Ask HN: Will they hate me?

[ThrowawayCoward]

I wrote the following letter to my team (only peers, no managers). Does it have a chance of being productive? If so, would I become hated or viewed as a superior jerk? Is there anything that would produce a better letter if it were eliminated?

http://pastie.org/8287095

[EllaMentry]

Honestly this does not read well. You have some decent points (not unique points to any professional software engineer but good points) - there are several grammatical errors "to(sic) much magic" and you have not provided any examples.

You state that you are new to the project - from the writing style it sounds like you are fairly young, so I would guess this is your first programming gig? Either way take a professional writing course.

Provide examples - where are there problems with code clarity? Where are the repeated sections? Simply stating there are problems help nobody - you need data - how many repeated lines? What is the overhead of this code? How many special cases etc.

Good luck!

[ThrowawayCoward]

Thanks for your review! I'm aware that they are not unique points. They all come either from hard experience or from things I've read from people I respect over the years.

It's not my first programming gig, and as for young, it's relative. I'm in my early thirties. Unfortunately, I am the most senior person there.

I am surprised my writing style would lead anyone to think I'm young. Are there any specifics there? I have taken professional writing courses, but writing is still something I try to improve. I will keep the suggestion of enrolling in one in mind, though!

That particular gramatical error was really a typo. Either way, if I decide to send it at all, I will definitely be proofreading it again. Even though I re-read it once, I missed that one and would probably have missed it every time.

I haven't provided examples for three reasons. First, I'm hoping it prompts discussion. Second, many of what I would include as examples are things that either I or they have pointed out previously. Third, it is already very long; details will bog down the pace. Given those reasons, would examples still help?

[EllaMentry]

That is interesting, from a reread, I would say the "youngness" is a product of the length and over-descriptiveness.

""Magic" is my term for "I don't know how it works because I haven't taken the time to figure it out yet" - it's an attribute of my understanding of a thing, not of the thing itself; that is, it's the Clarkian definition." - No one speaks like that - it makes it sound like you are trying to come across smarter than you really are - Every developer under the sun understands what you mean when you say "Magic"

At it's most essential it sounds like you gobbled up a "software best practices book" and then spouted chapter titles without any context - Examples definitely help!

e.g. <Module Z> is a great example of why we need to keep our code clean, since I joined the team no fewer than X lines have been added resulting in this module having Y different responsibilities - because <Module Z> is so bloated we have Q different tickets in our backlog which will result in code changes to it.

<Module T> and <Module R> are a perfect example of where we can utilize code reuse. Both contain a function Y, which we make heavy use of - we should consider extracting this into a common library function etc. etc.

If you make an actual proposal instead of a complaint, people are more likely to listen - it is a lot harder to argue against raw data than it is a personal opinion.

[ThrowawayCoward]

Interesting. My description of magic is there because I specifically see two programming-related definitions.

When used with Rails, it's usually derogatory, and it means something like "something tricky or obscure and complex that is done in order to provide an easy interface", and in that sense, it is usually derogatory and used when the someone believes there is a straightforward way that may be an even easier interface, or is at worst, not much more cumbersome than the interface that is presented.

The other definition - the one I used, is, for example, how many developers would answer "How does the Linux syscall mechanism work?" In that case, the syscall interface is very much needed to provide an easy interface, and it's not incredibly complex.

One often seems like the other until you look into it, but I still think they are separate.

[tjr]

A lot of how this is received may depend on the role you have in the group. You state you are new to the project; were you brought in as a junior novice? Or as a seasoned professional? If the former, then you might consider presenting these ideas as suggestions rather than as a critique.

I think, though, that either way, this is a lot to digest all at once. Breaking this apart into separate communications might be easier to grasp.

Generally, it sounds like your team needs some sort of coding / development standards. Until a behavior has been established as "correct" or "incorrect", it's all a matter of opinion.

[ThrowawayCoward]

I'm new to the project, but not new to the company. My previous role was a project lead, and I interacted with this project on a regular basis. I was brought in to this project to get stuff done because it's very behind and new work is piling up in the future.

I agree that it's too long. If I send it anything at all, it will have to be shorter or in segments. Based on this and other peoples review, I probably won't send it. Still, it helped my identify the issues for myself. There must be ways of getting results without sending this, and I'll continue to consider that as I work with the team.

I suppose I really would like these ideals to be a standard.

Thank you for taking the time to read that novella and to post your thoughts!

[bobfirestone]

Yes you will be hated and it won't change anything.

There are members of your team who will agree with some or all of the points you make you will still be known as the a-hole.

Additionally that is just way too long to be persuasive. Most people won't read past the first paragraph.

[gus_massa]

For context: When has the project started? When did most of your team come together? When did you enter the team? How many working experience do you have?

Most of your ideas are right, but in real world sometimes you need to make exceptions. For example sometimes you must add a special case. It makes the code worst, but sometimes something has to get fixed now and cleaned after.

I recommend you to read: "Getting Things Done When You're Only a Grunt": http://www.joelonsoftware.com/articles/fog0000000332.html (don't forget strategy 6)

[ThrowawayCoward]

The rewrite started a year ago. It's the _fifth_ version of the same product that is only four years old. I joined this project about a month ago, but have been working on a related project and interacting with them for about three years.

I agree that sometimes things have to be fixed now and cleaned later, but it should make you feel a little bit queasy when you do it, and you should be aware that you are adding future work to the project.

I skimmed that link and it looks great, I'll be diving in fully. Since I've seen the posts from others and was leaning that way already, it's likely I won't send it, but I really want to push this project to be better. Since you suggested strategy 6, I read that first, and it's awesome and difficult. Part of being new means that more than a usual amount of my development time is spent understanding the project. I've felt rather unproductive because of this.

Thank you.