Whatever it is, AvWeek says that the Obama administration is going to “stick with it.”
The fiscal 2010 NASA budget outline to be released by the Obama Administration Feb. 26 adds almost $700 million to the out-year figure proposed in the fiscal 2009 budget request submitted by former President Bush, and sticks with the goal of returning humans to the moon by 2020.
Well, the story doesn’t support the headline. What they’re sticking with is the goal, not the plan (which is a description of how the goal is to be achieved). It’s hard to know whether this is good news or bad. It depends on whether or not the “plan” (i.e., Constellation/ESAS) is going to be stuck with. We still have no information about the plan.
I understand your point. I have a quibble with your distinction. Plans consist of subgoals. Any goal can be considered a plan for some higher level goal, and any plan can be considered a goal for some lower level planning step. “Getting to the moon” is a goal, but it was also a plan for President Bush’s higher-level goal of redirecting the space program. Why quibble like this? Because it can be useful if you’re trying to understand what the author might have been thinking.
A goal is not a plan. A goal is never a plan. A goal is the intended outcome of a plan.
To say that a goal is a plan is to say that by deciding where you want to be at the end of your 3,000-mile drive you have done all the preparation necessary to get you there.
Let me know how that works out for you.
If you’re interested in this, I refer you to:
Roger C. Schank and Robert P. Abelson (1977). Scripts, Plans, Goals and Understanding: an Inquiry into Human Knowledge Structures. Schank and Abelson believed that there was a distinction between plans and goals, but when Schank’s students built AI programs inspired by the book, they dropped the distinction.
The bottom line is that when you try to build an AI planning system, you quickly realize that not only does every goal require a plan, but each step in the plan becomes sub-goal in itself. To drive across the country, you need a plan. One step in the plan will involve packing clothing. To pack clothing, you’ve got to open a drawer. If someone asks you “why are you opening the drawer ?” you’ll say “I need to pack clothes for my trip”. Packing clothes has become a goal ( a low level one) as well as part of your plan for your big trip. This is common sense, which is exactly what you’ve got to try to give an AI program.
If you really want to distinguish between goals and plans, you can say that goals are states of the world and plans are ways to change that state, but since one of the states of the world is what you are doing to change it, I bet you’ll find the distinction pretty useless, and more cumbersome than how we usually think about what we are doing.
Unless the piece is simply wrong, a commitment to return to the Moon by 2020 and more money for NASA than George W. Bush ever proposed strikes me as good news, if admittedly incomplete.
Although an Obama outright cancellation of NASA human spaceflight would be a win for future GOP political aspirations. Which is why Obama did a 180 degree reversal between November 2007 (give all the $$$ to education) and the Summer of 2008 when the Lori Garver influenced August position paper was released.
On the subject of reading tea leaves, I find it hard to imagine George Whitesides and Charles Miller going
“inside” if they thought Obama would stick with a brain-dead approach to the VSE.
Obama doesn’t take much interest in NASA, so he’s not likely to make any big decisions until he picks an administrator to conduct a thorough review and advise him on the pertinent issues.
That will naturally take a distant back seat to filling out the Cabinet, so it will be a while yet until we have concrete word on the administration’s priorities.
The new budget is neither here nor there in that respect, reflecting a general commitment to expand America’s scientific and technological endeavors.
Nevertheless, I remain hopeful. At bare minimum we will get the Moon program we were promised (insincerely) by the last administration, but the current president does not strike me as lacking ambition in matters of national pride and stature, nor lacking common sense in the more practical aspects.
> If you really want to distinguish between goals and plans, you can say that goals are states of the world
Wrong. Goals are DESIRED states.
> and plans are ways to change that state, but since one of the states of the world is what you are doing to change it
Shorter Bob – Nouns are verbs. That’s catchy, but wrong.
Citing failed AI as proof of anything is somewhat suspect. Moreover, even the cite doesn’t support Bob’s argument, but that’s his SOP.
I wonder if Bob is a professional journalist.
No.
Packing clothes is a goal if you had to make a plan for doing it. Otherwise it is a task.
Do you have to make a plan for packing your clothes? Does making that plan require prior planning?
Do you make a plan for inhaling, and then for exhaling?
Budget summary has been released
http://www.gpoaccess.gov/usbudget/fy10/pdf/fy10-newera.pdf
NASA on pages 103 & 104
Highlights, as I read them:
$18.7 billion for FY10
No mention of preferred launchers but the document says NASA “will stimulate” private sector development of crew and cargo capabilities to fulfill NASA requirements.
Orbiter retirement by the end of 2010 is re-affirmed.
Bob Says:
February 25th, 2009 at 7:04 pm
“………….Plans consist of subgoals. Any goal can be considered a plan for some higher level goal, and any plan can be considered a goal for some lower level planning step.”
Here goes Bob painting with broad brushes again. A project manager he is not — that appears quite certain.
Bob, a project plan consists of a stated goal. The project goal is outlined into a series of tasks which represent steps of actual work done to bring the project plan to completion. A series of meetings are held by a select group of stakeholders to decide upon the make up of all the work breakdown structures. In other words, the tasks in the project are put in chronological order and then arranged into summary groups of logical relationship. Some tasks are more important than other tasks in how they drive the project timeline. These tasks define the critical path to project completion. Not all tasks of a project plan lie on the critical path. Negative Lead times between tasks can even allow for some tasks to be run in parallel with eachother. There might be enough slack between summary tasks that some seemingly important task don’t lie on the critical path.
So, the project plan may state in its goal that we are to explore the Moon, Mars, and beyond. Well, the WBS committees sits down and decides if these things necessarily have to fall in that order, or can be run in parallel, or in fact may have specific components of the goal that might even fall outside of the scope of the current project. In which meetings would need to be held to redefine the originally stated goal. So far, I haven’t seen this type of discussion or debate held to outline specifically what order or which part of the VSE needs to be completed and when. If anything, the VSE that Pres. Bush originally outlined was rather open in its wording and intent.
The specific wording of these things are purely symantics as well. There are various conventions followed to project planning and plans and goals can often be thought of as negative and positive in electronics. Some follow the convention that positive flows into negative while the majority of of Benjamites agree that negative should in fact flow into positive. Ultimately, the device still spins around when the work flows.
We agree that going to the moon is a goal that requires a plan. So, what was Bush’s goal in giving his VSE speech? I’d say it was to give NASA a new sense of purpose, and talk of the “moon, mars, and beyond” were just a means to that end.
Bush had “a moon plan” which furthered his goal of revitalizing NASA, and Bush had “a moon goal” which was why the Exploration Systems Architecture Study came up with a more specific plan.
That was my most specific point, and it was intended to help Rand see a way to view the AvWeek article’s choice of words.
The cite I gave is an interesting introduction to a more general discussion of plans and goals. It needn’t prove my point – it is just an introduction into a discussion. The book, along with 1982’s “Dynamic Memory”, also by Schank, led to a 30 year history of AI research, which has indeed foundered on the problem of representing the sheer volume of everyday knowledge. During that 30 year history, researchers found it useful to see plans and goals as interchangeable: plans are just ways to achieve goals, but goals are usually just part of a bigger plan. If that way of thinking doesn’t work for you, don’t use it. These are just mental constructs, and arguing about it is as silly as the debate over whether pluto is a planet.
Although an Obama outright cancellation of NASA human spaceflight would be a win for future GOP political aspirations. Which is why Obama did a 180 degree reversal between November 2007 (give all the $$$ to education)
Obama never said he would cancel NASA human spaceflight outright or that he would “give all the $$$ to education.”
He said that he would postpone Orion by a few years, fly the Shuttle a few years longer, and give some of the savings to education. Whether you think that was a good plan or not, it was a far cry from what you state.
Were you talking about yourself when you predicted Bush Derangement Syndrome would be replaced by Obama Derangement Syndrome? 🙂
The budget released today reaffirms shuttle retirement by the end of FY10 and it affirms an intention to fund private sector crew and cargo capabilities for ISS.
That sounds like COTS.
And at $18.7 billion, Obama’s proposed NASA budget is larger than anything President Bush actually included in his budgets.
That’s an interesting line of thought, Bob. I’m vaguely reminded of the object-oriented programming mavens, with their contention that every imaginable programming task can be reduced to a collection of object methods. At the margin, the debate gets damn metaphysical.
I think an important distinction in the way people actually use “plan” and “goal,” however, is that a goal is irreplaceable, whereas the plan is not. That is, if the goal is getting to New York, and the plan is to drive, we use that terminology partly to express the fact that driving is not critical. If something came up, we could fly instead. But getting to New York is critical. If something came up, we would not drive to New Orleans instead, i.e. modify the destination instead of the means of getting there.
This is not irrelevant to the discussion. Part of the difficult we have with spaceflight is that we think, instinctively, it should work the same way. The goal is (say) getting to the Moon, or Mars, and the plan could be Apollo on steroids, or alt-space private ventures, or Orion, or international cooperation to build a space elevator, or whatever. Who cares? As long as we get there.
But actually, given that there’s not yet any compelling (= profitable) reason for people to go beyond orbit, and these things are driven (so far) more by political concerns than anything else, it seems much easier to build a consensus around a plan than the actual goal. That is, we can build a successful consensus around the Shuttle, or Ares, easier than we can build a consensus around what the Shuttle or Ares should do.
The budget released today reaffirms shuttle retirement by the end of FY10 and it affirms an intention to fund private sector crew and cargo capabilities for ISS.
In other words, a budget request submitted by an acting Administrator left over from the previous administration contains no radical changes. That’s not surprising.
That doesn’t mean the request couldn’t be modified by the new Administrator when he comes onboard. In fact, I expect it will. How he will change it is the real question.
And at $18.7 billion, Obama’s proposed NASA budget is larger than anything President Bush actually included in his budgets.
And Bush’s budget requests were larger than anything Clinton submitted. What’s your point?
Carl, it sounds like we are the same page – I agree with everything you said. Except, of course, goals do change, and part of being intelligent involves understanding how your goals fit into a larger framework of plans and goals so that you know when to change your goals. Otherwise, you suffer from tunnel vision. The guy who has the goal of getting to NY might say to his assistant “Getting to New York is critical – I don’t care how you get me there – just get me there – that’s the goal!” An typical assistant might interpret that as an order to investigate planes, trains, and automobiles, but a really smart assistant will say ” I know you need to talk to Jenkins in person, but it would be easier to catch Jenkins in Boston the night before.” And suddenly “get to NY”, which seemed like a critical goal became just a discarded plan. Again, this is just common sense.
And yes, of course understanding how our goals and plans fit into the bigger picture definitely applies to human activity in outer space. I remember my jaw dropping a bit the first time I read Paul Dietz convincingly debunk what had been an unquestioned assumption of mine for a long time: that space colonies would be helpful in the event of a worldwide catastrophe. Dietz pointed out that for many if not all of the catastrophes cited by space enthusiasts, Earth would remain a better place for a shelter than any other place in our solar system. Suddenly the goal of “protect Humanity” didn’t seem like it required the plan “colonize space” anymore, but of course, like all of you, I didn’t want to abandon the plan, because, in my tunnel-visioned way, “colonize space” was a goal as well as a plan.
I think the distinction between plans and goals is that goals are desired states. You’d like to at a location 3,000 miles away on a particular date. A plan is a structure on top of possible goals. If there are no decisions to be made, it’d be a sequence of goals (say, in our travel example, a list of places between here and where you want to be). Work and tasks is the means by which we transition from one state to another. To get to the next rest stop, we need to drive 100 miles.
What Bob’s example indicates is a peculiarity of programming and fast computers. Namely, it is possible to an extent to ignore both the work and planning aspects because the computer often can do them. If you give a goal to the computer and it achieves that goal faster than you can blink, then there is no reason to care at all about planning or the work involved.
Erm, I mean if there are no decisions and the outcome is completely deterministic. Obviously, you could otherwise have a flat tire, catch the flu, or get in an accident. These would change your overall plan.
Well, you’re talking about heirarchies of goals, Bob. The executive in your example would, if he is smart, say my goal is to talk to Jenkins. The plan is to go to New York to do so, but any other way of doing so works as well.
We can, as you pointed out, advance up and down the heirarchy, and call various things “goals” and “plans” depending on our focus at the moment.
I’m suggesting that this discussion is more than just metaphysical, however. We imagine that most of our painful public discussion revolves around the fact that we all agree on large “goals” (peace ‘n’ justice for everyone, prosperity, all children to be above average, whatever) and we just need to hammer out a consensus plan for reaching them.
I’m suggesting reality is far less pleasant. In fact, we can often only agree on methods and plans or short-term intermediate goals, and it is actually very often the high-level goals on which we find profound disagreement.
The disparate elements in the North all agreed, in 1862, that the war against the South needed to be won. But for what ultimate purpose? There the abolitionists and nationalists and so forth parted company, as the sad history of Reconstruction showed. We all agreed in 1942 on the goal of beating the Nazis — but for what larger purpose? After the war, the coalition dissolved in bickering.
Even now, we all agree the economy should be growing more. But where should the new wealth go? Into vast new government-supervised programs? Or into the pockets of “the rich” to be spent on frivolity like new houses and hiring new workers for their private business schemes?
Same thing with space. Many people agree we should be “in” space. But what for? The Planetary Society Mars freaks and the alt-space enthusiasts might kill each other if they had to agree on the goal.
I’m not saying this kind of thing doesn’t have some good purpose, often enough. At least by agreeing on the short-term goal, we keep moving forward, and sometimes events in the future prove out one vision or another. Moving forward is good. But it would be wise to undertand what is really going on, and not delude ourselves that we all agree on the goals towards which we work. I mean, except in the very broadest sense, that we all agree we don’t want our lives to be worthless and unrewarded.
Same thing with space. Many people agree we should be “in” space. But what for? The Planetary Society Mars freaks and the alt-space enthusiasts might kill each other if they had to agree on the goal.
Our problem is that the places we can cheaply get to (Moon, Phobos, Deimos, NEOs, Mars) are damned inhospitable and seemingly not more resource rich than Earth. Near-space is a dull place. It think the inhospitable part can be solved by automated ISRU (it was common during colonization to leave livestock in an island, let it multiply, so people who next visited the island would have something to eat).
People who say it is more expensive to get for e.g. platinum from a resource rich asteroid than mining it on Earth are IMO thinking it wrong. The cost is in getting out of the Earth gravity well. Once you stop needing that you can just rail launch payloads to Earth or whatever. Heck, comets travel the whole solar system and are just a block of ice.
As for the Mars or Moon or whatever debates I have this to say: before the Europeans colonized the New World they colonized places like the Azores, Madeira, Canary Islands, Cape Verde. These islands were deserted, and many are damned inhospitable too (try reading about Cape Verde to see what I mean). Madeira was made profitable by the Portuguese by growing sugar cane in it. Still, even the barren islands were colonized and people spilled blood over them. For a reason: they are in strategic spots of the globe. Look at a map and you will see why. Funny thing is most of the profit in these islands today is from, guess what, tourism! Regarding space, the Moon is half-way to Mars in delta-v. All, and I do mean all, the viable stepping stones should (and IMO will) be covered.
Goals and plans are distinct and separate things and I don’t understand the confusion.
A goal is a target; WHAT you wish to accomplish.
A plan is a method; HOW you hope to accomplish it.
This distinction is independent of good or bad, how many; their hierarchy or their stability.
Kept simple, there is no confusion.
> The book, along with 1982’s “Dynamic Memory”, also by Schank, led to a 30 year history of AI research
Umm, 1982+30 = 2012.
> During that 30 year history, researchers found it useful to see plans and goals as interchangeable
Since those researchers haven’t produced anything significant, it’s unclear why we should put much weight on what they found useful.
Andy,
1977+30=2007. 1982’s research was “along with” 1977’s. One might start to suspect you are just trying to be disagreeable for the sake of it.
The assertion is that what someone thinks of as a goal migth be also be a plan for some higher level goal. The usefulness of this assertion is to assist in interpreting AvWeek, and also to stimulate Carl and Karl’s interesting comments on the subject.
Ken Anthony and your own formulation that goals are targets or desired states, and that plans are methods, tasks, etc, isn’t at odds with what I’m saying. I’m just pointing out that the same idea (eg “Go to the moon”) can be a plan or a goal, depending on where it fits in a given plan-goal hierarchy.
AI’s contribution to today’s computational environments is a big subject which is completely off-topic, but it is worth noting in this forum that it can be useful to learn from past approaches (which sometimes become future approaches!) rather than just judge final products. There are countless examples of this in aerospace.
Until they prove to be untrustworthy, I will take my Obama Overlords at their word, and assume that they are going to do what works. If that means that SpaceX is going to take over shuttle flights, moon flights, and even mars flights eventually, so much the better. If SpaceX does not work, then I hope NASA eventually figures something out that does work.