I've been disconcerted by the speed of refactoring for a while. When the number of RecentChanges
by a single person are more than I can follow, I tend to:
- not bother to read them, and
- lose trust in the editor.
If I can't read fast enough to keep up with the RecentChanges
, I question the care of the person making the edits. This feeling is further influenced by my experience when I attempt to follow a spate of changes. Frankly, I get irritated by the quantity of broken links and lost content. I do not have all day to FixYourWiki
, so I've been voting with my attention span. I've stopped visiting and editing as often. I don't mind big page refactorings, but time and community consensus are two important parameters in Wiki's UnitTest
s. -- SeanOleary
Sean: right on. And beautifully expressed as well.
I agree. I don't have a specific problem with anything Francis has done, and I am glad that he is deleting a lot of empty/redundant pages. But whenever one person
starts making so many changes so quickly, there is a concern about bias and fairness. I don't know if the answer is to ask Francis to edit less, or for the rest of us to edit more. Unlike some, I don't think that reducing the number of pages is raising the quality of the content. Refactoring is an important activity, but it is not as important as adding useful information. -- KrisJohnson
Wiki should not be a speed-typing contest. If too many people are actively editing, and their edits are in conflict, everybody should back off until consensus is reached. -- DirckBlaskey
These are valid concerns. I must admit that the last batch of refactorings (in the last 4-5 days or so) have been too fast for me to keep up with as well. Although in principle I believe the best solution is for everyone concerned to participate more, there is a limit to how much time each of us has available to spend on the task. In any case, this kind of discussion is far more productive than attributing malicious intent to Francis is. Let's put away the name-calling and start figuring out what to do. -- RobHarwood
There is an element of "please GetaLife
so I can GetaLife
", but if the RecentChanges
are that important to me then the problem is at my end of the Cat5, not the editor(s).
Let's imagine for a moment an ideal world in which...
- everyone would refactor about the same amount of the Wiki each week
- people held off on adding or refactoring when RecentChanges was over a critical limit
- people had a way to achieve consensus on which parts of the Wiki to improve first
Now, what would we expect in such a world? We would expect the quality of the Wiki to grow, and the capacity of the Wiki community to improve the Wiki to grow as well. (I'm pretty sure someone could draw a diagram of effects showing how this would play out.)
We know the world isn't ideal. What is the actual state of the world, and what can we do that reduces the "reality gap" between this ideal world and the real world ?
I have a couple of thoughts on this:
- Wiki is very lopsided, with the vast majority of people being readers not contributors. Even among contributors, there are very few who actually refactor. Refactoring is difficult; not everyone really wants to.
- I think history has shown that consensus generally cannot be reached on these topics. At least, not in the traditional sense of 'consensus' where everybody agrees wholeheartedly. Wiki's contributors are too diverse. Wiki tends to arrive at solutions through a balancing set of forces, a constructive disagreement if you will.
Nonsense, all of it. RecentChangesIsNotTheWiki
, it's just your lazy window into it.
It sounded to me as if Sean was using RecentChanges as an
index into the quality of the Wiki - not seeing it as the whole. I don't think I made that mistake myself either, in my subsequent proposal. I might be mistaken about what I'm mistaken about or not, of course...
If you're saying "Wiki bad when Wiki change faster than I can see and assimilate all change", then you're wishing for a Wiki that doesn't scale to the Web.
Would a Wiki that doesn't scale be bad?
The complaining is all focused against page deletions and "refactorings", the latter of which are still fairly small in number, while no-one's complaining about the number of new pages being added. Aren't those also changes? So what does that mean?
I don't know... :)
Voting with your attention span is one option. Writing that you've voted with your attention span and using that to lobby for others to change their behavior is an attempt to manipulate. Reader/writer beware. -- WaldenMathews
Although I agree in part, Wiki is more than the pages herein. It is also a community of contributors and readers. The community part is what is speaking up right now. The issue doesn't appear to be a simple "There are too many recent changes for me" as the title unfortunately implies. Rather the issue is that these changes are too many for many of the contributors; too many to PeerReview. Without PeerReview, Wiki fails. Perhaps Wiki does not scale. Whether it does or doesn't isn't really important.
Could be I think scalability is important because Wiki is born of deep Web sensibility, building on a vision of TimBernersLee
. There are not too many changes to PeerReview
, only too many pages for a single reader
, but because there are many readers, that's not a problem. When a single reader experiences insecurity of this nature, the solution is for the single reader to solve his/her own insecurity problem. This requires some honesty and soul searching.
Let's call a spade a spade. Most of the content here is rubbish, and no-one complains as it piles up and up. People only complain when their darlings get maimed (see KillYourDarlings
). Please make your complaint personal and specific to you, and honest. "I'm losing control and I'm not the star I thought I was." You cannot control this medium, although you will repeatedly become addicted to the appearance that you can. Give it up, darling. -- WaldenMathews
Walden might have something here. I have control issues. My most satisfying edits are the ones that tie up loose ends (the blissful joy of refactoring MakeProgram
, ahh...). Also, the following line, "RecentChangesIsNotTheWiki
, it's just your lazy window into it." is the best line I've read all year. I'll keep an eye on both of these thoughts when I become involved in larger issues. ThankYou
, keeping his nose in his own WikiGarden?
, tending to his darlings.
However, I think Sean's concern is also related to the fact that if the edits come too fast - especially deletion seconding, which is impossible to undo unless you have an external copy - control over refactoring is ceded to people simply by virtue of them being there first. If you had to distill this to one sentence, you could simply requote Dirck's excellent statement above: "Wiki should not be a speed-typing contest." Personally, I'm going to try to do as much refactoring with less typing. We'll see how it works. -- FrancisHwang
Let's not confuse speed with decline in quality of decision. If a good page got deleted real slow
, the loss would be the same. I've seconded a passle [?] of Francis' deletions in the past week or so. It's easy, and it's fast. Click through to EditCopy
and see what was there originally. It's usually just a link to somewhere else in the internet with no value added. Then see if the page is heavily referenced. It's not. Bye bye. Whoops, should have checked RecentEdits
, oh well.
The only way to achieve the freedom to create is to not care if what you write gets deleted or otherwise maimed.
It's similar to JerryWeinberg
's claim about consulting, that if you care who gets the credit, you'll never be able to help anyone. Another lesson borrowed from Weinberg, I know that in his writing workshops he encourages students to write and then discard all they've written and begin again. If you're afraid to do that, you're clutching your rip-cord too tightly, and you're not ready for the WikiWikiPlunge
. We need a WikiBootCamp?
around here to get the troops into the right frame of mind.
Francis, beware of making changes to your process just to make sure everyone knows you're reasonable. You are, and you've proved it already. To your detractors, if I could speak for you, I'd say "PairWithMe?
if you have an issue", and let the actual pairing speak the truth. I believe Sunir said as much before, but it bears repeating.
I lost the courage to do any more refactoring, after receiving complaints whenever I've tried. (I can't even delete my own writings
without someone complaining.) I'm glad Francis is doing a job I don't have the courage to do myself. I haven't examined his refactorings, but if the complaints really are about people's "darlings", I do hope Francis will ignore them. -- KrisJohnson
To me, the proposed algorithm to "back off refactoring when RecentChanges
fills up" seems to have it all backwards. In times when there are more normal edits/contributions, some of them good, some bad, we need more
refactorings and deletions to keep s/n up, not less. -- FalkBruegmann
Can we tear this down now? RecentChangesIsNotTheWiki
seems to be the diamond in this rough. -- SeanOleary
Sorry to reincarnate this page, but it expresses much more than what is on RecentChangesIsNotTheWiki
. Plus it's relatively fresh and could use some more consideration before we pull out the good bits.
I see the Wiki as a place where people take on different roles. FrancisHwang
is not SunirShah
is not RonJeffries
; they each fill important roles based on their personalities. I welcome Francis and HisIncessantCleaningOfTheWiki
, because he's filling a role of extremely active refactorer. I can fill in my own role, with some refactoring and some writing and some WikiGnome
A world where every WikiZen
refactors about the same amount of Wiki is not ideal to my mind. -- BrentNewhall
Statistical sampling should be enough. Folks might check two deletions but wouldn't bother with twenty; however if they still bother to check two at random then that's 10%. I've said "CheckUpOnTheGnomes?
" before, and I do that when I feel like it. Every time I do, I feel admiration for the patience of someone who bothers to go through a list of really dull, dead pages in alphabetical order. -- MatthewAstley
Reaction to this page, which I haven't seen before. One of the things which I do to refactor wiki is to fix and enhance links, or to add a category. That may well mean a series of small changes to quite a few pages. Then I wait for 24 hours until the index is updated to check that it has happened as I expect. The result of that is a list of entries on RecentChanges
. -- JohnFletcher
Having looked at WikiPedia
(dozens of pages per second
), I find this debate laughable. Even at the height of this wiki's refactoring and edit wars, things changed relatively slowly here. I don't know when this page was started, but now (2006) edits are a mere trickle, and the days of the big refactorings seem to be long gone.