The Rise of Workplace Surveillance
theguardian.com
theguardian.com
This is what happens you try to calculate productivity with how much salary you get.
The poster make video player in excel to padding his clock time because he already finished his work in 1 hours and need 3 hours to appear busy.
> The concept of the design is to allow all (pan-) inmates of an institution to be observed (-opticon) by a single watchman without the inmates being able to tell whether or not they are being watched.
It's worse than a panopticon as in a panopticon prison you have one person surveilling 1000 for example but only able to watch one at a time. The idea is the prisoners never know who is being watched so in theory they all behave thinking they are all being watched. But math kinda destroys that theory as they each have at least a 1000:1 chance of being watched so the odds are in the prisoner's favor.
In regards to the math, I don’t believe that humans perceive risk with such objectivity. If the cost of being observed is high, then even a low risk may impose constant anxiety. Of course, perhaps criminal behavior selects for a group that tends to undervalue the costs and under estimate the risk.
Foucault was prescient about the sorts of harms panopticon would cause, and the metaphor is still an informative one. In particular, it was an insightful as to psychology under uncertain surveillance. But people are still calling it hyperbole when in fact we surpassed it years ago.
Panopticon had vulnerabilities: a brief or infrequent transgression could go uncaught, and an uncaught transgression could never be recovered. The current situation is that active surveillance can be applied algorithmically to everyone, and 100% of past data can be retained for later analysis. That's utterly unprecedented, and I don't think we've really done much to grapple with what it's going to mean.
The problem is that reducing negative behaviours in a just fashion is not the only, or even the most likely, possible dynamic for such a system. In a tyrannical regime (political, social, economic: this is not strictly related to politics), there's a great deal of power and control to be had by acting arbitrarily, without predictability, and without recourse, at any given time.
Take the example given in Paths of Glory, a 1957 film telling the story of the (arbitrary) execution of three French soldiers who had refused to undertake suicidal attacks in the trench warfare of World War I. The process draws on the ancient tradition of decimation, which is the arbitrary killing of one in ten of ones own troops* in order to instill on the remainder the consequences of failure to follow command.
The real risk of global surveillance is that it provides either the justification for, or the plausibility of justification for, arbitrary prosecution of any given individual at any given time.
That's an absolutely despotic state.
"Last year an employee at an IT services company sent a private chat message to a friend at work worried that he had just shared his sexual identity with his manager in a meeting and fearing he’d face career reprisal. Wiretap detected the employee’s concern and alerted a senior company exec who was then able to intervene, talk to the manager and defuse the situation."
What. the. fuck.
thanks "senior exec" for reading my messages, stepping in and "defusing the situation" whatever that means.
it gets worse:
"Or if you usually touch 10 documents a day and print two and suddenly you are touching 500 and printing 200 that may mean you’re stealing documents in preparation of leaving the company."
of course, that's the only possible logical conclusion.
What a waste of time, if I ever saw this in place as an employee or manager I would immediately leave and possibly sue.
What is wrong with people?
I can see it being argued both ways; transcripts of Bloomberg chat were a key part of the LIBOR scandal, for example. And I can also see there being some need to monitor internal chat in case it's used for unacceptable purposes such as sexual harassment.
Something about the automatedness of a very private situation gives me the creeps, though.
The key here is expectation of privacy. If you're posting on a JIRA ticket, you have no expectation of privacy within those with access to the board, nor if you send an email as it can easily be forwarded (although that could be contentious). If you send a message through a DM on slack though, you can expect that to remain private indefinitely.
Broadly, automated surveillance creates illusory offenses. It raises issues based on some imperfect set of rules, and does so independent of any report of harm. (Add to that the likelihood of unethical flagging - which happened right in the article!)
And that creates an obligation to act for employers. Ignoring alerts, no matter how silly, is the sort of thing that might get cited as negligence when a real problem does come down the pike. At best, the result is time-consuming investigations of fake issues. At worst, corporate makes a decision that summary firings are an easier bet than proper investigations.
That's not just a hypothetical fear - we already hear about things like Amazon warehouse employees being laid off on vague suspicion of theft, because investigations are harder than replacing staff. Expanding always-active surveillance will predictably hurt innocents who already lack job security.
You don't want simple monitoring. You want data stored in a way it can be checked after a complaint has been filed but only then. Encrypt it with something which requires one or multiple keys so random IT people don't have access to it. Log every access so an audit can be done.
Treat it like medical data should be treated.
...And I say that as someone who has had problems with remote workers in the past.
If it seems like someone is stealing documents, but you then discover they actually aren't, then no harm done.
If someone actually was stealing documents however, and you don't have the surveillance to realize this, your best salesman might have just took off with a bunch of client information to go work at a competitor. You might never find out, and if you did, you might have no way to prove it.
Workplace surveillance doesn't really have downsides, other than "workplace culture might be affected". However your culture can still be shitty even without surveillance, and just because you don't monitor employees doesn't mean they will become shining examples of company loyalty.
There's plenty of upsides though. Having data on what employees do, and when, and for how long, provides opportunities to discover where improvements can be made, as long as you collect the right data and use it for the right purposes.
If my company accused me of that I think I'd run a mile, so there's some harm.
There would be rampant abuse with this kind tool in place, as it give cause to fire anyone at any time, it would not be hard to find a reason to fire someone with this level of surveillance
If a company really wants to fire you they will find a way with or without surveillance.
Soon we'll have slavery again; except instead of your status being determined by your race, it will be determined by how much empathy you have. People who have empathy for others will be slaves of narcissistic, greedy corporate assholes.
Sure, he agreed it was stupid, but nothing he could do about it.
Sometimes, yeah, there are major red flags with this idea, but sometimes we can think that they are only affecting us, not the entire company. Schadenfreude? Oh you bet. But it still stinks for the local schools as their tax base is gone.
Also, the app does pop up with alerts every so often suggesting you take a yoga break so it is adding a lot of value. /Sarcasm
No-one plays Solitaire on a supermarket checkout till
(cue someone finding a photo of someone playing Solitaire on a supermarket checkout till)
Difficult in a supermarket checkout scenario though.
32.9% of U.S. employees "engaged" in workplace in February (2015)
http://news.gallup.com/poll/181895/employee-engagement-reach...
and from a 2014 report
http://www.aon.com/attachments/human-capital-consulting/2014...
There are signs that the employee value proposition is breaking down. Perceptions of internal brand alignment have decreased 4 percentage points in the last two years. More employees are saying positive things about their organization and striving to go above and beyond. Engagement overall has increased, but the global element signifying intent to “Stay” with their companies has shown no change. Employees are engaging more, but only a little over half see a long-term path with their current company and fewer see a compelling value proposition to keep their talents with the current company.
This is a case of paradox of composition -- whilst at a personal scale improved communications can increase trust, at a mass scale, the tendency seems to be to undermine it. This increasingly strikes me as a problem.
https://www.reddit.com/r/dredmorbius/comments/6jqakv/communi...
In a world without high-speed, high-bandwidth, rapid, and reliable communications, you have to extend, and rely on, trust between individuals. Cultures evolve systems (usually religious) to create and foster a sufficiently-reliable trust network.
As communications improve, reliance on that trust diminishes. You no longer need to be able to rely on a person working in your interest for days, or weeks, or months, or years. You can check on them at a moment's notice. You can monitor them continuously, across a wide range of metrics, without their conscious awareness.
A domain built on Trust becomes instead Panopticon.
(Further discussions of trust: https://www.reddit.com/r/dredmorbius/search?q=trust&restrict...)
The beauty of Bitcoin is an inherently-trustworthy transaction structure without the need for a central authority or trust between parties. The cost is exactly what you describe - trust is replaced by a massive expenditure of work to independently verify everything. It's an interesting microcosm of what's happening throughout society in general.
Decentralisation is the paramount feature in bitcoin, but it turns out that that's a bad idea that's really, really expensive, because it turns out that a tiny bit of trust saves you a fortune.
"Decentralised" isn't a useful buzzword in a lot of ways, because it turns out that you want to be a part of society.
From his interview with the Financial Times:
If the company can connect to the user computer and install software there, that means the user just opened up his computer to the company. And, if the company software happens to be "buggy" (to say the least) it could mean that company has teorically hacked the employee computer and now has access to every single file on that machine and if it wants it can try to connect to other machines on the same network. If the company software malfunctions, it could delete all the user files or share with others in the internet.
Having said that, its probably important to separate. As a last resource if the company really want to monitor the employee, use 1 internet for company (with 1 computer only for work related) and 1 internet for home usage (and home computers).
Edit: typo
Just because someone is paying you doesn't give them carte blanche to examine every minute of your time working. That was never explicit before because it was never possible before, but i guess we need to make it explicit now.
You see,I am at a point where I wish it stopped at work. $work these days spies on your off work activities as well. Essentially making you their slave.
If you don't experience this,then maybe your company isn't big enough to afford it(or if it is,please tell me more about this wonderful company that let's you do your job without 24/7 surveillance)
EDIT:Just search for " user behavior analytics".Orwell would be impressed
Ultimately its culture and the decision makers get to decide. Monitoring is of course downwards and those making the decision do not get to be monitored in this intrusive fashion.
Edit: it also helps when firing people, as you can make a list of metrics and then make goals that are just out of reach for bottom 50% of performers. Then when you need to fire people and deny unemployment, you point to the metrics they were failing and you have a valid cause to terminate for lack of performance, thus saving you money on unemployment insurance premiums.
If a worker is expected to make 10 pieces of an item per hour its easy for managers to verify this. For a knowledge or creative worker it's similarly easy to verify expected performance.
This is not new, workers have been measured for performance everywhere since manufacturing and corporate jobs began.
If we ignore that for a moment, the point you are making applies to management too, yet this tech is guaranteed to not be used for management where things and outcomes are left 'more fuzzy'. Why?
How about we apply it to governance with complete transparency to the electorate. Nope, not going to happen. Because its exploitative and intrusive and will only be applied to the powerless lowest who have no choice.
No it really isn't! Coding performance metrics are a wasteland of failed techniques.
"I'm gonna write me a new minivan this afternoon!"
Meanwhile the history of the field is one of judging programmers on bugs fixed, lines written, commit frequency, or a hundred other standards that are only perverse incentives. Most aren't even Goodhart's law issues, because they weren't predictive before they became targets!
How about we apply it to governance with complete transparency to the electorate. Nope, not going to happen. Because its exploitative and intrusive and will only be applied to the powerless lowest who have no choice.
I don't agree that creative workers have similarly easy to verify metrics, even now. However, with the ability to broadly track information, what is now possible is to view the distributions of output so that over time, an individual loses the ability to provide excuses, because chances are if your constantly in the tail end, then it's probably worth it for the organization to be rid of you.
I would love it if could apply this transparency to our governance, but obviously those with power will oppose it, and those without will have to fight for it. It's just the nature of things. Obfuscation makes it easier to arbitrage (make excess profits, i.e. have more power than others), so it is to be expected that should continue.
I'm not convinced this is true. Management may have been given more leeway on 'soft' outcomes, but there's a longstanding trend of (messy, failed) attempts to quantify that role. Every attempt to incentivize cost-cutting, scale pay inverse to raises given, stack rank between teams, or otherwise measure staff outcomes is in effect a performance metric on management.
It's true that most of those have been immediately eaten by Goodhart's law, and we don't know how to quantify actual success like we do for "widgets per hour". But I wouldn't suggest that a lack of will to do it has been the problem.
(I do think this argument applies to the top of the executive level. I'm fairly convinced that most non-founding CEOs could be replaced at a discount with minimal harm, and only market distortions are preventing that.)
-Allegedly true story from the DDR (GDR to you anglophones): Once citizens could start asking to see what the former East German state had on them, Erich Mielke (Head of the Stasi/the GDR KGB equivalent) asked to see his file.
Presumably, he figured that someone, somewhere, watched the watchmen.
There is another alternative - acceptance - something which is much easier on your mind. Acceptance isn't an option for everyone, however.
On which jurisdiction(s) does this apply? Readers (pref lawyers) should check if this is legal in their jurisdiction(s).
For people from NL I can recommend Arnoud Engelfriet's blog (a Dutch lawyer specialised in IT law) [1] and its search feature.
An ontopic example discussing whether an employer is allowed to see private files in a business OneDrive account is discussed [2]. There are many more examples, but the blog is in Dutch and generally applies to Dutch law, so YMMV. I'm curious if similar websites or platforms exist for other jurisdictions.
[1] https://blog.iusmentis.com
[2] https://blog.iusmentis.com/2016/10/17/mag-werkgever-privebes...
Wait, what? Are they trying to do ADHD diagnosis from window activity?
Uhh, no. Employees are adults.