The question that no-one in the job market dares to ask/answer

Everyone in the job market at some point or another is going to discuss an offer.
An offer is described in the context of a prospective hire.
The discussion around it is trying to pinpoint the right answer to the question:
 "for how much am I willing to join your company"

I think a more interesting question is instead a question that people don't ask (and would probably feel even more uncomfortable to answer)
 "for how much am I willing to leave my current company"

This "pivoting" of the question has the following interesting characteristics
 - It is only valid for people that currently do have a job.
 Thats ok - most prospective employers would rather not deal with people that never had a job or people that are laid off. They would be interested in talking to people that quit their jobs but people quite their jobs typically after having found what job to do next
 - It is also valid for people that are working on something - even though they don't have a job
 For any wannabe enterpreuneir  / not really getting paid startup coder, founder , cofounder or cofounding employee all these people that may be working but they don't really have a job this question still makes sense : it means for how much I am will move from my self-employed state to become (again) a hired employee.
 - It is agnostic about the new job : an offer to a person encapsulates two elements - the cost/benefit/opportunity loss of someone leaving a prior company together with that obtained by joining a new one.. Contrary this questions only capture one side of the equation and that makes a more useful/primitive knowledge piece.
 - It is an always valid to ask question: I can always ask and answer this question I don't have to be in job hunting mood. The answer to this question is the barometer of how much I like/I value/I benefit to staying to my company.
 - It is not a secret: the offers, how much you are being paid etc are all secrets protected with legal clauses in the employement contracts of employees. However, answering this question doesn't seem to betray any confidential information - even though in reality it does expose the true total $-ized benefit of the employee by their employer
 - People wouldn't dare to say it. In spite if the confidentiality clauses - people do talk about their salaries/offers etc.. Do they really open up to answer this question - hardly ever... It requires a significant sense of personal honesty to answer this accurately (and braveness to answer it openly)

It would really be cool if we could get people to think and answer this question....

Charging for free services that have computational needs

I love the way wolphram alpha does it....

Note that I have already done the search once - and got the right result ($17T approximately).
It seems that the service recalculates as opposed to serving cached results.
The service gave answers to several questions that seemed to require quite some time
(google+apple) gave me lots of outputs  - when I added every single mega sw tech company since the PC era that I could think of....

it then gave much fewer data points - still taking some time to answer.

So 

- point 1: the idea of allowing free usage but when your computation requires more than X - instead of , kicking the user out, penalizing everyone, making the user wait for ever, you provide the option to pay and use more resources...  

It reminds me the chess AI idea of greg to allow chess players to use an chess-player-assist program to play against their opponent (who also uses the same program...)  - and either people can pay to get more resources - run the alg at diff depth or both be limited by the # of funds that are being bet on the game - they can use the funds to get help - essentially reducing their potential earnout

 - point 2 : I really want to see the Market Cap of companies created the last 20-30 years as a % of the total over time since 1900 - do the same using the HQ address location etc... (e.g. bay area generating 2+Trillion dollars of market cap growth in 10-20yrs). 

The hardest challenge : finding jobs for people to do

An odd friend of mine told me several years as we were discussing about labor marketplaces - and whether the future holds a work-supply or work-demand challenge - that the problem is actually quite different.

He gave me a challenge :

I can give you as much money as you want (he meant it) to create jobs for them in the OD marketplace - but with one simple condition. The workers that get hired should get the impression that these are real jobs. You will soon realize that solving that problem in a scalable way is at least as hard as finding money in a scalable way.
The discussion continued with me saying things like

 - I will create jobs to translate every single wikipedia page in every other language
 - I will hire people to write ficticious jobs and then hire people to do them
 - I would post all generated issues in github as projects
etc..

He had a (possibly questionable) way to dismiss all of them based on the following logic :
 - the jobs produce no actual value in which case - people soon or later figure out that their jobs are pointless - at which point he would either say I violate the assumption above or he would say that the effort that would need to be added to confirm that a pointless job is actually well done is impossible or not scalable
 - the jobs produce actual value - in that case he would try to prove that the amount of value would decline over time - after million tasks/million people the value is too small - so he would use the prior argument  (pointless jobs) or that I have no scalable model to expand the solution for the society as a whole.

I think at the background of my odd-friend mind lied the idea that we are moving towards a matrix-like dystopia - where necessary resources (food,shelter,entertainment,health) are available for all - practically for free - as a result of the work of a tiny% of the earth's population (the "smarter" 1-10% of us) - still the only way to keep the planet in balance is to make the vast majority of people maintain the illusion that what they do for living is an important/necessary part of the value generation and attempt to attribute some reward <-> effort model to keep them do whatever they do - otherwise everybody wakes  up and realizes that they are in a sham world build by others and placed in front of their own eyes - and the system/society unfolds.

What brought that chain of thought - reading the title   Why we should give free money to everyone of this HN article... The article is on a different topic - but is still relevant - welfare / workfare are incarnations of the same problem as the one that my friend talks about.

How often do I google

I wanted to find out the distribution over the course of a day of a typical person's google searches.

A friend in the office said that I should look at real data ... I guess the only data I have was mine. I looked at my chrome web history - chrome doesn't have timestamps there...

Ok.. after a bit of digging I found it :

For a Mac user its

ithaca> cp ~/Library/Application Support/Google/Chrome/Profile 3/History /tmp

ithaca> sqlite3 /tmp/History

sqlite> select avg(cnt) from (

  select ddate,count(*) cnt from (

    select 

      date(last_visit_time/1000000-11644473600,'unixepoch','localtime') as ddate,       time(last_visit_time/1000000-11644473600,'unixepoch','localtime') as dtime 

    from urls 

    where 

      url like '%google.com/sea%' 

    order by 1 desc 

  ) a 

  group by ddate 

) b;

13.0923076923077

Interesting things there:

  There is a history file for each profile of yours as you would expect

  The history file is an sqlite db file! A very nice surprise

  The timestamps aren't unix epoch based - they start somewhere in the 17th century for some reason - so you need to do the adjustment (SO gave the right offset)

  I still can't believe that browsers treat web history as something disposable - no clue why mine is truncated 2-3 months ago. 

  A tool that is syncing the history file daily/weekly with a more persistent DB (from all my profiles) may be worthwhile

  On Average I do 13 searches per day (I was expecting that to be more given how frequently I use the chrome omni-bar..

  Reference to the previous post - Google is making at least $30/mo from me from my search activity..... 

If I knew R better at this point I would just show the histogram of these timestamps but given that I mostly use as a command line calculator ..... this this will take me more than a few minutes...

Think of this every time you search

Many people often have wondered how much money does it cost to google every time we perform a search. The Search that google will get everything perfect to respond instantly to produce 100 results, to search multiple times while you write the search etc etc... Even though we understand that the resources involved have to be significant we also understand that the cost of a web page request - even at an expensive site - it is small enough that you would need to batch a 1000 requests to really talk about meaningful numbers.

Starting from that point I asked my self the question
 - How much money does google earn every time I search.
   My first instinctive answer would have been :
       Nothing really - given that I never really click on ads ... but if you have millions searching... that quicky adds up

Which is completely wrong:
Google really charges per impression - not per clicks : given that the clickthoughrate of the ads are carefully regulated - and advertisers are penalized if the CTR is below 0.5-1%... we can make the approximation that average impression nets to Google about 1% of the average click.
What is the avg CPC?  I would definitely put it at $1  for US geos... which means that Google makes about 1cent per ad impression

Quick reality check
 - googling average ctr on google => 3%

 - googling average cpc on google => $0.35

ie cpc was 1/3rd of what I thought but ctr is 3x of what I thought - their multiple is exactly what I thought (smiling)

.

Ok So google makes 1cent per search of mine? No. Google makes 1cent for each ad that is impressed on my search results. How many ads are there on my results?3 at the top - 7-8 on the side and recently there are more in the bottom. Plus I will get more ads displayed as I am writing my query - and the page reflects results+ads on demand... So lots of ads.

If the avg # of ads is 10... then google makes 10cents every time I do a search - no matter whether I click or not!!!!

Unreal!

Cool node projects

Three related interesting projects - they open the potential for crowd-expanded education…in the form of multiple people adding problems/exercises that others can do/verify/learn
  1. https://github.com/maxogden/art-of-node
  2. https://github.com/rvagg/learnyounode
  3. https://github.com/rvagg/workshopper

They also make use of some new widgets I have not seen

Talking about widgets I also liked the license widget seen by substack at stream-handbook

My first HN post - Sunday Assembly

I did it rather impulsively - no preparation. It has  been falling like a rock since I posted it :-(

I really liked the idea of the "Sunday Assembly" when I heard about it on NPR on the way home yesterday. I liked their moto too "You can talk about Good without talking about God"
They seem to be running a crowdfunding campaign - which was not very successful 33K british pounds instead of 500K  (ends tomorrow..)
My hope was that a bit of HN publicity can go long ways in helping them get some funding...