#! Computer Generated Poetry | Nick Montfort | Talks at Google

Just another WordPress site

#! Computer Generated Poetry | Nick Montfort | Talks at Google

MALE SPEAKER: Welcome to Talks at Google in Cambridge, Massachusetts This morning, I’m excited to introduce Nick Montfort, Associate Professor of Digital Media at the Massachusetts Institute of Technology There, he directs The Trope Tank research lab, studying literary generators and computational art For over a decade, he served on the board of the Electronic Literature Organization and he’s principal of the naming firm Nomnym Today, Nick will read from his latest book, “#!.” “#!” presents computer generated poems alongside their corresponding source code His other works include “Twisty Little Passages”, a survey of interactive fiction, and “10 PRINT”, a 10-author collaboration on create computing Please welcome Nick Montfort NICK MONTFORT: Thanks Thanks very much “Form intends intense verse crease to tense form tense vent verse tone verse form crease form vent tends to crease to tends form form vent form crease tone verse tense Crease vent vent tends inverse tone into verse form verse verse form tone tense in tense vent crease verse tone tends verse tends tends tense verse crease form tone vent into tends to crease vent to crease vent verse verse vent to crease vent form tends vent crease tense form tends crease in intone To tone verse vent crease intends vent vent to tense in form crease vent crease form to to tends tone verse form vent tends tends to.” The work I have been reading, “Round”, is a very simple program and poem that’s available on the web at round.newbinarypress.com and, as with all my creative digital work, also available on my site, nickm.com It displays the digits of pi But instead of using the numerals 0 through 9, the program presents strings of text that correspond to each digit Eight of them are words that can stand alone, one is the prefix “in-“, and one is a line break, which shapes the poem into lines of irregular length So since 3 is represented by “form,” 1 by “in,” 4 by “tends,” 1 by “in,” 5 by “tense,” and 9 by “verse,” the first digits of pi– 3, 1, 4, 1, 5, 9– are represented by “form intends intense verse.” “Round” is a computational poem that is both non-interactive and deterministic It is computational in that computation is an essential aspect of the work, non-interactive because there’s no input accepted as the program runs, and deterministic because the text produced should be the same, each time, on any properly functioning computer The poem is also infinite, or boundless There’s no final line or internally specified condition that will cause it to stop “Round” does not contain any list of digits of pi It computes each digit using a spigot algorithm If it did store, as data, 1,000 digits, or 10,000, or any finite number, the program would be unable to continue representing the digits of pi after that point Instead, “Round” computes pi As the program runs, the production of text slows down as more and more steps are necessary to determine the next digit You may be able to see that in this Python instance of the program The multitasking computer running the program will also begin to run other processes more slowly and will physically heat up The computer’s fan will work harder as the processor labors to complete these computations “Verse tense tense tense vent crease crease form intone crease tense form tense verse tends vent increase

vent tends vent in in intone tends tense crease vent tends in crease tone inverse form vent tense crease in in tense tense tense verse to tends tends to crease crease verse tends vent verse tense tends verse form.” Of course, this is probably the audience I need to tell least about the materiality of computing, the labor that might be involved in scanning in books, the physical and material computation that’s required to do large amounts of symbol manipulation, data retrieval, and delivery But this is a way of looking at this and, for people whose experience of computing is maybe more ethereal, and for people who believe that digital poetry and the work that we do with text on the computer is actually immaterial, this is something of an answer to that showing that there is something going on in terms of computation, something almost mechanical, that is a laborious process And it also questions various ideas of the need to combine things at random– in a digital poem, for instance– which is not literally done in this case Although, obviously, it’s a quite unpredictable sequence unless you know that it’s the digits of pi that are being represented So I wrote this next poem, “Taroko Gorge,” in January, 2009, in about one day at Taiwan’s Taroko Gorge National Park It’s actually a pretty conventional poem compared to a lot of the things that I right, or a lot of the things that my programs output It’s meant to be a generator of nature poetry I figure that if other people had gone to beautiful, natural places– as, stereotypically, is the case– and they’ve written poems about them, why couldn’t I go to a beautiful, natural place and write a poetry generator about this place This one, also, is a boundless poem, running until it’s interrupted But as I think you can see, it’s not formless Each of the longer strophes begins and ends with a path line of a certain type representing motion along a path in Taroko Gorge Park And in between, there are zero or more view lines, representing stopping to look at the scenery at a designated viewing spot And then, these individual lines that appear in between the longer strophes are what I call cave lines, representing the tunnels that Chiang Kai-Shek had his nationalist army carve through the mountains when they came to Taiwan “Stones pace the ripplings Shapes sweep the flows Translate the objective driven Layer paces the coves Coves hold Mists trail the stones Shade the encompassing rough Brow ranges the flows The crags hold Shapes rest Layer ranges the stone Shade the encompassing, rough, sinuous straight.” Now I was pleased to do this exercise and to undertake this nature poetry generator One of the things that you see in the book is that, as with all of the outputs of these systems, they’re presented after the code for the program itself This is, in part, just a reference back to the days of basic programming, in the ’70s and ’80s, where one would find books and magazines that presented to code for people to type in They gave some sample output so you could make sure that what you were inputting was really what it was supposed to be And they gave an attitude of openness toward computing And of course, on the web, you can just view source and examine how things are done in JavaScript in the same way This is what Scott Rettberg, a friend and collaborator of mine, did And without telling me about it, he also made his own work out of “Taroko Gorge”– a piece called “Tokyo Garage.” And where I wrote a poem about a beautiful natural place that I’d actually visited myself, what Scott did was to write a poem about an urban place in the East that he had never been to, supplied only by his feverish imagination of what it would be like to be in Tokyo and all the stereotypes that we have in the West “Translator funds the libation

Dealers reverse Protagonist suffers the libation Indoor the electrical, banal blinking Teenager confuses the kids Earthquakes reverse Talk show host orients the sushi joints Harmonize the abbreviated.” Well, suffice it to say that other people started doing this as well J.R. Carpenter created this piece, “Gorge”– a modified version of “Taroko Gorge”– which describes another endless process, that of gluttony and endless consumption, and the disgusting and yet unstoppable processes thereof And J.R. did some additional pieces One of these that I certainly want to point out is Andrew Plotkins’s piece, in which he took all the remixes that happened up to this point, showed them remixed with themselves and with other remixes, and also displayed the code that corresponded to what was being printed at different times So this is really a tour de force And he doesn’t actually– only a small amount of my original code, obviously, is part of this It’s a different type of take off This is a remix of Eric Snodgrass’s “Yoko Engorged,” which is about Yoko Ono and her performance art And this is a remix of “Fred & George” and “Toy Garbage” down here And then, some people simply created work that– many people use the particular form that I created, of these path lines, view lines, and cave lines But some people actually discarded that They just sort of took the code as a starting point for making their own sort of project with their own form And Sonny Rae Tempest did that, creating these quatrains in his piece, “Camel Tail”, that are a patchwork of all of the lyrics from Metallica’s nine major studio albums You can see here, the production is something of quite a different sort But I think it rhetorically gets across the point that metal is more of a texture than a narrative or a story, right It’s that you can take these lines and put them together at random, and you still sort of feel what you’re supposed to feel when you’re listening to the song “Much too young to focus No more! Anger Master.” Pretty much, that could be a lyric All right So to end this most coherent section of my reading, I’ll present some of the output from my first Ruby program I was commissioned to write something for the book, “The Ill-Tempered Rubyist,” which is sort of code poems And it had this title And I thought, well, I’ll need to see what I can do, actually, in Ruby So the program is built with text from Edward Fitzgerald’s translation of “The Rubaiyat of Omar Khayyam.” And before I read the output, I’ll read you the program itself, not enunciating the punctuation marks well enough for you to type it in But you can see the book for that And I think you’ll get some sense of what animates it, even if you aren’t particularly familiar with Ruby, as, in fact, I’m not “Rubaiyat Shebang, user, bin, Ruby A phrases equal oriate, alternate, along with, argument, another, absolute, alchemist, arrest the, an empty, a Ruby, a little, a thousand, a book of, and peace to, a jug of, a loaf of, a distant, and out of, and lose your, a moments, amid the, are all but, awoke from, and by the, all this of, a shallow, as much as, as brings the B phrases equal beneath the, before the, but still the, beside the, beloved, better be, by logic, blaspheme the, before us, beset the Rhymes equal stare, there, care, dare, snare, bear, air, hair, tear, ne’er Non-rhymes equal sun, stars, shaft, cup, bird, wing, winds, dust, not, lamp, lip, dusk, balm, bowl, song Six times endings equal rhymes dot sample three Endings dot insert two non-rhymes dot sample Beginnings equal A phrases dot sample three Beginnings dot insert two B phrases dot sample Zero up to three I puts beginning I plus base plus endings I. Puts A little bare are all but there By logic shaft, an empty stare Absolute there amid the stair Blaspheme the dusk, argument bare And by the there, an empty air Better be cup A distant hair A thousand stare A moment’s care Beloved cup, arrest the snare

Awoke from tear a jug of ne’er Better be cup A distant stare A thousand tear A jug of snare Beset the balm A moment’s air.” And I’ll read you the version that is printed in the book “Amid the care and lose your air Better be wing Another there A distant care A loaf of ne’er By logic dust and by the dare A Ruby there An empty air Blaspheme the cup and out of hair And use your bare Absolute dare Before us wing Argument care As springs the care amid the care, before us stars another snare Along with tear and piece to dare Beside the cup, argument care.” Now in the book, you can see a few conceptual poems that refer to some earlier digital work One that is pretty simple in form is based on the permutation poems of Brion Gysin, which Ian Sommerville implemented computationally And so “I am that I am” is a sort of homage to that, that simply displays all the permutations of the five full vowels And there’s also a piece based on a work of Claude Closky, which is the first m numbers in alphabetical order So Claude Closky did a piece, in 1989, both in English and in French, where he presented the first thousand numbers in alphabetical order And of course, they’re alphabetized differently in French and English when they’re spelled out And so he showed these And to me, it was a very interesting display of the arbitrariness of our various systems of language when it comes up against number So I did a version called “The First M Numbers in Alphabetical Order.” And it generates and alphabetizes the first thousand Roman numerals So I’ll show you a little of the Concrete Perl series of programs There was one of these, actually, that I showed early on, before I got started And that was “Letterformed Terrain.” And this is a piece that uses a Perl special variable that has a value equal to 60 We actually did a– I worked with a Russian translator and we did a transliteration to Cyrillic of this piece, “Letterformed Terrain.” And we had a Russian title for it And for one thing, the landscape is denser, because there’s more characters in the Cyrillic alphabet But that value of 60 stays the same So one of the possibilities, even at the letter level, for moving between languages and alphabets, is to see how it is that the systems we use– Perl ASCII, Unicode– come up against different human languages, and how they work differently I’ll just show you the code for that, itself, which is that This is a 32 character Perl program And I have a set of these Concrete Perl programs are all 32 characters long And then, what I’ll show after, I’ll show a few examples of ppg256– 256 that series of 256 character Perl programs I showed some of these to an actual, professional, Perl programmer– not someone silly, like me, who’s playing around with Perl to do these sorts of things And I showed, actually, one of the ppg256 systems And this person was completely unimpressed by what the program did But he looked at the program and said, “Redo?” Because he’d actually never heard of that keyword But it turns out that the most– as far as I can tell– the most concise way to create a loop is to have a code block with “redo” at the end, in Perl So look at that Even I can learn something about Perl which probably shouldn’t be learned So let me ask if we can make, maybe, the screen– we can do, for the video, full screen for this And I’ll show you “Alphabet Expanding.”

So I find this mesmerizing It’s not at all a piece of video art You have to start it or have someone start it for you, I think, to really get the effect of it And of course, it helps to know the code that’s animating it as well These programs– these Concrete Perl programs– are written for terminals or consoles of any geometry So they’re meant to work no matter what typography and what size of window or console you’re using This one, here, is simply increasing the amount of spaces between the alphabet, which it’s printing repeatedly And it’s doing that very, very slowly And so that alphabet goes in and out of phase with the size of the terminal, producing these different types of effects So another one of these is “All the Names of God,” which is printing every one letter, and then, every two letter, and then, every three letter word, and so on, until eventually– because it’s trying to pre-compute them all in memory– it will crash And if you’ve read Arthur C Clarke’s “The Nine Billion Names of God,” you’ll probably be glad that it does crash, since you know what the outcome might be of continuing And the last one of these is actually a fairly simple one, but one that, I think, has a point It’s called “ASCII Hegemony.” So when we translated this to Russian or transliterated it to Cyrillic, this one stayed the same Because this is just all the printable characters in ASCII And actually, early electronic literature work in Russian was done in ASCII Because there was not support for the Cyrillic alphabet on the first computers that were popularly available there And even as we continue with amazing projects like Unicode, there’s still these questions of how different programming languages and different environments support different alphabets So you can get Latin alphabet in Perl by saying A dot dot Z But you can’t do that for Greek, or Cyrillic, or other alphabets Of course, you can work around that But your program isn’t going to end up being 32 characters long in those cases So I’ll finish up– this next section, the reading, there’s probably two more little pieces here But I’ll finish up by reading just a few of the ppg256 pieces Let me show you what these look like So here’s the first of these, for instance Besides the comments here, I made this file so that it could be run on Windows But the program itself is just the bottom three lines there And it doesn’t use– as with Concrete Perl, of course– it doesn’t use any external stores of data It doesn’t do anything online Everything that it produces is actually encoded in that 256 character program Of course, it uses the Perl programming language, and all the capabilities that that has, and all the ways that that animates computation But everything has to be packed into that short program And so for me, this is a challenge to try to explore things about Perl, and also, about poetry and language How could I make a very concise program that would at least look like a poetry generator– would be read that way by someone who was a generous, willing viewer of the program? And so this is what I came up with I’ll tell a little bit about this before I read from it So the first task I set myself was to generate a very, very large vocabulary– how could I create a large number of words? Well, there’s well known ways to do this So in Claude Shannon’s “Mathematical Theory of Communication,” from 1948, he describes the process by which, through using conditional probabilities and increasing n, one might be able to generate text at random, including words at random The problem with this is that if you want to use even just bigram probabilities, you have 26 letters

And then, there’s 26 letters that could come after it That’s 276 letters And so that’s going to be a lot of conditional probabilities for 256 characters of code They’re probably not going to fit in there very easily, if at all So I was trying to figure out a different way to do this And the way I came up with was– I tried a bunch of things And I was getting, maybe, dictionary words like 3% of the time It was very futile And then I decided, OK, well, what if I just take the first two letters of common four-letter words and the last two letters of common four-letter words? And I’ll just jam them together at random and see what happens Well, as it happened, even without any fix up, I was getting 60% dictionary words by doing that It was pretty amazing I had a chance to speak to a phonologist at MIT about this, and to describe this process And he said, oh yeah, turns out, from monosyllables, the initial and final bigrams are going to be pretty much independent And so my first thought was, oh well, he knew that already That’s too bad And then, I was like, well, he probably went to graduate school for like seven years to learn that And I was just playing around with a Perl program for a few hours So maybe that’s interesting And there’s other things that I learned about this as well So if you just produce a string of words, it’s not that recognizable as a poem to most people But if you shape them into lines and stanzas, ah, it’s starting to look more like a poem And then, I spent a good deal of those 256 characters producing these titles at the top of each of these, just indenting by one space And so it turns out that makes things, at least to me, look considerably more like a poem These framing elements and the overall shape of the poem are quite important to it “The car Tat to baed The bure Hoke no tosh Toes on hoes Cack on bar The pang Rall on mosh Bat of dear.” So I’ll show you the second one of these I tried to do some things quite differently here I was satisfied with a small vocabulary There’s only 89 words this generates, including all the close clasp words, the ampersand But I wanted to have a variable line length I wanted to have a shape of a stanza In fact, the line tends to become smaller, as you can see, in the pans here, as this continues And I wanted to have variability of word length as well And I was willing to sacrifice that large vocabulary to do this So I get things like this in this poem “The kits The pan– a kin A twin Of twit twits fit Shill Span– the skill Sit & skit To chin– a shin A spill To sit Fan The grin Grit The grins Twin & shill Of spit Chill shills the grin Spin To spill A kit Chill Twill.” A solution to generating the indefinite article is to not produce any words that begin with vowels or that begin with vowel sounds So I never have to produce “an.” I only have to produce “a.” So there’s a lot of things like that, which are not, certainly, general purpose solutions, but which given interesting– for me– an interesting engagement with language and understanding of their possible textures So there’s actually seven of these But I’ll read a bit from the third And then, I’ll move on to this final thing that I wanted to share with you And this third one, I decided, I’ll get rid of the titles And I’ll try to look at compound words– how can I compound a list of three-letter words that standalone as words into new words? And how can I make something that seems like, maybe, the beginning of a narrative– not a full story, but something that suggests a situation, something that’s suggesting narrativity? And this is what I came up with “One_red_manape and the_wax_apeape cut_on _hip_godbot and the__boyeel ran_off The_wan_manhat and __apeboy put_out One_botelf and on__elfnun jam_her One__godman and one_red_elforc set_on The__nunman and one_fat_orcbot hit_us One_red_hatorc and the__apeboy put_her.” So that’s ppg256 3 And it’s the third of seven of these very concise Perl programs which– I think you get the sense– I’m not

trying to use to express things about my life or emotion, but I’m trying to use to explore language in a poetic way, through the making of a poetry generator and the making of a poetic language– in some ways, the same type of investigation that people are doing when they’re doing linguistics or computational linguistics, but through a different, non-scientific means– that is, this poetic production So I wanted to share with you a book that I just finished and just had printed It’s called “Megawatt.” And it’s a book that– there’s some story behind it The seventh of these ppg256 programs is based on a passage from Samuel Beckett’s second novel, “Watt.” And after doing this, I made a Demoscene production called “Nanowatt,” which is a 3.5K VIC-20 program that generates 8K of “Watt” exactly, in English And then, it goes on to generate 8K of “Watt” in French And these are certainly out of the ordinary endeavors But the point of engaging with the novel in this way is to try to read it, to try to understand it through reimplementing it So I’ll show a little bit of– here’s the– So this is the book cover And this is a little from the preface, which is also the doc string And “Megawatt,” then, is the third computational project I’ve undertaken to try to deal with Samuel Beckett’s novel This novel has some more or less ordinary language in it But it also has these really bizarre passages in which every combination of three or four different elements will be described So it will be described, for instance, that Watt heard voices which sometimes sang, cried, stated, or murmured or that he felt calm, and free, and glad, but maybe not calm, and free, glad, but at least calm and free, or free and glad, and so on And so what I’ve done is to re-implement these passages from “Watt” computationally– created programs that would write what Beckett wrote And then, I extended them so that they would say even more So the 2 and 1/2 pages that are in “Watt” describing Mr. Knott’s physical appearance occupy, here, more than– the corresponding section in “Megawatt” occupies more than 200 pages And I’ll read you a little bit of that “With regard to the so important matter of Mr. Knott’s physical appearance, Watt had, unfortunately, little or nothing to say For one day, Mr. Knott would be thin, small, pale, dark, brown-eyed, ectomorphic, clean-shaven, and erect, and the next, thin, small, pale, dark, brown-eyed, ectomorphic, clean-shaven, and stooped, and the next, thin, small, pale, dark, brown-eyed, ectomorphic, clean-shaven, and leaning, and the next, thin, small, pale, dark, brown-eyed, ectomorphic, bearded, and erect.” OK And so on, for more than 200 pages And at the end of this book is all of the code that’s used to generate the main text of the book It’s a 350-line Python program And from a standpoint of reading from literary experience, it seems that we’re confronted with things that, although in terms of the amount of time it takes to read them, we physically have the time, in our lives, to read something like this But it’s illegible due to not being created in a human way, for reading But we can still encounter it computationally And so “Megawatt” is part of that project And “#!” is another aspect of that project, not to confuse, not to puzzle people, but to show that computation is something that can engage with poetry and language in new sorts of ways And of course, the focus is on these very simple programs, which, for the most part, can be used offline There’s only one part of “Megawatt” that even makes use of WordNet Because there’s the dish that Mr. Knott eats that seems to contain every possible food in it And so I use WordNet to enumerate that possibility So that’s some about the project of “#!” and this newly released book

And with that, I’ll open it up to questions AUDIENCE: So does your code come with a license? NICK MONTFORT: Yes So generally, all the code is ISC licensed So it’s all free software When you find it online, it has this license I have a note in “#!” saying that it’s free software and where to find it online But we didn’t repeat the license for each one of these very short programs, many of which– I mean, we have a 32-character Perl program It’s a little onerous to include even a very short license in that case It’s also a question as to whether a 32-character Perl program can be copywritten I’m not sure– short phrases do not fall under copyright law So it’s not obvious to me whether those would be afforded copyright protection AUDIENCE: Well then, then followup question is, does your code contain expression MALE SPEAKER: Oh, I hope– well, I hope not Yes Because most of it is a very conceptual project I mean, certainly, “Taroko Gorge” is expressive of a type of experience, a type of feeling It’s one that can be continued, that doesn’t have a particular end It’s an experience of nature So there are aspects of expression in it But to me, that’s not the point It is something that I’m using to investigate language, form, poetry And other people can do other types of projects that are more focused on expression But those are the things that interest me particularly, is really to be an investigator of language, which has such an amazing history and such a rich potential to connect and divide in various ways AUDIENCE: What about color? Have you ever played with terminal colors, like 256 color xterm, for example? NICK MONTFORT: I think, probably, just decorating things on the web is about the only thing that I’ve done along those lines With things like Concrete Perl, which are visual programs, I really tried to make them more or less generic to the color of text or the typography AUDIENCE: Thanks NICK MONTFORT: All right AUDIENCE: So the generators you’ve presented are all fairly minimal So you have any longer form works that aren’t quite interactive fiction, but are somewhere between these minimalistic examples and normal Z-machine programs? NICK MONTFORT: Yeah So certainly, the focus in “#!” is on these very short programs I think that almost all of them fit on a single page And some of them are much shorter And it’s probably– yeah, so “Through the Park” and “Taroko Gorge” go across two pages They’re quite brief In part, my interest in these short programs is just that there’s still enough to do in the very short amount of space– a very small amount of computation– that’s interesting that I don’t feel like moving on from that in many cases It’s also something where I can afford to more intensively go over this and understand more about it if I have something short I have systems– I mean, “Megawatt” is 350 lines, which is getting to be longer It’s not a large scale, enterprise program I do have this system, Curveship, which is a full interactive fiction system And it’s implemented in Python And it does the basics of what systems like Inform, TADS, that line of work will implement And plus, it also allows for narrative variation so that the same underlying events can be represented different ways So that’s a larger scale project which I’m pleased with But on the other hand, I’ve never really written– I have demonstration programs that I’ve done in Curveship But I’ve never really written an original work, in Curveship, that was an extensive original work And I think it’s possible that that’s because I spent so much time working on the system itself and building all of this out So my focus has been on this smaller scale

But I think there’s interesting possibilities for individual and collaborative authorship of medium scale works I mean, one of these pieces that is definitely larger– which is a collaboration of mine, with Stephanie Strickland– is “Sea and Spar Between.” So this is a poetry generator we published in 2011 And it’s a conflation of a Emily Dickinson’s– a lexicon of her poems and of the lexicon of “Moby Dick.” And it’s about 225 trillion stanzas that are arranged in this lattice So there’s coordinates, and you can go to them It’s actually a fixed text that you can go to But it’s so large that it’s meaningless to treat it that way You might as well just open it up, and look around, and encounter it the way we’re doing now And this is a larger scale project– I’d say sort of medium scale And for a collaborative project that spanned some time, it was pretty useful to do something of this size So you can go to– if I go to 0.500, what’s that? “Listen now, for hueless is the sky One air, one air, one each, one play Paradise! Spirit!” And then, I can go somewhere else I’m just navigating here If I go back to 1.500– oh, 0.500 is what I was looking for Oops I have to type it right– I should get the same thing Yeah “Listen now, for hueless is the sky.” So this is an example of a medium scale project I mean, there’s different things that come out So this has actually been remixed the way that “Taroko Gorge” was once, that we know of It’s been translated to Polish, which is an amazing undertaking And it’s also been remixed to create a piece called “House of Leaves of Grass,” which is based on Mark Danielewski and Walt Whitman And Mark Sample did that So one of the things is that, these very small scale pieces, they invite others to study and understand them, to manipulate them, to rework them to create their own pieces based on them So in that case, “Taroko Gorge” was, I think, extremely successful If I’d made it 10 times the size that it was– it fit on one– it’s less than 66 lines long It fits on one dot matrix printed page And if I’d made it 10 times as big, nothing that you saw there, along the right-hand side– all of those remixes and work– would have happened So there’s some significant benefits to doing smaller pieces And I think that you shouldn’t exceed the size of your concept in writing programs like this, poems like this, in this sort of code But if you have something that is a bit more expensive, then it makes sense to have a larger project AUDIENCE: Thanks NICK MONTFORT: If There’s no questions, I’ll just offer some company names for anyone who would like them MALE SPEAKER: Well, let’s thank Nick NICK MONTFORT: All right Thanks Thank you