Questions, Claims, and Evidence: An Introduction to the Discourse Graph Extension With Cortex Futura

thank you all for being here my name is ramses out i run rome stack which is a knowledge hub for rome power users today i'm joined by lucas and hope i pronounced his name a last name correctly cabarau yes oh i should have asked before sorry and we're we're going to talk about the discourse graph extension for rome so um if you are not really familiar with the roam extension um ecosystem it's possible you you can extend the possibilities or the capabilities of rome research with plug-ins or extensions uh one very popular repository of these extensions is rome.js built by david vargas and then um i think he collaborated lucas correct me if i'm wrong with jill chen to create this extension which basically helps you to digest academic papers to organize the questions that arise from the from your your reading activity it will help you to organize the claims that are made and then to find the evidence um to support those claims and i know absolutely nothing about this extension so i'm really happy that lucas is here lucas is also very important voice in the rome research community uh he has been creating beautiful courses actually i've been going through site to write version one today with and i've been learning quite a bit actually even as a two-year almost two-year user of rome research and today we'll be having a look at the uh extension that lucas will give a lot of attention to in the next version of side to right which hopefully he will also tell us a little bit uh about today so lucas enough talking from my side please take it away awesome thank you yeah and and thanks for for having me on i'm really excited uh to be here um and and to share um the extension with you all the really really cool thing um about this and the course which i can talk about later um as well is that this is actually part of a real research project right so um joel built this extension as part of his um work as a professor for a human computer interaction and he's really interested in discourse graphs and and that sort of thing and um he used grant money to pay david to to build this to to his back and we're actually going um to do over the next couple of weeks is that we're going to collect feedback from people um who use the extension and then feed that into actual research output um that that joel can can use so this is applied uh field work human computer interaction studies plus the awesome benefits um we all get just from using it so it's kind of a double bonus that that we get um here and so yeah maybe um a little bit of um background um on on me i'm lucas um i'm in germany i am just finishing my phd political science plus cyber stuff and i've been using rome for almost two years now and it's been kind of the the cornerstone of my work um during the the last um months of my of my phd i'm really happy that i had it and um the extension sadly came a year too late so everything you see uh today will be um current stuff i'm investigating looking into kind of um looking to to understand better so not stuff from my phd which might also be um good for you because maybe you are not interested in autocracies and cyber stuff um in whatever in a political context but we're going to do a little bit more of um widely applicable um stuff and so what i actually would would say that we um do is that we're going to jump straight in and here's basically um what we're um going to be looking at so i'm going to give you a little bit of a demo um and then later we can do a small interactive demo and the demo i'm going to talk you through kind of the the research loop as you start investigating um things whether they're academic or not does that make sense so far ramses any points yes you want to make the only point is can you please increase the text a little bit absolutely that is is that better for me that's perfect awesome cool um and by the way uh before we forget um we'll we'll start by looking at a question but um we're also going to do a little bit of an experiment right lucas uh exactly okay can we make a start with one of the questions you have so um please think of a question um no need to uh to prepend it with a question maybe you can prevent it with a research question uh but if you have a question for for lucas or for me prevent it with a question and then start to think about a research question we can actually start researching in uh during the session exactly yes awesome cool so um let me open up this in the sidebar so we have it as a reference and so as you do research or you're just interested in stuff which you probably are if you're here um you usually start with a question right something that you want to figure out um and so the example um for today that that i want to talk through is this question how can i increase my daily energy and motivation right complex page title and this is the first clue that we're using the extension is that questions question pages have as the very first element um a link to the question q u e page right and this lets the extension later know okay this is um a question that it can then um work with and when you have um a question um the first thing you do is that you look at a first source document a paper in this case a podcast um on how to or looking for answers to that question right so this is um a podcast episode um from the hubermann lab podcast highly recommend this really good um on the topic of controlling dopamine for motivation focus and satisfaction that came out i think a week or two ago um this is the the metadata um that i have from this through the zotero rome uh extension not a topic for today but highly useful um if you're interested in in that sort of thing and then i have here the the text and timestamps of the different chapters of that podcast and what's important to have as a frame of mind when you're consuming stuff is that most of what you're consuming is claims right so if andrew huberman who is a professor that's awesome says um caffeine increases uh dopamine receptors that is a claim right he says that's the case but that's not evidence right evidence would be the result of a study that showed this or evidence would be i have personal experience when i do heavy um fitness training the next day i'm tired right that's experiential evidence that also counts but it's different from if you do 10 squats a day with a hundred kilos on your back then you'll be tired right that first is a claim and depending on how strong you are that might be true untrue or impossible right so um that's kind of the the thing that we have to keep in mind that if you listen to a podcast you read an article um whether it's in the news or some blog post um usually what you're confronted with is claims and it's important to weigh what you're reading as a claim right but a cr claim requires evidence so how do you um get evidence um this is for example then a piece of evidence classified by the extension as a piece of evidence and the grounding context here is this is a study with 20 participants they did self-reports on how um caffeine made people feel and they did pet scans on how certain dopamine receptors um got activated um i'm not a biologist right so this is a huge claim like nothing of what you see here you should take as a medical advice or actually true because i have no idea what i'm talking about in real terms i'm trying to figure this out right this is why i'm using question claims and evidence to understand how dopamine and these things i know nothing about how they connect to what i want to to answer so we have um a question how do i increase my daily motivation and energy what can i do to to support that we've started um to look at an initial source which was um the podcast right and then we started um claims and i'm going to show you a couple of the claims that um hubermann makes right so cold exposure in the morning can help increase daily energy um by increasing the dopamine baseline throughout the day decreasing social media you significantly will increase daily energy motivating by increasing the available dopamine right so you don't deplete your your dopamine by all these reddit and twitter hits um to to your uh dopamine system right and so other activities have more dopamine they can they can kind of um excite you with and so all of these things are claims that um hubermann um makes so we've collected them and now i look for evidence in other sources than the hubermann podcast right and this is where the extension then becomes really really useful right because here um i basically just on my daily notes page for today right you see this october 4th um from 1 30 p.m to 4 30 p.m um i kind of did some some reading on and work on this right and i linked to this question i took some notes on the methods of this paper recorded quotes on the results right so this is a screenshot of the actual results um they they present um and then i formalized the result um into this this evidence page and i tagged it with um inform right so this informs now a question and this is the key thing you see here that i'm indenting right so this is the top level um question and i'm indenting a link to this evidence page and then i link to inform now if i go to the page here above the familiar linked references you see something called discourse context and you see it automatically pulls in this question is informed by a piece of evidence that caffeine increases d2d3 receptor availability right in some brain region which is awesome but what's even cooler is that if i go to this claims page right so this is again a page that classifies something as a claim i have here in the discourse context evidence that supports this claim right and so i um it automatically pulls together things that support that inform also that oppose right and all of that while being basically the same doing the the the same um amount of work right so all i did was copy in stuff but the cool thing is if i um type a new question here i select it i hit one shortcut um i type q and bam i have a new question right i indent as i would always indent within rome i select some new text right i type some new evidence shortcut give me some evidence now i have a new piece of evidence right and through the indentation something that we do in rome all day every day anyways and just by writing and hitting like two shortcuts i now can go in and say okay give me all the evidence that informs a certain question give me all the claims related to it give me all the evidence that supports or opposes a certain claim does that make sense to me it does but i want to uh to summarize it to see if i understand it so um what is extension basically does it gives you a few shortcuts to easily tag um pieces of text with different tags so for example for question it adds a q q-u-e tag for evidence it will add an evd tag so that way you can easily go through your notes and classify different parts as either a question you want to dig deeper into or something that's that supports something you're actually researching and then through the indentation it will basically act almost like a rom query where it will associate different parts so if you have a question it will pull in all the evidence and it will add some nice looking metadata um i saw some kind of attribute that it show probably something cosmetic but it looks nice that it it will show you okay this is actually uh this question um this is a question you have and here's the evidence you already have processed for this question right totally yes yes so i can picture this also not just as a tool for do it going through papers you could just use this for any question that you have like something very specific if you if you're if you're a programmer and you're looking you're looking at a specific uh library um to to solve a very specific problem like a question could be um is this library the right choice for to solve this problem and then you go and look at different use cases could be web pages could be anything but you turn those in like uh evidence and then as you go through that evidence maybe you find claims that say well here's a caveat if you use this library for this specific use case for example or this is something you need to keep in mind absolutely and the the really cool thing is that you can actually define your own uh grammar for how the extension um treats stuff right so if you're interested in um i want for example one thing i played around with is decisions right so i have a decision i have options and i have constraints right so i can say a decision can have multiple options but a an option has to um satisfy um a constraint right and so i can map all of these things together so the the extension is right now like in the default setting built for analyze texts and map their claims but you can totally extend it to whatever right if you want to map programming knowledge then you can do dependencies or you you could come up with satisfies use case or constraints in terms of hardware requirements software requirements what have you right you you can you can do all of that um as well um yeah yeah and by the way there is a question there are actually a few questions from the chat but um to tie it in with this uh melody asks where's the evidence uh being pulled from your own database question mark so i would say it's whatever you tag as evidence right exactly right so so here for example this is um a quote right this is how i um format quotes um here with um the the quote uh thing from from rome and if you read closely then you see here caffeine increases d23 availability what have you and where did i pull it from increases in where did i pull it from um in any case basically what i did was i had the quote i copied caffeine increases dt whatever and then i hit the shortcut and i made this copy-pasted text into a piece of evidence right so the evidence is not auto-generated in in any way but you have to make things um evidence you have to tag them in the right way um but what the what the extension does is it makes it really easy right it's literally just copy and then two two keys and and done right um and um in in a way what the extension gives you is default scaffolding for organizing your knowledge right you could come up with a system like this yourself minus the actually it pulls everything together part but what it gives you is it makes the the process of kind of actually using the scaffolding super easy i hope that helps as an answer yeah yeah and andrew asked what is the significance of the hashtag in form so i see for example here with the evidence i see hashtag in form why do you edit um so this is uh for um this is a one example you have some sometimes you have questions that are generated from something you already know right so some times you uh read um a piece of um of a study or whatever where you get it from you're like okay um that leads me to the question what do i do with this right and that's not support it doesn't support a question it doesn't oppose um a question right it informs a question right so this lets you kind of link pieces together that don't have any explicit um call it causal link and to what you're trying to do they're just kind of ambient information and this is what um the the extension allows you to to pull together by tagging things with inform so you have more of a granular control over okay this is evidence but evidence in the in the context of it it informs exactly it's not like okay um if if i drink caffeine uh my daily energy and motivation will skyrocket no that's not the case it's just like this is it's like a softer softer evidence almost yes exactly it's it's kind of the um the the ambient knowledge you have about a thing right where like okay based on all these little things i know if i think about them more closely i'll generate questions and claims and kind of letting you pull these things together and that give context to a question and that that's what the inform tag does but of course you don't need to use the inform tag you can you can come up with other stuff right it's um just one thing that that joel came up with um and that i find extremely useful yeah yeah by the way um when it comes to collecting um someone and i actually also myself triggered like how how did you get all those nice timestamps was that done manually did you actually write them down do you have like some kind of process that was um i don't do that for every podcast right that was um me really figuring out okay what um can i come up with a cool uh relatable example for things like this um but and the collecting actually was done by putting the mp3 file of the podcast into the script and then getting um a time stamped transcript out of that yeah that's what i did okay nice nice nice i mean we could spend an entire hour about collecting so let's let's not get too deep into it um and by the way i do want to remind people to to think of a question we're actually going to dive into but uh in the meanwhile uh if you're okay with it lucas unless you want to continue to show something else but there are a few more questions from the chat um i might want to to show one one more thing um sure all right for for people i'm not sure whether that's going to um [Music] go and help answer some some questions already but if we go to um playground energy and motivation this is something really cool that the extension also gives you and that is um open query drawer and now what i can do is i can find um claims where the claim has any relation to some sort of um question right so um let's say let's see if i actually have no no results um where did i let's find evidence where the evidence has any relation to the claim that um dopamine be careful about layering dopamine and now this pulls up all the pieces in this case just to the pieces of evidence that i have um related to this claim and now i can pin this and then i can add these to the whiteboard and i can start drawing um interesting things right so i can come up with a new question um what's the best way those aren't related but you could um now come up with new um connections in a visual way right so pulling together all the evidence all the claims um all the questions that you have um you can kind of have an interactive whiteboard inside rome where you connect and synthesize um and then what you can do is you can actually share this with others right you can export it and you can send it to me and i can import it and then i have um all your thoughts about how certain questions um and evidence um and that sort of thing are connected to each other um so it's not just i have to remember kind of what questions do i look in the linked references section of the discourse i can also have a really powerful way to query um for things even with has an attribute with a value whatever right so it's um extremely powerful um for for that as well and especially for for interactive thinking i find things are um like white boards are hard to beat and this is kind of the the closest thing you you can get um digital digitally i found um and that kind of then concludes kind of the loop right we have questions we've looked at a first source where we collected claims and we looked at evidence from other claims we put all that into rome just by taking notes as we usually would and here we can do some generative thinking brainstorming and then go into synthesis and and write these these things um together that was kind of the final piece i wanted to show and now we can do questions and go interactive this is really cool so this really this really gives you the the canvas to actually work visually and to actually start to see the the relationships or build the relationships between uh between them is this actually what happens here you can export it i saw it two different formats so if i i see so but you can also generate roam blocks okay exactly so what what happens now is it creates um um a page um where it kind of translates and for some reason that didn't work in this case interesting it's still in beta the extension but i can show you auto-generated let's do there's a space in between from playgrounds um let's do decision making um and here you see this was like a playground canvas as you just saw but what did um then created was something that reflected the necessary indentation um that you would need to actually represent this as we um where we're saying um earlier that connects these um different uh different elements right so you you're not stuck with like a nice visual layout but then you can't use it because you later can't again filter what's connected to what but it actually lets you generate the the output that that roam and the extension would need to to use it more and you can export it in csv or json format here you can actually put it into if you export csv you can put it into different graph databases like neo4j for example to use it there or if you use json for example i could send you the json file and you could just import it into rome um natively um and use it yeah in your role graph i loved how it added that that block in between a post by because that way you can actually filter on a page or you can uh use it in a query so you you can v you can visually uh find all the pros and cons or or like the the limitations in something or possibly you can you can uncover possibilities uh on the canvas and then export it and then actually just by using some filters filter down to your exact answer based on the different variables so that is extra powerful um actually yeah it i don't know i think there was a question about it but it this could work really well just to as also for a decision making um system like a algorithm of thought and i know you're you're big on those uh but the these could pretty much be uh applied to thinking through problems and then using evidence to actually support your your decisions absolutely absolutely yes um and that's something i'm super super excited to to explore in in the near future is is how that could could work through the existing grammar through building um a specific grammar for for kind of instantiating these relationships between what's a decision what's a claim what's uh constrained what's an option right and then um going through these and and generating something that leads you to to a desired result [Music] that is really interesting i can i can see if this if this plug-in you said it's still in beta but if it if it matures and if there's knowing david maybe there's some some way to export or import your your grammar uh then it will be really interesting to see how you can uh maybe create grammars for different domains different fields and you can actually actually on on the roadmap um so what you what what joel and david are actually working on although it's a while out is that you'll be actually able to subscribe to certain um information from different people right so you can will be able to subscribe to my thinking about decision making or algorithms of thought get it into your graph it will actually translate my grammar into your grammar of what's supported what's opposed how um are these maps that's still a little bit out right it is an active research project as well um and as lisa says yes a grammar marketplace um that's basically what's what's going to come um and that's going to be awesome yeah yeah i see so many possibilities with this um also just in discussions you know within companies like how do you how do you make a decision how do you actually uh come come to a solution so um there are quite a few questions so i think a few have been answered because jonathan said to summarize would it be correct if i said that this enables you to automate all sorts of decision trees i think it it it does i i think the question is what automate um uh means in in this context but definitely make them explicit right um really precisely see okay this is actually what we're thinking about um and do so in in a way that goes beyond let's talk about it but actually make it uh something something explicit using natural language yeah so it's not automate that it goes through your graph and it's going to tag stuff for you uh you still need to put in the work so uh but it's it's it makes it easier it's like it's just like a template right if you have if you have a template but then with the grammar you can actually define okay this is like this is the the templates or the the the tags that i want to add to different parts because this the way it's set up by default is really aimed at uh research papers to dissect them and to to distill the insights from them but with the grammar you can build basically any workflow you can if you want you can even create some kind of content creation workflow with it i don't know if you would want to but you can like you can take things through different stages with with the different tags um let's let's stick with the uh academic for a bit uh jagar had a question and um he asked if you had evidence that supports multiple claims to use block preferences to the uh to make that one-to-many link or do you just add the link because i said some ranking um so yes so because um evidence is like everything is a page right so um this is actually a full page so you just link to stuff right so you link to um different claims and then you indent links to the pieces of evidence so you don't need any um block references or that sort of thing um you directly link and indent the links all right um and then from bruno there was a there was a question is it easy to find unanswered questions or questions with a single evidence is that possible with a query because i saw like a condition you could add conditions in the playgrounds um that's a good question i actually don't know um so for for me personally i haven't done this um what i would hey i'll ask joel and b what i would probably do is look for so what i do on the um question pages actually is that i um collect sources that i want to go through for a certain question by linking to the question um on a source page right so here on the hubermann podcasting i have this this thing is relevant for the question and it's actually to process for the question how can i increase energy and so what i would probably do is i'd write a query that um looks for a question um that has nothing that is to process to integrate or integrate it um it's probably how i do this but i haven't done an implementation check so um uh come come back to me with that in a week and then i'll tell you to be continued all right um yeah a question from nikki was quite a long one nikki do you want me to read it uh or do you maybe want to uh just ask something um i'm halfway through your brilliant site to write course um lucas and um but i'm a week into studying after like 30 years so all of this is new to me like it's just so much i feel like i'm in a fire hose just taking everything in and starting to build a reading stack i think that's what you called it um so i'm looking at this and going okay i think tagging things q-u-e or e-v-d or you know what are the links that i should be thinking about as i start now when i don't fully understand this yet that will set me up for later when i'm ready and at what stage like i'm just at the stage where you're talking about all the different stages of reading so inspectionals and topic all of that um so what are the best times to be adding these so um that uh depends a little bit on kind of uh where how you get your sources right so if you're trying to do it like you do so through zotero and yeah sure i'm i'm i'm thinking about do you get like a syllabus from a professor that you kind of have to go through or do you have a question that you're trying to um to answer for your own own research because um i think um if you have a syllabus um i would um that you kind of need to go through i would probably kind of make a page for that syllabus list all the sources and then tag them off and if you do it in terms of i'm looking for sources for a certain question i would basically do do this so that you have you start with your question page um and um you'll get this in in the course but i can also share it with with everyone on my public graph is that you build a smart block that kind of builds you these queries right and then on every source page you just tag okay this is to process or to integrate for the question the course i'm working on does that make sense roughly i know if i'm getting there um i've i think i've got both i've got loads of questions but i don't know which one i'm not like chasing one yet for a specific bit of study but i'm quite open-minded about it and i've got a reading list as well so i've got both okay yeah and so um i what i uh would do is as you um put um your stuff into zotero um and you then decide okay am i actually going to turn this into something i have in rome right um then on that page um link with to process for and then either a link to the course or syllabus page or a certain question page and that will let you automatically collect all the different readings for for that thing and and have it ready to go and then you have to kind of uh take it off one by one i would say that sounds brilliant and i've actually got like six modules or with different bits of research so actually i can say to process four module one module two so that's fantastic thank you so much sure let's see there were there are few questions about the playgrounds um [Music] let me see which one to select well you've already sh yeah my class so is there an example where you use the whiteboard to good effect can we see the whiteboard output and so you you've shown that that page with uh that actually had the output but do you also have the i'll show you um um joel's examples because um his are better than mine um so so so mine are still very uh work in progress uh let's call it um but i think um in autogenerative from maybe it was in a different graph um i don't have a really nice example that i would be willing to share right now i have some that i'm kind of happy with but that are kind of private so i can't share um but um i'll do that it's definitely going to come in the course but i'll um have those ready to share publicly as well give me about a week or two um and and i'll share um hopefully good examples um of of playgrounds and and synthesis pages publicly all right um so we've seen uh going from a whiteboard to an outline is there also possibility denise asks to see on the whiteboard the relationship that already exists between blocks maybe within your graph or is it only a one-way street that you can can export out of the whiteboard but not see the relationships and get them into the whiteboard that already exists well and that is kind of what the this uh query drawer does right you don't you can't um it doesn't give you a full graph view of all your questions all your evidence all your claims and how they connect it doesn't do that um but if you're looking for relationships to a certain question to a certain claim then you can more or less targetedly look for those by having the the select be any relation to right um that's what what you can do but there's currently no way to have kind of a full graph view of this is what you've connected with each other that doesn't exist yeah all right um and actually you remind me of it we actually have a graph you in rome it's not that pretty but you actually can go to a question page and then open the graph and then see what other pages are connected to it so so what evidence is connected to it etc so speaking to go back to the marketplace lisa asks um she she already pictures uh a grammar word for for example academic workflows and some other uh workflows for some other more generic writing workflows is it possible to load and uh to export for example your grammar is that already possible um yes i have two two next to each other and so um it's not pretty right now um that that grammar sharing but basically the grammar is um built by these indented rome block relations right that looks very arcane um and what you actually do is that you um the the builder is actually a visual right so you can um select like uh if um a block references the page that is evidence and that block is in a page that is a question then this is evidence this is informed by right this is how the grammar works um and what that generates then is these um different um settings relations and those um can actually overlap right so you can have multiple things that are called in forms um you would probably want to call them differently to be able to distinguish them but you i could give you all of this and you could do copy paste in your own graph and then it works right so um that that i saw something i saw something if you scroll up i saw zero one is that is that basically yes and that's like a different grammar find a new grammar exactly you can have multiple grammars um that kind of mean the same uh thing this is i think the standard thing that comes with the um extension and this is something uh joel shared actually shared with me i did copy paste into my own graph and that is now a more elaborate grammar for something that in for evidence that informs a question so that's definitely possible so that way you could have potentially have two grammars at once you just go totally to your page and then you click the one and then you have a different grammar loaded they actually work in parallel so and this is um just visual distinguishing i think it isn't actually switching oh okay um but yeah you can have um multiple grammars um and unrelated grammars so if i um [Music] i can add any relation right so um i can let's let's do this so i want to have um a decision thing um and it's going to be decision and then the content of the thing and shortcut is d add a node and now i have something that i've defined as a decision page right decision pages start with the easy and now i can say okay i want to let's do constraint as well constraint and content or i don't know him um and now i can say please give me satisfies just satisfies and now i say okay i have constraint that um um is related to uh decisions right and now i could start um drawing um this is a b page or whatever i'm doing this would be not interesting right now but basically this is what you can do right you just start adding um nodes and um relations and then you can define your own grammar as you please and they work in parallel all right i realize that we we actually don't have time to dive into a question but i think we i mean question to to investigate but i think i think i think this is already um yeah this is already so so powerful and it's it's still in it's still in development um what kind of uh road map can we expect um jonathan is asking because he he's saying well he currently uses keyboard maestro for a bunch of macros who basically he has created his own workflow but what's what can we expect more with this uh well i'm i'm hesitant to speak for joel here because he's the the guy in charge um but what i know is that um he's for now very keen to collect feedback right so he's really looking for you to try and report your experience right so um you can download the extension play around with it try to to use it and then let him know um how you you think that works we're actually going to formalize that in in the course um coming end of week um but um the extension is free available on uh rom.js right so you can just download it and use it and i know he's um working on a way to um to to make the export and sharing through export um better i know he's working um on um doing the subscription thing um um and making that a reality i don't know what the actual timeline is how he's hiring not hiring david to to work on this again this is financed through a research grant um as far as i know so um that's um maybe maybe a limit in in some ways but um yeah joel is super excited to to share this i'm super excited to share this and he and i were looking for for feedback um and yeah that's what i can tell you about roadmaps right you uh for me you cut out a little bit i know you didn't cut out but was my side um and i don't know if so i don't know if you if you touched on this but your course is so side to right um is relaunching on friday right yes and also what you said at the beginning um of the session it's also like an opportunity first of all to not only become a better knowledge worker to actually uh use what you what you take note of and then use it to produce insights and to to share your knowledge it's also to get to know this extension at a way deeper level and then what you also said at the beginning of this session um that the insights that are gathered also in the in the course that will flow back to joel is that is that correct did i exactly yes yes um so we'll have um on the the lessons that cover the extension uh we'll have um prompts where we're asking for feedback um and if you want and that would be amazing you can record like a quick loom video of how you do things um and if we um manage to to get a few people this has the potential and to be one of the biggest human computer interaction field studies um if if we do this right and quick note for those of you who've already bought cite right possibly you'll get the update for free so um this is uh there's there's no upgrade pricing or something that you have to pay for and something i'm also going to officially announce on friday um if you're a roman scholar then you'll get the whole course for free as well so um that's uh something where um i think we'll we'll get quite a few people um to to use joel's extension because there's quite a few rome scholars and i'm really excited to to make this available but for everyone else i think it's going to be an incredibly useful source of learning to use rome and kind of a structure process for for investigating things right of course in an academic context but i think as we saw with the example of me going through that that podcast and and the claims for like a personal question how i can increase my daily energy um it's applicable basically everywhere um that's what i would say yeah so uh if you're academic you're uh you're in luck because uh well there's there's a lot of uh discounts to uh you you can get uh but obviously from a monetary sense because you still need to put in the work you still need to go through the process yes this this extension will help you make it easier um to go through the entire academic writing process but a sales process you need to to uh hone it's it's like anything it's it's a skill so um and from my own experience i really also believe that these uh note-taking and these writing skills aren't just needed in academia but also uh as what i call we go from knowledge workers to learning workers so basically every knowledge worker um faces some like at least some unknowns in their work it's no longer you're you're hired for one role and then you will do the same trick every day for the next 20 years that time is over so we all need to always keep learning and i think there's a lot we can learn from the the rigor from the the academy so even though maybe your your manager isn't looking for citations or whatever having those citations and knowing that what you the insights that you got that they actually have a strong foundation that in itself can already help you a lot in your in your work so um i've put two links in the chat um and again i will be sending an email after um also with with the the link uh to both the discourse graph extension but obviously that is not as interesting as lucas's course because you there you actually learn how to master it um and you will be sending out some free email lessons this week right so if people sign up they will get a few lessons already hit the ground running absolutely that's that's the plan as we're approaching the end of the week i'm going to send out um a little bit of material um that the kind of hopefully is already useful even if you don't buy the core so you're under no obligation to to buy it at the end i hope i can um still be be useful in any case but i'll be sending that out and then on on friday uh we'll take this into the real world and make it happen yeah so uh and if you're if you're not uh if this is like this this use case it looks interesting to you but you feel like okay my rome skills aren't there yet um lucas also has a i think it's a free course right yes to to get started with a bunch of uh actual use cases so instead of the the the normal beginner courses where you have like this is the sidebar this is the context menu actually lucas will take you uh through a few processes beginning to end so that you actually can get started uh using rome for for uh being productive in several ways so definitely definitely sign up for his newsletter as well yes and if you want to um check out um the the free course and if you want to brush up on that here's a quick link um that hopefully works um and so yeah roman context totally free will be free forever um uh six contexts of using rome uh seven videos each taking you from beginner to you know what you're doing and if you have any questions anything i can do otherwise for you reach out on twitter i'm very happy to help in any way i can yeah also add cortex for tura you're you're you're on point with your branding man i love it lucas for me like someone sent me a dm in the chat it's like okay this was this was a fire hose of information for me but it was very useful and i i must say for me it absolutely is a fire hose of information as well uh just looking at the at the extension page and looking at the claims and everything i was like oh my goodness i have no idea where to get started i now have the feeling i know where to get started uh there will be a lot of tinkering from my side so absolutely i'm not a master yet after this hour again if you're uh if you're an academic or if you do feel like you should uh be better at reading and learning from papers absolutely check out lucas's course and i want to thank you lucas hopefully we'll have you on in the future anytime and uh thank you man um it's it's been so much fun um and um thanks for for having me on this this was cool um i immensely appreciate it um and yeah anytime call me um and and we'll do this again great thank you and uh thank you all for your questions your participations the chat was on fire again i love that so uh hope to see you all soon bye-bye see ya

Channel: RoamStack

Views: 355

Rating: 5 out of 5

Keywords:

Id: CrIMC0cZKFI

Length: 60min 32sec (3632 seconds)

Published: Wed Oct 06 2021