4 ms·
I think the problem is that people value themselves in different ways (and often the criteria is skewed to be more favorable for themselves). For example, I ten
by fpig 9y ago
I think the problem is that people value themselves in different ways (and often the criteria is skewed to be more favorable for themselves). For example, I tend to extremely highly value people to whom I go for advice/help. These usually end up being people who are extremely smart and have a both wide and deep technical knowledge. They have almost always been programming for a long time and have a huge interest in the field, and have written all kinds of applications in various technologies.
At the same time, you have a lot of developers who started programming in college, read online about how to become a good developer, decided on technologies whey want to work with, read the staple books for that, wrote a bunch of applications in them, read some books on software architecture, on development process, etc, then got a job at a company which uses some of the technologies they are familiar with. Compared to group #1, they have a huge "knowledge debt" that will never disappear because they're focusing on learning things that are the most relevant to their current work. But this knowledge debt doesn't apply to everything that needs to get done at a company.
People from group #2 often consider themselves just as good as group #1 because they believe a huge part of knowledge group #1 has and they don't is irrelevant for the job. And in some ways, they're right. But they're also wrong, because group #1 adds business value by being able to solve the hardest problems and being the guys you go to for help when you get stuck.
At the same time, the people from group #1 often consider themselves much better than group #2. Which again, in a way might be right, but also wrong because group #2 might still do just fine in practice in most situations.
Of course these are not distinct groups, in reality most people are somewhere in between, being better in some areas and weaker in others, and this makes it even harder to compare 2 individuals in such a way that both of them will agree with the result.
Secret salaries allow the management to decide how much business value someone adds without having to explain themselves to 2 people who likely have skewed (each in their own favor) perspectives. It also prevents negative feelings between employees themselves. It is also good for the salaries of those that the company is willing to pay a lot, they don't have to worry about what others will say. It solves the "if I give it to you I have to give it to everyone" problem.
The downsides are that 1) salaries are likely a bit lower overall due to information imbalance and 2) people who are bad at negotiating get unfairly screwed.
- mirko22 9y ago> 2) people who are bad at negotiating get unfairly screwed. I am bad at negotiating, but I always felt like this is a skill like any other. I feel that saying people who are bad at negotiating get paid less it is like saying people who are not as good at their job get paid less. Well, yes... Why would it be wrong if you are good negotiator and good worker to be paid more then if you are bad negotiator and good worker, or good negotiator and bad worker, or bad negotiator and bad worker... To me this seems perfectly fair. I don't go asking for raises as my negotiation skills are awful, I get them offered to me cos I work as much as I can with highest quality I can attain. Otherwise I totally agree with your comment.
- fpig 9y agoEven if you think it's fair, people being paid based on negotiating skill is a market inefficiency so it is still undesirable. I mean, bench pressing a lot of weight or playing the flute are also skills like any other, but we don't want software developers to be paid based on that. We want software developers to be paid based on software development skills, or other skills that actually relate to doing the job.
- mirko22 9y agoBut those are not related to the software development, while negotiation is related to work in general. Someone who lifts weights is also responsible to negotiate a contract with his manager. I guess the question is then how do we decide which skills should be artificially governed based on how good they are for people? And lets take another point of view. Sales, for example, is based on negotiating skill. Do we say that is also market inefficiency and should we somehow subsidies sales person with bad negotiating skill cos that would be better for their salary? I don't know, I still feel like having the opportunity to negotiate is what counts (meaning no discrimination based on sex, race etc to actually be in position to negotiate) while the outcome is completely dependant on person. And if we really stretch the argument about which skills should be based for salary, we could say that it is not a software developer fault if he is not good at his job and somehow deserves to paid as the one who is dependant, for example, on years of experience. I think people should be thought how to negotiate rather then making some artificial rules that there should be no negotiation at all cos someone is not as good at it.