Why are we so unfriendly to novices?

Question

Retrocomputing has struck me, over the last six months, as being particularly unfriendly to novices. It's now reached the point where a new user (both to RC.SE and SE in general), probably a non-native speaker of English, comes in and gives the only correct answer to a question, and in less than a day it's deleted, leaving only information (in the question and other answer) that's greatly misleading.

My go-to example of the worst of this is "what is DMA and how does it work?". DMA invariably is used in modern computers, to the point where nobody even thinks about it any more, which is exactly why it's a good question in the context of '80s micros, where it would be a key selling point and worthy of discussion because it's rare. We managed to get that one re-opened, but only after quite the struggle against its almost-instant closing.

But there are plenty of other examples of situations where the context was so different from modern computing that the OP hardly knew what to ask, and all to frequently our response is not, "here is an introduction to the background you need to start to comprehend this entirely different universe of computing," but "You should have already had the background to have asked the question better for those who do already know all this."

Except, of course, as above when we start deleting correct answers.

I have no idea what to do about this at this point. Personally, I'm feeling like I should just move on to other places to document old computers. (You may note that after six months or so of fairly intense activity that lifted me from nowhere to #4 last year in reputation gain I've slowed down.)

Can this be fixed? Or has the community settled into into a state where the "in group" is going to do its thing and outsiders, even those with better information than anybody here, are simply not welcome?

And oh, yeah; I cannot undelete that one correct answer above because it's been deleted by a moderator. Maybe a mod could reconsider and undelete it? Though I suspect we've lost that user (again, the only one here who actually gave an answer anywhere close to being correct) forever.

---

Update: It's now a week later, another moderator restored the answer, and it's since gotten upvotes from people other than me (the OP) as well. So that's good, as far as it goes. However, the new user appears to have abandoned the site, and I'm also pretty disappointed that the moderator responsible for driving him away had not even owned up to what they've done, much less helped to fix the issue they created.

Answer 1

It's not just novices.

We have some documentation here on meta describing the site's norms and expectations (e.g. scope), but our culture has drifted since then. Our unwritten rules have drifted, and not always in (what I see as) a good way. What's worse, each user's expectations have drifted separately.

Right now, (by my reckoning,) the expectations of the top 36 users by reputation differ widely. Everybody wants a site focused on retrocomputing, but… what is retrocomputing? Early in the site, we had a lot of ideas about that, but one (the then-CM Robert Cartaino's) generally won out. Two-and-a-bit years later, and the perceived consensus had changed…

And while the most favoured answer back then focuses on the fact of real, every day usage of otherwise outdated systems, todays concensus incooperates a wider range of outdated systems, but especially theoretical and research questions about old software, hardware and culture in general.
— Raffzahn's answer

This is one of the best summaries I've seen, and yet I still disagree with it; for instance, the answer declares questions like SheepShaver won't open after reboot – which has received no close votes ever despite my repeated use of it as an example – off-topic. Are these on-topic? Well…

Disagreements about what's acceptable on the site can lead to users seeing completely different sets of rules applied, depending on which high-activity users happen to be around at the time. It can lead to one group closing a question while another group is trying to answer it, or a moderator deleting an answer that others consider fine. _{^{(Though I've never done this. *burns evidence*)}} Who's right? Who knows?

There are even places where the behaviour of (what I see as) the average community member is in conflict with not just network norms, but the Stack Exchange model itself. Take commenting, for example. If you'd all just use comments “properly”, it'd make moderation a lot easier – the tools are insufficient for how most sites use the software, but very insufficient for how we use it. But, as the comments to that meta post show:

[…] my little addition isn't what I would regard as full featured Answer, not even a partial. It is .. well .. more of a comment regarding one (of several) details. […] Hell, doesn't that very much highlight the issue with comment vs. answer? Not every valid information forms an reasonable answer nor should be handled so, but at the same time it might be a useful addition. — Raffzahn Dec 11 '18 at 17:28

^{(I know I keep quoting Raffzahn – that's a coincidence! Lots of other people besides Raffzahn disagree with me, and each other, about this kind of thing.)}

How can we possibly start treating novices nicely if we can't even agree on how we are to behave?

Canonical Q&A pairs

That being said, it might be a good idea to write some canonical Q&A pairs for simple things (e.g. “what is a floppy disc?”) so that we can point newbies there instead of just getting cross with them. Build the tools you often find yourself in want of.

But how do we co-ordinate producing these canonical Q&A pairs (that anybody could, and so probably would answer), so that they have one or two stellar answers instead of five or six mediocre ones? Same way as we co-ordinate tagging: we use meta.

To conclude:

Use meta more.

Both of these problems can be solved that way, and a lot more besides. It's not the best tool, but it's the best one we have.

Answer 2

As a novice user, who mostly browses the site, it seems to fulfil three distinct purposes:

1. Computer history, both documented and anecdotal. This might include systems, software, programming languages and social or business history.

2. Maintaining old systems in the present. This includes programming retro systems, or using languages that have fallen out of favour. It can also include adding to or replacing parts with modern equivalent and emulating.

3. Nostalgia, often for home micros, game systems, or games. Often trivial, but usually fascinating to read!

I note that each of these purposes could overlap with other SE sites, for example coding could fit into Stack Overflow and games or game systems into Arqade.

My own interests overlap as well. I was once a coder/developer, and have worked with VMS systems, DOS, Unix, Fortran, Pascal, C, Logo and other systems and languages in the past. I did use Zx Spectrum, C64, Amstrad CPC and PCW systems as a child. I still have and use a Sinclair ZX81 and CPC6128, though I rarely code on them. I often use emulators.

The breadth of the site is its strength. When policing the boundaries, it would be best to include whatever cannot be better fitted elsewhere. If the site comes out of beta, it should become a destination for questions from other sites. If it is hard to call, the presumption should be that questions remain.

The only way for the site to grow is to make it more inclusive. Edge cases should be allowed wherever possible. If the answer to “would this question get a better answer on another SE site?" is no, then it would be better to remain, even if it stretches the intended domain of RCSE.

Answer 3

You've picked two great examples.

In the first case, I may be as unhappy as you, as it may really have been a useful answer - if it wasn't for it's quality issue. Even if true, as of now, it isn't really helpful, as it fails to give any details or explanation. I would love if the author could add some substantial information. But I feel it's way too easy to blame some 'unwelcoming' nature. The way Stackexchange works offers only a limited set of tools to handle this. For an issue like here, there are essentially only three options:

• The author improves the answer
• Someone else, who can explain the correctness/background, does so
• It gets turned into a comment

(Important side note: it did not get deleted, but put into a comment - it's still visible to readers of the question.)

Here the last has been chosen. Maybe a bit premature, still the only valid way to handle it without the ability to add information. After all, while there may be many points moderation can be critizised for, not being all-knowing is not one of them. By the rules, it doesn't pass the threshold for a valid answer, as it just states something with next to no explanation, reference or background. Turning it into a comment is appropriate.

Don't get me wrong: I'm not happy with the moderation here, and personally may have tried to brush up the wording, but I can very well understand why it was handled that way.

Now, for the second example, I have to start by making clear that I do not understand in any way why this question has been reopened. Having said that, you picked a great example for this part - one that even contradicts your primary stance about RC.SE being unwelcome. A decision against a question because it's clearly off topic, has nothing to do with being "friendly". And this one is as clear off topic as it ever can get.

Try for yourself and replace DMA with CPU. Don't laugh, just do it. You'll notice that your argument about being 'so basic noone uses it anymore' works as well, after all, some mystic 'cores' are listed in tech specs of modern devices. The same can be done with next to any other basic item about computers. Like 'what is a bit'. Isn't it?

But again, this example hits an important point: Retrocomputing is a _sub_domain of computing (hence the need of an adjective) thus RC.SE wrestles with this basic dichotomy that next to all topics of computing happen in old computers and their usage, but not all topics are specific to old computers. It creates the need to distinguish every time between issues which are specific to the use with old computers and everything else.

Bottom line: It's mandatory to keep that line - unless we want to replace any other computing related SE area (including SO) - RC.SE for World Domination :)

So far the definition side. But then there's a spin of 'welcomeness' which in my opinion shows an assumed mission of RC.SE:

But there are plenty of other examples of situations where the context was so different from modern computing that the OP hardly knew what to ask, and all to frequently our response is not, "here is an introduction to the background you need to start to comprehend this entirely different universe of computing," but "You should have already had the background to have asked the question better for those who do already know all this."

First of all, I guess the second 'quote' is worded in an unfair way on purpose, to make a point. After all, the reply to the DMA question, as well as others alike, is not suppressing it, as implied, but redirecting it to a more fitting area. In this case one that is about generic computing.

Now this brings up one of the real issues RC.SE (and, as far as I can tell its moderators) wrestle with: The missing ability to reroute questions. As far as I understand this, they can't forward a question to a more appropriate area, as long as RC.SE stays in Beta. Leaving Beta may relieve this a lot. I think RC.SE is ready for this already since quite some time, even though it does not fulfil the numeric criteria. IIRC others have been promoted without passing the numeric criteria - so wouldn't it be a good idea if moderation (or whoever is in the relevant position) could make this their prime mission?

Back to order. Trying to give an answer to everything within RC.SE, like implied with the first 'quote', would mean to eventually tug every topic into RC.SE. As far as I can tell that's not what RC.SE is meant for. Since retrocomputing is a _sub_domain of computing, an area about it needs to restrict content to exactly that. Its content can only be relevant if restricted to its topic.

Or in more hands-on terms: A question needs to have a concrete RC specific anchor, even if it's answer may end up being a more generic one. Just mentioning that a term has been heared in relation to some old hardware does not satisfy it.