3 ms·

> data ownership and pipelines not ML. By your argument, the money is in data hoarding and brokering, and renting that data (with DRM) to ML outfits, not doin

by papln 7y ago

> data ownership and pipelines not ML. By your argument, the money is in data hoarding and brokering, and renting that data (with DRM) to ML outfits, not doing ML. Anyway, cleaning dirty data isn't execeptionally hard, it's just boring work; getting the raw data is the hard part.

killjoywashere 7y ago
Um, yes and no. Some of the data annotation problems get really hard and really expensive. Look at radiology or genomics. You literally need N people to die for more knowledge in some of these situations.
- jnwatson 7y ago
  Data cleaning and data labeling are two separate tasks, with vastly different workflows. For example, getting structured data out of Whois requires a lot of cleanup. Labeling domains as “bad” might take forever.
nostrademons 7y ago
This is an interesting observation that deserves to be addressed. I think in practice, what happens is that there's no easy transactional boundary that can keep the data ownership and ML in separate firms. The theory of the firm [1] states that firms arise when transaction costs are less than the economic inefficiencies of centralized resource allocation. There are some pretty heavy transaction costs between doing data cleaning and data science in separate organizations: 1.) DRM for datasets isn't really a thing, since to explore, visualize, and train on them, you need access to the raw data, and then instead of your machine-learning function you can just pass the identity function to get the raw data. DRM for consumers always relies on a publisher whose incentive is to stay in business (by not breaking any laws) rather than to obtain the raw data. 2.) You don't know whether a cleaned data set will be useful for your ML application until you've inspected it, visualized it, run some statistics over it, etc. at which point you've done most of the work for setting up your models. That means there's a big risk premium for buying data, and potential buyers usually want samples & statistics before committing a lot of money. 3.) At the price that good datasets go for, you're usually dealing with enterprise sales, which involves commissioned salespeople, face-to-face meetings, expensive dinners out, etc. 4.) The type of labeling you need to do is often intimately connected with the usage of the ML model. What constitutes spam? What constitutes abuse? These are questions for your policy team, which the data-collection organization would have no visibility into and no way to set up a one-size-fits-all policy that all potential customers would be okay with. 5.) Software markets tend to be winner-take-all, which means you're dealing with a handful of customers, and one will likely become dominant and able to acquire you. Instead of separate firms for data ownership, ML, and user interface, usually the service that makes the final product will end up collecting or buying the data outright, hire data-scientists to do the ML, and then offer a product. The end result, economically, is what you say: the money is in data hoarding. But it isn't apparent in prices of a sustainable customer ecosystem, it's apparent in acquisition prices for startups with data vs. salaries of data scientists. [1] https://en.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.wikipedia.org/wiki/The_Nature_of_the_Firm
- goatinaboat 7y ago
  easy transactional boundary that can keep the data ownership and ML in separate firms There is. Data broking is a well established business model in finance. Every bank and hedge fund is running ML models on data licensed from a provider such as Reuter’s, S&P, etc there are dozens of such brokers.
- littlestymaar 7y ago
  > DRM for datasets isn't really a thing, since to explore, visualize, and train on them, you need access to the raw data, and then instead of your machine-learning function you can just pass the identity function to get the raw data. Functional Encryption[1] is aiming at this exact problem. It's still really early stage, and it currently has a lot of caveats (the current algorithms are really expensive and only able to work with linear function AFAIK), but the field is moving rapidly, and we don't know how far it will get. There's also Fully Homomorphic Encryption [2], which is way cheaper and more versatile, but in that case, only the owner of the data can get the result of the calculation. None of this is currently used at scale, but their is a lot of research on this field and some PoC are being built at Microsoft and IBM IIRC. There's also Cosmian[3], a French startup working on this topic. (Full disclosure, I know the founders) [1]: https://en.m.wikipedia.org/wiki/Functional_encryption https://en.m.wikipedia.org/wiki/Functional_encryption [2]: https://en.m.wikipedia.org/wiki/Homomorphic_encryption#Fully_homomorphic_encryption https://en.m.wikipedia.org/wiki/Homomorphic_encryption#Fully... [3]: https://cosmian.com/ https://cosmian.com/

by papln 7y ago

killjoywashere 7y ago

Um, yes and no. Some of the data annotation problems get really hard and really expensive. Look at radiology or genomics. You literally need N people to die for more knowledge in some of these situations.

jnwatson 7y ago
Data cleaning and data labeling are two separate tasks, with vastly different workflows. For example, getting structured data out of Whois requires a lot of cleanup. Labeling domains as “bad” might take forever.

nostrademons 7y ago

This is an interesting observation that deserves to be addressed. I think in practice, what happens is that there's no easy transactional boundary that can keep the data ownership and ML in separate firms. The theory of the firm [1] states that firms arise when transaction costs are less than the economic inefficiencies of centralized resource allocation. There are some pretty heavy transaction costs between doing data cleaning and data science in separate organizations: 1.) DRM for datasets isn't really a thing, since to explore, visualize, and train on them, you need access to the raw data, and then instead of your machine-learning function you can just pass the identity function to get the raw data. DRM for consumers always relies on a publisher whose incentive is to stay in business (by not breaking any laws) rather than to obtain the raw data. 2.) You don't know whether a cleaned data set will be useful for your ML application until you've inspected it, visualized it, run some statistics over it, etc. at which point you've done most of the work for setting up your models. That means there's a big risk premium for buying data, and potential buyers usually want samples & statistics before committing a lot of money. 3.) At the price that good datasets go for, you're usually dealing with enterprise sales, which involves commissioned salespeople, face-to-face meetings, expensive dinners out, etc. 4.) The type of labeling you need to do is often intimately connected with the usage of the ML model. What constitutes spam? What constitutes abuse? These are questions for your policy team, which the data-collection organization would have no visibility into and no way to set up a one-size-fits-all policy that all potential customers would be okay with. 5.) Software markets tend to be winner-take-all, which means you're dealing with a handful of customers, and one will likely become dominant and able to acquire you. Instead of separate firms for data ownership, ML, and user interface, usually the service that makes the final product will end up collecting or buying the data outright, hire data-scientists to do the ML, and then offer a product. The end result, economically, is what you say: the money is in data hoarding. But it isn't apparent in prices of a sustainable customer ecosystem, it's apparent in acquisition prices for startups with data vs. salaries of data scientists. [1] https://en.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.wikipedia.org/wiki/The_Nature_of_the_Firm

goatinaboat 7y ago
easy transactional boundary that can keep the data ownership and ML in separate firms There is. Data broking is a well established business model in finance. Every bank and hedge fund is running ML models on data licensed from a provider such as Reuter’s, S&P, etc there are dozens of such brokers.
littlestymaar 7y ago
> DRM for datasets isn't really a thing, since to explore, visualize, and train on them, you need access to the raw data, and then instead of your machine-learning function you can just pass the identity function to get the raw data. Functional Encryption[1] is aiming at this exact problem. It's still really early stage, and it currently has a lot of caveats (the current algorithms are really expensive and only able to work with linear function AFAIK), but the field is moving rapidly, and we don't know how far it will get. There's also Fully Homomorphic Encryption [2], which is way cheaper and more versatile, but in that case, only the owner of the data can get the result of the calculation. None of this is currently used at scale, but their is a lot of research on this field and some PoC are being built at Microsoft and IBM IIRC. There's also Cosmian[3], a French startup working on this topic. (Full disclosure, I know the founders) [1]: https://en.m.wikipedia.org/wiki/Functional_encryption https://en.m.wikipedia.org/wiki/Functional_encryption [2]: https://en.m.wikipedia.org/wiki/Homomorphic_encryption#Fully_homomorphic_encryption https://en.m.wikipedia.org/wiki/Homomorphic_encryption#Fully... [3]: https://cosmian.com/ https://cosmian.com/