What you intend to do with your notes is the thing that motivates you to take notes in the first place. The act of taking notes can be enough when getting started, there needs to be something more to keep going year after year.

Sometimes I struggle with motivation to write notes. It takes effort. The gratification is delayed. It’s not that I don’t find it valuable (writing is enormously valuable), but that I can lose sight of why I do it in the first place.

The best way to remind myself of why I should take notes is by producing works using my notes. For example, writing an essay or a brief from notes surprisingly fast (maybe I shouldn’t be surprised at this point) and reveals new ideas and questions to explore. My notes are a treasure map with caches stashed all along the digital garden.

OpenAI GPT models (GPT 4 at time of writing) do not accurately or consistently parse or manipulate dates.

For example:

• Giving wrong year: when converting 2023-12-13 to a Unix timestamp and it returns the wrong year (2022 instead of 2023)
• Wrong timezone: when converting an ISO date, it randomly adds on hours of time even if instructed to use UTC midnight
• Giving current date: when asked to convert 2023-12-13 into a timestamp and it sometimes gives today’s date

Self querying by date using LangChain doesn’t work well. The default schema used for parsing natural language into the internal representation of langchain for querying a vector store does not work with dates because it uses the wrong type (it tries to use a dict but you can only filter using integers or strings).

To fix that, create your own “structured request” schema for use with a vector DB (like ChromaDB).

TASK_METADATA = [
    AttributeInfo(
        name="title",
        description="The title of the task, meeting, or heading",
        type="string",
    ),
    AttributeInfo(
        name="created_date",
        description="The timestamp the task or meeting was created formatted as an integer",
        type="int",
    ),
]

TASKS_SCHEMA = """\
<< Structured Request Schema >>
When responding use a markdown code snippet with a JSON object formatted in the following schema:

```json
{{{{
    "query": string \\ text string to compare to document contents
    "filter": string \\ logical condition statement for filtering documents
}}}}
```

The query string should contain only text that is expected to match the contents of documents. Any conditions in the filter should not be mentioned in the query as well.

A logical condition statement is composed of one or more comparison and logical operation statements.

A comparison statement takes the form: `comp(attr, val)`:
- `comp` ({allowed_comparators}): comparator
- `attr` (string):  name of attribute to apply the comparison to
- `val` (string): is the comparison value

A logical operation statement takes the form `op(statement1, statement2, ...)`:
- `op` ({allowed_operators}): logical operator
- `statement1`, `statement2`, ... (comparison statements or logical operation statements): one or more statements to apply the operation to

Make sure that you only use the comparators and logical operators listed above and no others.
Make sure that filters only refer to attributes that exist in the data source.
Make sure that filters only use the attributed names with its function names if there are functions applied on them.
Make sure that filters only use timestamp in seconds as an integer when handling date data typed values. If you need to convert them, translate the date into a unix epoch timstamp with UTC timezone and double check that it is the correct year. NEVER use the `eq` operator for timestamps, if the requested date is a single day, use the `gte` operator for the requested date AND a `lte` operator for the requested date plus one day. Be very careful with dates.
Make sure that filters take into account the descriptions of attributes and only make comparisons that are feasible given the type of data being stored.
Make sure that filters are only used as needed. If there are no filters that should be applied return "NO_FILTER" for the filter value.\
"""
TASKS_SCHEMA_PROMPT = PromptTemplate.from_template(TASKS_SCHEMA)

prompt = get_query_constructor_prompt(
    document_contents="My tasks",
    attribute_info=TASK_METADATA,
    schema_prompt=TASKS_SCHEMA_PROMPT,
)
output_parser = StructuredQueryOutputParser.from_components()
query_constructor = prompt | TASKS_LLM | output_parser

Unfortunately, this only works for a very limited number of cases because OpenAI doesn’t handle dates well.

The Lorca Model 1 is a romantic watch. It blends elements of several iconic watches that remind me of a time I never lived through. It’s a thoughtful combination of specs and details that make the watch charming and unique.

The dial is fantastic. The sword hands end in a fine point that gives the precise feeling to telling the time, which adds to the legibility. The GMT hand stands out nicely with its diamond head that matches the 12 o’clock marker on the bezel. The sector lines bisect the dial in a subtle way. The indices play a neat trick in the light where they appear darker than the hands and give off a slightly grayish-brown color that makes it match nicely with a leather band. The minute track markers add contrast to the indices, which further highlights the different color tones on the dial.

About the minute track—the longer I wear the watch, the more it’s clear to me that this is the most important element of the dial. In some lights, where you can’t see it as well, the dial looks plain, but in daylight, it pops out beautifully and brings the other elements together. The minute track is also why it’s impossible to place your finger on the exact combination of 1950s inspiration.

While some complain about the typeface chosen for the 24-hour fixed bezel, I think it suits the watch just fine. It appears like a superscript Garamond to my eyes, which I find pleasing and gives the watch a bookish vibe. The typeface and use of a diamond at the 12 o’clock and GMT hand also reminds me of a deck of cards.

The finishing is very good but not excellent. The applied index at 12 o’clock is ever so slightly off dead center with the vertical sector line. The date wheel comes across flat and paper-like (maybe the white needs some sort of texture to it?).

The lume needs some charging to be visible at night. This is probably a weird “me” thing, but I like to check the time when I wake up in the middle of the night. The Sinn U50 doesn’t need any special charging to last the entire night though there’s a lot more lume to work there.

Wearing the Lorca Model 1 fits almost any situation. The beads-of-rice bracelet dresses it up a bit (though not as much as the Cartier Tank), but putting it on a strap makes it look more casual. I prefer it on the bracelet, which also happens to be extremely comfortable.

Now that I’ve traveled with the watch, I can say that having a GMT complication is not that useful in practice. Converting 24-hour time to 12-hour time is still slow (for me). It’s easy to misread the hour when it’s late in the hour and there isn’t much space between ticks. The best way to read the GMT hand on the Lorca is to focus on the tip of the diamond shape, not the large plot of white lume.

The Lorca Model 1 deserves more attention. Of all the vintage-inspired modern watches, this one stands out to me. You can tell how agonizing the design process must have been when you inspect the tiny details all over the watch. By my favorite definition of good taste, the sum of all these details works beautifully together.

Making the most of practical applications of large language models requires creativity. It’s a blank canvas to be filled in the same way that early mobile application developers faced when a new set of APIs unlocked new possibilities.

I think it will take several years of creative exploration to find the best ways to integrate LLMs. A chat interface will feel competitively lazy.

See also:

• LLMs aren’t quite universal objects like computers but maybe they will be eventually or another API in the core operating system
• Chatbots lack affordances
• AI is the next great interop layer and I can see a world where it becomes more like python, gluing systems together

A recent analysis of 20 million research articles and 4 million patent applications found that distributed teams are less likely to have a breakthrough discovery than their on-site counterparts. They found that collaboration for remote teams was more likely to happen at later stages where technical tasks are more discreet rather than earlier during the ideation stage. The authors believe that this means remote teams fail to benefit from the shared knowledge of the team and therefore have less new, disruptive ideas.

Read Remote collaboration fuses fewer breakthrough ideas.

See also:

• Remote work creates fewer weak ties which is important for chance encounters and sharing knowledge
• Working in person matters at the beginning of a project
• Organizational support of remote work correlates with reported productivity

Here’s how to write permanent notes based on How to Take Smart Notes.

1. Write out the main idea succinctly and in your own words
2. Look for connections between this note and other notes
3. Ask questions about your understanding of the note:
   • What does this mean?
   • Why is it important?
   • What is missing?
   • What question does this raise?
   • What would disprove it?
   • What other possibilities could there be? If the same or similar note exists, how is it different? (Write a note about the difference).
4. Write a list of related notes and summarize the connection in a sentence (I try to link to the note inline or in a section “See Also”)
5. Review existing areas of inquiry and link them to the index of that note (structure note about a topic or question)

See also:

• Writing is thinking and linking the information is the difference between rote memorization and knowledge creation
• Writing down ideas can be the difference between emulation and replication because human knowledge is meme replication
• Writing notes and linking them together is accretive

If your investments are good only because of future optimism, you will have surprises in the downside case. If your investments are good without optimism required then you will be surprised in the upside case.

One way to avoid being surprised by the downside case is to avoid losers rather than pick winners. However, it might be just as important to protect against the downside by enjoying every bit of the upside.

From Howard Marks at Oaktree.

See also:

• Venture capital is different, you can only lose 1x your money but missing out on fund-returning gains is the real loss
• Startup investing is investing in options
• Consensus macro forecasts provide no value
• Investment strategies that only work in a bull market
• Skepticism and optimism in markets
• Zero interest rate policy (ZIRP)

Worrying about being sick seems to increase your risk of death. A recent study published in JAMA Psychiatry found that people with hypochondriasis (a psychiatric disorder characterized by a pre-occupation about having a serious illness) have an increased risk of death compared to those who do not have hypochondriasis.

See also:

• Everything arises in consciousness
• Stress and anxiety are cumulative

A business can get kicked off of their PEO or unable to use a PEO if even a single employee “overutilizes” insurance benefits. When it comes to health insurance this can mean one person has a severe illness or medical condition.

Why is that?

Since PEOs are an underwriting business that pools together multiple businesses to negotiate better rates from insurance carriers, a business within a pool can effectively raise the premiums of everyone else. PEOs make money by marking up insurance premiums so if premiums go up, they lose margin proportionally.

See also:

• PEOs offer better healthcare for small businesses…as long as employees don’t actually need it
• PEOs taking care of everything is a myth
• PEO was made up to dress up the image of employee-leasing

Most runaway successes we hear about often have some mythology of the moment the founding team unlocked exponential growth. Examples include Slack, Facebook, and HubSpot. The only problem is, most of these growth stories are not actually exponential but closer to “an initial period of quadratic growth followed by linear growth”.

In The myth of exponential hypergrowth from A Smart Bear, the author argues that this distinction is important because we need better mental models for how to achieve high growth. To do that, consider successful products and marketing campaigns follow an elephant curve (steap rise, followed by a drawn out decline) and the idea is to stack these over time. Only then can you create multiple activities that sustain high growth.

See also:

• The initial part of the elephant compounds, but compounding is unintuitive because the initial curve feels flat
• Stripe’s business applies an interesting strategy for growing with the elephant curve of startups
• The elephant curve also exhibits a power law in that most of the growth happens at the head of the curve

In business strategy you’ll often hear a competitive advantage described as a moat, but most moats are more like a long bridge. The moat is the thing that prevents others from easily replicating another business. A long bridge takes a time build and is effectively a moat if it’s impractical to catch up or simply makes it a schlep.

In a sense, generative AI is a bridge shortening technology opening up many businesses to greater competition. Of course, replication does not a business make, but what happens when a motivated group of smart people uses better tools to build a competitive offering in a short-bridge amount of time?

See also:

• The book 7 Powers shows that building large durable business has more to it than building
• Time is still the hardest to replicate advantage since advantages accrue to the leader and incumbants have been the leader for much longer
• Maybe managerial capitalism is the true moat destroying technology?

Sometimes the person that reaches out for a demo is not the decision maker. They may be a user of the product with a vested interest in solving the problem. They may be gathering information for others.

If your sales process typically involves multiple parties, some who you haven’t met yet, send a follow-up email that is easy to forward to the other people involved in the decision. This equips them with what they need to represent your product internally (at least initially) because you already wrote up what they need.

See also:

• Make it easy to say yes to making an introduction
• How to write a sales narrative
• Founder-led sales

If you do something more than once, write down the steps taken to complete the task. Chances are, if you have to do something twice, you’ll need to repeat it more times later.

Having a rule that everyone at the company writes things down when repeated has several advantages. The knowledge needed to perform the task or process doesn’t need to be rediscovered by the next person. Documentation is automation and opens up the possibility to fully automate it with software.

From The Great CEO Within.

See also:

• Automation reduces marginal cost of nonautomated tasks
• A list of tasks describes multi-party processes poorly

A simple way to deliver effective feedback to an employee is the use the like/wish framework. Say the specific actions you liked about their employees performance and then say which specific actions you wish they would have done differently (emphasis on specific actions!). This leads to productive conversations about how to adjust performance and clear up any misunderstandings.

From The Great CEO Within.

By far the best installation and running experience for using a large language model locally is llamafile. The entire model, weights, and a server are packaged into a single binary that can be run across multiple runtime environments.

It turns a complicated set of steps for compiling, fine tuning, and running into a single download and command sh -c ./path/to/llamafile.

This is ideal if you just want to try out a model to see how it performs. For me, it was interesting to see that my Macbook M1 laptop can get ~11 tokens per second out of the box and the llava model is good enough for basic tasks.

I hope they make more models available to try!

See also:

• This might make it easier for personal infrastructure applications where you want to avoid sending sensitive data over the wire but can provide the context for an LLM application locally
• Convenience is king, even for developers that might be used to yak shaving

If you’ve ever accidentally deleted something from a file and want to recover it using git you can search the full history for a string.

This will return any commit that includes the string in the diff:

git log -i -p -S "my search term here"

I’ve used this several times now to recover some previous code and prose I had written many commits ago.

See also

• This is especially useful not that I use git as the backend for org-roam mobile
• Working Copy is surprisingly useful for bringing git workflows to iOS

When building systems that require monitoring and alerting, a false alarm should be treated as a bug. If it’s not resolved quickly, it diminishes the usefulness of the alerts and trains responders to ignore it. When alerts are ignored, the potential for an actual incident increases along with the severity due to time to detection.

Thinking about false alarms as bugs is a handy idiom for defining the problem and the solution in a way that is easy teams to understand.

From my collegue, Eric Berglund.

See also:

• Analgous to “there is no software maintenance”, there are no false alarms
• Closing the gap between previous technical decisions and the current quality bar
• Talk then code

Being able to lead a project that coordinate multiple people, teams, and organizations is critical for career development. At a certain point, the limiting factor in one’s career is not skill, but the ability to do project management sufficiently well.

Project management gets a bad rap in software development. It’s associated with pushy, uninformed people that press deadlines.

However, the ability to manage a project well is a size-of-the-pie increasing activity regardless of role.

Not only can one contribute work individually but coordinate the work of many other people towards a common goal. Few things that are valuable can be done individually, no matter how skilled someone is.

See also:

• Running a pre-mortem is a way to identify risks before launch
• Collaborate at the contractual boundaries let’s people work together without stepping on toes and slowing things down
• How to do great work

Questions naturally arise when writing. Since writing is thinking, when we turn over an idea in our head, we can’t help spotting things we don’t understand or finding a topic we want to learn more about.

If you are in the practice of writing regularly, it’s the act of writing that makes you live curiously. Cultivating a practice of writing is therefore an act of cultivating curiosity.

From More people should write.

See also:

• How to Take Smart Notes talks about following your curiousity and letting the big questions emerge instead of trying to pick a topic for your thesis upfront
• Learning how to read shows that reading can also help us live more curiously when reading leads to connecting ideas together with existing knowledge rather than just consumption
• Surprise plus memory equals learning and there is no surprise without curiosity
• Curiosity is a technique for dealing with skepticism which helps us get over the obstacles of early work
• Business writing should always answer a key question

It should be repeatable, measurable, and profitable. Every product and business is different. This will focus on a B2B SaaS product.

Why inbound?

• Startup sales win rates decreased
• Running a startup during a recession
• Educating the market is expensive
• Lower ACV products < $20k can’t support outbound sales
  • Rule of thumb for inbound vs outbound sales

Sell the first $500k yourself to figure out what works

• Do founder-led sales
• Founder-led sales helps identify product problems
• § How to make your first sales before launching
• When should you make your first sales hire?

Position the product in the market

• When do you need to do this?
  • You have customers
  • You can identify who your best customers are
• Read Obviously Awesome
• Write it down and use it

Make sure customers don’t churn

• Customer success first then sales
• What does customer success do at a startup?
• Customer success overcomes technical hurdles

Build a playbook

• Making founder-led sales a repeatable process
• What is a sales playbook?
• Materials to support a sales playbook
  • How to make a sales deck (one big template, pull slides from it to customize)
  • 1 pagers
  • Testimonials
  • Media library (screenshots)
• Onboarding playbook

Generate inbound leads

• How to think about go-to-market
• Drive traffic to your website
  • How to do SEO
  • how to do paid ads
    • How to do keyword research
  • PR
  • Social
    • how to use twitter for business
• How to get referrals
• Capture leads Give users a way to get in touch (schedule a meeting) or stay in touch (join an email list)

Convert leads into customers

• Selling to inbound leads
  • Schedule a consultation is the top of the funnel
    • Enterprise businesses expect a consultative sales experience
  • Discovery questions in sales calls should feel consultative
  • Run the sales playbook
  • Run the onboarding playbook
• Self-serve
  • Is your product actually self-serve?
  • How to make a product self-serve?
  • Which customer segments are best for self-serve?

Automate using a CRM to turn prospects into leads, leads into customers, customers into successes

• Set up a CRM
  • How do you set up HubSpot for B2B sales?
    • How should you set up a sales pipeline in HubSpot
    • How to set up post-close processes in HubSpot
    • How to set up support in HubSpot?
    • How to track renewals in HubSpot
    • How to measure the sales funnel in HubSpot
      • Win rate
      • Top sources
      • Churn risk
      • Renewals
• Automate emails to drive prospects to schedule a consultation or sign up
  • Map out all of the customer touch points
  • Make a list of triggers
  • Write a list of N emails as a sequence
• Automate onboarding process - self-serve, sold

Chess players burn a lot of calories at the highest levels of competitive chess tournaments. Chess players can burn up to 6,000 calories in a day by playing a tournament (three times what a normal person burns in a day). A study of 21-year-old grandmaster, Mikhail Antipov, found that he burned 560 calories in two hours—roughly what Roger Federer burns in one hour of playing tennis.

It makes sense that intellectually demanding tasks leads to feeling tired at the end of the day!

Read The Chess Grandmaster’s Diet from NPR.

See also:

• Some brain functions might be quantum processes

Before COVID-19, the majority of B2B selling happened in-person and over email and phone calls (61%). After, only 29% of selling is being done that way and sales over video calls increased from 38% to 53%. Also, 70% of B2B buyers say they are willing to make purchases above $50,000 entirely remotely (27% would be willing to make purchases above $500k).

From Gongification of SaaS by Sacra.

See also:

• Being able to make remote work universal includes being able to buy and sell
• Zoom fatigue is still a thing but that doesn’t seem to inhibit sales
• At the same time, startup sales win rates decreased

An overlooked part of understanding information and not merely memorizing it, is to elaborate on the meaning of something you just learned. When taking notes, it’s easy to end up with a detailed list of things without actually understanding the content. Without connecting the ideas with what you already know, you can’t attach the ideas to any scaffolding that would be needed for generating new ideas later (or even recall them).

In How to Take Smart Notes, the author talks about the crucial step of pausing to ask questions of a fact or idea in order to elaborate on it. “What does it mean? What is this similar to? What is the difference between this thing and that thing?”

It’s the connections between ideas that creates meaning, understanding, and new knowledge.

See also:

• Human knowledge is meme replication and good explanations are how we learn
• Connecting notes together is the central idea of Zettelkasten and can help extend our working knowledge to a second brain
• Thinking better thoughts