A proposal to Google how to improve privacy in Google+

Hi there Google! It’s very nice how one can restrict postings to a certain audience.
But what would also be very nice is when you could restrict other people from seeing to what audience you posted something.

For example:
Alice posts something to her circle “Friends”. She’s got 34 people in this circle.
Bob, being a member of her circle “Friends”, has 14 friends in common in all his circles.
However, Bob also sees Charly, who’s only a friend of Alice, but not of Bob.

So:
I propose that Google adds a feature that Bob will only see the 14 friends he has in common with Alice, and a note “and 20 other people”. But Bob should never see the identitiy of those other people, means, he’d never see Charly.

Also, I’d like to see a feature saying “Prohibit this posting from being shared”. Sure, it won’t prevent people from sharing it by other means, let’s say copy&paste the text or downloading the pictures and resharing. But just the notice “This post was originally shared with a limited audience” doesn’t prevent anyone from unintended consequences.

Edit: While we’re at it, it’d be really neat if you could change the intended receiving circles afterwards.

Thanks!

P.S.: I’d like to see those features optional, but enabled by default, until the user decides to change it. Kthxbai!

Now SEGA took the lead, continuing the tradition of music games with special controllers.

Live, from Digital Signage Show, we present you:

Piss Hero!

Only this time, the controller is your wiener.

IPv6 Traffic at the German Internet Exchange

Interestingly, although IPv6 day is over, the traffic didn’t decrease. For the fun, I checked with youtube.com – and it’s still serving traffic in IPv6!

Now compare to the the montly stats:

DE-CIX monthly IPv6 traffic

Nethertheless, IPv6 traffic is still very little compared to regular IPv4 traffic:

IPv4 Traffic at DE-CIX

So while IPv6 peaked with 1.8 Gbit/s, IPv4 at DE-CIX is still a 3.2 TBit/s. Which is about 2000 times more traffic.

But what does that tell us?

Thinking of the IPv6 traffic stats I derive the following things. First, I assume that IPv6 day didn’t make user migrate to IPv6. I think – although I can’t prove it – that existing IPv6 users kept using the Internet like they used to.

But the statistics tells us one thing: That a lot of people were prepared and ready – the early adopters. Since youtube.com is still online with v6 – and the traffic didn’t change to much – I think that either youtube.com made up most of the v6 traffic, or, that most of the websites are still live with v6.

Whoever you are, running IPv6 since yesterday: Keep going! And let’s hope those traffic stats keep increasing over time.

If you’re interested in IPv6, and didn’t bother to try it, go and get yerself some v6 at the following spots:

• GOGO6 - http://gogonet.gogo6.com/page/freenet6-services
• Sixxs.net - http://www.sixxs.net/
• Hurricane Electric - http://tunnelbroker.net/

GET /WsIndexes/AZ1_EU.xml HTTP/1.1
Host: applicast.ga.sony.net
Accept: */*
X-WS-MODEL-NAME: KDL-46EX705
X-WS-CLIENT-ID: 54:42:49:B5:E1:28
X-WS-COUNTRY-CODE: DEU
X-WS-LANGUAGE-CODE: ger
User-Agent: WidgetSystem/2.0


HTTP/1.1 200 OK
Server: Apache
ETag: "dd74d85181fb035ebc97fe67fc242681:1303966227"
Last-Modified: Thu, 28 Apr 2011 04:50:27 GMT
Accept-Ranges: bytes
Content-Length: 3107
Content-Type: application/xml
Date: Thu, 26 May 2011 14:50:46 GMT
Connection: keep-alive

Why does my TV send it’s MAC-address to Sony?

Next on: Will try to change the X-WS-CLIENT-ID to some arbitrary value and see what happens.

Because I can. For science.

UPDATE: I just wrote Sony Europe a letter with a request for comment. Text (in German) below:

Bei einem Sicherheitsaudit meines Netzwerkes ist mir aufgefallen,
dass mein Fernseher jedes Mal beim Einschalten bestimmte Server von
Sony im Internet kontaktiert, um die Liste verfügbarer Widgets
herunter zu laden.

Das ist auch soweit OK.

Aber warum schickt mein Fernseher seine MAC-Adresse mit - das Merkmal,
was meinen bestimmten Fernseher identifiziert, obwohl ich in diesem Fall
noch keinen Kaufvorgang eingeleitet habe?
Das Deutsche Datenschutzgesetz sagt eindeutig, dass Daten nur dann
erhoben werden dürfen, wenn diese für einen Geschäftsvorgang unbedingt
nötig sind. Beim Einschalten meines Fernsehgerätes habe ich aber noch
nicht die Intention, mit Ihnen ein neues Geschäft zu tätigen.

Bitte teilen Sie mir bis zu Montag, dem 13. Juni, folgende
Informationen mit:

* Warum Sie das tun
* In welcher Form sie die MAC-Adresse in Kombination mit meiner
IP-Adresse speichern
* Wem Sie diese Daten weitergeben
* Was sie mit diesen Daten tun
* Wie lange sie diese Daten speichern

Ich bedanke mich recht herzlich im Voraus!

Alexander Janßen.

UPDATE: A few weeks later, no response from Sony. I wrote them another message over their webinterface.

Sehr geehrte Damen und Herren,

ich habe Ihnen am 30. Mai die Frage gestellt, warum mein Fernseher -
ein KDL-46EX705 - seine Ethernet MAC-Adresse zu ihnen mitschickt, wenn
er sich seine Diensteliste bei Ihnen abholt.

Leider haben Sie mir weder eine Antwort noch eine Bearbeitungsnummer
bisher zukommen lassen.
Um mein Anliegen noch einmal zu verdeutlichen, füge ich meine Nachricht
an Sie noch einmal unten an.
Bitte geben Sie mir bis nächste Woche Montag, den 27. Juni 2011
Bescheid, unter welcher Bearbeitungsnummer SIe mein Anliegen bearbeiten
und wen ich in Ihrem Unternehmen dazu telefonisch befragen kann - und
das bitte nicht unter einer kostenpflichtigen Nummer. Wir hatten schon
Kontakt miteinander und ich sehe nicht ein, mit meinem Lieferanten
gegen Geld bei einer Reklamation zu sprechen - einer Reklamation, die
auch noch meine Privatsphäre betrifft.

Meine alte Nachricht an Sie.
--- schnipp ---

Bei einem Sicherheitsaudit meines Netzwerkes ist mir aufgefallen,
dass mein Fernseher jedes Mal beim Einschalten bestimmte Server von
Sony im Internet kontaktiert, um die Liste verfügbarer Widgets
herunter zu laden.
Das ist auch soweit OK.
Aber warum schickt mein Fernseher seine MAC-Adresse mit - das Merkmal,
was meinen bestimmten Fernseher identifiziert, obwohl ich in diesem Fall
noch keinen Kaufvorgang eingeleitet habe?
Das Deutsche Datenschutzgesetz sagt eindeutig, dass Daten nur dann
erhoben werden dürfen, wenn diese für einen Geschäftsvorgang unbedingt
nötig sind. Beim Einschalten meines Fernsehgerätes habe ich aber noch
nicht die Intention, mit Ihnen ein neues Geschäft zu tätigen.

Bitte teilen Sie mir bis zu Montag, dem 13. Juni, folgende
Informationen mit:

* Warum Sie das tun
* In welcher Form sie die MAC-Adresse in Kombination mit meiner
IP-Adresse speichern
* Wem Sie diese Daten weitergeben
* Was sie mit diesen Daten tun
* Wie lange sie diese Daten speichern

Ich bedanke mich recht herzlich im Voraus!

Alexander Janßen.
--- schnapp ---

Ich hoffe, bis nächste Woche Montag von Ihnen zu hören -
Alexander Janßen.

News out of Italy suggests that seven researchers who did not predict the L’Aquila earthquake in April 2009 are under formal investigation and may be charged with gross negligent manslaughter.

I wonder if they will charge Berlusconi for aiding and abetting because he cut funds for basic research.

Next on: Politicians sueing meterologists for Kathrina. Politicians sueing vulcanologists for Mt. St. Helens. Politicians sueing petrologists for the BP oil spill. Errrr… Right.

Film at 10.

Lawrence D. Thomas has been appointed manager of NASA’s Constellation Program, which manages the effort to take humans beyond low-Earth orbit and develop the next generation launch vehicle and spacecraft.

Watch the emphasis (mine). We wanted to go to the Moon, Mars and also get back. Orion? Merely an escape-vehicle for the ISS, if at all. Ares? Canceled. Altair? Who knows.

Anyway: Congratulations to Lawrence Thomas!

It’s called the Arizona Financial Text system, or AZFinText, and it works by ingesting large quantities of financial news stories (in initial tests, from Yahoo Finance) along with minute-by-minute stock price data, and then using the former to figure out how to predict the latter. Then it buys, or shorts, every stock it believes will move more than 1% of its current price in the next 20 minutes – and it never holds a stock for longer.

TR points out, that analyses similar to the described algorithm exists since the 90ies. However, the new systems doesn’t actually parse all the data, but concentrates on some keywords which seem to be of relevance.

I see two very odd flaws there.

1. A good AI predicting the stock-market based on human-written text – which could be technically used by *anyone* who could afford it – would lead to a situation where stocks keep heating up. Speculation will grow rapidly and positive feedback loops will possibly run into an overdrive situation. I wouldn’t opt in for a ban on such a software but on full disclosure if this software was used on a certain bid. This could help in debugging situations and to give legislators something to think off when the shit already hit the fan.
2. If the algorithm actually concentrates on keywords in context rather than in the whole analysis of the text, I bet a fiver that it wouldn’t even take a few weeks until some clever consulting company analyzed the algorithm and makes up a process how to tweak your fiscal reports so that AZFinText favours this text. Think of the stock-market equivalent of a Google bomb.

Nobody in Technology Review’s forum seems to be worried about the real-life implications… I think I’m just pointing out the obvious and that the stock-market professionals already made up their own ideas.

We arrived at the registration desk at 13:20 sharp – Tutorial would start at 13:30. Registration was smooth. Give yer name and company, grab badge, WiFi-details, a map how to get from the CCH to the University and a schedule for all the tutorials.

First thing we didn’t like: The schedule was divided in tracks – there was a CUDA-track, an Infiniband-track, and so on. A lot of lectures in every individual track, but no timeline, when the individual talks should start! It was hard to decide what to do first.

We grabbed our stuff and had a really short walk of probably 5 minutes to the venue. The building was a typical: A huge 1900′s building, huge, massive, a lot of stairs – but eventually we got into the lecture room B, where we wanted to hear the CUDA tutorial.

Unfortunately, the CUDA talk wasn’t anything new. I really think that the slides of this tutorial were actually used in a NVIDIA webinar about CUDA I attended last year! It was a real Deja Vu, and I think they just changed the date on that slides. Gernot Ziegler started up with the tutorial – I thought this would be “big time”, cause we’d get the opportunity to hear someone who’s really into CUDA – he’s with NVIDIA after all. Anyway, he passed over the tutorial to John Stratton, who did quite a decent job, but he was pretty unlucky for he had to reiterate that presentation I already enjoyed last year in the webinar.

Ah, what a wasted opportunity. See, I’d expect more from the ISC when it comes to CUDA than reciting what was told earlier on less specific channels!

My colleague and me stuck to the talk until the long break – some cool things were said about how to abuse the texture-buffer in a clever way to to really local multidimensional computations, but unfortunately it wasn’t elaborated enough. For my taste.

After the long break we decided to hit the Infiniband-talk. Since we deploy large installations we thought that this gives us some insight about how to deploy Infiniband and how to make up migration-concepts of how to get from Ethernet to Infiniband. Sadly, the talk was just a roundup of the available vendors, their products and performance comparison. That’s not really what we expected, we could’ve just looked that data up elsewhere. While the talk was still going on (I think we were on slide 78 of 145) I decided to check the proceedings and see what this talk would be offering in the next hours. Unfortunately it was about to continue like that.
That was the time when we left the hall and went for the “Hybrid MPI & OpenMP Parallel Programming” lecture.

And wow, that was awesome. We got there pretty late, it must’ve been 16:30, but we were basically stunned by the ideas. Basically MPI and OpenMP both have their advantages and flaws. I certainly thought about combining both technologies, but never did for lazy personal reasons. And then those guys just did it: Awesome. In general, if we got out cluster of SMP-systems with multiple sockets and multiple cores we should be running MPI on the outer computing domain, and OpenMPI on socket- or core-layer. This ain’t not new, but they gave me some insight about why you should do it and what pitfalls my arise.

In the end, I thought I should have sticked to the last tutorial in the first place. Coming back to my first complaint, there wasn’t a real schedule, which is sad. We couldn’t decide to go to which tutorial first since we had no idea when all those lectures were taking place. If the ISC continues to give those tutorials, they should improve their schedule.

Then again, since my collegue and me were both attending the CUDA-talk, we already elaborated on the spot how we could use CUDA in our usecases. We just got rough ideas, but sitting together, listening to that talk wasn’t just a waste of time.
It brought us together.

Now I’m back at home and don’t feel too bad about the ISC lectures – Now I feel sad that I’m not able to attend the rest of the ISC. But I got to work on Monday. Which is in… omg, in about 8 hours.

Good night! I’ll be putting links to this posting tomorrow evening.

The PNNL just recently hired Adolfy Hoisie, the winner of the Gordon Bell Award back in 1996. He’ll be doing pretty interesting stuff: Desingning supercomputing from scratch, both in hardware and software. This is an amazingly interesting task since you can drop all burdens, all hardware- and software-restrictions you ever had. No more tailoring your software to fit the buffers into your cachelines. I mean, that’s just: Wow. Not that I’m envious or someting, I really like my job, but getting a grant for a task like this is like winning the lottery.
Adolfy, come on, show us how it’s to be done! Awesome.

Rice is one of the most common foods in various parts of the world. It’s in the interest of us all to find varieties and breeds of rice which are most nutritious or resistant against pests; the project’s goal is to find out which varieties of rice interbreed with others to give the best results so that we’ll get new strains of rice which are harder, better, faster, stronger.

Dr. Ram Samudrala

A lot of BOINC-users who contributed to the project (like myself) are now asking themselves a lot of questions. Who are the people behind the scenes? How much work is necessary to get a project like this into operation? What was IBM’s role? What will happen with the contributed results? And after all, who will benefit from the project?

I think no one can give better answers than Ram Samudrala, PhD and Principal Investigator of a computational genomics research group at the University of Washington. Rocker, scientist and Emacs-admirer – he was so kind to answer me some questions about the project.

Tell us a little about yourself and how you got involved in the Rice-project.
Ram: I’m a professor researching computational biology at the University of Washington Seattle. My overarching interest has been to understand and model how the genome of an organism (genotype) specifies its behaviour and characteristics (phenotype). We develop computational algorithms to this end that are applied to whole genomes and we work on many organisms. Rice was specifically chosen since our collaborators at the Beijing Genomics Institute had just finished sequence (and we annotated the refined version) and I also got a $1.9 million grant from the US National Science Foundation (NSF) to predict the structure and functions of all proteins encoded by the rice genome. We developed algorithms to do this and we applied it to all rice proteins. Then IBM came along and offered us the means to redo some of our calculations on the most difficult proteins using the WCG and then we ported our code over to work on the Grid.

When was the first time you considered using voluntary distributed computing for your project?
Ram: Since the days of SETI@home, and since we built our own local clusters to do structural computational biology, but porting our code to BOINC was always a inertial challenge.

Did you consider using other DC-infrastructures except BOINC, like distributed.net? If yes, why did you decide using BOINC?
Ram: No, we used BOINC since it was what was supported by IBM WCG.

Have you considered asking the NCSA for computing resources?
Ram: Yep, but it’s a cumbersome process, like applying for a grant, and again, porting software to work on different architectures. The barrier is that we get grant money to do research and not develop software. I have used NIST supercomputing resources in the past.

You said you would need 200 years of computing time using your available resources. Besides voluntary distributed computing and the University of Washington, were there other universities or institutes directly contributing computing-resources to your project?
Ram: Not for this project, no.

Rice BOINC Splashscreen

You were using algorithms from the Protinfo website. Which one did you actually use, how much effort did you put into customizing it for using it in BOINC? Can you tell us if those algorithms and implementation are released under a free license?
Ram: It’s the Protinfo AB algorithm, which is our ab initio or de novo simulation protocol. IBM spent a fair amount of time porting the code
to work with BOINC. The original algorithms/software are all freely available without any claim of copyright (i.e., in the public domain).

Could you explain “de novo” and “ab initio” for non-scientists, please?
Ram: “De novo” and “ab initio” generally are translated to mean “from first principles”. In the old days, this used to mean using pure physics energy potentials for protein folding. These days, to us, it means any set of general principles that is not biased to a particular protein or organism.

If the algorithms you used are under a free license, did you already manage to publish the modifications, if there are any?
Ram: The modifications involving the porting are with IBM and they are unpublished.

(Ed. note: Since the software was released in the public domain there’s no requirement to publish the modifications.)

IBM helped you out in customizing the protein-prediction algorithms for various platforms. Can you tell us how much they contributed?
Ram: All the customisation was done by IBM engineers. We just gave them the original software and ran sanity checks on the output. I’m a strong free software and anti IP proponent, to the degree that I encourage commercial use without restrictions on the software (people can always use the public domain versions if they want to).

Rice Terraces

How much time did you save by using the World Community Grid’s infrastructure compared to if you would’ve set it up all on your own, like other projects do?
Ram: IBM took about six months or so to port our software, so I presume it would’ve required that kind of an investment. Keep in mind that they had a lot of prior experience with BOINC. IBM now maintains the code and does the PR and runs the predictions for us. I’d say this would be a full time programmer/sysadm type of person and if I had that extra money, I’d rather spend it on someone doing the basic research.

If there are flaws about BOINC, which would you like to be addressed first?
Ram: I can’t think of any in the way we did it with IBM, but without IBM, the PR machine has to be powerful to get people on board. It’s more than just recruiting people, but also motivating them as IBM does with badges and giving them a sense of community and providing a support infrastructure. This is hard for a research lab to do on their own (it can be done, but is it really the best use of our talents is the questions).

Programming and debugging is an iterative process. Looking at your sourcecode-repository, how many releases of the software were necessary until you got the cow flying?
Ram: For this case, internally we probably had about 10 or so iterations in total, but the basic science part of the software is something that has evolved over 18 years.

How did you do beta-testing, did you use the publicly available beta-projects at WCG? Or, were you actually just doing it in your lab?
Ram: It was mostly in our group. We just submitted sequences for which we knew the answers and we did a dry run initially with the same sequences.

I’m curious there – were these structures predicted by other algorithms or was that done the hard way, using X-ray crystallography?
Ram: These were done the hard way, at the bench. These are our gold standard for when we know we’re right or wrong, so we benchmark our methods against all this. When we did the rice project, we did sequences with known answers to see how well things would work and that there was no chance of anything going wrong.

Dr. Ling-Hong Hung

How was is like getting in touch with the community? Was the feedback helpful? How many people from your team were actually dealing with the community?
Ram: At its peak, we had 3 people dealing with the community, our sysadm and project lead Michal Guerquin, our programmer and scientist Ling-Hong Hung, and myself. Opening our software to the Grid and the community definitely presented some challenges, which I believe will be the focus of our first paper. An interesting tangent of that is that we’ve had to port some of our analysis software to work on GPUs so we could handle all this data. So some good technological developments here that we’ll be writing about shortly.

Michal Guerquin

A lot of people are concerned about “Frankenfood”. Your project’s website explicitly states that this is not about genetic engineering, but about finding the most nutritious rice-strains for interbreeding with other rice-crops. Is there anything you’d like to explain to people who are still concerned?
Ram: We’re simply extending what farmers have been doing for millenia in a more rational way, and also what has been going on in nature for billions of years. The problem to us is scientific and all knowledge that is produced (which from our end will be completely free and transparent) can be used in various ways according to the will of the people. But we have governments and politicians to handle the deeper societal implications. What I mean by this is that people should petition their representatives, as they are doing successfully in many parts of the world, to decide where to go with genetically modified organisms, which I see as ultimately having a socioeconomic/political solution.

Your project is one of the very few with a fixed end, almost all other projects are handing out work-units for new phases. How comes that you’re finished now? Is everything from the rice-genome now analyzed from a computational point-of-view and nothing else left to do?
Ram: Not at all. We obtained a huge amount of data and we’re now pressed to analyse it. I honestly can say that we were overwhelmed with this data. My goal as a scientist though is not just to develop technical tools and produce large tables and graphs but try to come up with something tangible that is prioritised and can be tested at the bench that really changes the make up of rice in a desired manner. The computations and the Grid are the means by which we arrived at this step, but our job now is to figure out where the best low hanging fruit is in collaboration with rice researchers (which we are doing with researchers around the world including IRRI, Phillipines). [Ed. note: IRRI, International Rice Research Institute]

Focussing on the data: Now that you know how those proteins really look like, where do you draw a line and say “this protein is more nutritious than others”? My basic understanding is that the nutritious parts in rice is actually carbohydrates (starch), proteins and some fat. How do I have to imagine this analysis?
Ram: So the proteins we’re talking about are gene products, that carry out almost all the functions in rice (or any other organism). So we use
the protein to refer to a molecule that does this, rather than the nutrition use of the word “protein” which refers to these biological molecules broken down and aggregrated (see “Protein” and “Protein (nutrient)” in Wikipedia).

By nutrition we mean anything that leads to higher range of bioavailable substances like dietary minerals and vitamins. In rice, examples include elements like iron or organics like vitamin A. Incidentally the “golden rice” GMO is a product of Monsanto that has higher beta-carotene, a precursor to vitamin A (“Golden Rice” at Wikipedia). We’d like to get to something like that by crossbreeding without the use of genetic engineering, working on both micro and macronutrients.

So in the end, we need to be able to create a rice strain that does have enriched nutrients and is perhaps better than current strains in
terms of yield and/or hardiness. Before we go off and start crossing rice, there are a number of molecule biology bench experiments that
can be done to say whether predictions we make about the activity of certain proteins will be correct so we’d do them first.

Do you plan to publish all your results in an Open Access Journal?
Ram: Yep, that would be the ideal. Publishing in Open Access Journals also sometimes costs money. I’m not a big fan of the “pay to publish”
model—it’s not a lot of money and some scientists have grants to do this, but it’s not a good principle.

Thank you very much for this interview!
Ram: Thanks; I enjoyed the questions!

Dr. Ram Samudrala is a tenured Professor at the University of Washington, Seattle. He’s head of the Nutritious Rice For The World project and one of the inventors of protein prediction algorithms. He’s a notorious contributor of scientific papers and generally a very nice guy I’d like to buy a drink.

Dieses Werk bzw. Inhalt ist unter einer Creative Commons-Lizenz lizenziert.
The rice picture is copyrighted and CC-BY-SA by Flickr-user kadaoor.
The rice-paddy picture is copyrighted and CC-BY by Flickr-user ~MVI~
The pictures of the teammembers were used by permission of the Rice-team.
The BOINC splashscreen is copyrighted by IBM and the World Community Grid and was used with permission.