Hoorary wrote:Hi Stuart and Thorsten (CC'ing other wiki admins), I was made aware of the ongoing discussion on the devel list regarding articles to which I added quotes and/or articles that I created from scratch that are largely based on quotes, so given that I am the main person adding quotes to wiki articles, and just to be very clear about it (assuming that you will be/have been discussing this off-list anyway):
I am not the slightest bit "attached" to any of those quotes (or even the corresponding articles) - however, quotes are just a symptom of FlightGear, and espcecially ongoing developments, being de-facto under-documented, i.e. tons of useful/helpful information hidden somewhere in the archives, inaccessible from an end-user standpoint.
Often that applies to features/developments whose original developers have taken a hiatus/backseat, so that ongoing discussions are intended to help others learning more about something without having to spend weeks searching the archives. Concerning the wiki article that Durk mentioned, that's a good example, because we've literally seen dozens of discussions on the forum/devel list, so that it simply made sense to provide a "goto place" (or even just a FAQ) for those wanting to learn more about the functionality and restrictions of the system.
Honestly, Erik and Stuart pretty much hit the nail on the head with their statements:
https://sourceforge.net/p/flightgear/ma ... /34950336/Erik wrote:I see why he does this. And sometimes it's
nice to have the discussion available somewhere.
The way I see it he does this since he feels documentation is lacking
and he adds the bits to be replaced with something more useful in the
future.
It's not the best way of documenting things by far, but it is slightly
better than nothing I guess.
Still, quotes are ugly, and I agree with that very much. I have been discussing with others to change their appearance and to rewrite them automatically using the piece of code that is currently used to create them automatically (Instant Cquotes).
However, quoting remains an extremely powerful, and a really straightforward, way to bootstrap new articles/documentation, and even features that materialize over time (see below for more on that).
I think some of you (e.g. Stuart, Erik, TheTom) may be more aware of that than most others are, e.g. because Stuart actually bothered to help rewrite the HLA article on the wiki containing tons of quotes. Most others do not bother to do that, i.e. to actually document their ongoing efforts.
Apart from that, that's the whole point of having "quotes" on the wiki: bootstrapping new articles, i.e. to help document features/developments that would otherwise be undocumented.The really power of quoting comes in when people can easily look up who's made certain statements, who's supportive of certain developments and what the context of the original discussion was. For instance, rendering related comments coming from Thorsten/Stuart are more relevant than comments I make
With the obvious exception of fellow wiki admins, and Stuart and Thorsten in particular, many of those who have expressed dissatisfaction with the state of affairs on the wiki, are not even regular wiki contributors (and also haven't been prior to the use of the corresponding script/technique to create/update articles, despite now making their potential involvement/contributions dependent on quoting being generally discouraged).
In fact, most others have literally taken a backseat when it comes to using the wiki to document their work/developments, and that even applies to $FG_ROOT/Docs/README.* - the legacy mechanism to document new features/developments.
In other words, this is a chicken and egg problem (totally unrelated to the wiki or my involvement there, and quoting in particular) - we would not need to have
any quotes whatsoever if more people bothered to document their work, no matter the form/platform, even if people were directly contributing to the LaTex docs or $FG_ROOT/Docs.
That being said, there's a wiki template to veto the addition of quotes to certain articles (i.e. those that people actually, and actively, create/maintain and update articles), and it would be trivial to easily disable quoting template whenever that template is encountered.
However, you will find that most articles using quotes either haven't been updated in months (or years) by their creators (think Rembrandt), or were even bootstrapped using quotes (often by myself or others tracking devel list discussions, trying to perserve valuable discussions, information and pointers).
For instance, features like Phi are basically undocumented on the wiki - equally, the Qt5 launcher article on the wiki is entirely based on quotes that were rewritten over time:
http://wiki.flightgear.org/Integrated_Qt5_LauncherFor the record, the Qt5 article hasn't ever been touched by the developer of the corresponding feature (Zakalawe).
Equally, the HLA article was bootstrapped, and updated, using quotes:
http://wiki.flightgear.org/High-Level_ArchitectureI don't like quotes, and I hate their appearance (which is trivial to change by editing a single template, which will affect all quotes/articles automatically ...)
But there's a certain loss of information involved if quotes should be removed altogether, unless people are willing to roll up their sleeves and do something about the lack of documentation, which is obviously appreciated - and the whole point of bootstrapping articles via quotes.
Which is getting us back to the points that Erik and Stuart made independently, despite certainly also disliking those quotes.
For some background, I suggest to refer to the article on the wiki documenting the script we've been using to create/add those quotes:
http://wiki.flightgear.org/FlightGear_w ... nt-CquotesLike I mentioned previously, I won't be spending much time on FG matters anytime soon, so feel free to do with those quotes/articles whatever you think is appropriate, but I would strongly suggest to consider looking at them without looking specifically at the degree/nature of my involvement, and maybe looking at the potential merits of preserving valuable information - especially keeping in mind the typical latencies involved in contributing to FlightGear, and the shelf life of "cool" ideas.
Quite frankly, certain features we simply would not have today in FlightGear, if it was not for articles summarizing community debates/consensus and popular feature request - e.g. the Canvas system, which was basically suggested by Mathias and James in early 2008, before being prototyped by someone external/new to the project at the time, 4+ years later, based on a wiki article and the momentum it helped create/maintain:
http://wiki.flightgear.org/index.php?ti ... ldid=21408Likewise, you could say that Stuart's current HLA/AI related work is in line with the collection of long-standing discussions "quoted" at:
http://wiki.flightgear.org/Decoupling_t ... _RationaleThe main difference here is that this is not using automatically-created "quotes", but manually assembled - and nobody seems to take any offense by that article apparently, probably because its appearance is not as obnoxious.
Thus, I would respectfully suggest not to underestimate the power of "aggregating" long-standing debates/discussions and presenting those on the wiki - while the style/appearance leaves a lot to be desired, that's easy to change (or move elsewhere, like Stuart suggested).
People may disagree with my persona and the degree/nature of my involvement, but once you spend 15 minutes reviewing such articles, you will see that I didn't/don't have any personal/particular agenda other than summarizing such debates, i.e. that these are actually long-standing issues that have been going on for years literally, so that it does make sense to preserve such discussions, and provide pointers.
In fact, referring to the "Decoupling AI traffic" article I mentioned before, you will find that Durk is the one primarily quoted there, and that Stuart's current HLA/AI work aligns well with his original remarks - and most people would not even be aware of that (probably including even Durk), if it wasn't for that particular article.
For two more recent examples of articles, and actual code/features. that were entirely bootstrapped using "quotes", refer to:
http://wiki.flightgear.org/Graphics_Card_Profileshttp://wiki.flightgear.org/Howto:Using_ ... r_%28IG%29Other examples like the FGPythonSys article may seen unfortunate at best, but they still represent ongoing work, and contain tons of useful information/pointers for people wanting to get involved:
http://wiki.flightgear.org/FGPythonSysAgain, there's no agenda here on my end, simply because I don't have any use for Python myself, I just took the time to document this to ensure that things will be preserved for the future.
No matter what you end up deciding here, I suggest to tread carefully with the degree of information that is currently preserved in the form of "quotes", especially if you are considering to remove them altogether, without providing a feasible replacement/alternative.
And I would particularly recommend to look at who is actually going to get involved to improve the state of affairs, rather than just those who criticize the current situation, which is really just the result of people not documenting their works - i.e. are people willing to be part of the solution or just part of the problem ?
In a perfect world, we would not need to be using quotes - or they'd just be an interim thing, until someone comes along to help rewrite such articles (like Stuart has done in the case of the HLA article, and like others are offering to do now) - but realistically, most people do not even remotely care enough about documenting their work properly (excluding obviously Thorsten and a few others, who're apparently writing at least as much documentation as code).
(feel free to discuss/forward/re-post this as you deem appropriate)
PS: Speaking for the future, and in general, just like core development matters are generally discussed on the devel list, wiki matters are generally discussed -in public- on the wiki directly, or at the very least in the Documentation sub-forum on the forum, and the corresponding feedback/opinions are weighted based on the involvement of those voicing their concerns - also not unlike core development matters I guess.
In summary, whenever someone is willing to roll up their sleeves to do something about a particular problem (e.g. getting rid of quotes), that is highly appreciated. And quite frankly, those quotes have accomplished quite something if there is a consensus to improve the state of documentation on the wiki, even if that means to get rid of quoting and to re-introduce the practice of requiring developers to document new/updated features in the form of additions to $FG_ROOT/Docs