Please summarise the following transcript as meeting notes. Don’t miss anything out if possible ensure all relevant information / ideas / topics are retained:
Is has the role of the tree, and you have the role of the human interviewer, and you play along. You you bring in conversation. And the tree, of course, has a way of working because of the system it’s running on, but also, it’s continuously grounded on the things that it it really experienced, things that really happened. And if you you have experienced the same things from your perspective. Is it is it helpful to think of it I mean, I’ve been thinking of it in terms of it’s if data can talk.
So we’re feeding data into a system, and the system animates the data. And because the data is essentially everything that’s happening to the tree, this is the tree’s experience as relayed through the all the data that we can essentially either glean from the tree or decked on. So That’s okay. And that and then it’s telling the avatar or no. The tree is the tree is the skeleton, and we’ve kind of fleshed it out with the data.
Does that make sense? It it does make sense. And an important detail is that we’re not feeding just the raw data. The data is processed by, what what we say, tree comfort model or a tree health model. So all the data means something for the tree’s physiology.
So the data is just now clear of understanding, how the tree must be feeling. Yeah. But yeah. It’s the best we can do. We can’t take the 3 input in it on the lab and and and measure it through goals.
But so this is the best we can do to be fair and to use, situation. Yeah. Yeah. And and because then because we think we’re always grounding what what the LLM writes on this data, We hope that this project is better than just talking to edit to to GPT and telling it to role play it play it. This is actually part of a human situation of a real world situation.
Yeah. Yeah. There is also, though, a human, like, in in the trees, personality. Because data is one thing and it’s one, like George said, makes it different for from sensitivity. But there is the part of we want the truth of personality, and that has to be human guided.
So someone, maybe you, should dream, like, what these three things about the world or Yeah. What are the the goals and the motivations and the thoughts. Yeah. Which which which is what we’ve been doing, and I think we had an interesting written read through your text, trying to to read the behind the lines or between the lines and Sergei, this sentence indicates this type of personality maybe. Yeah.
Well, that’s been something that I’ve been sort of the the last couple of drafts of this week, there’s been a sense of I mean, one was a general draft. 1 was a kind of wintery, more stuporish, sort of quite lo fi draft, and one was a more up spring draft. And to some degree, it might have looked like I was rearranging deck chairs as opposed to sort of, like, giving you all new material. But I was thinking of it in terms of flow and as a reader and potentially as a, you know, large language model, the sort of algorithms that you’re using, if that’s what you’re using. Hi.
Not really a tech person. Then the kind of flow of it was the sense of we’re hitting these kind of, like, is this a positive? Is it a negative? Is it a neutral? And so you’re going all the way through to try and get the the tree avatar that I’m sort of building, I suppose.
There’s kind of, like, what would we glean from this? What would we read through the lines? What’s emotively there? And the spring one, which I thought was kind of the most positive both in the sense of the actual vibe of it, but also the sense of becoming a sense of kind of forward movement, was potentially also the most sort of poetic. And we can, you know, that can be all filtered out or the poetry can be still there, but less is a big chunk of it.
That was more of a experiment. But the idea of kind of, like, the personality of the tree, I completely get that. I’m very, very keen to kind of, like, help in any way I can to kind of give you a workable model that allows you to extrapolate out in the directions that you think most interesting based on real world data as opposed to me going, you know because, you know, the idea that maybe it’s a thin tree so it’s shy, that’s all well and good but not actually based on anything. Whereas maybe it’s kind of maybe it’s a thin tree, and so it responds to the wind in a certain way so it has a certain sort of voice that could, you know, that I’m not saying that’s any more or less woo woo, but it could be actually based on some sort of, you know, flow dynamic of the of the wind or whatever as opposed to let’s just give this a personality. And everything I’ve been trying to do is kind of try and glean with a bit of research into what these silver furs do, how they grow, how they move, how they, you know, potentially think as a unit.
So I’m drawing all that out, but that’s this is very early day stuff, so I’m still very keen to be guided by you. Does that make sense? Yes. I I think that’s exact I you know, to maybe simplify it and maybe make them seem stupid, which is not my intention, but I think FCB are thinking, like, fir tree has spiky leaves, and therefore it might be very spiky and to the point kind of thing. Whereas I think you’re what you’re saying is more yeah.
When you look at a fir tree, actually, they’re super elegant. And when the snow sits on them, they almost look like they’ve been doused in fabric or something. Already, the kind of the fact that its name means bright. And I and I think what we almost need to do is have this the the the 500 words. And alongside that, this kind of, like, almost maybe almost bullet points like I am this, I’m this, I’m this, I’m this.
Mhmm. We We we were discussing with George that there are various formats that Dan could write about the tree. It could be like writing in a journal. It could be these bullet points that you’re saying. It could be a short essay.
You know, there’s there’s different types of text that could form a tree’s personality, and I think it’s it’s viable to see it also in a in a maybe in a longer period in time. Yeah. And if we see it in in terms of a dialogue structure, it could be an internal monologue. Like, that that text might be the 3 just not talking to anyone, not talking to the humans, just thinking out loud. And also.
Or good structured format with bullet points that describe the personality in a almost technical way or, like, psych like a psychiatrist trying to to pitch you in a category. You know? Yeah. Thinking of what’s what’s useful to the LLM in creating this persona and the the longer format internal monologue sort of trail of thought style writing is very useful for essentially guiding it in terms of tone and style. And then the bullet points are giving it kind of very objective things to or more objective things to kind of stick to.
So you’ve got, like, we’re giving it a reference of this is this is your voice. This is kind of how you sound and how you talk. And then the bullet points are more like, here are some almost, like, hard truths to bear in mind when you’re responding. And and wouldn’t those bullet points then be wouldn’t we then say, I don’t know, the bullet point is optimistic. We say that x, y, zed links to optimism and puts them on a spectrum of optimism.
X, y, zed being data. Do we is that something we’re seeing happening? Yeah. We we could create, sorry, Dan. We’re just throwing ideas here.
Alright. We could create some sort of conditional, like, paradigms where, say, if the tree is in its comfort zone because the temperature and moisture are in an ideal range, then its its outlook on these factors should should be more like this example, and then we have an example of writing. Whereas if it’s in the danger zone, its reaction should be more like this. And and that’s also, a way we can instruct large language model by giving it examples. Wouldn’t that be if sorry.
I’m just playing with your thought. But wouldn’t that be if Dan has already written some text about how the tree reacts to high temperature or low temperature? Taking this example. And we use this text to inform the personality. I’m I’m not sure we even have to tell it.
Okay. Now the high temperature use this, it it should automatically Yeah. Yeah. Yeah. Abstract from these Yeah.
Claims and And then exercise for them would be write about an amazing day of the tree and write about a horrible day of the tree and about, A boring day. A boring yeah. Nothing happened. And, also, I had I have something I haven’t shown the guys yet because we didn’t have time. But I was reading I was plotting, the temperature data for the 1st 2 months of 2024, like hourly data, and I was seeing some patterns there that made me me think about different events in the tree’s life, like small scale oscillations where the temperature is along a small range.
And you think it’s gonna go like that, but suddenly, you have a cold spell. And we think half a day, you go from plus plus 5 to minus 23. And that’s an event that the tree, might remember, but it’s only a day in a a bad day or in a good month, for example. Things like that Mhmm. Are examples that we could give the LLM, and then it would do a great job blending between them.
So I I really like this idea of where you can come up with a few different situations that Dan writes an example paragraph like, a couple of paragraphs on. Maybe for every paragraph, there’s also a corresponding list of bullet points so so that you’ve got the the style and the tone, but you’ve also got the kind of almost easier bits of information for it to access. But And then but at what point does this spiral out of control and we suddenly end up with That’s what I was just thinking. 500 different possible, reference paragraphs. And then down that as weeps.
But that’s that’s how we direct the what? Where is that? We got Lauren’s. That’s how we direct the the LLM in fine tuning it. So, essentially, we’re we’re not doing the fine tuning thing because it’s takes a lot of time and a lot of resources, and it’s not necessarily giving better results.
But what we will do is we use an amount of text to inform its response. So we can draw from the more we give it, the richer the context it is, the more, you know, fine grained the response should be. Yeah. So we have to read between the the references. As I to take maybe just one step back because I think all of that’s right, I think, in terms of the workflow that Dan’s been doing so far.
So we have these 4 drafts or kind of 3 drafts. Should we, as a next step, say, okay, let’s just now let’s have one final and let’s call it something, whether it’s internal monologue or whether it’s, whatever. Maybe we would call it biography of the tree. And it’s like so there’s then, like, general thing about the tree, and then there’s maybe also, like, a very small bit each where it’s like, this is how I feel in the winter, this is how I feel in spring, etcetera, etcetera, so that we capture a range. And that’s one chunk of text with the kind of bullet points next to it.
And that that then forms the deliverable for client to say, cool. Proceed with this tree. I’m not I’m not even sure the clients should okay this. No. I think I don’t know if we can avoid that.
Well, I think this is giving us this is giving us the material which we’re gonna give to the LLM. It’s not forming the whole tree. And it’s only them up with additional data because it’s that’s going to spiral. I think what what we can say is what you’ve just said. We have we have hero biography.
These are 500 words of this is the best summary of what this tree is and who this tree is. Then we have, high temperature, low temperature, perfectly happy, hazardous. Like, we come up with a list of of smaller scenarios for its reference, and then we present this to the client as this is the information which we are feeding it. And in in our ongoing tests, we’re gonna work out how best the LLM can utilize this information and combine it with the data. So the There’s no approval.
The way this technique works, which I’ll actually forget the the abbreviation stands for, retrieval augmented generation. So, essentially, you have this additional information that it goes through the moment you ask something or you’re prompted to write something. So it goes through all of this information. And what it does behind the scenes, it it enhances your prompt to a richer prompt that’s informed with this data. That’s what we can expose to the client potentially to show them, okay.
You know, this is how we enrich the person with the personality and not the the data itself. Yeah. So to in that case, the sort of hero biography can be taken from the work I’ve already done, and, basically, you can stitch together the things that you like best. And then I can give that a final polish. And I I can do 3 scenarios, And I think, basically, good, bad, boring is good.
But, I mean, you know, best case scenario, tree under stress, which is bad. So whether that’s an oscillation of temperatures, whether that’s a kind of, you know, storm or something like that, I can think about this. It would be really, really good to actually have some of that temperature days, the oscillations that you were mentioning because that would seem like a really good thing because temperature is interesting because it can go fast and it can go slow and it can go creeping, and you have this sort of, like, existential dread potentially, and then you have a bold snap. So in terms of from my side, I suppose, I’m talking about dynamics because I don’t want this to kind of, like, you know, it put it another way. It seems a good opportunity to suggest that the tree’s life, far from being this sort of, like, aggregation of slow things that befall it, it’s far more dynamic than you might think.
So a good day is actually not much happens. A bad day is far more dynamic. And the personality of the tree is potentially this tree that lives in deep time and really just wants to be the kind of, you know, sort of not flat line, but it wants the seasons to flow as the season should, which gives us an opportunity potentially to talk about climate change, to talk about, you know, the world is far more busy for this tree than it might like. And, again, reading between the lines and the fact that there is an idea which runs through all of this project, which is that Hyundai are interested in this idea of long term stewardship and how Free is potentially very much up for that because the world is far busier than it would like because of these oscillations. So the fact that a good day, very little happens.
A bad day, you have all of these oscillations that, you know Yep. And then we have a boring day, which is, similar to a good day potentially. I mean, I wonder if a good and a bad day might do. A boring day is kind of like this kind of because I don’t think the tree actually was at if that makes sense. It’s quite good, with consistency.
I could do you another day that’s essentially this kind of, like, a day in the life, if that makes sense, as opposed to these paradigms. You’ve got paradigm good, paradigm bad, and paradigm average. So rather than boring, which probably is is unhelpful, this idea of, you know, what’s going on, what happens on the average day, and that could be far more sensorial. This is what I hear. What I this is what occurs around me.
Maybe that maybe that one day in the life. So good, bad, average day in the life. Maybe a good starting point would be to use March 21st, the first official day of spring, as your day. Okay. Let’s see with that.
Just because that’s the launch of the project. I’ve I’ve had a thought as well, but I don’t want it to I’m conscious that it could, again, spiral into a 100 different variations. Going back to, like, primary school plant, could we also structure it as plant wants 3 things? It wants light, water, and nutrients. We’re measuring kind of soil health or or the kind of soil capacity for those things.
We’re measuring the amount of water that’s in the atmosphere, and we’re measuring how much light is available. Could we similarly do tree’s response to, I’ve got loads of light, but no water, or I’ve got loads of water, but no light, or I’ve got loads of this and let know that. I’m realizing that’s a matrix of, like, 9 different possible combinations. This is the con this is the comfort model we’ve been we’ve been saying that Yeah. These 3 will have a different combination of those in an ideal range.
And, I mean, we can find find that out for the for the silver fear and everything. The other reason So it’s it’s good good, bad, and neutral is enough then? I think yeah. Map that to to to our numbers. So a good day is a good day.
And for history, a good day will mean different things in terms of data, but we can do it in reverse. Like, Dan writes about a good and a bad day, and then we tap into that to when we get the right data Yeah. Or the or the wrong data. I would that was what I was gonna suggest. If I write these kind of models, you can map onto the models the more specific things when you have the data.
Lawrence, when was the day of spring? 21st March, did you say? Yeah. Yeah. Okay.
Would it be possible? International Forest Day. Well, that’s delightful. You know, they think of everything, these guys, don’t they? Is it possible to have some data for the 21st March?
Last year. 2020. Yeah. I mean, obviously, obviously, this year, everything’s gonna be on fire because that’s just our luck. But, last year might be quite helpful to just sort of just so I’ve got a sense of what’s going on.
So this is last year from from January to end of February. K. It’s just the sample I I I built yesterday. And this is the like that weird dip I mentioned. What does it mean by 2 meter?
It’s 2 meter from the surface of the earth. It takes into account, some wind effects. But you can also get the skin temperature, which is the ground and the heat and the high ground as well. It’s based on my experience, where my head is, Lawrence. Ah, okay.
Okay. I wouldn’t understand that. Mhmm. So that’s Chris For you, everything’s a bit warmer. Mhmm.
So done by looking into this, I think one can can even imagine a story of that tree just basically the intensity of this value, especially if you correlate them with with other charts that could be overlaid that is, solar humidity, and wind speeds, and, sun sun radiation. And then we have a list that we asked, we asked an LLM to to help us with this. So we got a list of pollutants of anthropogenic emissions that we can measure from the Copernicus dataset. But we didn’t know what each one means, so we got a list from, GPT telling us what kind of harm its does and for like black carbon, which would come from a wildfire that can sit on the leaves of the tree and block the photosynthesis because the the leaf will be less efficient capturing, capturing, CO 2. You got we have other things that literally destroy the cells, the cellular structures, or you got other things that, alter the microbes on the soil, and this, in turn, infringes the tree.
And we got a list of all that, and we will give this to you. Okay. I mean, with regard to the mapping over the top, if I’m I’m obviously taking a small, dataset to make up a a day or kind of, like, you know, alluding to things, and then it can be mapped over the top. So if I don’t use all the data that you give me or I don’t allude to it all, that’s not to say there’s not space, obviously, for it to be mapped on. In terms of what I need, I probably you know, any data that you’ve got that can be overlaid that I can kind of read through is really, really helpful, and that’s probably in the main document that’s online already.
But if not, if it could be there, so I could just have a read of it, sort of read down what’s going on in this this particular week in the run up to spring as of last year, but with a kind of idea as as Lawrence was saying, kind of like, you know, the the story of it. I think that’s a good way forward. So we’ve got good day, bad day. And in those, perhaps, I can allude to the pollutants. You know, it was a bad day when black carbon blocked out my ability to photosynthesize.
You know, we’ve all been there. Great. That’s a the I think we had a comment, on on the on the stuff you you produced. It’s not really, talking about the tree as an alien species that doesn’t necessarily have names for things, the way we do. Like, carbon, something we we if it’s a name we give to this substance.
So for free, black carbon might mean actually the smell of its siblings, because it’s literally burnt, burnt wood. Yeah. So I’m I’m I’m you you know where I’m getting. Like Yeah. No.
I do. But there’s a kind of I think there’s a I mean, I was talking to Lawrence about this last week or maybe this week. Who knows? This idea of a difference between kind of, like, we we need the tree to be articulate. That doesn’t mean it’s it’s able to you know, it’s it’s it’s generally articulate, but that doesn’t mean it knows everything.
So, you know, I didn’t I don’t think it’s helpful to kind of, like, give it an age, but, you know, we’re talking to a kind of, like, intelligent child, effectively. We’re not talking to a, biological scientist. So whilst it experiences things and it tries to convey what’s happening, I think it’s a good idea that it doesn’t have all the terminology. It doesn’t have all the words. The tree, as far as I’ve been writing it and as I understand it, is it’s reactive rather than proactive inevitably.
So thing it’s done to, and it processes. It’s essentially a processing machine, but it is there to be stimulated by things going on. And it keeps a record of the things that have occurred to it in its rings and in the story that it’s essentially telling for us. So I completely get what you’re saying, that it won’t say, well, this is black carbon. This is it.
But it will be talking about the effect that this has on it, and it will be essentially, I suppose, as with everything, trying to make sense of it. So it will say, you know, this happened before, or we have a memory of this happening, you know, when half the forest disappeared. And, you know, so I’m I’m worried about that or that this has some, you know, in the collective memory, this is a red flag. Words to that effect. So I think this idea of the tree having knowledge of time, the tree having knowledge of kind of, like, past events, and the tree having knowledge of kind of consequence, both that personal and also for the forest.
That’s the thing that’s been guiding me as opposed to this idea of it giving us some, you know, blow by blow account in technical language of everything that’s going on because this reading between the lines thing, I think, is important. But there is that balance. So I think with what I do and what you’re doing is trying to make that balance so that this can be read by, again, a relatively intelligent reader who’s interested in the subject and has picked it up. And it’s it’s you’re able to pass it, I suppose, if that makes sense, which is to say that I’m, sort of fulsomely agreeing with you. So, you know, that’s none of that is an issue.
I just I’m I’m kind of like this is my view, and I think I think we’re on the same page with it. Yeah. We had this idea that I think we never had before. Like, it seems the tree hasn’t doesn’t have eyes. It can only I mean, we know it can sense vibrations.
So we could don’t know. We can tell The wolf. That’s trees. Whereas other animals. Sorry.
I’m gonna stress you. Or if they do, they do it with our cloud. Sure. Is that better now? Yeah.
Yeah. You’re good. Better? Okay. Go again.
We’re saying that We we know the tree can’t see. Yeah. So we know that a tree we know from, from research that the tree can sense vibrations. So from the point of view of a tree, a human walking past it and a wolf walking past it might not be very different. They might have a different vibration profile.
But humans tend to touch trees, whereas quadrupeds don’t really tend to touch trees. Or if they do, they do it with their clothes because they want to climb or something. So a tree can identify another biological species by the way that that species affects the tree’s vibration or if if it tries to eat the tree some in some way or if it urinates, on the tree. So where I’m going with this is that the tree will ultimately have a different way of describing things that surround it from us. And I think there’s a a real beauty in this, like, literature beauty, because it I believe most people will will never have thought of that point of view of a tree, that that sensory Yeah.
Point of view of a tree. Well, this idea potentially of, you know, having once some senses that are not, heightened and some that are very heightened. Yeah. Maybe I go away and I have a look at the kind of the literature of writers who are blind or something like that. The fact that what is what would be, you know, top and what would be dialed back.
Equally again, I think within this project, as as the rights are in you know, as you’re dealing with the datasets and you’re dealing with the meetings. But in terms of the there are things that we can do that make our lives possible, and there are things that we do that poison our own water and make it very difficult to, you know, for for this going forward to connect with people. If I part of this is obviously, if black carbon covers all of the needles and it’s from your fellow burning nearby, I can write a whole, you know, treatment that deals with ideas of forest fire and holocaust. I’m not going to do that because I I don’t think it’s necessarily terribly helpful going forward. I mean, that would be a completely viable thing to write, but I think you need to keep that balance of, you know, I suppose, the balance of tone so that it’s not so bleak or it’s not so specialized.
And there’s a kind of, like, mid ground that we’re trying to tread here. And, inevitably, when you go back to the client, you don’t want anything that’s too, you know, in this zone or in that zone. You want something that the majority of people can connect with. So I can write about how this tree is essentially blind and goes through the world and responds to this kind of chemical profiles of things and the, vibrations of things. And I would like to do that, but it needs to be balanced out with this general kind of voice, which is approachable rather than potentially either missing key characteristics that a reader would expect to to too much of a degree or negative to some degree or just weird to some degree.
And I like weirdness, and I don’t mind negativity. But equally, we don’t want it to go one way or the other. So if I take the I’m taking long long way of saying, I’ll take all this on board, but some of this might be readable between the lines. Okay. By its absence rather than me flagging the fact that the tree can’t see, if that makes sense.
Yeah. So you might talk about things in certain ways where you think, well, that’s a long way around or that’s an odd way of saying it. But, you know, I’m not gonna, for example, begin the piece with, I’m a blind tree that lives here. The I don’t know. It begs more questions than actually it solves.
So, which is to say, I agree again with all of you’re saying, but it might just sort of, like, be filtered in such a way that this sort of, you know, is less what’s the word? Anyway, it’s not so in your face, I suppose. But yes. I agree. Yeah.
Of course. I I I enjoy this discourse, right now. What what I’m thinking is that it is not better than the voice of a robot. George George occasionally embodies an AI himself. I’ve been spending too much time with LLMs.
So Yes. I do like this discourse. Am I even real? I think there are some surprising facts, that are are worth a reader thinking about them. Like, the the tree does not see the sun.
It actually eats the sun. So so some some inversions, of what the sensory input means for us and what it means for the tree. Mhmm. Are not a scenario. Agree.
It must not be about that. Yeah. There has to be a bigger story to tell this. Elements of synesthesia too, isn’t there where it’s like? It’s it’s taking one sensation and describing it in another way.
Yeah. But almost at the I mean, this is a filter that you put over a main narrative, which is built on the kind of, like, the factual things that we need or the personality things that we need, and then almost that happens at the end. That’s almost the Instagram filter of the words that are slight tweaks to give your reader a sense that you’re dealing with this creature, this this, you know, you know, this this presence. That’s not the word I mean. A kind of, you know, this awareness, I suppose.
This this creature that there’s awareness. This this, you know, that perceives and eats the world and in a in a different way. And I feel that some of this can be, like, when we’ve got the main scaffolding of everything we want, I can then go over with this kind of, like, filter and go right. So I can tweak that, and I can tweak that, and I can tweak that, which makes it slightly weirder and more arresting for the reader whilst not losing the essential approachability that we need to make this fly as narrative. Does that make sense?
Makes sense. And I guess there are some aspects of our differences from from the tree that are superficial like a filter, and some of them might be more structural for for the piece itself. I mean, we’ve already discussed the way we perceive time and our our timelines and our general life expectancy, which feels like when we when you were talking earlier about the slowness and the the different way of perceiving life in general that might be aligned with the client’s, sustainability goals Yeah. Is this a key difference between us as species and the frame rate that we perceive information, essentially. Yeah.
I mean, I tried to allude to that with the kind of the 7 ages kind of idea, but I think there’s still stuff that can be worked on with that. Because now I mean, it was interesting thinking about this last night. We wouldn’t deal with an eagle. We’d deal with a Greenland shark, you know, which wasn’t known at the time that these things live for 300 because, actually, I think also that fundamentally underestimates how long a yew tree lives. We know it’s 1000 of years now.
So agreement chart, let’s say, is 400 years. So suddenly that we can change. Would our tree know about agreement chart? No. But there are ways of dealing with this, I think, and ways of giving that information.
I’d the word I meant earlier was consciousness. So we’re dealing with this alien consciousness, and I completely get that. And I think some of these interesting kind of, the way it perceives the world in terms of time, in terms of sense of threat, potentially, in terms of kind of its existential being in the forest, the sense of crisis, we deal with that, and that’s actually that’s absolutely cool to who and what this tree is. The other things potentially will be will be filtered, you know, towards the end, which is also my way of saying that whatever I deliver to you next, if it’s not featuring the things that you want there, hopefully, it will at least be in a state where those things can be added or swapped in or out more easily. Because, I mean, everything that I’m writing the reason I’m writing in these short paragraphs, the reason that I’m writing in a certain way is it’s essentially modular.
And the last three things I’ve delivered are modular in some ways, and they are just been moved about so that the flow of it is positive, negative, or the hero biography. It would be great if we can pick and choose in that way, and we can point the LLM to specific phrases even that Yeah. That it might need to employ or not employ directly, but actually just make reference to. So I think the modularity is really important. I agree with everything that’s been said.
I think it’s all really valuable. Speaking slightly more practically, we’ve got 3 I’ve got a list. We’ve got 3 sites. Are we planning on Dan, are you able to do this for each species of tree? I mean, we know you’re only coming to Czech Republic.
We’re fighting tooth and nail to not go to the other sites at the moment. So Yeah. I mean, potentially, yes. I can I would have thought I would have thought that 2 is better than 3 in terms of the quality of it because there’s only so much time? But I can you know, I’m not saying no because I think I think yes.
Czech Republic, do we have this is a different question. Do we have dates? Is it still 3rd to 6th or whatever that was? Yeah. 3rd to 7th.
Then you’re you’re joining us post are you not traveling before or something? Can’t remember. Then I don’t know yet. Yeah. So the the window of that is you’re arriving on 3rd and leaving on 7th.
Yeah. Is that okay? And the the dates haven’t been officially officially confirmed by client, but we have already booked accommodation with you in mind as well. Right. So have a look at your schedule.
I mean, we’re pretty confident that that’s gonna be the time frame. Like yeah. With with regards to the 3, I mean, I completely go yeah. We we need to try and keep the quality, and I think we’d be, yeah, willing to bring you in for for more more scope than we’ve originally talked about. But, I think we sort of we we kind of need all 3 to be on a similar level, basically.
Like, if if if we leave the Brazilian tree high and dry, we’ve got these really nice personas for the for Czech Republic. And, I mean, I sort of expect Czech Republic to be that much more informed because you’re gonna be going. But I think if we can make a best guess or at least get the material that we need as reference for the other two sites, I Well, I think let’s say I mean, practically, I think you can have hero biographies for the other two trees, but you have the majority of the data for the check one. Yeah. I think that’s something to aim for in the first, and I think that’s possible.
The other does that make sense? Yeah. Yeah. I I I wanna make sure that we think about it in a bit of a phased way. I mean, I think we are, but just to be aligned so also on on timing, on Monday, we’re presenting the end of this phase to the client.
Now what starts one phase and ends another is a little bit flex transient at the moment. But, essentially, this is our discovery phase, and in that, Czech Republic features more heavily. Yeah. So I wonder just to focus on what we’ve done so far and, obviously, have a look. We could go through all the notes in that document I shared with you, but maybe it’s just better for you to digest them as soon as when needed.
Would it be possible for Monday to have what we discussed for Czech Republic by way of general biography. Maybe, again, you tell us on timing. Okay. Yeah. General biography and and maybe, if possible, with timing.
The good, bad scenario thing. No. That’s fine. I can get you the main biography and all 3. Okay.
Amazing. And then I think from that point on, we look at Yeah. That makes sense. Out. Because otherwise, you’re kind of Yeah.
Yeah. To focus on that makes sense. I don’t have things going on this week, so I can focus on that. Are we, and then, I mean, I’m yeah. From a practical point of view, it’s sort of like, I’m keeping a kind of record of days so far.
So are you kind of in theory happy for me to have, you know, the rest of this week working on that to get that to you by Monday in terms of the day rate discussed? Yeah. Let’s we said we said 500 a day, didn’t we? Yeah. Yeah.
And so where do we sit right now? Wednesday. So 3 3 days because that’s updating their biography and then a day on each of the sort of good, bad average sprint. Does that work? I think that works, but let us, I think we also because we’ve had all these other things, then we need to look at the budget.
So maybe I’ll come back to you on that in a little bit. That’s fine. I think also if you could think about so there’s a few other things. Think about the idea of co authoring that we talked about before and what that might mean, in terms of your time and role and whatever. That’s a tree as in as in As in, like, yeah, article written by Tree and Dan Richards as one scenario.
Just so that we have that prepped for client in terms of Dan is happy with this. He thinks it’s credible, and he’s probably gonna need this in terms of In terms of time? Time and money. Time and fee. Yeah.
Yeah. Okay. I mean, from my gut, sort of I’m thinking about in terms of what of sort of getting stuff to you for Monday. Yeah. So that’s the I mean and also, I wanted to be transparent about about this at this point rather than, you know, waiting till the end of the week, take you down, you know, which obviously I’m in a great position to do.
So, yeah, that’s fine. I think in terms of writing the article and, you know, being in conversation with the tree, it depends on how long how many words they want. If they want a 2,000 word article and, essentially, it is a weird inevitably, I’ll be in that I’ll be in a headspace where it’s less weird, but it’s a bit of a weird commission. So in terms of this will take me x amount of time, I’m not sure at this stage. But since all of that is probably gonna happen after the Czech Republic Definitely.
I mean, this might be for for for the actual article writing, we might be talking end of Feb, early March Yep. Kind of through March. But it might be quite iterative because, ultimately, the client and the publication is gonna have to sign off on the final edit. Yeah. So it will be In terms of sorry.
Go ahead, Chris. It it will be you working with whatever our end system is, to to produce that, basically. Yeah. But on on terms of I don’t know. I I mean, I I’m, you know, and I know you weren’t asking for yeses or noes.
But, I mean, with all of that in mind, it’s difficult for me to say beyond writing these three things for you for Monday, anything else is a bit up in the air. But, yes, I’d be happy to do it. But I can’t That’s great for now. That’s great for now. Yeah.
Yeah. All we need is yes. You’re happy to do it. Yeah. But in in terms of how many words, though, I suspect we might benefit from you saying, we I think this is worth this many words because I don’t know if we either us or FCB can say that.
Obviously, the publication might say Yeah. I think on the the publication will will will give you a brief format, because and, also, it depends. They’ve bought some space. How big is the space? You know?
Do you want a page of text and a page of photographs? I mean, editorially, they do certain things, magazines, and they, sir, and they don’t do other things. So you need to be editorially led on that rather than me saying, yes. It’s 7,000 words, and it’s 7 grand from my side. That’s not what I’m saying.
But, you know, if I come in like that now, that doesn’t actually make any sense because it’s about placement. So I am going to make you make you build you. We are going to build the thing that the publications actually want and will print because, otherwise, this is a waste of everyone’s time. So it doesn’t make any sense for me to be too prescriptive now other than to say, yes. I think it can be done, and, yes, I’m happy to do it.
It almost needs to be reverse engineered from the publications. Yeah. Yeah. Well, I mean, maybe just an example of what you think we might get. So just to give you insight in terms of buying space, I think they’re hoping that they don’t have to do a lot of that.
What they are buying is just 1 page adverts that we can turn into an article. So a one page is what they’re buying. 1 page is generally, sort of 400 words. 400, 500 words. If they want photographs on that page, obviously, it depends it depends on font size and everything like this.
I you know, you will know that. But, depending again, it’s the layout and the number of words will be based on what is possible and what people want to print. I would say a conversation and interview with a tree, you would be looking at probably a 1,000 words. Let’s say that as a ballpark, which is double the length of the hero bio. It’s really a profile.
So if you think of, you know, when you have a magazine article and you have a day in life with or 10 questions for, these are the formats that we’re probably gonna be dealing with, and those are about a 1000 words depending on what the what you want as a what you want as a result of this. Do you want, sort of something that gives you the personality of the tree and its situation? Do you want something that’s more scientific and actually reads like a site survey? Do you want depending on what you want, it’s a different number of words. I mean, you know, if you want an expose on how the tree, you know, is bad with BDSM and consent, in the New York, whatever it was.
I think that Neil Gaiman article was about 7,000 words. You don’t want that. But, you know Hard to read. Hard to read. Hard to read.
But it’s it’s that thing of kind of, like, there is depending on what you want, there’s all the words in the world. You know? But I imagine we want to keep this pithy and on point because as much as anything, as a reader, you want this to be impactful. And so in that case, from my side, I would suggest that, also, the client would want that to be pithy and probably a 1,000 maximum 1,500 words because, otherwise, you’re dealing with a kind of, like, odd concept, and you the fear is you’ll lose your reader. So you need it to be bang, bang, bang.
You don’t want to to be too cerebral other than, as has already been pointed out, the strange use of language, which alludes to the fact that that you’re dealing with an alien consciousness. Which drags you to the website and the wider projects. Exactly. Yeah. Because all of these sort of things all of these are gateways.
If you think about them in terms of gateways rather than means and ends to themselves, all the way through, I’ve been thinking that, actually, the project is the project. And the fact that the MacGuffin is that we want the tree to be writing these articles, that’s great. But in and of themselves, it doesn’t have as much value as the project itself. So these all need to be sort of gathering the reader and directing them. So if you think of it as a big arrow, there’s no point having a 3,000 word arrow.
Does that make sense? You’re just kind of, like, the one as pithily as possible to get people from a to b to engage with the project to come away thinking, wow. Hyundai are a forward thinking sort of, like, zany company with a with an eye to, you know, conservative conservative ship, whatever, of nature. You know, that’s what we’re going. You know, they’re using technology in an interesting way in an analog world with trees.
Gosh. I’m gonna buy a Hyundai. So if you think of it in terms of an arrow, you don’t want it more than 1500 words. Blah blah blah blah blah. Sorry.
Not at all. Perfect. No. Perfect. That’s already super helpful.
So another couple of things then, to think about. I know we already spoke about, like, are you able to help connect us with journalists and things like that? Low more local writers, journalists, follow-up. And I know you were you were a bit like, maybe not. Do you know anyone that you could also connect us with?
That that has sort of expanded to local would be great, and we’re still trying to pursue locals. But if you know any cool, high profile, journalistic people, Can we drive them into the The person I was thinking of is Patrick Barkham who works for The Guardian. So he’s The Guardian’s nature correspondent, but he’s also written a biography of Roger Deacon, who is the the Wildwood chap. And he’s got and he’s got bark in his name. He does have bark in his name and and paps, which the trees would feel because of the vibrations.
And I yeah. And I think he is a good friend, and he would be interested. The danger with all of this is that depending on who you’re working with, the fact that this is a Hyundai project makes it suddenly you know, are we dealing with an actual article, or is it editorial? So he might have questions about it. But as somebody to sort of bring on board to make aware of the project, he’s a good one.
And he would be my first choice because this feels quite sort of, like, guardian y. You know, it’s not there’s enough going on with the project that it has a, serious, you know, it has a serious intent, but at the same time, it’s odd enough that you’ve got a couple of things going on where it’s approachable, again, these big arrows. I have his regarding it, doesn’t it? Yeah. I have these I have I have his details and can introduce.
Keep amazing. Thank you. Wait. What was the next thing I have? So the next thing to be aware of is at some point, depending on how we define your involvement in this, at the moment, it feels like your involvement could be massive or we sculpt it down or whatever.
I suspect they’re gonna want to maybe do an interview with you for a behind the scenes video. So just something for you to be aware of. They they definitely they definitely are in the Czech Republic, so wear a nice hat and bring your best. Yeah. You’ll be stood in front of some sort of nicer looking tree than the one that we’re gonna have the actual stuff on.
Yeah. Sure. And, you know, basically basically just say say everything that you’ve just said in this call because I think it’s all super valuable. We’ve recorded the call, shouldn’t we? Oh, I have recorded the call.
Also okay. Dan, it might help you for it might help you from for for what you’re right from now on, but they say that we don’t have to use the little trees. And as a matter of fact, the snow cover might be so high that the little trees might be completely hidden by 1 meter. No? That might be a good idea.
So they pretty much gave us a free pass on what tree we choose, and we can choose what whichever we want, whichever is at the best location for our sensors or works from a narrative point of view. Yeah. So don’t blame yourself on on the tiny tree that you saw on the image. No. This had been mentioned, and and I think a couple of the newer iterations suggest that it’s not, you know, your couple of feet high sapling anymore.
Yeah. Yeah. Yeah. Yeah. Do we have snow, I don’t know.
What do you call it? Snow cover is a data thing that we can We’ve got everything. That’s great. Okay. We we we’ve right now, I’m I’m downloading really high resolution, free canopy heights.
So we will have the exact heights of of the entire area of the entire Czechia, actually. So I think for most of the things you can think of, we will have data, at least for the Czech Republic. For Brazil, I haven’t started looking yet. No. No.
Sure. Some of them are global, but we might have less resolution. George, can you put somewhere in the in the Google Doc we’ve done the stuff in? Could you just put the links or or pictures or whatever is relevant? Clearly, don’t put everything everything, but just a few things so that we don’t Yep.
Blow blow down up. The final thing I had was just as you’re pulling that together and writing this final bit, if things come to you, it’s not necessarily an exercise that needs to be done, but it would be maybe useful to have almost titles or synopsis of articles that you think might be valuable. So things like and just on a general level, because at the moment, I’d have to bring up the document. The the client’s done mock ups that say things like, have you got you guys got any to mind? Of the articles they’ve done.
The the Jesus. Yeah. The headlines. Like, the actual headlines. Yeah.
They’re saying, like, I’m, yeah, I’ve got plenty of water, but some trees aren’t so lucky. Yeah. It’s shit like that. So if there’s anything that comes to mind that you think is slightly more valuable, in in in in a few respects. So, like, whether that’s what I said about you writing the article about the experience and the project or what a tree might write about.
Just in a general term, obviously, we need to then localize it and things like this. But if things come to you, like, what because I think, yeah, we need it would be great to give them that for their mock ups, actually. Sure. Okay. No.
That’s fine. And, again, you know, not my existential crisis. Fuck. I’m a tree. Help.
You know. I’m I’m covered in the skin of my siblings, and I can’t breathe. Yes. Exactly. So, yeah, absolutely.
I can do that. I’m sure. You know, I fear that my, you know, my chances of being a Bolshoi ballerina have passed because I I’m a I’m a fir tree. Yeah. No.
We can do this. Absolutely. And I think the idea that the tree is curious, the idea of the tree is, you know, trying to make sense of things is helpful in this idea of it would have a conversation because it’s a kind of questing force. So all of this as we are building it, I think, means that we have lots of possibilities. I think the idea of writing, you know, I I mean, some of them inevitably will be probably quite sort of, like, clickbaity.
It kind of, like, I have, you know, I have loads of water and other trees don’t you know, the here’s why. I don’t know if that’s it’s all possible. It’s just, you know, if you think that this is a kind of I’m talking to a tree, but I could be talking to whoever. You know? But if if if I imagine it as a person, then anything’s possible.
I just guess you need the big what I’m trying to say is the big who cares hovers over all writing. So I completely understand that we need to make this kind of compelling in some way. And that is quite a delicate and difficult and weird thing to try and do when you’re interviewing a tree, but I will bear it in mind. And if I come up with an telling, I will jot it down. That makes sense.
The way I’ve been framing it in my head is, like, what’s the article that The New York Times publishes versus The Guardian versus National Geographic? Yeah. I think I mean, it might it might be that Patrick Barkham at The Guardian has a conversation with the, you know, large language model instead of me. And then, actually, I talked to Patrick, and we kind of, like, you know but it would be quite an interesting thing to sit a journalist down with what we’ve got and then have them ask the questions of what it’s like to be a tree here. And, you know, what do you think about the fact that, you know, if you consider me the translator or I’m the kind of, you know, I’m this tree’s literary manager and, you know, and sort of like my my my guy over here, and then I just make sure that he’s not letting himself down or said anything embarrassing.
You know, you could have Patrick do that. So that would be a different way of approaching it where rather than everything going through my filter, I’m just essentially almost the, I act as some sort of interpreter or, you know, manager for for the tree, and you get Patrick to have a chat with it. And then that article is framed more as going forward in the Anthropocene. This is the kind of, like, witness statement of a tree that’s being profoundly affected by human climate change or whatever it is. That’s this is the first opportunity we’ve had to actually get it from the horse’s mouth or whatever.
That’s an idea. But, you know, this is a this is a month down the line, but I’ll keep a note of it. Sorry. I feel I’m doing a lot of talking. That’s why you’re here.
It’s all good. It’s all good talking. K. Now I’ll do some writing. Yes.
Right. Good. Excellent. I think I ticked everything off. As I said, I think, like, go I I think you’ve seen the notes that we all collectively put on that doc.
Yeah. I think we covered we covered some of it already. But if you’ve got any extra questions, you can either put it in there or you can text me, and we can do it like that. That’s fine. Unless there’s any one of the notes someone wants to talk about specifically.
No. No. I think yeah. I just yeah. As I was reading it, I kept thinking, like, this would be a really lovely for it.
Like, the wood Wildwood barometers stands out as, like, there are certain phrases and certain things in there, which, yes, we can have this hero paragraph, and we can have these 3 kind of states, but we could also have, like, a a little dictionary of terms, which are particularly reson particularly resonant or, like, oh, yeah. Okay. Yeah. Kind of, yeah. What would you call them?
Sort of glossary. Yeah. Yeah. Kinda like phraseology. I think at one point, I said, like, tree slang, like, sort of little terms that feel particularly relevant.
Yeah. Okay. Well, I mean, I’m happy to factor it all in, and I’ll probably be in touch with you, Laurence, on text and everything. But, I’ll go away and write this, and I’ll try and get it to you, obviously, before Monday. So I’ll aim for it’s now sorry.
I mean, Edinburgh literally is a man in a kilt just walking down the road, which is quite is quite unusual, but it just reminds me, Riley. I will try and get all this to you by Friday, end of play Friday. So let’s say 2 days on that. And that’s all that’s the 3 with an updated hero biography. It might go into Saturday depending on how things go, but that is my plan because this is kind of now what my this is what I’m doing this week.
Does that make sense? Yes. Then do you do you need any any specific data right away? Is this I would I would love the you put the oscillations of temperature. I would really like the sort of, humidity and water, sort of snow depths, things like that.
Can we can we extend the time frame to, say, 1st January to March 21st last year or, like, January through March so that we can say this is the window that we’ve looked at. Yeah. Cool. I mean but that that’s the kind of, like, winter into spring, I suppose, which makes sense as a kind of transition. Mhmm.
Okay. I’ll get as much as as I can. I mean thank you. I mean, inevitably, it will only be alluded to or there will be individual bits. I don’t want you to think that I’m just ignoring all of the kind of, like, massive amount of data that you’re giving me.
It’s just inevitably I need to cherry pick because there’s only so many. We will have to do it anyway because it’s not really about numbers. It’s about the bigger story, as they say. So And I guess sending you images of them with with the legend and, scales is enough. Yeah.
That’s really helpful. Alright. Nice one. Thank you so much. Alright.
Pleasure. Chris, Chris, can I get 2 more minutes of your time? Is it about his hat? Yeah. This is made for me by a friend.
Oh. Isn’t it? Bye, guys. Bye, Robert. Hello.
Is it about my heart? No. It’s not about your heart. Ain’t no conscious partner. God, it’s impossible to keep a lid on this project.
Isn’t it? It really is. It could like, the desire to make it as good as it can be Yeah. Has to be suppressed by a budget. Yeah.
And I think we need to go back to Atlanta with the whole journalist thing and say, look. We haven’t got a budget to activate the journalist. That needs to come from there. Yeah. I can’t believe it’s called marketing.
I don’t know the activation budget. On the journalist already, I’ve just put a thing in the Figma that can be a collection of media platforms. So I’ve got a general one of what we already put Patrick Barkam in there. And then I also have a list of platforms in Korea from a friend. Mhmm.
And I maybe have a lead on someone who can help, but I’m gonna explore that more. But, that’s something we can already start to then list out and put them through the top. So that’s helped. Well, I think and have a call with Elena today about and how we’re gonna pursue this and how we need budget to fund it because, ultimately, we can’t how much did we put in the budget for journalists consultants? Like, a grand?
Not very much. Yeah. No. If you will send it yeah. Like, we we accommodated for consulting, not for, like, them actually writing articles.
Yeah. Exactly. We can’t feasibly expect to go to Patrick Barkham and say, for free, could you write us an article that’s gonna go on? The Hyundai. The Hyundai.
Yeah. So that’s conversation we need to have. Yeah. When should I arrange this call with Elena to go through that, the travel stuff, and the what was the other thing? I I’m currently available between 9 AM and 11:30 AM her time.
Okay. So, your 2 o’clock onwards. Alright. And then I just wanted to touch on Dan’s budget. So 1500 was what we talked about.
Obviously, there has been a fair amount of work, and so he’s now saying, like, I’ll just take a couple more days to finish it off. How do we feel about saying, okay, 2,500 to finish discovery and to finish check? Yeah. If that includes because it’s 2,000 if 2,500 includes the check trip stuff and him interviewing and everything, then fine. We’re paying for his travel accommodation and stuff.
Right? Yeah. Yeah. Yeah. Yeah.
How much should we how much should we I don’t know if it’s phase 1? Yeah. I mean, not I’m just reading Alyanna’s message. Sure. Sure.
I mean, I completely agree. She sort of needs to just remove herself from meetings. Yeah. That’s the case. Like, she doesn’t have to join the meet all the meetings.
In discovery, 3 calculator 3 sat oh, 3. We had 1500. Yeah. But then but then we also had 2,500 for hardware specialist. For what?
Sorry. We also have 2,400 for hardware specialist. Yeah. Which we spent on hardware and not No. The hardware, we already had as well.
Yes, sir. So center we had sensor test, 5 grand, and then actual hardware, 10 grand. 10 grand. Could we could we turn to 3 grand then to include the check stub? Oh, I’ll ask.
Yeah. Or is he expecting a day rate per day in Czech Republic? No. He’s he’s not. He had actually originally been, like, don’t necessarily need to charge for that.
Okay. But it’s obviously what he does within that thing. It’s like so you can come with us. He’s gonna be I think he’s in I think he said he’s in Munich or something. So, like, we pay for your travel.
You come with us, and we’ve and we call it $3. But if then he’s also writing an article about that, which obviously needs to be talked about with FCB, then that’s obviously a new budget. Yeah. But, yeah, let me voice that to it. Yeah.
It’s getting it’s getting it’s getting tight. I mean, ultimately, if we have to start creeping into profit margins and things to get this over the line, that’s what we have to do. But for now Yeah. I mean, ideally, we don’t know. So now let’s try and stick to either what we’ve budgeted for, or we need to press Elena a bit for extra stuff because extra stuff is coming up.
Like, I’m I’m afraid interviewing Dan and Dan charging for an interview can’t come out of the budget for consulting. Like, that has to come out of their that has to come out of their film production budget and con and con for sure.