What you need to know about the new Tcl/Tk Consortium
Sun-sponsored group showing signs of promise after rocky start
On October 29th of this year Sun's Tcl Evangelist Brent Welch announced "we have launched the Tcl/Tk Consortium that will promote Tcl/Tk and its extensions." The initial reaction was less positive than expected. Skeptics comfortable with the laid-back habits of the existing Tcl community wondered what they'd be losing. Others pointedly speculated about where the membership fees-- starting at $500 annually--would end up. Chris Cox, a developer in Dallas, put it most harshly: "So... either the Consortium is a neat way to make money for specific people... or it's a way to pay Sun to listen to you who are wealthy enough to buy your way into the exclusive club."
The dust has settled since then. As December begins, Welch says the Consortium is doing "fine." The critics have been mollified. As concrete action starts to unfold, the real work begins.
Tckl/Tk Consortium takes root
Tcl probably holds a smaller mind share than it deserves. Or so went the recurring lament at last summer's Tcl Workshop in Boston. Brian Kernighan, famous for his definitive authorship of The C Programming Language, labeled Tcl "the best-kept secret" in the software industry. By the close of the conference, the consensus was that companies and individuals with a stake in the scripting language should set up a free-standing promotional organization.
In early August, Sun employees and other insiders formed a temporary board of directors. A few days later, they constructed a Web site for the Consortium. Public activity quieted, until board vice president Welch announced ten weeks later that Sun had enrolled as a founding sponsor.
More recently, director pro tem Peter Salus has guided the Consortium through its incorporation, and initiated paperwork to have it recognized as a tax-exempt organization.
Launch of the Consortium
The Consortium took flight with momentum from the Boston Workshop. Sun's enrollment was a potent endorsement. Its board members boast impeccable industry standing and academic credentials. What went wrong in November? Why did so much hostility erupt?
The controversy was predictable and even inevitable. Most of it played out in the comp.lang.tcl Usenet newsgroup. Several dozen follow-ups appeared during the first three weeks after Welch's announcement. The frustration and hurt feelings they evinced were sobering for the normally placid Tcl community.
On the other hand, that level of emotion wouldn't be noticed on a hot morning in comp.sys.mac.advocacy. More to the point, the antagonists have returned to their collegial habits. Cox has recently written that the Consortium "probably" will benefit non-members, if only indirectly. Donal K. Fellows at first argued that the Consortium neglected the interests of overseas and academic organization. However, he writes now, "my primary concern was allayed" when Tcl's founder, John Ousterhout, explained that he will "continue to listen to everyone in the Tcl community as we plan the evolution of Tcl and Tk."
Repair of personal suspicions and affiliations mirrors the tensions characteristic for a consortium. Consortia are like their organizational cousins, joint ventures, in having a troubled history. They're notorious for looking better on paper than in practice.
The Tcl/Tk Consortium, in particular, faces a textbook management problem: How does an airline, for example, keep first class and "cattle car" customers happy at the same time? The challenge for the Consortium is to maintain the healthy life of the existing Tcl community while addressing the concerns of commercially-oriented users and prospective users. Linux faced the same conundrum. It began with a collection of individuals united by their love of a free software package. These individuals saw a band of companies moving in on their territory, and they feared they'd lose out.
Salus in particular has walked this tightrope before. He's already had success with the work he's done for Usenix, the Sun User Group, and the Free Software Foundation. If the flurry of contention in November is all it took to settle its issues, the Tcl/Tk Consortium should be a success.
What's the Consortium doing now?
After its rocky start, the Consortium is now drawing a clearer picture of where it's headed.
The Web pages themselves remain skeletal, with a "low-prestige" URL subordinate to a Dallas-area consultancy (a sign of how compact the existing Tcl world is: The Web host for the Consortium is coincidentally one former critic Cox uses). Curiously, neither Sun nor Lucent, the other top-level charter subscriber, have publicized their memberships at all. Behind the scenes, though, Board President Michael McLennan has worked out milestones for the next several months:
In the future, McLennan foresees salaried support staff, "a few magazine ads," snowballing membership growth, and even overseas expansion.
McLennan emphasizes the Consortium will be a leader in such work as standards development, not a steamroller. "We'll involve the entire Tcl/Tk community when we draft the actual standards," he pledges. He's aware of Tcl's diverse constituencies, and promises to explicitly target things like user groups to benefit individuals and small companies. On the high end, he has plans to strengthen the "economic infrastructure" of the engineering community. The Consortium aims to build customer/supplier relationships through a system of discount coupons for Tcl products and services.
What's in it for you?
On its current trajectory, the Tcl/Tk Consortium will be, at least, a modest resource for working developers. Publication of the CD-ROM, for example, will go a long way toward easing the evaluation and acceptance of Tcl where it's now being considered. Several groups had tried to publish reliable binaries in the past; always without success.
Tcl developers, as well as those who just want to stay on top of the technology, will want to keep an eye on the Consortium's home page to track industry acceptance. The Web pages will give clear evidence of how well the Consortium's achievements match engineers' needs.
One of the costs of using Tcl has always been justifying it to the unconverted. Long-time users will recognize a few homely indexes of this resistance:
So how will the Consortium measure its accomplishments? Among the targets McLennan mentioned to SunWorld are:
About the author
Cameron Laird and Kathryn Soraiz manage their own software consultancy, Network Engineered Solutions, from just outside Houston, Texas. Reach Cameron at email@example.com.
If you have technical problems with this magazine, contact firstname.lastname@example.org