Such architectural endeavours aren’t done by volunteers in their free time, nor in a design process that is open to the public. The kind of stake various participants have in the outcome versus their say in the process (and how it manifests) is also very different.

I don’t think you can draw any sensible direct parallels between the processes, only maybe some metaphorical ones, and even those will be stilted.

Makeshifts last the longest.

When an architect is designing a new, novel building does he endure lots of people saying "It'll never get built!"

Yes, and no. Usually somebody else has decided to build a building and the architect's new, novel design has won a bidding process to be that building. Architects may have a few sketches of new, novel buildings on which to base the designs they're going to propose, but architects do not usually go around with a building design looking for somebody to fund its construction. Designing a building is a lot of work; sketches (such as Frank Lloyd Wright's mile high building) are pretty pictures. Designs can run to tens of thousands of pages and take many tens of thousands of hours of labor to produce.

After the architect's proposal wins the bidding process, the real work starts. Architects do not design buildings by themselves! Civil engineers design the structure, mechanical engineers the plumbing and air conditioning systems, fire protection engineers the fire protection systems, electrical engineers do the electrical and lighting design. Every item in that building has to be designed or specified and ordered, with serious penalties for changing one's mind after the order is made. Regulatory agencies have to be satisfied. Builders have to be consulted to make sure that something is possible, or if doing something will require a special, unique, piece of construction equipment that is booked for the next 4 years. (a bridge at the train station in New Haven required a special, near-unique crane to install. Three cranes in the world had the combination of reach and capacity required. The alternative would have required shutting down Amtrak's Northeast Corridor and west-bound commuter rail service out of New Haven for several weeks).

I think one very real difference is that people have been building physical artifacts -- like buildings -- for far longer than we've been building software. Another difference is that the idea of a "dictator programmer" may be alive, but the heroes in engineering are famous for managing projects, not just designing them. Checkout a biography of George Washington Roebling, Kelly Johnson, Isambard Brunel, or even Thomas Edison.

Well, there are things that take way too long and come in way over budget, such as Boston's Big Dig or the Sydney Opera House. Often you hear the complaints because they are justified.

Perl 6 generated quite a bit of excitemenat at first and had a tremendous amount of progress through the RFC process. Now it's over five years later. I don't think it's unreasonable for people to wonder what's going on. We might not know the answer, but we also know that things took a lot longer than everyone expected.

The problem comes in when people try to ignore valid questions about the timeline. We can't just pretend that everything is going exactly as we planned because it didn't happen like that.

Sure things took longer than expected, but I get the impression that the general populace view this as "unacceptable" or "bad" or a sign that perl6 is doomed or something. Haven't people heard that old saying about how battle plans never survive first contact with the enemy?

The way I look at it is this: we made a plan and while executing that plan we needed to make a slight course adjustment and then again and again as more of the reality of the process unfolded and continues to unfold. This is a good thing as long as we don't lose sight of the goal (I don't think of the people doing the work have lost sight of the goal). What boggles my mind though is people who think we should have already arrived at our goal of perl6. It's still being designed! Even completely specified, fully funded projects take too long and go overbudget and they expect a minimally funded and partially specified project to be completed based on initial estimates?

I'm starting to feel a rant rise in me now, so I think I'll stop :-) But I get a general feel that Perl6's detractors all suffer from "can't you just ..." syndrome. See Andy Lester's weblog.

duff

This isn't a a battle plan, and there is no enemy. :) I bet a lot of people haven't heard that metaphor either (well, at least it's a metaphor in this area of my life, whereas in my other life it's literal :). Even then, it's been a long time since initial contact and we still don't know what's going on.

Sure, things change and they don't turn out as people expect. The problem comes in when you downplay or ignore the valid observations of people who wonder what's going on. When Perl 6 started, we thought we'd have it in a couple of years. Maybe three. It's been a lot longer than that, and many people have joined, and left, the project.

This isn't a "Can't you just..." problem, because I don't see anyone saying that. I don't see anyone suggesting magic bullets, cure-alls, or anything of the sort. I do hear a lot of people simply asking "When?"

A lot of the problem is the unwillingness for some people to admit that Perl 6 is taking much longer than anyone expected. They try to invalidate anyone who points that out. It makes it look like they live in as much of a fantasyland as the project managers they like to make fun of.

If you want to complain about minimal funding, that's fine, but then you have to ask yourself if the open source volunteerism model can work. You'll have to answer "Of course it works" because many other big projects use it. Why isn't it working here? Partial specs, poor management, worker churn, etc are things that those involved should acknowledge though. Reasonable people see the problems and know that they're there. Comments like yours seem to want to pretend that they don't exist. That just irritates people more.

If we're still designing Perl 6, we're in pretty bad shape. Without Pugs, I think there would be no hope.

Reasonable people can accept quite a bit when they know you're being honest with them (and yourself).

--
brian d foy <brian@stonehenge.com>
Subscribe to The Perl Review

As [id://Aristotle] noted, Perl 6 is being created by volunteers. When a building is contemplated and a decision is made to build it, people have invested their money into it. Someone had to convince them it could and would be done. Few people are going to invest money in a project that is doubtful. So I suspect that not many engineers or architects actually run into this problem at this stage. They might encounter it before anyone has decided to use the plans. Some famous architects, such as Frank Lloyd Wright, were told their plans were impossible.

As for Perl 6, I couldn't say for sure that it will be done, but I'm betting that it will. One advantage of Open Source software is that there aren't deadlines to get the product out the door to beat the competition. I have faith that these volunteers are working to create a fine product, and aren't going to be pushed into releasing something less than that.

when it is being built does he hear crys of "It's taking too long! How come it's not finished yet?"

Well, yes, even if the project is on-schedule and under budget. And forget home remodeling projects. In general, they're so predictably over budget that one should always add 20% in cost and a few weeks in time to the builder's estimate.

I was watching the same show, and I got the impression of them saying " It's costing way more than we planned for". :-)

They had all sorts of cool innovativel features in their original designs, but then they had to be scaled back because it went over original estimated budget.