you shouldn't have to be mad or pushed beyond a certain emotional threshold to organize. organization shouldn't be a form of retaliation. this should always have been the status quo.
People always paint these as some conspiracies against unions but there's plenty of genuine discontent with unions amongst the public and there's some legitimate concerns with them.
You can see it today with police unions being the biggest hawks blindly defending any police abuse and protecting them for ever being fired. This elephant in the room often gets ignored in public discourse. And the police aren't the only ones.
I grew up in a small town where unions were used like mini fiefdoms riddled with nepotism where you had to know someone to get a job. Theres also a long history of racism in unions, the original forming of them in the US were to exclude low wage black workers from 'stealing' white jobs.
It seems to work well at first before slowly morphing into a rigid protection racket for current employees, locking a company down to its current structure (which only makes sense for a few slow moving homogeneous industries like factories - until of course they simply automate or move to Asia or shut down).
It's also notable people always use historical nostalgic examples to hype up some golden age of unions for jobs that no longer exist for some reason...
They've also long been covers for corruption and back room political influence.
I personally believe unions are long overdue for reform and modernization. I'd maybe support something like unions-light, but today they always come with tons of other baggage like preventing bad workers from being fired, adding tons of bureaucracy and hoop jumping, crippling companies ability to adapt to market changes, forcing people like comedians and small indie bands to hire 30 different people for individual jobs (like paying a guy $40hr to pull a cord to open the curtains and nothing else which happens in NY clubs) creating inefficiencies which hurt company growth and customers and therefore employees/economies/taxes, completely shutting down small businesses who simply can't afford to support unions, not applying well to companies with a variety of different job roles, etc.
It's like a sledgehammer when most people just want to make more $$.
> It's like a sledgehammer when most people just want to make more $$.
Except the Epic dispute is not about money, it's about working conditions.
Epic workers are fairly well compensated. They're not demanding to be paid more. What they want is to not be forced to return to the office in the middle of the pandemic, a very reasonable demand.
> I grew up in a small town where unions were used like mini fiefdoms riddled with nepotism where you had to know someone to get a job.
Yeah same. I grew up in a town with a large union presence (auto workers) and it was unbelievably corrupt. I’m not against unions conceptually, but they can be implemented very badly in practice.
The problem with unions is they reward seniority over merit.
The problem with at will employment is most industries remove the reward for seniority in favor of merit.
There are still industries that value experience over cost for performance but they tend to have their own walled gardens (looking at you American Medical Association).
I once thought unions the bane of the free markets existence but I’m now starting to believe that unions are the best tool available to keep Capital in check.
We forget how out of control capital distribution has gotten these days. It can help to use time instead of dollars to measure things.
> The problem with unions is they reward seniority over merit.
I've never seen a company, union or non-union, that doesn't have a relatively narrow salary range and progression for non-management positions. Of course the base salary tends to be much higher in tech companies, but how much non-seniority "merit pay" do non-manager software engineers actually get? The bigger the company, the more well-defined and bureaucratized the salary progression, even when there's no union. AFAICT it seems largely a myth that large amounts of non-seniority merit raises exist in software engineering. Some engineers can advance through the "rankings" more quickly than others, but overall it doesn't seem much different from union scales. The differences due to "merit" are overall small in the whole scheme.
The two best ways to increase your salary in a tech company are still seniority and promotion to management, which is no different than union shops.
Of course stock compensation can be a big factor in tech companies, but the stock price rises or falls due to market forces rather than individual employee merit.
I agree with more or less everything you've said. However I still think adding a union to the balance of power is the lesser evil in a good number of cases.
> I personally believe unions are long overdue for reform and modernization.
Are you thinking legislative or otherwise? Also, what do you think of the German model (labour representation on the board, etc)?
this is still drastically better than no union which results in massive exploitation once there is a decent amount of qualified workers available. We see this less in software in general because the demand for competent developers is so high, when unions first started coming about in the US they were met with bloody resistance, people were killed just for striking, so your bringing up of nepotism and unfairness seems rather out of place when it was addressing a system that literally murdered people for asking for more money. One of the few unions I can't get behind is police unions but there is no history behind police unions like you do with most workers unions. Cops just picked it up because they realized it gives them a massive amount of power, organized strikes by police will result in almost all their demands being met, city with no police makes people uneasy.
I hope you're right (although I know you're not): the easier it is, the better. What I'm suggesting isn't meant to be impossible.
(of course, it would be easier just to turn an existing company into a co-op, but if co-op start-ups weren't at a massive systematic disadvantage, that would be great)
how do I ascertain that a software product is made by a co-op? marketing, probably, to be sustainable. what examples are there of well-marketed co-ops? it's not really something that is in the forefront of the collective consciousness. a few knowledgable grocery shoppers, sure, know of Welch's which is owned by the National Grape Coop, and the Green Bay Packers are as a publicly-owned American football team are a stretch for being a coop as owners only have voting rights and not profit reimbursement. I say this as someone who volunteered to run the finances of non-profit co-op housing communities, way back when, that it's a foreign concept for most.
Someone needs to differentiate products that don't abuse workers. I could see twitch, etc being an avenue for promotion, of course, but getting a groundswell of streamers into this is another matter.
Just to be clear: I agree that for most people they can earn more money at lower risk and effort working for an established company than starting their own company or cooperative.
>There is a difference between the legal possibility of doing so, and the practical possibility.
Yes, because the existing employer has actually created a lot of value by investing their capital in the business, because providing investment is actually incredibly valuable
It is a form of retaliation though. People who are treated well don't often see the need to organise. I've never heard of a "let's organise now - everything's awesome and we're happy, but who knows about the future" push.
If everything is hunky dory there’s no need to put any energy into labor organization.
And to be slightly cynical: if all is well, an organization designed to drive opposition is going to need something to complain about just to justify its existence.
(The second point is often used by anti labor folks to claim that all unions are malign. I am Not one of those people — in fact I’m pro labor pro union. Just pointing out there’s a legit reason for a union not to have been formed)
Hmm, I feel like unionisation have the potential to push up salaries a lot. But I suspect this can depend a lot on the power dynamic between employee and employer of the particular market. If the jobs can easily move abroad local unions are not as effective.
So far it seems to me that software engineers already have a pretty good position on the market. Would it be improved through unionisation?
Software engineers in the "regular" industry indeed are better off without unions. There's no obvious problem that unions would solve, and there are costs (unions in general add inefficiency).
In gamedev, though, salaries are low and work/life balance is traditionally very poor. I can see why some Epic workers decided to consider unionizing.
Software engineering is a global market. Unions are local to the US. If unionizing becomes a thing, it will be a huge boost to developers in Eastern Europe, China, India, Mexico, but a big blow to the US.
And it's not about offshoring a-la 1995 or 2005. It's about building the whole company outside of the US, only having minimal presence here for fundraising / legal / whatever. There is already a number of startups executing this strategy; unions will immediately make it a no-brainer for all new companies.
The salaries paid to skilled US tech workers in the tech hubs without unions are already so much higher than anywhere else in the world, even after adjusting for cost of living, that I don't see the impact happening the way you say it. The workers who have skills and experience would unionize at some tech workplaces so that they don't have to job hop for fair treatment, but in any case they already make more in the US than elsewhere, because employers know they have to go to the US tech hubs to have a high concentration of great tech workers, and unions wouldn't change this. (I know there are many great tech workers elsewhere, but not nearly as concentrated.)
Most of the high-skilled tech hubs I'm aware of outside the US are in countries with stronger worker protections than the US, and often more unionization too. (This does vary, absolutely - worker rights are greater in Berlin or Paris than they are in Zurich or London.)
Now, for low-paid low-skilled workers, the situation might be different, but those aren't determinative of where a tech company is based long-term, and the union would be best if it spans all skill levels.
Unions aren't just a US thing, by the way. Hello from Quebec where unions are far more common than anywhere in the US. Even undergraduate university students are typically unionized here, complete with strikes and cross-university federations with elected leadership. Much of Europe is also far more unionized than the US, though certainly not the part of it you referenced by Eastern Europe (maybe aside from Finland).
I doubt it. Since it is a sellers market, devs might be drawn to places where they have benefits. I doubt that many US or European devs would want to work for wages common in India and I don't think these particular startups are too relevant.
Information transparency and negotiating assistance with respect to compensation, better handling of harassment and discrimination claims, and better protections against arbitrary termination without either chance to improve that's more genuine than most PIPs (if bad performance or mild misbehavior is claimed) or a right to compensation (if it's something like a layoff or restructuring).
Very real problems in the tech industry, all of which a union could help with, and none of which require the clearly mismatched "you can't touch that socket - call the union electrician" contract provisions from industries like trade shows that the anti-union people like to overfocus on.
> There's no obvious problem that unions would solve
- No more stack ranking or other absurd professional advancement schemes
- Refusing unethical business practices (to take an example in the spotlight again, reusing 2FA phone numbers for any other purpose)
- Refusing unethical business relationships (China, ICE)
- Higher liquidation preferences for more employees
- Working hours (end of "unlimited" vacation; on-call restrictions; prior to COVID, WFH)
- Racial and sexual harassment across vertical hierarchies (unions alone won't solve these problems in tech - they can even make horizontal harassment worse if union leadership is not committed to this goal - but could have ousted e.g. Gundotra or Drummond much sooner)
You can already pick your employer based on these criteria. What makes you think unions will push these harder?
Also what about people who have different preferences? You might get lucky and the brave new world of software unions will push for exactly those conditions. Unluckily for me, I have very different preferences.
Because currently there's a employer PR, services like glassdoor, etc. but still not enough power to influence or avoid future bad faith executive decisions. I think the case of google's work on a defense project counts. When I was a fresh grad, Google was the great, good go-to place, it has changed. Given how big it is I don't know what effect Unions can have, and I hate the negative nepotistic and other effects of Unions, but an argument could be made, google would still be true to the "Don't be evil", if there were unions.*
P.S: Admittedly it's just an argument that may not be tested/testable.
"If you don't like the working conditions, you can leave."
Why does that only apply when people are trying to dismiss the value of unions or discourage their colleagues from forming a union with their coworkers? Surely if you didn't want to work in a union shop, but a majority of your coworkers did, it'd be on you to just find a new employer.
Though one caveat: that simple analysis only applies for as long as employers are legally allowed to refuse to hire union members.
Then anti-union me can go to some anti-union company. We write in our employment contract that I won't join a union, and that the company will never hire any union members, and we live happily ever after.
You can try to think through the consequences of giving unions various legal privileges. Eg if for example, a company can't legally discriminate against union members in hiring nor fire an established employee who joins a union despite having promised not to, things might be different. Especially if you add in common extra privileges like banning employers from firing striking unions members or if picketing is legalised etc.
In terms of salaries, the only effective tool unions have to increase them is their ability to exclude people from employment. You can see this having a great effect with the ALPA, AMA and Bar Associations, who have all succeed in making becoming a pilot, doctor or lawyer incredibly expensive, difficult and time intensive. Regardless of the true underlying motives, their success in keeping people out of the professions is why they’re all so highly paid. This is also why union organisers believe they should have the right to prevent non-union members from seeking employment in a company they are striking against (and call such people scabs).
The more ordinary trade unions are typically much less successful in having any significant impact on salaries, because their only meaningful tool for constraining supply is the temporary impact of striking.
Unions also usually serve as a levelling mechanism, where CBAs aim to reduce the difference in remuneration between the more productive and the less productive members of staff. So whether that’s effective for any particular person would depend on how productive they are, potentially having a negative impact, positive impact, or little impact at all.
The single area where unions can easily have the most positive impact is in enforcing compliance with employment laws. It’s much harder for employers to skirt employment law when employees have union representation (unless the union in question is one of the many “yellow unions”, who primarily protect the interests of employers over employees). The flip side of that is that it is of course much harder to fire poorly performing employees (try firing a corrupt cop for instance).
Just imagine how good it would be if the CEO of Walmart were to give his entire compensation to his employers instead. Each of them could be earning an additional $9.50... per year, or about 0.45 cents per hour. For Amazon you’d be looking at about 0.09 cents per hour. For Microsoft you’d be looking at an astonishing 7.9 cents per hour. The idea that CEO compensation has some sort of impact on employee compensation is just a myth.
It’s irrelevant anyway, because that’s not how wages are set. Please explain why you think a company would choose to pay a union member more than a non-union member would be willing to do the job for.
Market economies set prices by supply and demand. If enough candidate employees are willing to work for less than the union is demanding, then the union has almost no leverage to have their demands met. That is unless they can use regulation to exclude non-union members from employment, like lawyers, doctors and pilots have accomplished.
Striking can have an influence on supply, but it’s influence is very limited by the fact that striking can at best represent a temporary disruption to business. How much an employer could potentially afford is not relevant as long as there is somebody willing to do the job for less.
Why's that? I used to work at a place where engineers were unionized and there were some downsides, but I think it was overall a positive. Probably the best benefits of any place I've ever worked
Because the need for unions arises when there is a severe power imbalance between employer and employee. If I'm ever put in a position where I feel as though I am simply unable to negotiate a wage that is aligned with how I value my labor, and I feel as though this is through an exploitative advantage that the entire industry has over me, then I would start seeking other individuals in a similar situation in order to collectively gain leverage. I have not found that the software engineering industry remotely resembles this scenario, and that it is actually us engineers that hold most of the leverage advantage.
So while I recognize that there is a potential value to unionizing, short of a need for it, you're just introducing a middle-man into the relationships you establish, and unions eventually begin to take on all sorts of self-serving interests of their own. Many industries have developed to the point that the power imbalance doesn't exist between the employer and the employee, but rather the competent individual and the union.
I know this is going to be unpopular, but I just want to think about this companies perspective to add a bit of balance to this thread.
I know a lot of people on here feel like work from home is just as productive or even more so than working in the office, especially in the case of open office layouts, but it's just not universally true.
Some people are really good with responsibility, and some people really are not.
As an illustrative example, you might have 20% of workers working at 120% efficiency, 50% of workers working at 80% efficency, and 30% of workers working at 10% efficiency. This example comes out to about 67% productivity overall.
But you are still paying 100% of the salaries.
It would be nice if you could just tell the 30% who have terrible efficiency to come in while everyone else can work from home, but you can't. It immediately going to be seen as unfair. "Why does X get to work from home and I don't?!". Many people don't even have the self awareness to tell that they are terrible at working from home.
Another option would paying people for the amount of work they are actually delivering. Yeah, that isn't going to happen. I mean, for a start, it's hard to even quantify most of the time. A good manager knows who is delivering and who isn't, but you can't exactly point to number of bugs closed, or lines of code checked in to prove it.
And aside from anything else differing pay is a hard pill to swallow even without WFH involved. The guy earning 200k might be 4x as useful as the guy being paid 100k, or he may be no better or even worse!
My point is that there is no really satisfactory option here. I mean, you can make the argument "It's a pandemic, they should be happy to pay the same amount for 67% of the work" and sure, maybe that's just the way the cookie crumbles. I mean, lots of people are being totally forced out of business due to this pandemic so it's better situation than those guys I guess?
But you can at least understand that it's a sucky position to be in.
"As an illustrative example, you might have 20% of workers working at 120% efficiency, 50% of workers working at 80% efficency, and 30% of workers working at 10% efficiency. This example comes out to about 67% productivity overall."
While I agree with your overall point, this is a terrible way of making it. Of course I can argue any position if I can just make up numbers to support it.
It relates the overall principle to a concrete example. The particular example isn't important. As you said, we can tweak the numbers to go either way. What's useful about it is showing an example of how the numbers could work to logically explaining Epic's decision. I find concrete examples like this really helpful for comprehension.
Paying employees according to the true value of their labor is tempting at first, but you better be careful-- Apply the principle too consistently and the C-suite gets a huge pay cut.
This sort of fundamental misunderstanding of engineer productivity is precisely why we're so tired. Our emotions and our ability to produce aren't simply under your control. We are fundamentally about maintaining systems, and productivity is the wrong way to measure maintenance.
I wonder about people blocked on a decision, or folks working on a project that will ultimately be cancelled, or... a lot of knowledge workers literally work on "research and development" which might be hard to map to widgets shipped.
I live in Wisconsin and I applied to work there twice after finishing bachelor's degree about 8 years ago. Both times I was rejected. I was thinking by working at Epic I would be solving real-world problem and helping humanity so I was excited. The way I was treated says a lot about their culture and how much they value their employees. They are a terrible company inside and outside.
Took the online assessment twice. It takes 2 hours to complete it. Never received any phone call or email or anything saying that my application was denied or accepted.
I'm glad it wasn't worse. It seems like employers ghosting candidates has become a norm to the extent that I know if I apply for 5 or 10 jobs, I'll be ghosted at least a couple of times. It's stupidly unprofessional -- IMO, if a human looks at your application, you should get a personalized email saying whether or not they intend to move forward; and, if no human looks at the application, the candidate should get at least an automated email explaining why, or a timeframe in which a human should be able to get around to looking at it.
This ain't rocket science. Sure, individual humans who are part of the process can make mistakes sometimes, but when it's coming from a group that says "we can't hire enough people! skills shortage!", I have very little sympathy.
For those interested, Google’s Madison office has some openings for software and hardware engineers. Our office is downtown in the same building as The Sylvee, and Google has publicly announced that voluntary WFH has been extended through June 2021 to allow employees the confidence to commit to personal arrangements until then.
