The Whiteboard

►

2:14 PM

they just can't stop using Programmers as a toiled bowl!

This isn't a programming question and as such doesn't really belong here. Try programmers.stackexchange.com. — glts 18 hours ago

@glts quoting comment of Programmers moderator from here (10K only link to deleted question), "Please do not recommend sites, especially those that you aren't an active participant in. This question is not appropriate for Programmers and has not only been cross-posted, but closed there." — gnat 17 secs ago

Q: How to survive the wrong job and bad management

@jozefg Fair point, const != seq, I suppose that's the difference between reader and K

@gnat A bowl that has been worked on for a long time? I think that's an appropriate treatment...

0

I'm a long time googler and have been here many times, but this is my first post. I was inspired to sign up and post after I saw this post on Ars Technica. I'm struggling with my current job. I feel I was promised one thing but delivered another. I'm a developer with 7 years of ERP, SQL and ...

Good lord, 18 months it takes him to realize he was expected to be a BA? What has he been doing all this time?

2:44 PM

Poor guy. Sounds like he got the bait and switch so damn common in this industry.

Jeroen Vannevel

I'm rather curious as to what he could have possibly spent his days with

user55340

Random thought - chess game via github. Branching for different lines of play.

Q: Allow diamond moderators to comment on deleted or locked items

3:03 PM

wonder why this question was deleted - Is programming experience/education important for QA developers/engineers? (10K-only link) Neither comments that are there, nor revisions history (10K-only) give an idea of what might cause this (guess there were flags). Not that I am challenging the decision, merely trying to understand the reasoning that led to it.

If anything, it would be helpful to see mod comment explaining this for 10K-readers

109

Moderators should be able to comment on things after normal users are unable to, such as when a post is deleted or locked. As a moderator on Programmers, I try to leave a comment (if necessary) before taking the appropriate action on the post. However, in the time it takes to write a comment (whi...

feature-request status-completed comments moderator-tools moderators

> Educational aspect of deleted content has been specifically stressed in the rationale for a recently introduced change: Turbocharging the Roomba: solutions for premature deletion...

@MichaelT are you planning on taking all possible paths?

user55340

@Ampt See, thats the fun part of the thought experiment.

user55340

Or the git one... would it be feasible to use Git for this? How would one play a correspondence game via git? Are other games also suited well?

user55340

3:18 PM

There's also the realization on move 20 that you goofed at move 15... so branch from that commit...

3:34 PM

so basically play out all the possibilities of the game by hand

interesting

user55340

The comments on a commit may also be a useful resource.

@MichaelT That's where I was thinking you would put the move

Unfortunately actually the set of possible moves in chess is an open set

So you could never do the full combination

well you could find those sets that are unending and call those draws

because that's how they would be handled in the real world

user55340

But there are the interesting lines of the game... and historic games.

Q: examining history of deleted file

does SVN keep old versions of deleted files?

user55340

3:40 PM

90

If I delete a file in Subversion, how can I look at it's history and contents? If I try to do svn cat or svn log on a nonexistent file, it complains that the file doesn't exist. Also, if I wanted to resurrect the file, should I just svn add it back? (I asked specifically about Subversion, but I...

svn version-control bazaar

@Ampt the single set of combinationrs of moves is open I'm saying; therefore it's not a matter of draws or predicting, it's a matter of the fact that one possible combination of chess moves is that both players move their knight back and forth endlesslessly, but you can't know if it's a draw or not until you infinitely try it because maybe after the 5 millionth time moving knight back and forth they start to play and one of them wins so it's not a draw.

or maybe they never do that

it's undecidable

so you set an upper limit on the moves

and say anything beyond that is a draw

because no human would play the same move 5 million moves in a row

@Ampt Now that is possible. However that upper limit would have to be high, and the segment of actually completed games in the whole space you set out would be miniscule compared to the size of the available combinations set

@MichaelT my question is because the HW on our firmware changed so that I no longer have any outputs but I'm pretty happy with the file that runs the outputs

so it seems like a waste to just blow it away and lose all of its history

user55340

You haven't lost its history. Only if you delete it from svn itself is it gone. It is often more damaging to keep stuff around because of its legacy.

3:50 PM

yeah

I'm alright with deleting it, so long as it's not like completely gone forever

may have use in the future

4:02 PM

man it must have been a slow weekend, I've still got stars on like half of the messages on the right

user55340

Well, it was a holiday weekend... and chat is rather slow on weekends in general.

user55340

If you look at the room info, you can see the work week pattern to it - chat.stackexchange.com/rooms/info/21/the-whiteboard

oh for sure lol.

always more busy during Central time afternoons

just extra slow this weekend

4:24 PM

I would propose you should keep one simple rule in mind when asking a question: If you need to instruct people on how to answer, it's likely the core of the question is lacking the qualitative attributes that would attract good answers; and no amount of instructions on answering can really correct the fact that you're asking a question that fails to meet the quality standards of the site. — Jimmy Hoffa 16 secs ago

user55340

He really needs to read peopleware. The things it addresses and the answers it gives haven't changed.

Q: Good subjective question on hold

who does?

user55340

-4

Do programmers need quiet? Do they get it? Re: "subjective Stack is better than no Stack" In response to my question being placed on hold I would like to add: My question exhibits many of the relevant qualities of a good subjective question , including: A, B, D, E ... "Some subjective question...

discussion subjective

user55340

5:15 PM

o/` Welcome to the GitHub California...
"Relax, " said the octocat,
"We are programmed to receive.
You can check-out any branch you like,
But you can never leave! " o/`

Thomas Owens

@gnat It was deleted by request. I'll leave a comment on the post explaining that as well.

user55340

(I kept trying to get a 'version' rhyme in there... they were too forced... we are programed to version -- But never conversion? subversion? incursion? inversion? hmm... burden? assertion? )

Brant

I started working my way through Learn You a Haskell for Great Good while on vacation last week. I made it most of the way through the chapter on higher-order functions before my head started spinning.

I can't even fathom how something like I/O could even happen in Haskell, based on what I've seen so far.

Of course, I have quite a long way to go...

well I think your first mistake was assuming haskell can do anything useful like I/O

@MichaelT You can check out any branch you like, and you can always merge?

personally I find merging on SVN to be one of the most pain inducing things a VCS can do

user55340

5:31 PM

Hmm... a few good rhymes for merge...

I can see it now; GIT Song goes viral

user55340

o/` Welcome to the GitHub California...
"Relax, " said the octocat,
"We are programmed to emerge.
You can check-out any branch you like,
And you can always merge." o/`

programmed to diverge?

user55340

programed by an urge?

svn is a scourge?

user55340

5:37 PM

Could go to a near rhyme... source forge

@MichaelT ..last thing I remember, was commenting on the core, had to find the reversion to the phase I was before

ermahgerd serce ferge

@JimmyHoffa Don Henley himself would be proud

@MichaelT this is awesome

user55340

How they code in the cubicles, sweet fluorescent pale.
Some branch to remember, some fork to forget

user55340

(hmm... need a rhyme there instead...)

user55340

5:43 PM

Ohh! I got it...

user55340

How they code in the cubicles, piles of diskette.
Some branch to remember, some fork to forget

user55340

(one could also do octet, null set, reset, not yet)

eating a baguette?

hitting the reset?

defining the offset?

How they code in the cubicles, parsing the closed set.
Some branch to remember, some changset to forget

@Ampt do you guys actually work? :P

@enderland what could possibly be more important than adapting a classic song to a GIT metaphor

user55340

5:48 PM

@enderland He's trying to avoid testing how not to drop a snow plow on people.

@MichaelT why would you avoid that! that sounds wonderful

well its a long story but to keep it short theres this guy I know who just needs a plow dropped on him.

user55340

@enderland It depends heavily on if you live where it snows.

@MichaelT dropping snow plows on people though? you can do that even in hawaii :)

@MichaelT So I called up the director, "please pay my overtime", he said "we haven't counted that here since unix epoch tiiiime"

3

user55340

5:50 PM

Aug 29 at 18:08, by Ampt

I write the code that keeps the 6 axle snow truck from dropping it's side plow on top of your car when you drive by :D

@MichaelT oh that's a bit less awesome to have a plow dropping on people through, I suppose

jozefg

6:02 PM

@JimmyHoffa Also that the reader monad has >>= to allow sequencing, the applicative version of Reader is somewhat closer to your intuition

It boils down to, "Operate on stuff inside a lambda, without being syntactically in the same lambda"

@jozefg That actually makes a lot of sense that the applicative would be more like const than seq; functors inherently avoid the effectfulness monads give you

user55340

@JimmyHoffa "we aren't counting that here 'till it flips its sign"

@MichaelT unix epoch time has the nice coincidence that it means the same thing as the actual lyric

oooh I didn't even notice that

"We haven't had that spirit here since 1969" - Original lyric

user55340

But if you wait until the overtime accumulates to the point it flips its sign bit, you have to pay them.

user55340

6:13 PM

Oh, btw, new codeless code out. Sweet Tooth

I understand the correction - force them to use what they think is great, but how would you put that into context of the monk with the candy? What should his master have done? Asked him to eat the entire jar as Banzen did to the nun?

user55340

The monk with the candy kept snacking on neat ideas slowly without realizing it was bad for you. Lots of 'neat little ideas' off in classes isn't a good thing and causes headaches when its time to refactor. It is necessary to sit down and realize that the 'neat idea' just isn't that good.

user55340

The other thing to realize there is that sometimes you do need some 'candy' in your coding to have fun, lest you grow tired of it and unproductive. These two things must be balanced.

ah I see. Maybe I missed that due to my lack of experience with SQL

user55340

> If a database object were suitably annotated, I could develop a framework which would acquire the form input, validate it, convert it to SQL, query the table, and present the search results to the user in tabular format... all requiring the most minimal effort on the part of the framework’s consumer!

6:20 PM

the top half was mainly nonsense to me, I just gathered that he was doing something unique and different

user55340

He's describing a single table solution. Which works for simple problems.

@JimmyHoffa Huh? Nested parenthesis can be done without backtracking using an LR parser, for example. Just not a state machine (like a pure regex). en.wikipedia.org/wiki/Parsing_expression_grammar

user55340

@psr Its in the context of working with parens in a calculator.

@MichaelT I don't know parsers enough, he's probably correct

user55340

5 * ( 4 + 3) - 2 -- to handle this properly, you need to stash the 'at here' and process the contents of the parens.

6:23 PM

@psr Though I'm not completely convinced, you can do it in a single pass to parse and execute without generating an intermediary structure like an AST that needs a pass?

@psr I know how to generate a structure with a parser with a single pass that appropriately denotes nested parens with precedence, but I can't imagine how to execute as I go without backtracking

@MichaelT I have at least one product in the field using a LR parser to parse nested parenthesis, more or less like a calculator. So, one, I'm pretty sure, and , two, you would think someone would have mentioned something by now.

user55340

5, got it. Times ok, parens, oops... push the 5 and times onto the stack and start processing again, four plus three, thats 7, close parens, pop the stack and get the operation, pop it and get the value, we've got a 5 * 7...

I could even see how to generate many intermediary structures that get collapsed to evaluations as I go, but parse+evaluate in a single pass without any intermediary structures and no backtracking on a nested parenthetical calculation I cannot imagine

@MichaelT Exactly. The stack keeps track of that for you. PEG parsers could potentially use infinite memory though.

user55340

The real part was the poster who we were discussion was trying to do a simple string scan to process user entered polynomials to get the coefficients.

6:27 PM

@psr So if you have done this help me out; what am I missing that allows you to execute it in place without an intermediary structure (which is almost like backtracking only more intuitively might be pictured as forward tracking because it's simply a reordering of the operations to fit the LR parser you're working with)

user55340

> In my app the user will be inputting a mathematical expression as a string, such as n*3/7+9-5, and I need to be able to get the polynomials from that string, like 4, 3/7, 0, 0, 0, 0 is there a way to do this or a library that will do this in javascript? I am using node, so preferably something that is node compatible.

@JimmyHoffa MichaelT has a brief example. But you have a stack (or something similar), so you don't evaluate immediately but you do as soon as the sub-expression is done. It isn't exactly a backtrack because you don't re-parse the string. You do have sort of like an AST but just on a stack, not a tree.

how would you parse and evaluate in place with a simple accumulator that didn't need a tertiary pass to evaluate the following expression: 3*(4*(5*(4+(4-2))))

@psr My whole point was you can't do in-place evaluation of it, sorry I wasn't clear I guess

user55340

3, times, ok... (, push '3, *' on stack. new frame. 4 *, ok, (, push that on stack....

The guy wanted the value with "a single pass" I was thinking of that as in-place evaluation, single pass parser could easily generate an AST of some sort, but then you need a secondary pass that evaluates the AST (or whatever intermediary structure)

6:30 PM

@JimmyHoffa No, not in place. But less than another pass. I suppose it may effectively be fairly similar to backtracking with memoizing.

@psr Yeah that's like i said, using multiple intermediary structures, evaluating them as enough terms are present; almost like currying

@JimmyHoffa LR parsers are usually considered single pass even with the stack. But it's a matter of terminology, they are less single pass than a pure state machine.

Alright, that makes sense. I always assume a parser should generate an intermediary structure; only in the case of an outright interpreter do you care to evaluate in-line, 99% of parsing tasks are going to be totally fine using intermediary structures; stacks or AST or what have you

user55340

Btw, would it be a 'good', 'not bad', 'meh', or 'bad' idea to have The Codeless Code RSS feed here like the blog posts?

I think it's quite apt to refer to all of that as "forward tracking" because backtracking is going backwards in the stream, where as all of that is like consuming and appending to the end of the stream as a matter of term reordering/structure normalizing

@MichaelT I would opt for XKCD before The Codeless Code, but it's a Meh from me. You could poll meta so you don't annoy anybody with particular reservations (though meta doesn't really exist for chat issues)

Not sure if that is an appropriate meta poll..

user55340

6:37 PM

The argument for the Codeless Code over XKCD would be that CC is always topical to the white board and P.SE while XKCD not always.

we need a meta.chat.stackexchange

then we need a meta.meta.stackexchange to keep track of the metas

user55340

Since MSO has its own chat, would that mean we need a C.M.M.SE?

user20683

afternoon

Oh hey there. Do you have 2 accounts now? Weston and World engineer?

or do you get to choose

@WorldEngineer do any of you folks work! :P

user20683

6:49 PM

@Ampt neither

user20683

@enderland yes, part time whilst I look for full time

@WorldEngineer ahhhh. you guys are just alwyas here lol :)

user20683

@enderland Not always, this place dies on the weekends for the most part

user55340

Being here does not mean thinking. And sometimes it means trying to avoid thinking.

user20683

being in here usually means I'm either helping someone or learning something

6:51 PM

ahhhh. I'm probably going to do the same

Welcome to the github California, such an agile place, such an agile race. We're forking it up at the github California. Such a nice surprise, such a nice surprise, when you're hard drive dies. reflection and annealing, you give your code a price. And she said we are all just developers here, of our own device.

@WorldEngineer then how does that work? does it just choose a name at random?

@Ampt you can have different names on different SE sites - chat uses your "main" one iirc

@enderland he'll sometimes show up as Weston.h though, as opposed to WorldEngineer

@Ampt oooh. well.. who knows! ask a mod ;)

6:56 PM

I did :P

@Ampt hence ";)"

I feel like I should follow this project I'm working on here for the next 6 months or so ha

user55340

I believe its the one associated with your parent user (if you go to your chat.SE info page you can change that to a different 'parent user')

I think this will become a wonderful place to ask questions, or, cry...

user55340

And you can have different names on different sites.

Probably cry @enderland. That's my initial reaction

user55340

6:57 PM

For example...

user55340

Will, United States

67.3k 21 153 243

c# .net wpf asp.net asp.net-mvc workflow-foundation-4 mvvm

discussion feature-request support stackoverflow status-completed moderation flags

@Ampt I'm GONNA BE OPTIMISTIC

user55340

Won't, United States

25.7k 10 72 187

user55340

Same guy... different names.

user55340

And chat is associated with his MSO version:

user55340

6:58 PM

chat.stackexchange.com/users/466/wont

@enderland good luck! Nah, it's not all bad here

Aug 30 at 13:30, by Ampt

and just don't listen to @JimmyHoffa or @MichaelT too much and you wont be so depressed

@Ampt i thin it's funny that is a room owner starred comment :D

user55340

@enderland Could have been done by a mod... or the other room owner.

user55340

And I don't dispute that statement.

I think I'm going to get into "having the moon asked of me" territory

c'est la vie

user55340

7:02 PM

@enderland Just don't pull down your pants.

@MichaelT .... too late :\

Q: What javascript framework will become alot more popular the next 3 years ? (2013 - 2016)

If you find someone who claims to have the answer @punkouter, I have this bridge I want to sell you. — Ampt 10 mins ago

in response to

-5

As a .NET developer who just recently is using more and more JavaScript I played with knockout and recently learning angularJS. I like angularJS because it seems to pack everything in oneplace rather than chain together multiple javascript libraries to get something done. Anyone have any predic...

javascript

I'm going to have such a screwed up voting ratio here if you guys consistently post bad questions into chat lol

you and @gnat will get along fine hahaha

user55340

@enderland Oh... I bet you don't want to look at the cliff...

7:15 PM

@Ampt we enjoy The Workplace pretty well already - though he is a lurker

@MichaelT "the cliff?"

user55340

Questions that are a vote or two away from deletion.

basically a great place to just downvote them off the edge

300 style

@MichaelT oh. bwahahaha

user55340

data.stackexchange.com/stackoverflow/query/133171/…

2

7:23 PM

would a quesiton like, "how to effectively have project meetings when you are the sole developer?" be good here - my manager and team member(s) have asked ME to define what meetings we need over the next few months (which is totally wtf awesome)

user55340

7:33 PM

@enderland Possibly, though realize that you'd get it from the programmer side. Given a more managerial side, you might want to check out PM.SE

user55340

@MichaelT no I want the programmer side - my boss is actually asking me that from the"what works best for you" perspective

user55340

I'd still glance at pm.stackexchange.com/questions/tagged/meetings

7:49 PM

@MichaelT looks like most of those tagged "meetings" there are more "how to facilitate meetings"

user55340

Well, if you do ask it here (and I wouldn't toss an off topic close on it)... make sure you include what you want to get out of the 'team (of 1) meeting' --- I think you're likely to end up with essentially regular status reports as emails, since thats what most meetings are.

@Shauna that may be, but I neither participate in workplace, nor know anything about what's on topic there so I prefer to err on the side of closure rather than suggesting people move their questions when for all I know it might be terribly off-topic there and I'd just be suggesting garbage be migrated there which I'd rather avoid doing. — Jimmy Hoffa 21 secs ago

@JimmyHoffa we generally close questions which are that specific

(I'm the 3rd hightest rep user there :P)

"tell me what to do for career dev" generally are bad questinos

@enderland Then perhaps you should pipe in on that question to suggest it would not fit on workplace before the poster see's Shauna's comment and decides to cross post

user55340

And flag the comment as "don't need to propogate suggestions to repost junk to The Workplace" so it gets deleted.

Q: How to discourage people encouraging reposting?

7:53 PM

@JimmyHoffa there's also not a constructive question

user55340

When someone reposts junk from SO on some suggestion, I try to comment back "Please don't do this, check the site on topic before suggesting anything and flag for migration instead of reposting" and then flag the comment that suggested reposting to P.SE... so far, every comment I've flagged has been deleted.

user55340

And then there's this...

user55340

19

I'm mostly on Programmers.SE. All too not-uncommonly we get a question that is multiply posted - once on Stack Overflow, and once on P.SE. It is almost standard practice to check to see if there is a newly asked SO question by the user whenever casting a close vote. Often when looking at the o...

discussion migration cross-posting

Q: How can we get SE to stop migrating questions which clearly need work here

@MichaelT and this:

2

How to act mature and professional at work? This is going to be closed (hopefully) since it's a vague complete mess. It's not even remotely appropriate for Stack Exchange in its current form. After being closed, it will then get locked and eventually auto-delete, in spite of having what might b...

discussion migration

@Shauna this question would almost assuredly be closed on The Workplace in its current form, there isn't really a question here and it's basically "my job sucks" followed by "everyone chime in!" — enderland 1 min ago

user55340

Ok, I found one another declined flag after digging a bit more that read "declined - Server Fault would burn us to the ground if we migrated this." And thats a good thing. Crap wasn't migrated. — MichaelT Aug 28 at 20:00

8:08 PM

I need a clean way to display python errors

without all that unit testing stuff

@Ampt Yeah, unit test shmunit shmest, you're enterprisey, that shit'll never work for you!

well.... its just that my tests aren't exactly independent

@Ampt just write perfect code and you don't have to handle any errors amirite

user55340

@JimmyHoffa Hey! Cut that out! He needs to make sure he doesn't drop a plow on my car. That would be 'bad'.

Blog - Stack Exchange: Podcast #52 – We Didn’t Need Headphones

without resetting a whole module theres not a good way to start from the same state

Stack Exchange

8:10 PM

Alex Miller on September 03, 2013

Welcome to Stack Exchange Podcast #52 with your hosts Joel Spolsky, David Fullerton and Jay Hanlon. Today’s show is brought to you by Marmite Yeast Extract – you either love it, or hate it! (You probably hate it.) Joining us today are Careers 2.0 Marketing Coordinator Bethany Marzipan, er, Marzewski and Careers 2.0 Product Manager Will Cole.

Site Milestones: Space continues to be all around us, everywhere. It’s also a Stack Exchange site, but we’ve talked about it already. …

new podcast wooooooooo

@MichaelT Don't look at me, I'm not the one here who's too enterprisey for unit tests

user55340

Aug 29 at 18:09, by Ampt

@MichaelT nah, I write perfect code the first time remember? We're putting the hex together to ship it right now

you guys are right, I don't really need to test the production units at all do I? I mean our vendor are the most reputable of places

@Ampt Atomic mocking unit tests? No? nadda?

8:11 PM

not testing the code, testing the hardware

the last batch we received from our vendor had a pass rate of under 60%

basically I'm making a testing platform in python

@Ampt Well, if you want them clean and atomic be sure to use fusion, not fission.

user55340

►

@MichaelT that only happens 40% of the time remember?

user55340

Hardware test.

@psr fission, not fusion. Got it.

i guess I could use unit testing and literally reset the module for each test during the set up

that just seems less than efficient

8:17 PM

@Ampt Unit tests aren't supposed to be efficient

well I'm not exactly unit testing either

It's a common (though controversial) practice to make your unit tests run in independent transactional contexts so when each unit test ends, a rollback occurs undoing everything

its more akin to production testing

@Ampt with unit tests, these are just...hardware units...

it is production testing

8:19 PM

How long would it take to run through a full suite with resets between each test on a single piece of hardware?

user55340

@JimmyHoffa Well, you need to get the Mythbuster team out, and a new car...

well right now it takes about 15 seconds with no resets and resets could take 500ms+

I would guess somewhere around 45-60 seconds

@Ampt HAH. that's nothing.

it is when you pay for it :P

we pay for their time in testing

Why are you paying for the time spent unit testing?

wth lol

nevermind

I don't even want to know how this business model works, and it's none of my business

8:20 PM

we have an interesting relationship with our vendors

yeah, its complicated

so the consensus is that I'm doomed right?

@Ampt I would still do it, make the case and explanation to your overlords, let them tell you yay/nay (probably nay) and then at least you can stop thinking about it and call it a day

AMPT2-D2

user55340

amazon.com/William-Shakespeares-Star-Wars-Doescher/dp/…

user55340

From the comments on that page, an excerpt...

user55340

8:25 PM

HAN: Pray tell, what shall the cargo be?

OBI-WAN: -Myself
The boy, two droids, and ne're a question ask'd.

HAN 'Tis what, a touch of local trouble here?

OBI-WAN Nay, let us simply say it thus: we would
Imperial entanglements avoid

HAN Aye, there's the rub, so shalt though further pay.
Ten thousand is the cost, and ev'ry bit
Shalt though deliver ere we leave the dock.

LUKE Ten thousand? Fie! We could our own ship buy
For such a sum as this.

HAN -A goodly jest!
For who should pilot such a ship -- shouldst thou?

user55340

(You know that in 40 years, they'll torture highschoolers with reading this in "Modern Hollywood English" class)

I find this absolutely amazing and am compelled to read the whole thing.

@Ampt I think they made a play or a movie recently, not sure which

heard about it on the radio

though I may be remembering for crap and maybe it was just a book

@JimmyHoffa I think I'm goign to buy that for kindle

user55340

I think its a bit more 'accessible' than the Klingon Hamlet.

user55340

8:32 PM

amazon.com/The-Klingon-Hamlet-William-Shakespeare/dp/0671035789

2

@ThomasOwens perfect - thank you!

@MichaelT Personally, I'd rather see the Klinger Hamlet

user55340

One man show (They know I'm crazy then!) Including the part of Ophelia (I get to show off my dresses...)

I'm voting for Sith and Sensibility, the tale of a young noblewoman whose dashing suitor with 5000 pounds a year turns out to have a dark side...

user55340

8:49 PM

Jr. High english teacher did a one man show... just him and taped voices.

@MichaelT was this someone who desperately needed a captive audience?

user55340

@psr It was more a "complete control over the production" thing. As the author and (only) actor... everything else was pre-recorded. There is also a challenge to being the only person on stage. Doing costume changes as part of the show and getting the timing right.

Q: How to go about creating a prolog program that can work backwards to determine steps needed to reach a goal

Is this good enough to keep open with an SO migration? It seems it to me, but perhaps it's more open-ended than I think

0

I'm not sure what exactly I'm trying to ask. I want to be able to make some code that can easily take an initial and final state and some rules, and determine paths/choices to get there. So think, for example, in a game like Starcraft. To build a factory I need to have a barracks and a command...

artificial-intelligence prolog design-decisions

@JimmyHoffa sounds like gimme teh codez!

It sounds to me like he's creating constraints in prolog and asking how to query the rules engine for available known valid combinations that meet that

which is a valid question; I don't know how to query the rules engine for a full combination off hand myself but I'm relatively certain it can be done

9:03 PM

@JimmyHoffa I'm not sure what exactly I'm trying to ask <-- this also is a bad sign to me :)

it might be require creating base set and iterating choices recursively to flesh them out by asking the rule engine the questions

@enderland I agree, though I feel I can tell what he's asking... yeah I'll just hold off I suppose

though having reread the question it is actually fascinating

I won't vote for migration or closure because it feels like a decent question to me, if off topic, I'll leave it up to the community to decide

@JimmyHoffa it was migrated already, lol

@JimmyHoffa I think he would accept a Haskell solution.

9:12 PM

@psr The problem with Haskell for rules engine work is the rules engine is the type system and as such it's in the compiler, makes it harder to interact with; in prolog the rules engine is a realtime interactive constraint space which is genuinely cool.

In Haskell you make a run time rules engine in like 4 lines of code.

If you're going haskell at that point you may as well just go full-blown crazy and go with Agda so you can have a complete verification that your constraint space is valid

I'm assuming.