, there should be a nice little collection of them there by the time you read this.

-- M.S.

Terms of Service | Privacy Statement | Copyright © 2024 UBM Tech, All rights reserved.