What's left of the forum?

[Troy Martin]

IMHO, only about 20%.

I've noticed that a lot of the questions, posts, etc. about OS development lately are either related to tutorials like JamesM's and bkerndev and bare bones, and not as much on harder, more thought-inducing questions. For example, look at the first screenful of the active topics page. A few on basic booting, a few on bare bones, a couple of the usual things like triple faults on paging, and cross-compiling. Only a few of them are good, wholesome questions like "PCI Header Types" and "Exec call in userspace library."

Why don't we have a forum for beginner questions, and one on advanced questions, or a separate forum site for hobby OSes and advanced topics?

[purage]

You keep failing to understand that they don't want to keep a forum for beginners. No one here wants to teach people how to program. There are other forums for that. What they should do is open up the wiki to anyone with an account and allow them to make suggested changes/topics which could be applied by a moderator. They should close the forum and make it strictly an archive (which would eventually be phased out) so to focus more on developing the wiki further. And, I think the wiki should be a simplified version of the official technical guide with little to no code, used only to illustrate a point and not just a less detailed copy of the original source.

Things to consider:
1. The forum fails as a community
2. The wiki is potentially a better resource
3. There are often more appropriate places to find info outside this forum

Just my 2 rupees.

[Colonel Kernel]

There are still occasionally good design discussions on the forum. The signal to noise ratio is getting worse, but for me there is still some value in visiting here. I just end up ignoring a lot of stuff.

IMO the best set of resources would be a heavily-modded forum for advanced topics in the style of Lambda the Ultimate, the wiki, and the beginner questions can all go to a larger community like Stack Overflow that is designed to filter good questions out of the general soup.

[01000101]

They should close the forum and make it strictly an archive (which would eventually be phased out) so to focus more on developing the wiki further.

That's a horrible idea IMHO. This forum has its task of allowing people to discuss ideas and get help with things (which is often), the wiki is the reference for everything. They depend on each other.

I think some change does need to occur here, but not that kind of change. I think more events should be active (like the 512-byte contest). Those stimulate the community into further discussions and competitions are always a good way to see some great ideas in practice.

[pcmattman]

I've noticed that a lot of the questions, posts, etc. about OS development lately are either related to tutorials like JamesM's and bkerndev and bare bones

Put simply, once you reach a certain stage in development it's not worth asking questions anymore - you can figure things out yourself for a start, and often questions are so linked to your own kernel that it's impossible to ask without having to explain how half your kernel works.

and not as much on harder, more thought-inducing questions.

It's also difficult to have a decent design conversation because everyone has an opinion on how things should work. It's too easy for a simple discussion to become a multipage argument which gets nowhere.

My personal opinion?

Put the time in to your own kernel, learn what's happening, ask questions when you're not sure about something. But when you get to a point further than most, come back to the forums and put some time in for the newbies. You were one before (NOTE that I didn't say noobs - there's no place for someone here who doesn't know any C let alone how to write an application).

Basically, what's left of the forum is precisely what this forum is for:

The Place to Start for Operating System Developers

[Colonel Kernel]

Put simply, once you reach a certain stage in development it's not worth asking questions anymore - you can figure things out yourself for a start, and often questions are so linked to your own kernel that it's impossible to ask without having to explain how half your kernel works.

That hasn't stopped me from asking pretty advanced questions. Back in the Mega-Tokyo days, there were maybe 5-7 people who could actually answer them. Now there are maybe 1-2 (and one of them is named "Brendan" ).

It's also difficult to have a decent design conversation because everyone has an opinion on how things should work. It's too easy for a simple discussion to become a multipage argument which gets nowhere.

I once participated in a 12-page thread that was frequently just a shouting match about Exokernels between two other forum-goers, but I learned a lot from it. Mature flame wars are educational; immature flame wars are just noise.

[Troy Martin]

Yeah, so I made a forum for advanced stuff at http://aosdev.freeforums.org/. Join if you want, it was basically my boredom taking effect.

[JamesM]

What they should do is open up the wiki to anyone with an account and allow them to make suggested changes/topics which could be applied by a moderator.

The wiki is open to anyone with an account. You merely have to ask a moderator (in the osdev wiki forum subsection) to add you to the "wiki" group.

[purage]

No, that isn't what I meant. I mean that anyone who signs up to the wiki could make changes/add topics without approval which would be placed on queue and approved by moderator later; without ever asking for approval to do so. You see my idea is designed to phase out the forum to be replaced by a better wiki; this idea solves many problems and relieves many stresses. I would want to archive the forum until the best of it has been accounted for in the wiki, then eventually the rest of the forum would be gone forever. Think of it as cutting away the fat.

[JamesM]

No, that isn't what I meant. I mean that anyone who signs up to the wiki could make changes/add topics without approval which would be placed on queue and approved by moderator later; without ever asking for approval to do so. You see my idea is designed to phase out the forum to be replaced by a better wiki; this idea solves many problems and relieves many stresses. I would want to archive the forum until the best of it has been accounted for in the wiki, then eventually the rest of the forum would be gone forever. Think of it as cutting away the fat.

It certainly solves the problem that noone here respects you because you entered the community like a total prat.

The reason we have a forum and a wiki is because there are two entirely different use cases, both of which cannot be solved adequately by a single solution.

There is a need for a targeted encyclopedia - a reference. This role is currently filled by the wiki.

There is also a need for those who have problems, don't understand something, or would just like peer review on their latest project to post and have a community respond to them. Question - answer, as opposed to the question - lookup paradigm of the wiki.

Both have their place. We get annoyed when people refuse to read the wiki first, and expect to be spoonfed. But conversely I certainly would not be in the position I am today if I could not have queried topics I didn't understand.

Your motion to remove the forum will never succeed.

Cheers,

James

[steveklabnik]

To be fair, OS development is a pretty niche subject. You shouldn't expect there to be a ton of people here. Not very many people are trying to write new operating systems from scratch, and often those that try it on their own can just do it, because if you're motivated to write your own OS, you're motivated to read up on things and figure out how they work.

That is not to say this place has no value. It is to say that I can see why there's not a lot of activity here.

[Combuster]

The wiki is open to anyone with an account. You merely have to ask a moderator (in the osdev wiki forum subsection) to add you to the "wiki" group.

Even that is not necessary. Subscribing is sufficient.

You see my idea is designed to phase out the forum to be replaced by a better wiki; this idea solves many problems and relieves many stresses.

Which takes away people's ability to ask for things that are not present in the wiki. As in, I have this bug that I can't seem to fix... As JamesM said, the need for a forum will not go away.

Yeah, so I made a forum for advanced stuff at http://aosdev.freeforums.org/.

And imo, its not a good idea to split the community like that. As mentioned, there is little need for an advanced forum as people know very well how to google by the time they want to ask those questions, which doesn't make your set of forums an attractive place to go. <troll> Amusingly though, guess who was the first to join? </troll>

Joking aside, I kindof expect this to become a learning moment on not rushing big things.

[Solar]

Things that don't improve a community and its offering:

• Telling the long-timers of that community that they fail. (In case you haven't noticed, purage, those are the guys that can actually answer detailed questions that might come up!)
• Turning a Wiki into a moderated newsgroup. (... )
• Forking the effort. (Troy, do you really think that a forum "just like OSDev just more advanced" will attract enough people to become a community, and actually provide a better signal-to-noise ratio than this merger of the two biggest OS development ressources of their time?)

Sorry, guys. What we see here is the usual "departure of the old ones". They get out of university, or get a life, or both, and their hobby OS projects go dormant. This has happened to many great ones. Pype/Clicker or Candy come to mind. I myself could contribute more if I had the time, both to continue my own OS project and actually hang out here.

But this is the way it goes, and there is only one solution: You yourself must strive to become a "great old one". Yes, that means that there is no longer someone there to lean on, no higher authority, because it is you they come to for answers.

It also means that you can no longer afford to sling best guesses from the hip, but have to do research and be precise. It also means that, if you want something to be done in the Wiki, you have to do it yourself - and if you get it wrong, you'll make a mess of things.

Responsibility.

It's called "growing up". It's a painful process, and it's quite the same as becoming a parent: Those who come after you will cry and ***** and sulk and listen to your words much less that you would like them to, and probably hate you most of the time for being so know-it-all. But if you don't give them the good advice and guidance to get started, they won't have a happy time afterwards, so you give it the best you can.


...

Erm.

OK, so I got a bit carried away on the rant. But I think you get the idea.

It's not about how a Wiki works or where this forum is located. It's all about you.

[JackScott]

Yet again, Solar has put us all to shame.

Procedure for the beginner's forum:

1. Create forum.
2. Newbies *slowly* and *painfully* learn to post there.
3. As soon as a sufficient number of newbies are posting in beginner's forum, the regulars stop reading it.
4. Newbies get no answers to questions, get fed up, and leave.
5. Less newbies turn into regulars.
6. Death of forum.osdev.org imminent.

[Troy Martin]

Newbies get no answers to questions, get fed up, and leave.

That, in my mind, is a good thing. If they don't get answers, there's a 50/50 chance they'll google. If they do, the amount of questions will fall, and the newbies will become regulars, posting with the rest of us. It either turns arrogant/pissy/whiny noobs off from attempting this, or makes them stronger.