(This is a provisional tutorial to help get you started using the language)
This is an unusual project, in the age of so-called Artificial Intelligence. It's a project about knowledge--your knowledge, not that abstract knowledge of Mankind. How how can you compete with AI and with other people in the information age? It's a bit like going to the gym to get fit. No one can do it for you, but there are tools to help you. So here is a methodology with tools, to improve the user experience of learning.
If you're here, you're likely a programmer or an IT person, so you're looking for simple answers in code, as software. Software is much more than code, of course. On one level, there is a graph database here, but that's missing the point. There are plenty of graph databases, but people use them poorly. This is not that.
It's the usual story: garbage in, garbage out. Putting data and information into boxes is easy, but knowing how to find it again is hard. But, surely that's why we have databases!? Before search engines studied search more carefully, people thought they could just use logic to find data by Random Access. That only works when you know what you're looking for, but we're become stuck with that model. That's partly because we are cavalier about stuffing things into data models we feel are orderly and logical. Yet our thinking is far from orderly and logical later on when we're trying to find things again.
To make a good way of encapsulating stuff, we need to understand the process of thinking. Instead of trying to order information (with hopelessly ambitious ontologies), we need to think about how to connect the dots of our thinking for later retrieval. This is what authors and teachers have to think about when producing material. Everyone knows the difference between a good and a bad teacher.
- How we think is quite personal. If we try to make the ultimate database of knowledge, it won't suit everyone. No one can feed knowledge into you. It's more like tending a garden of your own thinking.
The usual way of working is this: stuff everything into a database as quickly as possible and then search everything from the database. The source data are quickly thrown away, and we rely on an often hastily thrown together archive. We even call these data warehouses or lakes--landfill. The user has to know what field to search in the database, an often how to write queries in a special language. It's quite far from how we think in the moment.
All this commodity thinking leads to a canned soup knowledge cuisine. No wonder we end up with a culture of soundbites and hearsay. If you want fresh organic knowledge, served up just as you like it, you have to put in the work of tending your crop yourself. Knowledge, after all, isn't knowledge if you don't know it. No one can know it for you, so it's up to you to curate it and organize it to suit yourself--perhaps collaborating with friends or colleagues (but only in small groups).
If you subscribe to the vision of replacing humans with "AGI" (Artificially Gathered Information), you'll be shocked and disappointed by this project. If you're a teacher or a writer, you might quite like it.
With SSTorytime, the source files are your main focus, and the database is just a convenient aid to remembering, because retrieval sometimes needs help. You will spend most of your time writing and editing your notes, written in N4L. You adapt the language to suit yourself, with a couple of simple principles to follow. Then you regularly upload your notes into the database and see how it looks when it comes out.
You start with a simple text file, in your favourite editor. Somewhere you like to jot down notes, but as plain text (not a special format like Word or Open Office).
- my notes # you give it a title
# and you can leave comments to yourself.
IF YOU WRITE IN ALL CAPS, YOU WILL BE REMINDED OF THE NOTE LATER!
Mostly you just write notes
" (e.g.) This is a simple example that illustrates the line above
The >ditto symbol of inverted commas has a special meaning
Other symbols can be defined with your own meanings, like >"special meanings"
You can also refer to the previous line
@mylabel foot (note) important concept! # will refer to this label below, defined with @
# english to hanzi to pinyin & back to (english)
hand (eh) 手 (hp) shǒu (pe) $THIS.1
# references are referred to with $name.position
$PREV.3 (e.g.) nǐ de zuǒ shǒu (hp) 你的左手 (he) your left hand
$mylabel.1 (eh) 脚 (hp) jiǎo (e.g.) nǐ de yòu jiǎo (ph) 你的右脚 (he) your right foot
You can save this as a text file. It's helpful, but not necessary, to use a suffix .n4l.
This file is already available in the distribution:
$ cd SSTorytime $ make $ cd example $ ../src/N4L tutorial.n4l
When you run this, you'll see something like this:
If you choose verbose output, you see more of what's going on:
- First N4L reads a configuration file that's called
N4Lconfig.inwith lots of customizations. - Then it reads your file and chops it into parts that are related.
- N4L thinks that each line is an event, or an item.
- If you out something in parentheses, it treats it as a relationship or an "arrow" that points from one item to another. You can define your own arrows, and the idea is to use them to find things more easily.
- If you use the "ditto" inverted commas under an item, you don't have to type it again.
- You can define special symbols like = >, etc in the configuration to automatically annotate words inside a longer piece of text.
That already covers a lot of possibilities!
Eventually, there will be tools for scripting the search in simple ways, because the most powerful ways to search are to use a programming language. You can see examples in the demos and proof of concept directory under src/demo_poc. But as the project progresses, you can use the searchN4L tool to play around with the result:
$ src/searchN4L notes
$ searchN4L -chapter notes $ cd examples; make # add a set of examples $ searchN4L -chapter chinese -explore meat food $ searchN4L -chapter chinese browse, arrows=pe,ph $ searchN4L -chapter brain -browse arrows=occurs,freq,role
You can play around with a prototype web interface. Install your data, then go to the src
directory and run go run http_server.go. Then you can connect by loading page.html or connecting to the
address: http://localhost:8080 in a browser.
mark% go run http_server.go Listening at http://localhost:8080
If you load the test page.html into a browser, you should see a webpage, something like this:
There are several text areas and several buttons. Not all of text areas are used with all buttons.
- Orbits yields a description of a "thought" and its cloud of nearest relationships.
- Paths yields a trail starting from a "thought"
- Browse,Previous, and Next, are used when reading through notes from start to finish in a systematic order, page by page.
To search for anything matching a substring, enter a search string into Thought. If you want to limit the search to a particular section of notes and any specific context terms, you can add those to reduce the number of hits. For Orbits and Paths, the largest number of hits will come from leaving Chapter and Context blank.
To read systematically, enter chapter and a set of arrows by which to select the kind of item you're searching for. For example, if you want to read notes, then enter "notes" into the arrows. If you are searching Chinese language with an emphasis on pinyin, then enter "ph,pe" or "ph pe" into the arrow ties text field and hit Browse. Use Next and Previous to walk through the different contexts within that chapter. If there is only one context, then
When you make notes, you should think about what you want to see when you look back at your notes. For example, suppose you are learning French.
- French phrases
petit-déjeuner (means) breakfast
" (e.g.) Je voudrais commander le petit-déjeuner (means) I would like to order breakfast
" (note) Don't forget to say please!
- Notice that you can use accents and Unicode characters freely.
- Notice that you can make intuitive short names for arrows like (e.g.). You can define what these mean in the configuration. More on that later.
- Notice you can define many different kind of arrows with different meanings, e.g. (e.g.), (note).
You start to see a pattern in the notes: usually, if you're trying to remember something, you want to see the raw thing, like the word for breakfast. You also want to remember how to use it, so you naturally add a couple of examples just after the item. N4L will connect these dots to show you related things later. But, more importantly, you don't event have to do anything with N4L except write stuff down. These notes are already your potential knowledge in the making--and this simple structure helps you to be systematic in writing things down. You will spend a lot of time just curating these notes, altering, editing, improving, and most of the value is actually there.
You don't learn French by putting it in a database. You learn by revisiting it, and by remembering relevance and context. Just writing the notes is 80 percent of the job.
- The N4L compiler can help you to find errors and make a good structure.
- When your notes become long, it's hard to keep an good overview.
- Once inside the database, you can present the information in different ways. When you upload it to a database, you can still find things quickly, even when you're not sitting in front of your text editor--perhaps using your phone.
From here, it's up to you how you want to proceed. If you're feeling perverse, you could add more languages:
- French phrases, and other languages
petit-déjeuner (means) breakfast
" (e.g.) Je voudrais commander le petit-déjeuner (means) I would like to order breakfast
" (note) Don't forget to say please!
I would like to order breakfast
# let's add Norwegian..
" (betyr på norsk) Jeg vil bestille frokost
# let's add Mandarin
" (中文意思是) 我想订早餐
Writing notes isn't all that easy. It takes a certain self-discipline, but it gets easier over time. Forcing yourself to start is often the biggest hurdle. The news is that you can drip new notes into your working files occasionally over a long time. You don't have to sit down and study for hours at a time. On the other hand, it's only when you do make time to sit and study that you actually learn.
Once again, the message is: writing it down is nice, putting it into a database is cool, but it's all wasted effort if you don't look at it yourself regularly. No one learned French by writing in their school book, or even by cramming for the exam. You only learn by using knowledge. It isn't knowledge if you don't know it.
It's not rocket science, unless of course it is rocket science.
--rocket science
rockets are powered projectiles
Rocket science in finance (wikipedia) "https://en.wikipedia.org/wiki/Rocket_science_(finance)"
HOW TO SPELL VONBRAUNS NAME???
Werner Von Braun (developed) V2 aircraft
" (developed) NASA early rockets
ASK FRIEND AT NASA...
Apollo Program
Mercury Program
Gemini Program ...
Space Camp movie ..