Data Sportability

It’s nice to see everyone joining the portability and open standards bandwagon, and with such breakneck speed. It’s also great to see a complete 180° turn for many people involved — it wasn’t too long ago that many of the newly outspoken proponents of these concepts were the ones maintaining and enforcing the walled gardens of social media with an iron fist, holding users and third party services hostage at their leisure.

All the traction this movement has been gaining has made it very easy to get excited — quite a lot has happened in the last few months, and people are often still lauding some open technology just as its successor or complement is released. It can be quite difficult to keep up with, and amidst all this celebration a few important concepts have been lost or sidelined. I think we’ve come to a point where we all need to slow down for a moment and take a good look at the recent technologies and instead of blindly celebrating them, start asking some tough questions: like what are the implications of this, and why are they being created so quickly in the first place?

If you chat with industry pundits, read the blogs, even open up a major newspaper, you’ll quickly notice that data portability is all the rage (the general concept, not the ‘DataPortabilty’ working group). People love talking about the new efforts in interconnectedness on the internet — the ability to meld content and relationships across social media outlets, and the positive strides the industry is making. If you talk to a product manager or blogger ‘in the scene’ for a few minutes minutes, you’ll note how similarly they mimic an eleventeen year old girl talking about puppies, ponies and rainbows ( and ice cream! and unicorns! and rainbow sprinkles! oh my!).

Looking beyond the gloss of portability, there is a much more pervasive méme to the industry — the new marketing culture driving the open initiatives, Data Sportability. Before people can even hash out what products really are, we see fancy marketing jargon around corporate initiatives, an endless array of shiny new products and services launched one-after-another, and a figurative pissing contest between large technology companies trying to ‘outdo’ one another in terms of portability options and open standards. While data portability might be about getting people to join the bandwagon… Data Sportability is everyone trying to have the fanciest and flashiest wagon in the train, with hopes people will flock to it more so they can go off in their own direction.

When people can step back from the celebrations of technology, it is painfully clear that the new push for open data initiatives are not so open, well intentioned, or in many cases not even well thought out. I think it’s about time we put an end to Data Sportability and end this trend before any real damage is done.

Over the next few days I’ll be releasing a series of articles based on dispelling Data Sportability, using my company FindMeOn as the context. In 2006 we released and ‘Open Social Network’- a consumer site and set of open standards promising secure, privacy minded social network integration… which was really just a testbed for our ‘next generation’ social media advertising platform. In late 2007, Groups like Google, Yahoo and Microsoft started releasing similar products, under similar names, and going through all the same ‘realizations’ and growth spurts as our products.

Common Sense Truths Behind Portability

A preview.

The Open Movement is a Shell Game
Open Source and Open Standards are absolutely meaningless when their point is to sell in proprietary platforms and services. Google, Facebook, Windows, MySpace, etc. are all promoting *their* platformed versions of portability. This isn’t a goodwill effort, this is an arms race for technology, users, and a market leadership position. Corporations are also very much focused on what THEY get out of being open — not end users; they’ve done a 180° on portability for a reason, they figured out how to monetize it.

Beware the Wolf In Sheep’s Clothing
Being backed by, or creating, a non-profit is nothing more than a PR stunt. Lobbyists have been doing this for yers to mislead consumers on the behalf of the Tobacco and Oil industries. Being a non-profit doesn’t mean that you’re searching for the cure to cancer, or developing cross-platform software under the MIT license — it just means that your organization *as the organization* is not focused on creating corporate profits for itself. Groups like “OpenSocial Foundation” are industry associations of large social media stakeholders and advertisers – they’re not coming together to save children, they’re coming together to optimize their businesses.

The road to hell is paved with good intentions
A lot of portability people mean well – but its important to view these innovations as the struggle they are — corporations are looking to monetize, technologists are looking to quickly adopt and push forward with every new innovation. When the dust settles and the novelty begins to wear off, a mess is often left behind. In politics, people talk about ‘inside the beltway’ — a disconnect between the political system and the people it represents, created by secluded and self-reinforcing culture of its members. Technology is the same way — technologists have the mindset of early adopters – and people who try to live their lives outwardly across all mediums; the average internet user is vastly different.

I recently got into a heated argument on the DataPortability group’s mailing list, when I was incensed by the lack of discussion covering user privacy ramifications of OpenID adoption — especially in a so-called ‘what could go wrong?’ panel discussion that was scheduled. From a consumer and corporate perspective, I am *deeply* troubled by the conflation of online account information and relationships hastily integrated systems promote.

Most people failed see any issue where a large one exists; a small subset of people saw the issue and replied something to the extent of ‘well some of us do care enough, and have implemented privacy constraints in our software’. My response: why is this not the default? why isn’t everyone adhering? why am I the only one questioning this?

More than meets the eye
People are pushing data portability for a reason – monetizing openness. The monetary layer in data portability isn’t in being an open service, or providing a platform service… it’s in analyzing and applying cross-network user intelligence for internet advertising. Open platforms aren’t about providing a service to users, they’re about making money off of users.

They’re also about creating new commerce-driven standards. While MySpace, Google, Facebook are big names in Social Media, they’re the groups behind the world’s most powerful networks – not the people interacting with APIs and building new social media projects. While its nice to see these groups offering some ‘standards’ for interaction, they’re also saying “ok, but you have to play by our rules now!”. How different would these Open Standards look like if they were mandated by the widget developers, social media startups, or ad companies / corporate brands who are constantly building new online media projects? These ‘standards’ are concepts and hooks defined by benevolent dictators — not by the people who make applications.

PostgreSQL Rename Sequence Bug/Feature Fix

In PostgreSQL ( < 8.3 I believe ), if you rename a sequence you don't fully rename it This is recorded in bug 3619

While the sequence will be renamed as an object, the metadata won’t…

PG stores the sequence name both as the object , and as ‘sequence_name’ in the sequence itself ( viewable via ‘SELECT * FROM sequence’ ; i think its in the postgres db catalog too )

Most people won’t run into an issue… until you use an ORM which is likely to pull the old sequence name from PG.

after much searching and trying, i found a fix. note: this only works for SMALL databases:

$ pg_dump -U$owner $db > $db.sql
$ psql -Upostgres
psql> DROP DATABASE $db;
psql> CREATE DATABASE $db WITH OWNER = $owner;
psql> q
$ psql -U$owner $db < $db.sql yes, that's the amazing fix -- backup up, dropping, and restoring the database. i'm sure there's a better fix out there somewhere -- i just couldn't find it with 2 hours of searching and syntax attempts. If you can find a better way, LET ME KNOW and I'll update this reference.

New Category – Scalability & Performance

I’ve decided to create a new category on Scalability & Performance

I recently learned that not enough people talk about this – and there are a lot of independent efforts that don’t cross communicate.

Since I’m plugged into a lot of misc communities, I decided to start talking about the different efforts in each — and where there is overlap.