It is all about me!

you have nothign to prove, except theorems

These days heavy things are hitting the news. Google and Telco want to define how you access internet, facebook is hitting its peak times and twitter still hasn’t figured out how to make money.

Facebook the dominant system creation continues to hold number one spot as web app. People make money from it, through marketing and flash games so now many people have a an interest in supporting the facebook platform.

And it is not like there is competition.
Where are these guys at?

Someone just asked about that on HN.

How hard is it to setup a crypto system and some very scalable simple web app?
I mean why not?
Weekend project ;) Anyone care to give me 50k and watch me try?

Someone should do it! Because the longer facebook goes unchallenged the more difficult it will be eradicate it as a social platform… Just look at how entrenched windows is despite being a shit OS (relative to Ubuntu and OSX). It is not really windows that matters so much as the apps for that platform. Even if you build a better platform, you might still fail because of the number of apps for the old system…

The big selling point of an OSS and self-administered facebook is privacy. But people don’t seem to care about privacy all that much.
Every time the privacy debate comes up in relation to google or facebook the conversation is steered towards the following assumptions:

1. The attacker is another member of the website or an anonymous web user.
2. The platform (gmail or facebook) is a trusted third party.

This is not the debate I want to be having. We don’t agree on the assumptions. Any “privacy policy” I can fill out for my data will not prevent the company from accessing my data. They own the database and the file servers and even have authority over my login credentials for that website.

What about the privacy in the following paradigm:

1. The attacker is person on the internet.
2. The software runs on a distributed platform of hosts run by untrusted VM contributors.
3. You run your own credentials server which you can trust.

The last item is still beyond the technical abilities of mom and pop who barely figured
out how to login to facebook so maybe this paradigm isn’t that relevant.
It is relevant to me though. And I think to the generation of people who use the internet
even more than I do.

How do you secure the web?
How do you make the web immune to eavesdroppers?
(Oh noo…. oh… waiiiiit… boay. whatyou tawking bout? if we can’t listen in on your social conversations we won’t be happy — say the letter agencies. )

This guys has lots to say about this.

I spent the entire Wednesday night preparing for the grandiose launch of minireference.com and the book version. I prepared a “pitch” letter that invites students to come to the website and test the product.
I also printed out 8 copies of the book — all 60 pages of current content printed on half-of-letter paper and stapled together on the side to give the book-like feeling. I was quite impressed with the print quality actually. \documentclass[10pt]{book} works magic things out of the box…
Everything was ready by 8:10AM and I went to the classroom and placed the pitch letters on the exact spots where people
tend to sit (I had previously attended a class to size-up how many copies I will need).

One problem. There was no class that day!

To be honest, the launch is not a total fail. I saw two students going into the class to use the space for studying and one of them must have looked at the pitch. He/she bought it and gave me this nice trace in access.log.


[05/Aug/2010:09:53:34 -0400] "GET index HTTP/1.1" 200 2292 "" "Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.3) Gecko/2008092414 Firefox/3.0.3"
[05/Aug/2010:09:53:49 -0400] "GET lessons/fundamentals HTTP/1.1" 200 655 "index"
[05/Aug/2010:09:54:12 -0400] "GET index HTTP/1.1" 200 2292 "lessons/fundamentals"
[05/Aug/2010:09:54:14 -0400] "GET lessons/cal2/index HTTP/1.1" 200 78 "index"
[05/Aug/2010:09:54:27 -0400] "GET lessons/cal2/riemann_sum HTTP/1.1" 200 100 "lessons/cal2/index"
[05/Aug/2010:09:54:34 -0400] "GET lessons/cal2/techniques_of_integration HTTP/1.1" 200 193 "lessons/cal2/riemann_sum"
[05/Aug/2010:09:57:03 -0400] "GET calculus/series HTTP/1.1" 200 3318 "lessons/cal2/techniques_of_integration"
[05/Aug/2010:09:57:40 -0400] "GET calculus/formulas_to_memorize HTTP/1.1" 200 915 "calculus/series"


I captured someone’s attention for 4 minutes!
That is pretty good by internet standards no?

Visit analysis: Came in, checked the fundamentals page and didn’t find the link to “solving equations” to be too interesting so I guess she is a calculus student — she knows her basics. Then from the main page she started the cal2 lesson sequence where she spend:

1. 30 secs on the calculus intro page
2. 7 secs on the Riemann sum page
3. 3 minutes on techniques of integration
4. 30 secs on series

Ok. So more visitors needed. More content quality needed. And definitely some software help for log analysis!

I am reading this guide about the file specs you need to interface with a printing company.
They recommend using CMYK color scheme and resolutions of about 300dpi. As high as 800dpi if you have line art or grayscale drawings.

What you see on your monitor is not necessarily a reliable representation of the tone or density of the final printed product. While your monitor uses pure Red, Green and Blue light to display your images, the printed piece will be made from Cyan, Magenta, Yellow and Black inks which are inherently impure. While your monitor reflects light, ink absorbs it. Monitors tend to have a bluish cast to them, while paper tends to have a yellow or grey cast. Paper will soak up ink, causing your image to darken, and the grain of the paper can cause some unevenness, mottling or flecking in the image that will not be present on the monitor.

In addition, every monitor is different – different make, different model, different age – as is every user. Your monitor may appear different in the morning than it does in the afternoon light. The color of the walls surrounding the monitor and the quality of the lighting source will alter your perception of the colors on the monitor. Your perception of color can even be affected by how tired you are, whether you have a headache, whether you’ve had too much coffee (or not enough).

lol…

I am not wearing shorts yet, but I have been spending a lot more time outside in the sun. Looking at the calendar I see that 15 days of summer have already expired without much productivity from my part. Not on projects, not on research and not even on the resting front.

Now is the time to sit down and make a little plan for the coming months to see how things will go down. The main plan is to work hard this summer. Yeah. Seriously work, work and work. Try to get to some high-energy level where I am waking up early and getting 4h of intellectual labour per day and 4h of other stuff.
Come september I get on a plane and ship myself to Bulgaria for a couple of months. Productivity in-a-de-homeland? Vacation too. Go to the cottage. Visit P. in Berlin. Visit D. in Austria/London. Why not an Amsterdam trip while we are at it.

But lets get more specific on the work side. What work absolutely HAS to be done by the end of this summer.

1. PhD thesis topic settled
2. Interference channel paper
3. minireference content (en: and fr:) be written and organized
4. minireference website
5. liblda.py must be started

Apart from that there are these projects that I would like to work on,
but they are not mission critical.

• The kronos project with A., more specifically the web-text editor, the scheduler and the pdflatex renderer…
• Non-binning information theory for neuroscience research
• Writing up papers
• minireference iPhone/iPad app in Objective-C

So the TODO list has been set down. Now the harder part of actually doing all that is in the list comes about. I have to be frank with myself — none of this will get done without effort and without getting up early in the morning. I have to get myself to a higher energy level and then things will work out.

Vamolos !

it be the last week of the semester
coming up

what you gonna do?
how you gonna play it?

roll with the trusted avoidance technique
or change posture and motivate yourself?
be productive and shit
like for real

the main objective on the block is the comp 610 final exam
and the interference channel project

yet exactly at this moment I have another two prominent ideas bouncing around in my head….

1. Professor pad
A web service for profs to store all their documents — scan in, 2d barcodes and all that jazz.
I can charge for this service for sure…

2. liblda
A web service business that receives DocumentCollection objects and returns TopicModel objects … ;)
Privacy of submitter?
Parallel computing and efficient message passing?
CUDA?

now go to bed bro

I have two assignments to work on today. This means that, naturally, my morning reading of hacker news will likely get protracted indefinitely so that I can avoid actually sitting down and starting to work.

Two interesting articles piqued my attention and I want to share them with you. The first one is about how education creates cogs for the system. I.e. the high school — CEGEP — University chain combo has as its purpose producing employees for the corporations. I always knew it, but the author puts it in historical context and make some other good points also.

Then there is an interesting analysis of the Google Buzz fiasco. The main point the author makes is that the buzz launch wasn’t an accident or an deployment fuck-up by google, but rather an intentional strategic action by top management to capture market share. Many analogies with Microsoft-style leveraging of existing products (in this case gmail) to launch new products are made.

Ok… now that we have reached the bottom of this post there is nowhere else to go than to the assignments….

I don’t get the point of twitter. Why do people tweet? So they can share their exciting lives with the rest of the world? So that they can keep in touch with friends? So that they can exercise an influence over others? So they can feel as part of some #community?

I think a good portion of the users are blatant advertisers and self/friend-promoters. Check out my article… check out this website (which my friend told me to twit about)…. check out this product….

Then there is the self-interested group of people. “Just had a meeting with so and so” ok… good for you bro, why are you telling the whole world about it? Why don’t you tweet every time you go to the restroom. “I just had a wonderful piss, it was necessary too because of all that #beer”.

Now since I don’t have a need to promote any website (yet?) and I don’t feel like blogging about the minutes of my daily life, I just don’t get twitter.

There is another cool aspect though and that is link-sharing, or if you prefer the more buzz-word like name “collaborative filtering”. If you find a cool article that you want to share, you can post it to twitter. Real time news fit into this. Links to good news articles too. Links to thoughtful blog posts also.

It would be interesting to have some “twitter app” that can separate the wheat from the chaff.
You still “follow” someone, but filter their vanity-tweets (a.k.a i did this and that…) and their promotion tweets.
What remains should be only quality material no?

One could even rate users based on their Signal-to-Noise ratio.

This article has some good advice about entrepreneurship on the internet:

1. Make a product that a billion people will fall in love with and use for the rest of their lives.
2. Make it easy for a single-digit percentage of them to pay you a few bucks a month once in a while.
3. Make sure your variable costs are low enough that you can make a mountain of profit if you get #1 and #2 right.

Which of my projects are capable of doing this?

What are the other web-friendly business models?

Page description languages are things that represent documents in their final state before being printed.

Typesetting is the process of taking text and other objects and placing them on the page. I want to be part of that too.

Document editing is the yet another step where UI is most important. Is the user looking at a WYSIWYG editor, a textbox or something else? How does the user specify advanced object placement like where the images should be (.css, GUI drag and drops + anchor property, auto placement).

The last part is the most difficult since it is “user facing” and users are tough. A simple web-app can do part of the job, but will it be rich enough for people to use on a daily basis?