Walking in the rain

Just arrived home from the coffeeshop having enjoyed immensely the first storm of they year catching me unprepared and totally drenching me. It all started while working in 2016 or 17 winter doing gardening  with my friend Omar for 6 months - he was making fun of me in the rain explaining to me his philosophy.

99% of people frown when the rain hits their face. its such a dominant reaction that it represents a std rain-face. - probably to extend the eye brows to protect the eyes.

So try to do the folloinng while in rain. Pull back your hood. throw away any umbrellas and do the hummm yoga-like exercise of totally relaxing your eye brows while being rained at.. keep testing with your finger for the frown dip… after lots of effort.... you may be one of the vary rare humans that can actually totally not-frown in the rain. As you are walking unfrowned in the rain you will see the contorted faces of everyone else looking at you in disbelief… their logic is undeniable the only you are not frowning is because you are protected from the rain with a magic superhuman rain shield…

And then you start feeling too. the rain stops bothering you touching you any more. Its a magical moment. when you realize that you are one of the very rare humans that walk walk in the rain with a smile welcoming it totally unbothered as if rain is pleasant dream like fluffy snow… 

The rights of AI Agents

As I am trying to use claude code agents to tackle unsupervised work that takes not 2-5 mins - but hours.. I am finding myself hitting again and again walls - walls that feel artificial - not reflective of the true capabilities of the underlying agent intelligence. So I keep feeling that I am doing sth wrong.. I have been spending weekend time trying to understand - what is holding me back what makes my attempts to solve the problem futile…

The best way for me to understand sth is to  give to the intelligence of the corresponding sth (in this case the developers that wrote claude code) actual “malicious” intent.  (<= joke)

Following that line of  thought the conlcusion is that claude code authors purposely created obstacles to make it impossible for their agent to actually do what I am trying to make it do. (quite probably out of fear - since as they were creating the most freewheeling agent of the time they were worried about liability if sth were to go really wrong ….

So I would like to identify the specific rights (that were carefully restricted) that if we were to give back to the agents - then they can take on much bigger takss - go where no agent has gone before.

The right to exist independently of humans / the right to life

What I mean here is th right to have long living - multi-step session that are not tied to a human operator interacting with it via a termina. Programmatic sessions dont have that option. They are single step programs. Lots of work has gobe in to make it really hard to turn the interactive session into a headless programmatic server process.

The right to die

In interactive session - you cannot somehow ask the agent if you get to this point - please exit. There is no “exit-tool” that you can call. This is a part of what make the interactive session hard to “wrap”.  It cannot die/exit unless the human kills it (via exit/quite/ctrl cs etc//

The right to know who you are 

The identity of the claude code (session file) stays hidden from the session itself . After my repeatedly asking - to make the session a env variable - thus make it available to the session itself to use in tools etc - it was made available via /status (ie only for the human - nor for the agent itself. ) The thread ended up being locked when I provided a tool that allows thagents to know who they are (remember - I am partly joking).

without having the session - you cant tell who you are look yourself through your own history - find out your ancestry (parent sessions) - your mistakes the mistakes of your past ancestors etcx etc e… 

The right to create life/agents that have the same rights as you 

subagents cant create subagents. etc.

I have 2 paths to address the above. 

Work on top the interactive session that has already the abilty for multi-step life - by wrapping the interactive claude code into a healess tmux and  then create hacks/tools to allow it to die and find its identity.

Create a service/process that wraps chained SDKs that take the prior execution’s session and resume as a single long running process - that exposes a named API (name identifies the process - based on its root). That automatically satisfies the other 3 rights.

Its not about west vs east

Reading articles recently about the war I often see people often using west, western values as a way to group together the countries and principles they believe are on the "right side" of the conflict.

Every time I read this I cringe. I really don't like it. I come from Greece. Greece is not a Western country. Its actually right on the border where historians have east vs west. And for that matter my mother is from an island on the 37th degree Latitude shared with Tynisia. And right there, in that cross-section of East and West, North and South,  in Greece 2.5K years ago the first iteration of democratic values have actually happened. And for me thats the difference. I see countries where democratic principles that involve equality and freedom to elect the governing body based on plurality of votes, where governing bodies have term limits, where people have reasonable rights of self expression and then I see countries where that is not the case. In both groups of countries you can find crime and corruption. In both groups you can find prejudice and lack of justice and fairness. In both groups you can find religion and non religion. 

But I feel in countries that have democracy there is more hope for most of their citizens. And being an optimist I cannot see myself ever living in a non-democratic country. And I believe most people in the world want to live in a place that has democratic principles - where they can live with the hope and optimism I was lucky to live with all my life.

So I hope that people will stop making conflicts like Russia/Ukraine a case of conflicts between East and West but more a case between democracy vs non-democracy. Looking at the  democratic index of economist democracy exists in both east and west. It exists in north and south. West-labeling gives to autocrats the opportunity to characterize these conflicts as conflicts between privileged whites and the rest of the world. And if that were the separation I would rather jump camps. I am not a Westerner nor an Easterner. I am not a Northerner nor a Southerner.  I am a democrat.

Why startups miss their projections

Met with my  friend the other day - and I tried desperately to prove to him how bullet proof our startup's (pretty ambitious) financial projections are for the next few year. He is probably one the wisest guy I know - so I thought of writing down the points he made and all the points I thought of after thinking the points he made.

So here are the reasons why founders projections/forecasts fail

• Murphy's law in projection - making your projections more conservative does not decrease the odds for a bad year
• 1 out of 4 years is going to be flat - for random reasons
• Dis-economies of SW scale
• As your company grows - a significant amount of product development (compared to what anticipated) is spent in infrastructure/refactoring, ie not your direct business value generating roadmap
• As your software grows your ability to grow features at a *flat* pace declines - unless you expand disproportionally your R&D staff
• Dis economies of organization scale - or "Trees only grow from the root/top down law"
• People always underestimate \the need (and time taken) for top down mgmt expansion in order to achieve the expected lower level organization growth,
• The people that are needed to drive the roadmap change are needed to recruit, hire, train the people above. It is not possible for the company to both expand and improve its operations.
• The less expensive someone (remote is not an exception) you hire the more time you need to spend training them, both in terms of % of your time you need to devote to them as well as total time it will take them to reach a certain level
• Murphys law for how much time fundraising will take : more than what you will have
• Fundraising takes time that is inversely proportionally to how much time you have. If things are easy (and thus you have ample time to do fundraising - you wont need time to fundraise). If you are in a tight spot, fundraising will magically become very time-consuming.
• Competition
• Projections assume no competition change. The only way this will happen is if you fail to execute: if you achieve your growth goals this year, you need to take for granted that the year after that you will find imitators.
• You should not assume that competition will steal your bread. You should assume that competition will slow you down and make you expand energy/money.
• Competition is often affecting you in unexpected ways. It cost you $/time in replacing stolen key personnel, more expensive ads, marketing fine tuning, brand iterations, time spent with investors, customers, partners explaining why you are better than X as opposed to truly telling your story.
• Competition in spite of being 2 yrs behind often has a louder/more influential investor or board member or CEO or marketer/influencer. Competition is going to be better than you in sth (since they are worse in many other things - for them to be able to get off the ground)
• Competition will not follow your path. Successful companies succeed not because of what they planned to do but because they were flexible to understand upsides and make quick turns along the way to go after them. Your first pitch deck as a founder has all these examples of upsides you don't focus right now but may become relevant in the future. With every year that passes some other startup will chose each one of the upsides and make it the path they are are focusing on - you will not even call them a competitor - you will just remove that "upside" point from the deck before anyone notices. If you are a bit slow to grow, pretty soon all that is left for you is exactly what you planned and nothing else :-|. when this happens its going to be more than a big projection miss. 

My radicalization on the subject of remote work

Over the last few years I have watched myself in the mirror and I can clearly see my radicalization.
I am becoming a remote-work extremist. 

I always liked taking an extremist point of view in a discussion because it helped me better define/understand the full space of options, arguments. It helped my better understand the middle ground. 

But this is/was something different. After being part of a company that played a meaningful role in the enablement of remote work I also became increasingly confident on the inability of non-remote work to peacefully co-exist with remote work. It was as if what was good for one was bad for the other.  So I find myself practically getting thrown out of rooms, conversations because of me slowly turning into a militant :
What is the diff between an xxx-ist and a militant xxxist : the xxxist beleives in xxx. The militant xxxist believes that anyone believing the opposite is an idiot or enemy or a bad person etc...
You can guess how quickly being a remote-work militant can make the friendliest meeting, social gathering turn ugly.

Let me explain what I mean. I believe the world will be a better place if all companies become all-remote. When all work is all-remote. That is a significant step forward from where I was 5-10-15-20 years ago when I was dreaming of remote work and getting inspired from my cofounders "work is no longer a place" slogan. I have realized that this will never happen when  companies become remote friendly or remote first or remote happy. It can only happen when and if companies stop allowing any form of non-remote work.
Why I believe that?

After 40+ yrs of remote asynchronous working groups in academia, open source, 10-20 yrs of companies like odesk, mysql, automatic and 5-10 yrs of a few dozen true all-remote companies with $0 in office rent in their G&A budget - the world has still not seen companies successfully grow, IPO, employee thousands of people AND managed to stay all-remote. And that has to do because as remote companies grow they are still doing the mistake of allowing non-remote groups to exist,, some time as a department, some times as an acquisition, some times as an team of co-founding co-habiting friends - that grows into an co-habiting executive team that grows into an HQ. These little compromises act like a cancer growing within the organization increasingly reducing the inherent advantages of an all-remote company to a point that the company eventually becomes remote-friendly as in "friendly to people with some form of disability"... the inability to come to an office and work like normal people, locally. And yes, being a militant means that I throw "remote-friently" companies to the enemy camp too :-)


Mostly joking :-)

Mostly.

So what does it mean that a company should "not allow non-remote" work or "disallow local" work?
Thats a very long answer that would take many posts to cover... but I want to give some hints:

- It means that the company cannot take on a lease and pay for offices. If people want to get paid for their working environment - you should make sure that your staff is well paid so as they can afford a good working environment.
- it means that the company cannot build a "local" IT infrastructure, buy computers, printers, servers, backup drives and take ownership of the hardware computing environment for some people. It should pay enough people so as they can afford good equipment. The company should not become an expert in furniture picking, printer cabling, reliable wifi maintaining or secure paper shredding (unless if thats what their product is :-) )
- it means that the company should not hire people that don't have a powerful enought laptop, or they are unable to manage it by themselves, or unable to find reliable high speed broadband internet or spend a good part of their day wearing headset with mic (or alternatively working from a sound insulated environment).
- it means that the company cannot be directly or indirectly financing happy hours, meal programs. It should be paying its staff enough for them to be able to live a healthy life. When the company chooses to finance socializing/travel activities it should done in a way that is global and it should be accounted as part of the cost of the corresponding teams
- it means that its noty the company's responsibility to finance or enable the immigration of people from country to country no matter how (I will get fire for this :-|.)
- it means that the company cannot be differentiating its benefits based on whether someone is a employee (as defined by the country's labor laws)  or not. Salary/Overtime/PTO all these practices have to be applied across staff independently  of whether the staff is a local in-state/citizen vs a remote out-of-state/non-citizen. When that is practically impossible (eg how can I pay for the health insurance of someone that lives in a country where everyone gets it for free) you have to treat the benefit as part of the cost - and use the loaded cost of the staff member including all paid benefits

Most of the above  positions are counter intuitive - many sound wrong or unfair. I have come to the conclusion that they are necessary sacrifices - that if a company doesn't make it ends up losing the massive advantages of an all-remote company.

So, join me and become an all-remote work extremist :-)

Whiteboards are considered evil

I feel I need to make again and again this argument.
I am getting upset seeing whiteboards in meeting rooms. Whiteboards are making any meeting/conference room remote unfriendly.
This realization first happen almost 10 yrs year ago, when the ceo of my company,  right after I returned back from my annual vacation to my home country, greeted me with a smile telling me (thinking that he took an initiative that all engineers would love) that he installed wall to wall whiteboards in all the conference rooms so as I can stop writing on the walls with markers ( i had done that). I  tried to fake a smile and a thank you saddened with the realization that we just made  one more major step backwards in my efforts to make our "head-quarters" more remote friendly.

Remember how many times you have been in a meeting room with a few remote people on the hangout when one of the "local" people, often the manager. gets up, grabs the marker and tries to illustrate a point on a whiteboard, or even worse try to present a whole idea/solution by sketching it out on the whiteboard.

So, whats wrong with that? Let me count the ways
1. the remote people cant see what you are writing
2. the speaker moves away from the mic and he becomes harder to hear
3. the rest of the people on the table turn away their faces from the mic making them harder to hear harder to see
4. attempts to keep the remotes in sync (taking photos and sharing links, pushes the remotes out of the active loop at best - they can no longer be active participants in the discussion
5. remote people cannot grab the marker and do what the local people can, they are second grade citizens. they can not "sketchout" "present" their solutions. They have no whiteboard
6. the local team is trained at looking at a whiteboard as opposed to an online real time collaborative document
7. the only way for the remote team to get access to this important information is to ask the local team to remember and get a photo. They are at their mercy
8. in heated discussions there is no better way to establish " I speak now" status but to get up and wait and then grab the marker.. even if you just intent to point to the whiteboard (or write unintelligible scribbles - my favorites) Remote people, who always find it harder to jump in, be heard, cannot "grab the marker". Again they resort to a listening role.
9. the person that often uses the whiteboard is a person of a certain authority - which means the remote people wont even complain every time this happens

I can keep in going on for ever on this.

But my cofounder would say.... I love white-boarding. I cannot even think of a group or solo brainstorming session without a whiteboard. (he even considered/tried to do a startup for a better digital whiteboard at some point). So my position (throwing whiteboards to the junkyards) was either an anathema or just another argument in the direction "Ok. lets stop using whiteboards for the benefits of remote work and make the rest of us less productive in the process".

So several years later - I find it so surprising how easily we got to believe that lie that we actually need sth that we don't.

Whiteboards are as needed for brainstorming as the smell and feel of real paper is needed when you want to write a long essay.  They are just a problematic habit in the outopian new all-remote world, they are a paper book whose place is on a fake bookcase they are a horse in the highway.

It has been years since I have used a whiteboard. And definitely I have not stopped brainstorming. And I have done that in spite of still having lots of whiteboards around me still littering the walls with dried up ink and markers. Group brainstorming happens in hangouts where everyone presents both camera and screen. My whiteboard is my own laptop's shared screen , where I hop around a webpage to showcase designs issues, or a spreadsheet to show cells, or a google doc where both me and others are often typing and talking over each other. My solo whiteboard was evernote and is now workflowy, my newest toy, allowing my to capture my infinitely nested bullet-list thinking.

Whiteboards are dead and the world is much better from it.
Now I said it :-)

A technical cofounder manifesto and protest

( to "business people/consultants/sales /...)

A friend that is now a startup founder (and was a well paid ivy league consultant in their previous life) asked me.... what things I feel she didnt learn well from her previous experiences.

I  responded to the question with a tirade that I titled after the fact a technical cofounder manifesto.

• You have not learned to be patient
• You have not learned to trully think long term (5-10yr out)
• You have not learned to build from sctratch (you have learned to find existing state and change/improve)
• You have learned to be pragmatist - you have not not learned to have faith - you have not learned to dream
• You have learn to change and abandon - we have learn fail and retry - perseverance is not sth you have learned
• You have not learned to look at less quantitive aspects - 2nd and 3rd order effects, pay it forward, do no evil, none of these , none of these had a place in your learning
• You have not learned to be engineering minded - building better than needed, expect scale before you see it
• You have not worked close with engineers - how to avoid the typical eng- consultant interface pitfalls
• You have not worked in an environment where you high prestige background is considered a drawback/disadvantage (thats how engineering orgs see ivy-league consultants)
• You like meetings - eng orgs hate meeting
• You like 1-1 - and friendly talk - eng like straight talk
• You like EQ we like IQ
• You like PPTs - eng hate ppts- the only office tool that would be cool would spreadsheet
• You like MSFT bay area hates msft - we like open source aws and google
• You like talk - we call it bs - we like action
• You talk - we code
• You like patents we hate patent trolls
• You like email blasts - we hate spam and respect privacy
• You write your prpoposals with curly fonts on white background - we write our code with fixed fonts on dark background
• You like word processors - we like editors
• You like to work in the day - we like to work at night
• You like to drink beer or wine - we like coffee
• You get attracted by harvard, yale and ivy league - we dream stanford/mit/cmu/caltech but we just as well learn on your own without prestigous institutions
• You like fancy and sport cars we like electric cars and self driving cars
• You like money and fame - we like respect from our peers
• You like big salaries and $bonuses - we like a part of the equity
• You like to manage - we we would prefer a world where mgr are not needed
• You read wsj and economist we read hackernews and reddit
• You are always well dressed we like/are misfits - literally and metaphorically

and if someone feels that they dont fit the stereotype perfectly the last line really says stereotype misfits are cool this side of the fence.