We can't speak symbolically because that makes us inaccurate

We can't lament because we're not providing solutions

We can't provide solutions because they will not work

We can't refer to society or trends because that means we're generalizing

We can't speak without providing facts and evidence

We can't be expressive because someone will be offended

How then can we talk in this Kenya?
To get any conversation going in this Kenya, you must be limit yourself to a vary narrow topic, you must present the facts from confirmed sources, you must stick to dictionary and literal meanings, you must not betray any emotion, and everyone must agree with you.
The point of conversation in Kenya is not truth or knowledge.

It's consensus.

More from #LandIsNotProperty Mwalimu Wandia

I have said all I have to say on #BBIreport, which is:

1. It's a waste of Kenyan's time and precious intellect to spend it on the tantrums of malignant narcissists. We should be working where our talent and skills call us to, but we're doing damage control and #bullshitjobs.


We have to have a country where Ndii's talent is used for our economy, not on campaigning to protect Kenya from Muigai's latest tantrums. Or where doctors are employed and working, not doing locums or being jobless while Cuban doctors get cars. #bbireport


2. The political class is useless. It has no talent or skills, and it cant work. It is always looking for ways to make Kenyans work and performing rituals of power and relevance that kill enough of us to truamatize the rest of us. That's all #BBIreport is.

3. Our youth are are either underemployed or in #bullshitjobs, or being told to be entrepreneurs, only for KRA to invent new taxes and GoK to invent laws to protect cartels. #BBIreport attributes youth's problems to ethnicity and says solution is power sharing by the dynasties.

4. I don't think I can say anything more about #BBIreport than I've already said. The BBI-stans have no argument, and have been resorting to force, emotional blackmail and ultimatums. They have the force of the state but I have what they crave and will never give: my acceptance.
I will use this thread to explain the concept of "administrative bloat."

Administrative bloat is the employing of so many managers and supervisors, paying them excess salaries, while the people who do the actual work don't get employed or paid.

That is the government of Kenya.


Since the time of GoK's mother the colonial government, and its grandmother the British East Africa company, the role of the state is to control workers, control our work and its produce, and send it up the foodchain through chiefs, paramount chiefs all the way to London.

One of the cultural lies these parasistes have sold to us is that the only work that can be exploited is where there is a tangible product like tea or coffee. No. These exploit EVERYTHING, EVERYTHING. Health, education, faith, religion...EVERTYHING.

Elites are parasites. They can't work, won't work. How do they live, then? From taking away the work that we do. They use the state (especially weapons) to threaten us, and they get bureaucrats and managers to distract workers and push the loot up the gravy train.

That is why the less the elites work, the more bureaucrats the employ to protect them from direct contact with us "peasants." That's the function of civil servants. To steal our work on behalf of the elite, while conning us that we're working out of national duty.

More from All

How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like