Text archives Help


Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns


Chronological Thread 
  • From: Peter Cranstone < >
  • To: Mary Hodder < >
  • Cc: ProjectVRM list < >
  • Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns
  • Date: Fri, 14 Jun 2013 04:01:05 +0000
  • Accept-language: en-US
Title: Microsoft Word - UC3P-Portal Services-130313.docx
Just spotted a typo..

I'm sure you get the idea. Everybody is different and yet for VRM to work we have to look at the accelerators, limiters and dependencies – if we can't extract new value from the circle of value then no one will sign up. But if we can then it's a game changer.

From: "Peter J. Cranstone" < "> >
Date: Thursday, June 13, 2013 9:58 PM
To: Mary Hodder < "> >
Cc: ProjectVRM list < "> >
Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

Mary:

What I mean by strategically.. is that we make a strategy for advancing a VMR-y system.. ie: let's identify 2-3 key entities that would be willing to do a trial.. and then strategize how this would benefit them as well as us.  Use cases would have to be made that fit into this strategy.

Agreed.

What I mean by tactically is that then we identify who outside the entity can help with implementation, to get it done.  Use case(s) would then be built. 

Agreed.

Well.. I think part of your diagram in the early thread of this email was getting the data.. so we need data portability and controls for the user.
And the other part is getting that data into personal clouds for people now, with value just from that

Agreed. Not a problem – ignore the cloud for a moment. Lets keep it incredibly simple. We need a way to store the data, a way to allow the user to control the data and a way to transmit the data. The cloud just adds complexity we don't need at the moment (we can add it back in later).


Depending on the type of intent.. this will vary.. and yes.. context is key.

Agreed.

Interesting.. I was thinking of use cases for the user.. then work backwards to providers that would pick up "intent" and "receptivity"...  but you are right that we need use cases for the vendors..

Here's how I see it (challenge away). The user has something to trade – their real time data with more context around their intention. However it's the vendor that needs to make make a profit to keep his service going which invariably is free and supported by advertising. So what we need to analyze is that and see how we can add to it. If the vendor can't make money then sharing my intentions in return for a better experience doesn't create the value VRM needs to drive adoption.

Wow ! Thanks for the new job!

I thought you'd like it :) So to help you with your new job I've jotted down some ideas you can use as a starting point to build the use case…

Every day, Mary, who teaches architecture at NYU heads out with her smartphone. On the days she takes the train into work, she loves to watch the trending YouTube humor video of the day. Mary loves tea, fine red wines and sushi, and is currently looking to purchase a Victorian style house in the next 90 days. Today she takes a new way home from school and searches for tea. She views the tea houses listed in order of proximity to her current location. One tea shop actually offers sushi and appears at the top of the list. The search is sponsored by a real estate agent specializing in Victorian home sales within the burrough. She clicks to call the agent and schedules a meeting at the tea house for tomorrow at 2:00. She accesses turn-by-turn directions, walking the route, to make sure the place is not a dive. The agent confirms via email and Mary has the option to add it to her calendar. Mary also has an online photo album, filled with inspiring buildings from her walks and summer travels. She often geotags pictures of unique architecture, uploads them and automatically overlays them onto speciality map services - creating both a photographic travel journal and customized architectural walking tours to more actively engage her students. These are saved and stored to her journal, so she can edit them with an online word processor and print them once she gets to work.

So what are we going to need for context?
  • Location and device data because the portal needs to return contextually relevant search results formatted perfectly for the best viewing experience. Advertising when paired with user-entered data is both local and relevant, even to temporary behavior such as house hunting. Location data paired with photo blogging and mapping services creates a powerful educational and social networking tool. 
  • My advertising preferences – custom fields that can easily be changed
    • Tea, Red Wines, Sushi, Victorian style homes for sale
  • The app has to have the ability to transfer pictures directly from her device already geo-tagged.
  • Ideally this should all be done in the browser (because it works the same on every device)
For a portal the holy grail is real time geo-location aware local advertising. That can only happen if you can share data in real time with all the parties. What no vendor wants to do is for you to leave the portal sight to go to another operators site. If that happens they can't close the marketing loop. So ideally everything stays in one place and services are delivered 'mashed – up' via a Web service.

Google et al are ready to launch sales people into the 'local markets' but for vendors to 'write the check' they have to be assured that they can derive value from that search request. So what we have to ensure is that we clearly understand the 'value chain' and how we can leverage the infrastructure to add more meta data for that value to be realized in A) a better service/experience for Mary and B) a ROI on the Advertisers dollar based on Mary getting a good experience. I think of it as the circle of value. 

BTW – you've just hired a new assistant (Peter) He's new to the area, is moving his entire family and needs to look for rental properties that are pet friendly. He's more of a coffee drinker, a white wine aficionado and really loves Thai food. So whatever works for you above has to also scale to Peter – he's also in the market for a good 'used' camera so his advertising preferences are going to be different.

I'm sure you get the idea. Everybody is different and yet for VRM to work we have to look at the accelerators, limiters and dependencies – if we can extract new value from the circle of value then no one will sign up. But if we can then it's a game changer.




Peter
_________________________
Peter J. Cranstone
CEO.  3PMobile

From: Mary Hodder < "> >
Date: Thursday, June 13, 2013 8:27 PM
To: "Peter J. Cranstone" < "> >
Cc: ProjectVRM list < "> >
Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

Hi Peter,

See my comments in line ..


On Jun 13, 2013, at 7:53 PM, Peter Cranstone wrote:

Mary,

I'm going to combine two emails with one here…

[Email #1]

>> but most likely i believe eventually it will come out of a personal cloud app with apps inside (maybe a wallet there) that submits my intent and receptivity.

Sure – the how it's sent is not as important as the value it generates. The how will be resolved fairly quickly when you have to scale it to everything, but for now lets ignore that piece.

>> I also get that we know, here at Project VRM in this discussion, that this all has value. But does the other side know that? Really? In a measurable way, and see that it's more value than they have now? I don't think it's enough for us to assert there that this has significantly higher value.. we need to strategically and tactically show it to them. 

I'm one of those people who can only talk about doing it for so long before I just want to build it. So what's the CORE problem you've defined above? How do we strategically and tactically show them it has value?
  • Strategically implies it has to extend value across their ecosystem
What I mean by strategically.. is that we make a strategy for advancing a VMR-y system.. ie: let's identify 2-3 key entities that would be willing to do a trial.. and then strategize how this would benefit them as well as us.  Use cases would have to be made that fit into this strategy.


  • Tactically implies that the cost to implement it (technically) doesn't consume the value it generates i.e. there must be 'sustainable NET NEW revenue' from this for people to engage
What I mean by tactically is that then we identify who outside the entity can help with implementation, to get it done.  Use case(s) would then be built.

>> I think it might be better to wait until there are some numbers that matter to those on the other side.. like a million personal cloud users, with a significant apps use.. 

Well that's the chicken and the egg situation that causes us to consume cycles talking about it versus doing it. So how can we accelerate this? Simple – lets use COTS – Commodity Off the Shelf hardware and software. We don't need to have a million personal cloud users to validate the value of VRM – we need it working first so we can iterate it, and demonstrate both strategic and tactical value. And we can do that today.

Well.. I think part of your diagram in the early thread of this email was getting the data.. so we need data portability and controls for the user.
And the other part is getting that data into personal clouds for people now, with value just from that (ie: get your taxes done faster and in more organized fashion because you have it all in one place, that you can share with your accountant or, organize your records for your expense reports, or whatever...)


[Email #2]

>>  VRM_REQUIREMENTS=["LOCATION","TIME","PREFERENCES","COMMUNICATION_METHOD"]

Ok, lets parse that out a little.
  • Location – for that I need real time – what may also be useful would be location where I was 30 minutes ago so we could build a trend line
Yes and no.. one use case: I want to buy a car, and no matter where I do the research, I probably would like to find something within a couple hundred miles of where I live..
another use case: dinner needs to be where ever I am around dinner time.. and that could be different than where I am in, say, the morning as I realize I want to plan dinner.

So . . context..
  • Time – easy enough
  • Preferences – needs more context i.e. what are your preferences
Depending on the type of intent.. this will vary.. and yes.. context is key.

  • Communications method – we really only NEED one here. The absolute simplest way to communicate with a Web service and for that we have HTTP. But if you wanted to add more you could add HTTP_VRM_MOIBLE="Yes" or HTTP_VRM_TABLET="No"

I use a few devices and depending.. might want to specify this or change it.  Sometimes I only have my phone.. but maybe this won't matter.. i would suppose my cloud could track me down.. 


Now for the next part - but also changed depending (I need a car in the next two months and to communicate through my wallet on my laptop, I want dinner tonight and let's do this on my phone but I'm traveling so my location is different from where I'll buy the car, 

My first question is this – would what you're asking above result in value for the VRM vendor? Remember this is almost a 'goods for services' transaction so by complicating the message we complicate the vendors ability to drive value. This is why a use case is vital – what sounds good in a chat room has no incremental value to a vendor in the real world.

How about I suggest a Portal Use case scenario and then you pick it apart to see where we can add real value to both the consumer (by wanting to share data) and the vendor (by generating more value from advertising) by delivering an improved service.

Scenario…

Portals such as XYZ are learning that to distribute and maintain every new Web service as an application on every mobile device is too complex and expensive. In the case of XYZ, they have committed to a web-based versus mobile app approach. It has proven more difficult than anticipated for them to deliver optimized mobile Web services to every mobile user regardless of device or carrier. It has also created both optimization and monetization challenges for the XYZ portal, limiting advertising revenue opportunities, forcing them to build in tracking “hooks” that may soon become illegal in the US, and making it impossible for them to test the real-world performance of their services over wireless carrier networks. They need a Web-based solution to help them overcome these challenges and fully monetize their mobile initiatives and Web services.

Interesting.. I was thinking of use cases for the user.. then work backwards to providers that would pick up "intent" and "receptivity"...  but you are right that we need use cases for the vendors..


So strategically we have to show them how we can drive net new revenue across all their Web properties. And then tactically we have to show them how it can be implemented. Finally we have to put real numbers to the value that it generates. In the portals case it's free 'services' in return for bombarding you with ads (attention economy). What new pieces of 'contextual metadata' can turn that into the Intention economy? (Think local here). 

The answer should begin with… 

A Day in the life of Mary who teaches at a NYU university and heads out to work with her smartphone…

Wow ! Thanks for the new job!


Now lets get creative.

Ok.. :)

From: Mary Hodder < "> >
Date: Thursday, June 13, 2013 6:01 PM
To: "Peter J. Cranstone" < "> >
Cc: ProjectVRM list < "> >
Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

Peter,

I think a use case is a great idea.. or a series.. we could put them at the Project VRM wiki and send those folks to ask them to join in a demo of this new way of thinking.

I like the data model you suggest below. I would add particulars for a specific demo, for example requirements the user might have around their desire.

So maybe:

6.    VRM_REQUIREMENTS=["LOCATION","TIME","PREFERENCES","COMMUNICATION_METHOD"]
This might refer less to the specific sector or product and more about the user's requirements for the item or service.  These would probably partly be defaults in a wallet, but also changed depending (I need a car in the next two months and to communicate through my wallet on my laptop, I want dinner tonight and let's do this on my phone but I'm traveling so my location is different from where I'll buy the car, for example -- that might be internal to the app in a way, but vendors will likely need some specifics in order to get their side of this right). 

And yes.. then mapping the sector specifics would be next.

essentially making the use cases and microformat would be super cool.. i LURV it.

mary

On Jun 13, 2013, at 6:34 PM, Peter Cranstone wrote:

Mary:

How about a 'gedanken (thought) experiment'. 

Why don't we build via email a use case that can value intent and receptivity from what is currently 'unstructured Web data' (cookies and other nonsense). If you were a vendor offering a service what kinds of structured data would you want to know about in real time to drive more value to that customer?

What about using the following data as part of your 'Web wallet'? You can add anything you want to the list, you can change the values of what's already there – literally anything that works for you.

2.    VRM_RECEPTIVITY_TO SALE="HIGH"
3.    VRM_SHARE_MY_INTENTION="YES"
4.    VRM_CURIOSITY="HIGH"
5.    VRM_ENGAGEMENT="HIGH"

Next you pick the service that you're familiar with and we can map out what kind of extra data that the vendor could value and leverage as 'long as he has your permission'. The thing to bear in mind here is that what has to happen with the Intention economy is the following. It must be cost effective to extend 'existing profitable business models' to ALL devices and you have to be able to capture real ($$$) value by taking on the this new approach – so it MUST be easy for users/consumers to use. 

To me this is the underpinnings of the semantic Web – the ability to put context and value (integrity) around unstructured data. 




Peter
_________________________
Peter J. Cranstone
CEO.  3PMobile

From: Mary Hodder < "> >
Date: Thursday, June 13, 2013 4:08 PM
To: "Peter J. Cranstone" < "> >
Cc: ProjectVRM list < "> >
Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

Pete,

I get there is high value in the intent and receptivity structure data.. and the form you submit below. 

I wonder though.. if the receiving parties will ignore it or use it.

My hope is that they do use it productivelly, but the alternative, which most big data collectors have taken, is to amp up the arms race and route around whatever we do..
in 2007 browsers added the feature to delete cookies, and by 2008 or 09, to block 3rd party cookies.. or cookies generally.. so those "best and brightest of our society working on advertising" decided to invent flash cookies, beacons and machine fingerprinting.  Now we have DNT.. and they are further working on getting around it.

My concern is many of them will go to the next level of spy ware, rather than take the metadata we give them and make better use of it.

So your structure is good.. but I think we need to work with a couple of key partners who want to innovate and do the right thing, to help them implement their side of this. Then, we and they report publicly on how *much better* it is for *them* when they take that structured intent and receptivity and make something of value with it.

But it's a great start.  

Thank you!
mary

On Jun 13, 2013, at 4:02 PM, Peter Cranstone wrote:

Mary:

Thanks. What happens next is the ability to transmit two things:
  1. Intent
  2. Receptivity 
Imagine that you're able to transmit the following user (controlled) information to a VRM enabled vendor in real time. Think of these as 'headers' leaving the device in a 'format' that the VRM's current infrastructure can easily read…

2.     VRM_RECEPTIVITY_TO SALE="HIGH"

Now the second the vendor sees that 'context' he knows that he has a 'live prospect on the line' and the 'quality of the experience' goes way up. In the absence of that information it's 'business as usual'. All you need to do is send the signal that you're interested, once the other side sees it then there's an incentive to deliver more value.

Think how valuable just those two pieces of Metadata would be to 'Big Data'. 

It would change everything.



Peter
_________________________
Peter J. Cranstone
CEO.  3PMobile

From: Mary Hodder < "> >
Date: Thursday, June 13, 2013 12:26 PM
To: "Peter J. Cranstone" < "> >
Cc: ProjectVRM list < "> >
Subject: Re: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

Peter..

That's a terrific diagram showing how things need to work so that you control your data.

I don't think the change will take that long, before we are collecting our data and putting it at our personal cloud.

Then the question is, what follows from companies on the right side of your diagram?
I hope, it's that they realize they need to provide good services, whether we pay for them, or get them free...

in order to interact with us.

mary


On Jun 13, 2013, at 12:28 PM, Peter Cranstone wrote:

Mary,

So how is "personal" or personally directed services (VRM) style different than general cloud services? And you answered the question perfectly - it's not "personal" unless the individual controls their own data, not just the use of the product  - Brilliant!

Which means for all this to work you have to control the collection, flow and use of your data from the device that you're using when you interact with the product/service. So the schematic becomes the following. You can store and control your data at one of two possible points – either directly on your device AND/OR through the Personal Cloud. They KEY though is EXACTLY as you've described it – it MUST be the user who transmits the intention behind the data and NOT the Content provider who ASSUMES the Intent.

From here it's easy to build a solution into the browser.

<304C4B61-922F-4EEB-8344-E6DAA5DC3368.png>


Peter
_________________________
Peter J. Cranstone
CEO.  3PMobile

From: Mary Hodder < "> >
Date: Thursday, June 13, 2013 9:43 AM
To: ProjectVRM list < "> >
Subject: [projectvrm] Sweden Bans Google Cloud Services Over Privacy Concerns

http://www.privacysurgeon.org/blog/incision/swedens-data-protection-authority-bans-google-apps/

"The ruling (By the Swedish Data Inspection Board) – which bans Google cloud products such as calendar services, email and data processing functions – is based on inadequacies in the Google contract. A risk assessment by the Board determined that the contract gives Google too much covert discretion over how data can be used, and that public sector customers are unable to ensure that data protection rights are protected.

"The assessment gives several examples of this deficiency, including uncertainty over how data may be mined or processed by Google and lack of knowledge about which subcontractors may be involved in the processing. The assessment also concluded that there was no certainty about if or when data would be deleted after expiration of the contract."


It's going to be a PR struggle to convince regular people that "personal" or personally directed services (VRM) style are different than general cloud services.. because I bet that Google would argue that Google apps are personally directed.. nothing happens unless the individual uses the services, from Google's perspective. But the individual's data  isn't controlled by the individual, VRM style.

So I think this will be the pivot point.. convincing the public, as well as the companies and governments, that it's not "personal" unless the individual controls their own data, not just the use of the product.  

And that struggle to be heard above the company's PR about what is VRM and what is not, what is "personal" and what is not, is key to this whole mess.

mary

Archive powered by MHonArc 2.6.19.