OSDev.org

People are on this forum because they write an OS. Do you code everything alone or you have a development team that works with you? I have small team, there's two people working at code (look at commit history, there's me and k1-801), one man worked at website, but he made some mistakes and I rewrote it, and one man made a website design that I used. So do you have a development team or you work alone?

I work entirely alone. I write everything myself, but that also means development stops when I'm too busy, and it means I haven't really had time to have a website or anything. I don't think my OS would ever be interesting enough for others to join anyway :p

I work alone.

In Russian social network VK there's lots of technical communities, I have an ~450 people auditory after promotion in them. Also my team member bought a domain name for his other project and hosted my website on his server, but it's down for now.

max wrote:I work alone.

There's a big advantages of both work types

For working in team advantage is that you can go e.g. to a place without computer and Internet for couple of months and leave all work on team member for this time. For working alone advantage is that you can do everything without discussion, so there's more freedom.

MollenOS wrote:I work entirely alone. I write everything myself, but that also means development stops when I'm too busy, and it means I haven't really had time to have a website or anything. I don't think my OS would ever be interesting enough for others to join anyway :p

I don't allow anybody to join if I don't know them for a long time or if I think that they are untrusted. There's been an incident three months ago.

tyndur was started as a community project, so I've technically been part of a team for the past ten years. That team has never been officially dissolved, but in practice it's been more like sporadic activity bursts of individual persons (mostly myself) for the past few years.

max wrote:I work alone.

Ah, yes, I still need to finally convince you...

While I do hope that I will eventually have some collaborators, I am working alone for now until I actually have something which might draw someone else's interest. I am not expecting this to happen, but hope springs eternal

Mind you, as I have said many times before, my primary goals are learning and research, so it really isn't the sort of thing that gets done for practical application purposes. I would love it if actually got to the point where it could be used for such, and am trying to write it as if it might, but I don't expect it to.

Exactly me!

Why do you think, why is Windows developement so way slow? It is being worked by too many unorganized people.
Most people are not organized.

I think that working in team of 2, 3, maybe 4 people will only speed up development, but if there will be too many developers development will slow down.

The thing is if people are not naturally making an "ecosystem " you need to manage them. There is no need of social management when people are complementary and/or know each other for long and have a common goal , and are not redundant.

This is true for small OS and is also true for professional development teams.

Back in the day, we (Pro-POS) tried to set up a team. Quite a few people were interested. But we got tangled in endless, and unfortunately not too seldom pointless, discussions on what to do, how to do it, and how to organize it.

And when I buckled and said, "well, fine, you want to start coding before we hammered down the design, let's start coding", the whole thing basically collapsed overnight. (PDCLib is what remained of that project.)

Getting people to the table is easy. Making them work as a team is hard work, and requires a whole new skill set. Especially if you don't have "prior art" for them to align to.

Yep, very soon one needs to somehow organize the people, so they follow the design/plan (made by the architect or whoever has the most suitable experience) or first create one, agree on it and then follow it.

From experience, I have to agree, getting people to actually work on a project is hard, especially when they don't get to take it in whichever direction they want. In my opinion, the only solution seems to be to either compromise on the design, or get so much of the base of the project up and working that changing those parts becomes out of the question. Thus my conclusion that to have a stable development team, most of the base OS functionality needs to be present, and the team should be involved in simply the upkeep of the OS and development of the software it will run (after all, no matter how elegant the design is, an OS won't sell/gain users/developers if it doesn't have attractive software).