Slashdot Mirror
Data Mining Moves To Human Resources
theodp writes "Just when you thought annual reviews couldn't get worse, BusinessWeek reports that HR departments at companies like Microsoft and IBM are starting to use mathematical analysis to determine the value of each employee. At an undisclosed Internet company, analysis of (non-verbal) communications was used to produce a circle to represent each employee — those determined to generate or pass along valuable info were portrayed as large and dark-colored circles ('thought leaders' and 'networked curators'), while those with small and pale circles were written off as not adding a hell of a lot. 'You have to bring the same rigor you bring to operations and finance to the analysis of people,' explains Microsoft's Rupert Bader. Hey, who could argue with what Quants did for finance?"
That's why their stuff is so bloated and slow.
Often, the problem with companies is that information doesn't get spread around. People work on their own projects in secret, never bothering to spread their knowledge. Perhaps this will urge some of the those to pay some attention to being useful for other employees as well, as opposed to just getting their own little project done.
Save your wrists today - switch to Dvorak
I believe quantifying employee "importance" by the number of email conversations they had and who read what they wrote is pretty silly. Soon they'll fire all their network admins because they all are represented by small-ish pale circles that usually reside in some dark basement bureau.
Can business get any more dehumanizing? I don't think so. I at least wouldn't want to work at a company like that. From TFA:
"You have to bring the same rigor you bring to operations and finance to the analysis of people," says Rupert Bader, director of workforce planning at Microsoft
Can you say fucking stupid, kids? Humans are not machines (at least not yet), they have bad days and bad weeks and some have bad decades (imagine your child dies). Evaluating them through "rigorous" methodological measures is pure idiocy.
I thought this was common practice. Sorry, I never worked H.R.
The cost of that cleanup, of course, will be borne by taxpayers, not industry.
I notice that it is always the large companies that try to do stuff like this, not the smaller companies and businesses. If this concerns you, reevaluate who you work for. May be tough in these economic times but there are still job opportunities for if you look hard enough (and really, smaller businesses to not hire from Monster.com/Dice/Local Newspaper).
so, if you don't actually do anything productive, but just forward other people's information all the time, you are "a valuable worker"?
Ah. Yes this would be the 'Brazil' solution.
"I'm sorry Mr Jones, our database says that you are a statistical outlier and that you should be dead by now."
*pulls out gun*
"You must become compliant."
It may be interesting as an approximation, but people really should know who their good workers are without these tools.
and live in perpetual fear of being outsourced. Seems like a lose-lose proposition to me.
At an undisclosed Internet company, analysis of (non-verbal) communications was used to produce a circle to represent each employee â" those determined to generate or pass along valuable info were portrayed as large and dark-colored circles ('thought leaders' and 'networked curators'), while those with small and pale circles were written off as not adding a hell of a lot. 'You have to bring the same rigor you bring to operations and finance to the analysis of people,' explains Microsoft's Rupert Bader
This is rigorous analysis!?
So in other words they want schmoozers and suits, not people who are busy.. working?
Great. So just stand on the throttle until you hit something. Because that worked so well for the economy.
Given the widespread deployment of technology to filter resumes, HR is ripe to accept any new technology that is thrown its way.
Surfacing "thought leaders" over others amounts to rewarding a personality type. I don't think companies have a problem rewarding the people who influence others. The people who do the heavy lifting are rarely recognized.
I would joke that this could be a good thing, in that, we'll just game the review system to get raises.
The reality is, though, that the more corporations seek to control and monitor their employees, the more they will crush the entrepreneur in them. Corporations work best when they motivate people and you do that by creating a positive, team culture that gives its participants a sense of mission. Take that away, reduce people to cogs, and you are going to get cogs as a result, and you'll get an inevitable decline. What enterprising person would want to work as an anonymous cog, coming out of college with a degree and history that says they are anything but, when they could make a real difference at a startup.
Actions like this doom large corporations, and frankly, this sort of thinking was what alienated the big 3 for a lot of people, and now they want to do this to the computer industry?
Stupid, stupid, stupid.
This is my sig.
I knew my boss was an idiot. Now I can tell all his emails, run them through the HR analysis program, and prove it. Yay! :-)
"I disapprove of what you say, but I will defend to the death your right to say it." - historian Evelyn Beatrice Hall
those determined to generate or pass along valuable info were portrayed as large and dark-colored circles ('thought leaders' and 'networked curators'), while those with small and pale circles were written off as not adding a hell of a lot.
So loudmouths that brag (non-verbally... ok) every time they managed to piss without getting too much on their pants get promoted while people who quietly do their work get the shaft. Anything else new?
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
I love how they say that only a certain percentage of employees can be meeting, exceeding, or failing to meet expectations.
I have a relative working for a large appliance manufacturer that pulls this crap. Even if every employee did a perfect job (hypothetically speaking), there has to be a least a certain amount of them that are labelled as underperformers.
"Oh. Well, yes, Jimmy, we agree you did a perfect job but since you didn't wear matching socks on August 14th, we'll have to put you in the underperformers category."
Let me tell you about why QA exists in the first place. You see...at one point in time in every large company, there was no QA department. Then, a bunch of sales managers, temps, underperforming execs, obsolete trainers, and lazy HR personnel were all about to get canned for their uselessness. Then, they go to the company president and say, "Hey. You know...we really need to do some 'improving' here at XYZ Corporation. My associates and I have come up with a plan to monitor and track employee productivity and customer satisfaction based on this rigid set of next-to-meaningless criteria. There's always room for improvement, you know. And guess what? The best part is that since we have a plan to apply quantitative figures to qualitative matters, you'll be able to screw most people out of raises and bonuses as often as you'd like! All you have to do is let us keep our jobs."
And, of course, the higer-ups fall for it hook, line, and sinker. Idiots.
Why don't they just fire everybody with less years of formal education, or degrees in non-technical fields? That's what big companies use primarily to screen hiring candidates.
Because people with advanced degrees are surely going to be better employees right?
Seriously, I've seen only an inverse relationship between years of formal schooling and job performance. Yeah yeah, the ones who got selected had to be that much better to overcome that hiring criterion.
Shows you that metrics applied to people are only as good as the people who design them. Generally, they're dangerously bad.
How can you form any kind of social bonding in a company when your worth is distilled down to the results of some fucking mathematical formula? I'm not naive to think there's any concept of loyalty or trust in the modern business, but man, things just keep getting worse.
Forget even referring to us by name anymore, just give us numbers if you're gonna stop treating us as human fucking beings.
Does it bother anyone this is the same type of gadget analysis that got us into the current economic situation? Your most valuable employees aren't always the most communicative.
We have one developer who shuns any type of contact, doesn't have a phone on his desk, rarely sends an email longer than two sentences. Yet he's the most heads-down, dogged and prolific programmer I've ever worked with. I suppose the gadget developers would argue that would be accounted by how often his code fragments turn up in other projects but how do you really account for the source of a code fragment? Especially one that is later modified for other uses?
I can see a lot of bad conclusions coming from this kind of analysis. Where the most outgoing employees are valued over those actually meeting deadlines. So you end up with a company full of a lot of talkers and lay off all the actual doers. Which, come to think of it, is pretty much how we got in the economic mess we're in.
That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
you have no need for this kind of "rigorous" (bah) methods.
Swedish plasma phys. PhD student; MSc EE; knows maths, programming, electronics; finance interest; seeks opportunities
BUT I HAVE PEOPLE SKILLS!
OK. So by this metric, people who forward chain letters and jokes will get a better rating?
Let's say you have a company that makes widgets. One of the people in the widget design office is a bit of a dork. He's a musician, a quiet and not very sociable bass player. And this is his day job. He works hard enough to keep his job but not much more. One day, he comes up with an idea that is dead brilliant, and then goes on tour. The idea saves the company millions of dollars. And let's see - he always comes in late, frequently hungover, kind of smells, and tries to leave early. He doesn't do that much when he's a round, and he's often not around because of his band.
But, in one afternoon, he has been of more use to the company than all other employees in Widget Design combined, ever.
By the metrics described, he would have been laid off upon return from tour.
Typical fuckwittery by HR bozos.
The best companies don't have HR, except in terms of processing new hires, dealing with benefits, and assisting people on the way out. The rest is left to the departments and managers. It makes for a flatter and faster organisation - ideas M$ has no clue about.
RS
Shoes for Industry. Shoes for the Dead.
All this means is that the people emailing links to porn sites will get the first promotions.
Deleted
There's a huge and erroneous misconception that centralization makes a corporation more efficient. I think centralization is a cancer. How often do mergers actually work? How often do governments actually execute well. The biggest failing of the free enterprise system as of late, is that, after hearing all of this about how government is inherently wasteful and inefficient and choked with slackers, that corporations set themselves to be operated just like governments. Just look at the result!
The fact of the matter is, that the thing that matters most in any corporation is time to market. It doesn't matter if you are centralized and more "efficient" if it takes you two years longer to ship a late product out the door, because while your smaller competitors were signing stuff and building things, your own design was going through committees and signoffs to make sure that you weren't doing what someone else already did.
Like, the stupidest thing GM ever did was to try and share so much data across so many divisions. What they should have done is just run each different division as a separate company, responsible for one thing - the bottom line. If they don't produce, then close them down. But instead, they have a huge corporate system that makes it very difficult for them to bring a new car design to market. And, by the time they get there, what started out as an award winning design is so late that they get slammed for making a mediocre product by the trade magazines first and the consumers second. All that's left of that company is Bob Lutz heroically pushing through car designs, but once he's gone (he's retiring), that company is screwed.
I think the larger story is, really, that management education in the United States is a colossal failure. There is no reason that a large and previously successful company needs to decline and fail when other civilizations created empires and institutions that lasted for hundreds and thousands of years. But as it is, in America, as soon as a founder leaves a company, the MBAs get in and these "professional managers" slowly sink the ship. It doesn't have to be this way, but it will be this way until we get some serious curriculum changes at our management schools.
That's right: HARVARD, WHARTON, YALE AND OTHER MBAS : YOU F---- SUCK!
This is my sig.
Use of the tool promptly gets stopped when it reveals that upper management adds no value.
My wife just took a new position, because her last boss was an idiot. He was a passive aggressive micro-manager, puffed up with his own self-importance, *at least* 15 years out of date technically, and long since regulated to the most irrelevant corner of the company.
By the metrics discussed here, though, he'd have looked like the hero! *All* had to run though him - customers, suppliers, management, co-workers - if you talked to someone without including him in the conversation, he'd flip. He threatened to fire my wife (and a few more people since) for doing their job without his constant oversight. Unfortunately, while everyone knows about the situation, my wife was the first to report it to HR, so they can only now start to think about taking action against they guy.
Counting the number of communications makes the people who send one word, no value added emails and attend a lot of meetings they don't need to be at look good.
Also, it completely misses your crack team - the 3-4 people who you can hand a problem to, and know that they'll have it solved by next Tuesday, no questions asked. When those people shut their office door, you leave them alone, because you know they are working miracles, and you'll only get int their way.
Web analogy - Google and page rank. Rule number one is that you never trust the page to tell you how important it really is. Pages with all the right keywords and a bunch of links are one of two things - the best of the best about a topic, or an SEO linkfarm. So you take those things into account, but you do so with a *huge* grain of salt. To augment it, you go looking for other supporting metrics - what do other people think?
The HR department has just automated a human approach to the problem - they took one piece of evidence that the human brain can wrap its head around, and made the computer count that. You want to do informatics and data mining right, you need to learn what the computer is good at, and start looking for deeper patterns that are hidden by masses of data too large for the human mind to encompass.
-V-
Who can decide a priori? Nobody.
-Sartre
Unfortunately, employee quality is something that is currently NOT quantifiable.
Co-workers all have an intuitive sense of who is useful and who is not. But trying to measure it with today's crude tools is an exercise in futility. Kinda like measuring productivity by lines of code, or scientific value by number or articles published.
This will end up benefiting those who skillfully learn to play by the new rules, and punish those who may be excellent, but don't fit the standards expected by the measurement methods.
6-month review is exactly like a 'math equation'.
Certain employees produce chunks of data - whether words or software code - that later pop up in other messages. The people copied most often, Cataphora concludes, are thought leaders.
In my experience the code that is discussed in emails is just as likely to be because it is bad as it is because it is good and I'm sure the examples at http://thedailywtf.com/ often "pop up in other messages".
Good managers already know the value of their staff by talking to them, talking to their colleagues and assessing their work. If a manager has to resort to analytics like this at least a corporation knows where their management problems lie.
Engineering only works because you still have people vetting the numbers. However, even there, there are problems that you just need a human opinion, because the engineers can't figure it out. One example - engineers called in to calculate how much you can cut a pile of earth back without shoring it up. None of them got within 50% of the actual number derived by subsequent tests. The solution is simple - call someone in whose work is excavating, and they'll give you a more accurate answer just by eyeballing.
Bottom line: If your boss doesn't know how much your're contributing to the company, then your boss is deadwood and should be fired. No need for statistical analysis to replace common sense (which is what created the toxic CDOs and SIVs, etc)... but the deadwood boss will like this, because now it's not their job to know what you do any more - they can point to a chart.
Short any company using this method.
The above rhetorical question implies, the submitter/editor disagree with mathematical methods. For Slashdot, that's quite a shocker... From the linked posting:
A particular math theory may or may not be flawed, but do we really prefer "subjective beliefs" (a.k.a. "hunches") around here?
In Soviet Washington the swamp drains you.
On the face of it, these methods' reliance on machine-readable communications introduces an incentive to spam colleagues with messages in order to inflate one's score. It also penalizes any form of off-line communication.
Calling a meeting to discuss an issue that could have been resolved in the hallway is rewarded, while taking a minute to share information with a coworker in their office is not.
org.slashdot.post.SignatureNotFoundException: ewg
So sending out funny emails to large groups of people all the time is going to get me a promotion? Sweet!
Theodp's ignorant comment about "quants" and mathmatical ignores one of the primary thing I have heard repeated about the blame being put on mathmatical modeling for the financial crisis. Namely, that it is not that mathmatical modeling was used but rather that only one mathmatical model was used by everyone.
Mathmatical modeling is a tool, like a computer. It can be used properly or improperly. And, in the case of the financial systems, it was used improperly, just like theodp's use of his computer and his post to spread FUD.
There is no "-1 offended" or "-1 you don't agree with me" mod options for a reason.
That was my first thought, too, Clover Kicker. Especially if the metric becomes hated. I wonder if there may be ways around that though, with a well-done staff orientation about the intention behind the metric (what kind of behaviour is it recognizing/promoting and what will that do for the company). Sometimes if you appeal to people's integrity it (amazingly) can work. Also, some kind of cross checking would be needed, to make sure the metric is not just being implemented mindlessly.
My love of work has been very bad for my career.
The cost of that cleanup, of course, will be borne by taxpayers, not industry.
I guess everyone is going to hammer this, so here's a counterpoint. I'm all for better tools to help people development. For far too long, HR and management of people in general has either been too 'robotic' (think Taylorism) or subjective, as in 'your boss likes you, so you get a better raise'.
There's no reason why stuctured approaches, that have worked well elsewhere, should not be used. The real problem lies with lazy, incompetent managers who are everlastingly seeking the holy grail of the 'quick fix'. Now, if a person identified by this method is a high performer but a low communicator, then either his/her job does not require endless emailing, or they are a candidate for some coaching or training.
Unfortunately, interesting and potentially useful tools tend to be abused and hence get a bad name. If some middle-management dick just starts firing people with low bubble-count, all that will happen is that you'll turn the place into an email factory...
When I read things like this, I'm always reminded of Frederick Taylor. If you've never heard of him, he's probably the guy you should thank for such quackery.
This idea of mathematically determining the value of each employee fits very well with his ideas. Face it: in the modern corporate world, humans are part of a system that is, overall, far more important than the individual. It is increasingly a scientifically-managed system, so it should come as no surprise that such dehumanising practices should take place. Business does not want humans; it wants workers.
It is quite a logical outcome of our increasing reliance on scientific principles to explain and analyse our world. I find it ironic that many /. members would hate this approach of analysing workers, yet its roots lie in our reliance on science to breakdown, label, categorise, and figure out how we and our world works. In the same way psychology, neuroscience, and other mind-related fields were bastardised to figure out how to manipulate the human mind to makes us consume, the computer sciences will be used in a similar fashion to make us behave a certain way: if you don't want to get fired, you need to make sure what you do conforms to their model.
Sadly, figuring out the "optimal" and "perfect" workers will, like my .sig says, make us realise just what it was that made us human, instead of just robots.
'If Christ had tweeted the sermon on the mount, it might have lasted until nightfall.' - John Perry Barlow
> The people copied most often, Cataphora concludes, are thought leaders.
If your company adopts this system, stop working so hard and start sending out funny emails and internet memes. Then sit back and enjoy your pay raises as everyone forwards your emails.
"God bless the meek." "Oh, that's nice. They have a devil of a time of it." This shouldn't help.
Actually, what's really scary is it's five minutes in the future from Cory Doctorow's Down and Out in the Magic Kingdom where a person thrives by his cumulative online "Whuffie".
I've worked in management positions and I don't really need advice on who my most valuable employees are. But I wouldn't mind having this data to show to underperformers. It's sometimes hard to convince individuals that they are not as good as they think they are.
This sig is just as redundant as the rest of this posting
Am I the only one very, very disturbed by this type of analysis? To reduce a human being down to a statistical average and use that in hiring and termination practices is just utterly ridiculous. A person is worth more than the sum of their quantifiable parts! I agree that there are far too many people that are either deadwood within an organization or have no business being managers of areas that they have little to no background within, but there are people out there that excel at areas they have no background in (mostly because their thought processes function differently than others, i.e. thinking outside the box) and have the ability to cut across quantifiable boundaries and contribute to an organization's goals in immeasurably positive ways.
This is a horrible idea and will backfire on those that implement it. Mark my words, the first person to be wrongly terminated because of this practice is going to ream the hell out the company that does it. You cannot quantify the human element in an equation. They will ALWAYS surprise you!
Humanity will destroy itself not by the bomb, but it's compulsion to micro-manage every detail of living.
to those who lack understanding.
The point of a company is that it is a system. Productivity is an emergent property, and individual productivity is dependent on putting that individual's talents to best use. True, if you can't figure out how to do this, you should let that person go. But that doesn't mean you should use statistics to run your business like a fantasy football franchise.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
I currently do chemistry work that I report on paper to get entered by others or tell verbally to someone by phone. I think I've sent three emails at work so far this year.
By this measure I am of no value and a temp who does data entry is a national treasure. (That may be true, but it doesn't follow from this analysis.)
Guess I'll have to start responding to those weekly email tag fests of "who is going to bring what to the Friday pot-luck lunch". It may up my stats, but it'll probably add to my waistline.
What id the system concludes that the CEO or several senior managers are worthless?
Those results will be just ignored or...?
please also consider my valuable contribution to this discussion:
Elephants are large land mammals of the order Proboscidea and the family Elephantidae. There are three living species: the African Bush Elephant, the African Forest Elephant and the Asian Elephant (also known as the Indian Elephant). Other species have become extinct since the last ice age, the Mammoths, dwarf forms of which may have survived as late as 2,000 BC,[1] being the best-known of these. They were once classified along with other thick skinned animals in a now invalid order, Pachydermata.
Elephants are the largest land animals.[2] The elephant's gestation period is 22 months, the longest of any land animal. At birth it is common for an elephant calf to weigh 120 kilograms (260 lb). They typically live for 50 to 70 years, but the oldest recorded elephant lived for 82 years.[3] The largest elephant ever recorded was shot in Angola in 1956. This male weighed about 12,000 kilograms (26,000 lb),[4] with a shoulder height of 4.2 metres (14 ft), a metre (yard) taller than the average male African elephant.[5] The smallest elephants, about the size of a calf or a large pig, were a prehistoric species that lived on the island of Crete during the Pleistocene epoch.[6]
The elephant has appeared in cultures across the world. They are a symbol of wisdom in Asian cultures and are famed for their memory and intelligence, where they are thought to be on par with cetaceans[7] and hominids.[8] Aristotle once said the elephant was "the beast which passeth all others in wit and mind"[9]. The word "elephant" has its origins in the Greek á¼ÎÎÏαÏ, meaning "ivory" or "elephant".[10]
Healthy adult elephants have no natural predators[11], although lions may take calves or weak individuals.[12][13] They are, however, increasingly threatened by human intrusion and poaching. Once numbering in the millions, the African elephant population has dwindled to between 470,000 and 690,000 individuals according to a March 2007 estimate.[14] While the elephant is a protected species worldwide, with restrictions in place on capture, domestic use, and trade in products such as ivory, CITES reopening of "one time" ivory stock sales, has resulted in increased poaching. Certain African nations report a decrease of their elephant populations by as much as two-thirds, and populations in certain protected areas are in danger of being eliminated[15] Since recent poaching has increased by as much as 45%, the current population is unknown (2008).[16]
Contents
[hide]
* 1 Taxonomy and evolution
o 1.1 African Elephant
o 1.2 Asian Elephant
* 2 Physical characteristics
o 2.1 Trunk
o 2.2 Tusks
o 2.3 Teeth
o 2.4 Skin
o 2.5 Legs and feet
o 2.6 Ears
* 3 Biology and behavior
o 3.1 Social behavior
o 3.2 Intelligence
o 3.3 Senses
o 3.4 Self-awareness
o 3.5 Communication
o 3.6 Diet
"Not an actor, but he plays one on TV."
Theodp's ignorant comment about "quants" and mathmatical ignores one of the primary thing I have heard repeated about the blame being put on mathmatical modeling for the financial crisis. Namely, that it is not that mathmatical modeling was used but rather that only one mathmatical model was used by everyone.
That's just the tip of the iceberg and in part is meant to deflect blame. They want to make the problem seem diffuse and systemic (spread the blame thin enough and it lands on nobody) rather than sharply defined and individual.
The problem was that the mathematical model had only limited historical data to draw from and so, only limited predictive power but it was ignorantly applied as if it's predictive power was limitless by people who didn't actually understand how it worked in the first place, much less why (or even THAT) it could go spectacularly wrong. This in spite of the model's inventor himself cautioning that it wasn't strong enough to be used the way they were using it.
Each and every person who used that tool had a duty to understand it well and to use it properly. Each and every one of them individually failed to do so.
Their managers each individually had the responsibility to weed out those who didn't use their tools properly. Each and every one of THEM individually did the opposite. They weeded out the guys who DID understand the limitations of the tools because they kept saying unpopular things like TANSTAAFL and you can't make a silk purse from a sow's ear.
Some of them did that because they are/were incompetent but well connected boobs (exactly the sort of people that the model in TFA will select for retention) who never should have been in charge of other people's money and some because they were ethically challenged and figured they would be rich and long gone by the time it all crashed.
Naturally, they would much prefer that we blame systemic effects that would be difficult for them to have individually seen for the meltdown rather than blame each of them individually for failing to heed clear warnings and for putting their individual greed ahead of the entire world's well being.
Does anybody doubt for a minute that the first thing managers will do is exempt themselves from being evaluated this way? The only thing this "tool" will be used for is to intimidate employees at evaluation time, or when they're looking for a raise.
I've calculated my velocity with such exquisite precision that I have no idea where I am.
The above rhetorical question implies, the submitter/editor disagree with mathematical methods.
Actually, I think the basic objection may not be so much modeling itself as putting models in the hands of "professionals" who routinely do things like ask for 5+ years of Java Development experience... in 1998.
There's an argument floating around that what caused the financial system mess wasn't so much the mathematical models and the quants who invented them, but the higher ups who built on them as tools without understanding them. It doesn't take a lot of imagination or, unfortunately, a ton of experience, to visualize suits and mid-level management who *may* get the broad details (more likely may not) of a model but be completely lacking in the expertise to really wrap their head around the stuff.
With HR? I guarantee you it will be worse. Human Resources is a barely skilled profession, and the kind of people who work it are nearly to a person EXACTLY the kind of people who simply do not have the equipment to understand the principles behind and limits of a mathematical model. They're more the kind of people who are likely to say "these are the rules, that's what the system says, here are the dictated consequences."
Libertarianism is rich wolves and poor sheep playing gambler's ruin for dinner.
Please Read The Black Swan, for reasons why you are wrong.
Arguing that you can just mash people together and get "synergies" is like arguing that you can put Van Gogh and Rembrant on a team to make a better painting, can put the Rolling Stones and Beatles into a single band and get a better album. They are just -different- things and the whole idea of synergies is really about simplification and eliminating those important local differences that serve to identify products. I'm not the one that looks for simplicity - it is the merger and mba types that just slam businesses together as if they can so easily be put together.
For some good examples of colossal mergers that failed, read up on the merger of PRR and NY Central. Railroads can't be different, there's synergies on the routes... all wrong. There's been plenty of disasters since then.
This is my sig.
My answer to any employer who tries these shenanigans on me: "Treat me like a number again and I'll treat you like an animal." I only work with friends.
Its profit, dumbass.
My real problem with this approach is that since it is an objective measurement, it implies that it is meaningful. But if it is useful as you said, it is merely a blunt instrument used to force employees to shut up and sit down without relaying how they fail.
I expect more from management, and yes, I have written my fair share of performance reviews.
Isn't letting computer programs "assigning" numbers to things what led to the last few stock market collapses?
Having worked for a number of big companies, I've noticed that managers tend to keep tight control over information flows into and out of their groups. The productive employee who doesn't have an interest in corporate ladder climbing stays in their cubicle, doing the work. The boss sends the group idiot or management wannabe off to the meetings to disseminate all of the information. That way, if a competition between managers for the most productive groups arises within the company, managers can prevent their key employees from being picked up by competing groups. This policy is also supported by HR departments to prevent internal markets in employees from developing and bidding their wages up.
Scott Adams discussed an idea related to this with his Dilbert Principle. The best employees are kept on the shop floor where they do productive work. The screw-ups are promoted up the management ladder.
These practices have been in place for all of the decades that I've been in the working world and, certainly, since the beginnings of the corporation. As TFA states, the data mining techniques fail to capture the wisdom exchanged at the water cooler. Assuming that the source of knowledge is anonymized in official company communications, how do they identify its true origins?
Have gnu, will travel.
It's partially the job of the management to establish communication and to create a proper working environment. Data mining is just another attempt to make decisions without having to understand the projects which the company has taken on. Throwing a dice would be much more straightforward and less disruptive to the moral of the workers.
"Certain employees produce chunks of data - whether words or software code - that later pop up in other messages. The people copied most often, Cataphora concludes, are thought leaders."
- quote from the article
"Oh my god! Did you hear what Larry did? He told Mark that our blue-spotted widget is garbage and that hiding behind the good reviews that Mark paid for isn't going to keep our investors in the dark for long. Can you believe it? Do you know if Sarah or the q/a team has heard yet? Can you imagine what the boss is going to say when he gets word of this? Wow!
- email from A.N.Y. corporation
more nonsense! Yes a single model was used by everyone, but more importantly thefools thought that the random error could be modeled when they knew the error wasn't random. They knew that they were modeling the number defaults by unqualified people with models with that assumed they were qualified and that the default rates were identical. The "quants", ie., fake mathematicians and statisticians, were in it for the money and the h**l with what happened
but even Slashdot has a mathematical modeling system to judge the quality of postings. I'd say /.'s is one of the better within the context of 'forum posting judging systems' because it takes a combination of member feedback over time to judge karma (reputation) and specific feedback on a post to judge a particular post. Plus, members can disagree and, to some extent, override an initial bad mark. On other 'forum posting judging systems' you either get points for ANY post, no matter how innane, or no points or recognition at all for a post that may have taken you hours to prepare. I prefer slashdot's method myself, even though the alogorithms are shrouded in secrecy.
The fact is, I would LIKE some sort of objective evaluation system to get away from the subjective prattlings and insecurities of management that does not appreciate, understand, recognize, or even WANT good work. I would love some hard data to present which proved my contribution to the company exceeded most everyone else's contributions. Designing a system that truly did this is, of course, the major issue.
But here's a situation where one has been used successfully in court suits to prove discrimination. In the case I remember it was a female academic who had not been promoted into a tenured position where certain males had. The metric used in court was the number of times the female's papers had been cited by other papers and sources compared to the males. Given that academia works on a 'publish or perish' model of promotion, this was deemed an appropriate metric to show her contribution to her field as evidenced by others who had cited her work in their own efforts. I believe there was enough evidence to show discrimination and she won the case.
So, don't throw all metrics out because someone can design a bad system.
How about a moderation of -1 pedantic.
I am so glad i No longer work for That largE empLoyer that would probably use this method to rate their employees. The method that they currently use is already screwed up and manipulated enough by conniving employees, I could just imagine how honest hard workers would get screwed over by this bs.
and the guy who thinks he is "adding value" by sending the me-too replies and other nonsense shows up as the bigge4st gilded turd in the company.
i constantly see crap like
"Thanks [coworker]"
"did you see this?" (addressed to coworker)
"[coworker] can you look at this?" (the dude is oncall and already investigating!!
"I think [coworker] is working on this" CC:[coworker] even though we are ALL ON THE ALIAS THAT THE MAIL CAME IN ON!
yeah, go on and add value, you ass-hat!
Wow, so the first round of people subjected to this scrutiny get screwed. But come the next round of reviews everyone will have figured out how to game the system and cluttered inboxes with drivel to get a big, dark circle. Makes sense to me.
If you do what you always did, you get what you always got.
For me, half a dozen in a lifetime of consulting. Usually, their managers were intellectually hobbled in multiple dimensions, so their contributions were wasted.
Almost always, their manager's manager, the VP or Directory, were worse.
Given the frequency of good managers in the population of managers, the amount of superior training of such managers (HP and IBM are standouts here), companies can't possibly scale.
> The best companies don't have HR,
Wrong. IBM and other serious companies are very systematic in collecting info and opinions from subordinates and peers about a manager's effectiveness. Not many morons make it to Director level in these companies, while I continue to meet CEOs here in Silicon Valley who couldn't manage a kids sandbox.
I'm no fan of Microsoft's products but they employ a huge number of very smart and highly compensated employees. They have very smart people in HR too.
Understanding how all of those people contribute and interact is an interesting challenge. There is nothing wrong with studying it. So much depends on how you apply the data. And those smart employees will game the system.
... is this the first example of an email worm implemented in wetware?
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
Stephen Baker wrote a book about this:
http://www.amazon.com/Numerati-Stephen-Baker/dp/0618784608/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1237149821&sr=8-1
So I guess this means that that complete idiot in the building across town who insists on bccing 2/3 of the organization to let us all know there's leftover cake in room 3b will shortly be promoted to head of innovative activities.
Oh, and that I'll be fired because 90% of the ideas I'm working on began as (completely undocumented) conversations with people from random other departments at the local bar, not as email conversations.
Anyway, time to start polishing the CV and looking for another place to work which isn't driven by total idiocy..
... blame the people who abuse them.
A statistic is just that - in this case it is a summary of one aspect of how valuable a human is to the organization.
I think in spite of all this negativity in the discussion, we should take a bigger picture and be a bit more open. This approach appears to be in its infancy, and there is no guarantee that a data mined statistic might not be useful in the end - there is nothing wrong for HR to experiment.
So long as HR managers can keep this in mind and remember that a number is just a number, and not the end all of deciding a person's worth we will still be in good hands. If not, maybe you might want to work for a different company :-)
I work as a systems administrator for around 300 computers.
I don't talk very much, and stay in my office most of the time.
I am not very good at non-verbal communication, and people have always seen me as odd. I once heard a coworker comment to another "Look at him; it's like he's autistic."
The person I replaced had left his office stacked to the ceiling full of boxes, and had left everything disorganized. On the day I started, literally half of the office was inaccessible.
I've only been working at my job for a few months, but in that time, I've implemented several systems to automate work, and set up remote desktop so I can manage all of their computers remotely. I've made my office basically paperless and have helped other employees change their offices to be paperless, too.
At this point my office is almost empty, and because of the systems I've put in place, there is a lot of downtime in my job. I've been complimented many times for keeping everything working so well.
Are employees' non-verbal communication skills a good metric when employees are entirely capable of doing their respective jobs without much communication? I don't think so.
I work for apple, and this has been something they've done for a long time. All employees are graded on a 1-10 scale. Those in the bottom 20% are encouraged for firing, and all of them below 5 are glared at with scrutiny.
Apply the same tools from low level management to C*O, Boards, politicians... and the proof will be of far greater value to US, EU....
When everything is FUBAR economics, politics, education, religion....
Who had no obvious fracken value for good economics, government, dogma....
Who are the same fracken fools waiting at home-plate to claim a touchdown.
We are victims of rape, the same serial rapists of many past economic blunders.
We are victims of abusive plutocrats that see themselves as victimized.
We are victims of entitled aristocratic calling welfare-economics capitalism.
We are the victims of economic credit/debit lynchings, not capital.
The entitled plutocrats are corporate welfare-economist not capitalist.
When the same parts/dogma continuously fail our economic engine for decades,
causing economic catastrophes for posterity maybe the dogma/parts are just
bullshit-spin and rhetoric that need to be replaced NOW!
Con-gressional politics as always usual and sad.
CPAC = Corporatist Political Action Committee
DPAC = Dogmatist Political Action Committee
Unaccountable leaders are masters, and unrepresented people are slaves. How do US and EU fare?
This model makes the CEO's secretary just as or more important than the CEO.
Hasn't anyone every heard of Hyperion. The top corporations in the Silicon Valley all have charts with peoples heads on them. Either your produce or your out. No bullshitting allowed.
I have no aversion to mathematical models: I'm an AI researcher with a large focus on machine learning, so it's of course in my interest that lots and lots of people use such models.
But as someone who develops them, it also shocks me what people actually use them for. The hardest part of a complex problem is not usually applying a statistical algorithm, of which there are a lot, many of whose properties are pretty well understood. Rather, it's figuring out how to pose your problem as a statistical one in a way that is remotely valid. Also, getting good data is a big stumbling block.
It's remarkably hard to do this for complex problems; the level of complexity at which modeling is really well understood is not all that high. And your results are worse than useless if you don't even have a data set to begin with whose properties you understand.
In this case, I'd say lack of any data to start with would be the first stumbling block. Is it actually understood what properties of workers are good for the long-term interests of an organization? How do you even define "good", what is your notion of "long-term", and does this vary by organization? Sure, represent worker X as a 20-dimensional vector of numbers; I don't inherently have a problem with that. But how are you predicting the response variable likelihood_of_excellence(X) from that? Do you have some magical labeled data set of 10,000 workers and their long-term excellence values?
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
My experience interviewing in the recent job market is that HR is now HC, or Human Captial. As I was told by the last young lady who corrected me when I mistakenly called her department HR instead of HC: "it is a modernization change that will allow the organization to better measure the output of each individual resource at all levels." I declined a second interview when she wasn't able to provide the metrics under which I was going to be evaluated as a resource because the new changes were still being implemented and refined.
-- Wondering how long until the internet becomes fully corporatist, like television.
This reminds me of a story from a bank few years ago:
A software company has developed a trading system for the bank. One of bank mid-managers was very keen of the product and supported it strongly because it was going to have a nice feature: reports about all the trades on per employee basis so that at the and of month, year, whatever it would be very easy to see who make the most money for the bank (and who the least).
His backing for the project ended abruptly right after the first month of the deployment of the product when the stat results were for him ... well, not very favourable.
hany
"promoting the guy with the right hairstyle"
Well, I recently paid $350.00 for a Trump-doo and now I'm waiting for the moola to start rolling in. Been practicing my "you're fired" in the mirror too! Oh, where do they sell those gaudy ties?
-- Wondering how long until the internet becomes fully corporatist, like television.
Am I the only one who remembers the Voyager episode where the Doc was stolen and everybody on that planet has a "value"? it looks like we're gonna have to steal the Doc to teach the companies some sense!
Where you're not just a name.. You're a number!
Computer Science is all about trying to find the right wrench to bang in the right screw. -T.Cumbo?
So if I work for IBM or Microsoft, every day I need to send an e-mail to the entire company saying what color socks I'm wearing and ask for others to tell me about their socks too. And then starting about 10:00am we can start planing who is going out to lunch and who has a car large enough for the group.