Hacker News Reader

That seems pretty unlikely for a mid sized, independently operated/capitalized Japanese company tbh.

Why? Because they don't have enough users for this to be worth it you mean?

Unless they want easy money. Show me a modern mid-sized, independently operated/capitalized corporation that doesn't want it.

Having started numerous tech companies myself, interviewed hundreds of others who have, and started a community that instructions others on how to do so, I will say that HN's perception of how easy, profitable, and common it is to have a business model focused on selling data is overblown. I'd wager the vast majority of data collectors (e.g. Google Analytics accounts) just want it for their own internal decision-making and analysis.

As much as I would like to believe you, I cannot. With literally almost every company around me gathering analytics and with every average person's operating system, web browser, social media, phones, televisions, smart locks, smart fridges, graphic card drivers and graphic tablets sending this data out in bulk, you are not going to convince me that the authors of all that software are siphoning data out of us all for nothing.

> you are not going to convince me that the authors of all that software are siphoning data out of us all for nothing.

That's not what they are saying at all, and suggesting it's for nothing proves you don't understand the value of using the data for internal decision making. Simply put: the main use of this data is for that reason: internal decision making. Answering questions like:

* How are our customers using our products? * What errors are they experiencing? * What features are they using? * Where are they confused? * What features cause the most problems? * What feature should we work on next?

These are all regular questions that are answered by collecting these types of metrics, including the one described in this post. Selling the data to third parties isn't easy. The data is generally gathered to inform product decisions, not to sell, so it's not in any easy format that makes it easy to sell. One has to go out of their way to sell this data, and the cost to put together this data in a way that's useful to sell would almost certainly cost more to setup and manage than they would get from such a small number of relative users.

The simple fact is, everyone sends data back to their servers for collecting and parsing, including Apple, the company everyone puts on a pedestal for privacy.

Simply put: show me the evidence they are selling the data to third parties for profit. Anything less is speculation.

Maybe you're right, maybe you're wrong.

But trusting companies to do the right thing is untenable. That trust has been broken far too often, by far too many companies. The only rational position a concerned user can take is to assume that anybody collecting such sensitive information (particularly in a sneaky way) intends to monetize it or use it for purposes other than product improvement.

And even if the data really will only be used for product development, getting the user's informed consent -- and refraining from data collection without it -- is critical.

Further, using GA automatically means that the data is being used for Google's purposes as well as the application creator's.

Can you point me towards the marketplace where I can sell data I've siphoned off? Thanks!

If you have large data-sets/streams, companies like Acxiom and Innovis are interested buyers. Are you trying to imply that brokering consumer data is a marginal or non-existant business?

> Are you trying to imply that brokering consumer data is a marginal or non-existant business?

For small/mid-size datasets (which is what we're talking about), yes, that's exactly what I'm implying. It's not actually easy for most companies to sell user data for a quick buck like is being claimed.

I know this is the popular conclusion - but from a developer perspective, data such as what software your product is used with is INCREDIBLY valuable for project planning and prioritization. People hate web tracking, but knowing what browsers or devices are visiting your website can be enormously helpful.

Also I don't know all the details here - I know that Vizio TVs where collecting data and explicitly kept the IP and other personal data with it. I don't know if wacom is doing that.

Now that said - I don't like that they're handing this data to Google through Google Analytics. I also think they should be far more up front about what they collect, what they use it for, etc.

> I know this is the popular conclusion - but from a developer perspective, data such as what software your product is used with is INCREDIBLY valuable for project planning and prioritization. People hate web tracking, but knowing what browsers or devices are visiting your website can be enormously helpful.

What happened to actually communicating with users to learn more about how they use the product?

That assumes a great deal about both the resources available to do it and the ability to get responses from a representative sample of customers.

It's not challenging to see why someone might choose a one-time cost in software engineering over an ongoing cost in communication.

It's not challenging, but it's wrong. If you want to explain it to them, making a binding promise of what it would be used for, and have it be opt-in, that would be another thing.

It's not hard to see how to do it right.

Morally and ethically, you're absolutely correct in every single possible way.

I was attempting to illustrate the decision-making processes that may have led to this juncture and what happened to communicating with customers. Please accept my apologies, as I have plainly failed to be clear that this was not an argument as to the moral or ethical questions concerned.

Again, you're completely right. It's not at all difficult to see the morally and ethically correct way to go about this.

I really like the approach that the Steam survey uses which blends engineering and communication:

1) Pop up and ask for permission to scan the machine.

2) Show the data collected that will be sent back and give a second chance to decline.

3) Allow everyone to see the aggregate results.

Being mostly automated, it's lower friction than a manual Q&A survey. But it also feels way more respectful that trying to snoop around and then clandestinely exfiltrate the data. It's one of the few cases where I'm willing to opt-in to data collection.

This. You aren't asking for vague blanket permission, nor are you asking for the user to manually fill out a survey. And you give them the opportunity to review what they're about to send.

Great point! And being able to see the results in aggregate is also interesting. It inclines me to share, because it becomes a two-way share, even though I don't actually have use for the information.

Being able to see the aggregate data isn't just interesting but publicly helpful.

From a game developer perspective, looking at it right now tells me that (simplified):

* Most gamers have at least a GTX 1050 and 8GB of ram or higher. Perfect now we know where to aim our medium settings.

* 74% use Nvidia GPUs, 15% use AMD - now we know where to focus driver optimizations

* English, Simplified Chinese, and Russian are the top languages (where to focus translations)

* 72% play on 1080p, 14% on 1440p, etc. Tells us what resolutions to make sure our UI works on.

Users stopped communicating. They install a thing, they uninstall it. 99% don't leave any comment at all.

(and yes, the obverse inference is also true. If you see one person complaining, there are probably 99 more who have had the same issue and have said nothing)

That's not an excuse, but a reason to pay for a focus group or study.

Slashed in pursuit of VC money and profits.

Wacom's a public company. There's a lot of things they might be chasing, but VC money seems unlikely to be high on that list.

> What happened to actually communicating with users to learn more about how they use the product?

https://www.nngroup.com/articles/first-rule-of-usability-don...

"watch users as they attempt to perform tasks"

Like a survey? I'm not sure the information would be all that accurate.

If you want to put resources into "hey folks like to use this product with ours" you need accurate information.

Do what companies have always done: pay for user studies and focus groups.

Big companies maybe.

It's a confirmation bias problem. Customers who fill out surveys and give direct feedback back to vendors are frequently lower-usage enthusiasts that aren't representative of your overall userbase.

I work on a product, and we include some telemetry. I'm also a strong privacy advocate, and I believe I've done my best within the corporate realm to ensure that the data we're collecting is extremely scoped AND useful for decision making and prioritization. In my experience, there aren't that many of me, but I implore folks to realize that as PMs and engineers, we absolutely do have a say in making sure that blanket data exfiltration and aggregation doesn't happen in our products.

Communicating with your customers proactively about what you're collecting and why is important too. And not buried in some privacy policy legalese: publish a blog post, explain what you're gathering, give examples of how it's driven decision-making for you in the past, and what you're hoping to learn in the future. It goes a long way.

As a user, even if you're up about these things and why they're important for product development, my answer is still "No."

Because I will only say "Yes" if I'm being paid or otherwise compensated specifically for that input.

I think it's important companies pay for usage testing so that they value that information and are more likely to hold it closely since it represents an investment and is perceived as competitive advantage.

That's why a sane data collection is aggreggate, upfront about the data it collects, avoidable, and most of all, explicitly opt-in. Alas, this requires eternal vigilance, as the pushback is neverending.

I don’t think anyone is denying that the data is valuable, but what is usually missing in the implementations (not sure if true in this case) are 1. Transparency about what is being collected, 2. Requesting consent from the user and 3. Providing control to the user in the form of an opt-in or opt-out.

Transparency, consent, and control.

If every company addressed these three issues, we wouldn’t be having this conversation about privacy and data collection over and over and over.

I fully agree that they need to address those problems. I really like the way Valve does it with their Steam Hardware Survey.

What I'm addressing is that I feel many people see a company tracking data, and assume this data is valuable enough to sell, and that the data is for sure being sold.

My point was that the data isn't just valuable to sell (maybe), but is legitimately valuable in making a better product/service.

> I know this is the popular conclusion - but from a developer perspective, data such as what software your product is used with is INCREDIBLY valuable for project planning and prioritization.

Maybe if it were only used for that it wouldn't be so bad. But I don't trust a company not to take another bite at the apple by selling customer data if they think they can get away with it. Matter of fact, refusing to do so is leaving money on the table and could get a CEO fired for not making the company as profitable as it could have been. Once companies have the data, they are almost certainly obligated to use it in ways to their benefit and your expense.

GDPR helps a litle with this.

Not really, it depends on the revenue strategy. Shareholders would likely be more displeased at initiatives that could end up breaking user trust and harming core revenue (e.g. the sale of peripheral hardware). Blindly leveraging everything in a way that doesn't align with vision or strategy generally leads to disappointing returns. Smart leaders know this.

> from a developer perspective, data such as what software your product is used with is INCREDIBLY valuable for project planning and prioritization.

I don't think anybody is disputing that. But that its very valuable to devs does not excuse collecting it without getting the user's informed consent first.

It appears to be this communities' commenters unpopular conclusion. That every piece of tracking is part of a conspiracy to sell you advertising.

> In section 3.1 of their privacy policy, Wacom wondered if it would be OK if they sent a few bits and bobs of data from my computer to Google Analytics, “[including] aggregate usage data, technical session information and information about [my] hardware device.”

What wasn't upfront about this? That they didn't add more details about what the session information was? Legally why would they? The post includes an image the section where they legally disclosed it. People not reading the privacy policy before using a product is not Wacom's legal problem.

Can you ask them to put this section on a separate screen? Sure. Will they do it? Who knows. I'm sure they'd want to know if you are a customer giving ideas than a low priority non customer as would any person.

How many blogs or websites disclose the use of Google Analytics in their privacy policy?

You could talk to many customers and this is the least thing they have on their minds. Paranoia displayed by commenters here is amazing.

As the post concludes, if you are a (prospective) customer who does not like what they collect then there are other brands. I might add who probably have a hidden, more intrusive way to track you because they are smaller, have smaller volume/margin and have the incentive to build and sell your profile like other small companies not in the field.

> IMO the more simple explanation is they want this data to sell to data aggregators

Seems the obvious answer, yep.

It's certainly the more cynical explanation.

I believe wanting it for product development is just as simple.

Can you even export this information with PII on Google Analytics or only aggregate of that information?

Only aggregate

Then if the goal was to sell, they wouldn't use Google Analytic.

It would be "simpler" to say, but that doesn't invalidate the other reason.

My tablet behaves differently per application. If I typically have one app open only on one screen I can limit the tablets "workspace".

Context-specific buttons based on app.

And if you're doing that _and have build sufficient app infrastructure around it_ as Wacom has to support fairly custom per-app behavior, the more realistic conclusion is that they're trying to get more info on that - now you can argue about opt in on the "share experience data" privacy setting - and I would agree, absolutely.

But "more simple to say that they're just selling data for money" is a pretty reductionist argument that jumps solely to the most negative possible motivation. "What's the worst they could be doing with it? Selling it? That's probably what they're doing, not making their tool more useful."

> IMO the more simple explanation

For me Occam's Razor points towards internal analytics.