User talk:WillowW/Archive6

From Wikipedia, the free encyclopedia

Contents 1 Wikipedia 2 You're going? 3 Prod (and some begging) 4 Begging the question 5 Thanks for sayin hello! 6 Textile arts 7 We need you! 8 Sweaters and the curse 9 Bias 10 Eclectic tastes! 11 3RR on Wikipedia:Wikipedia is failing 12 Ideas for a new bot 13 Note to O. regalis 14 Hi, if you have a minute please read my essay. 15 Essays and such 16 New WikiProject 17 Thanks 18 Help! TA template 19 WikiProject Unassessed Articles 20 MCB worklist question 21 You've been busy 22 Return visit 23 Beverage production 24 Well, well... 25 This may be helpful 26 Ice cubes that fly 27 Minor thiols in little-known organisms 28 Oops! Slight mistake! 29 Thanks

[edit] Wikipedia

Hi, er... sorry for viciously tearing away at your hard work, but I'm trying to balance out the article and while the section is needed I think it was a bit too long, and there were some other concerns (see the talk page). I think in its original form it was an excellent comparison of Wikipedia and traditional encyclopedias which could stand alone, but which didn't really fit in an article that is predominantly about one and not the other. Opabinia suggests "A lot of this material is very well written and would make an excellent essay in project space comparing and contrasting the advantages of the two models", so feel free to use an older revision of the page as the basis for such an essay. The article needs re-organizing and re-thinking, and I've made some fairly significant changes in pursuit of this, not just to this section. Opabinia has also left a big list of concerns on the talk page which I'm trying to address; Chris Cunningham wanted to think up a new layout for the article but seems to be short of time at the moment, so I might have to do it for him. Thanks again for everything you've done on the article – Qxz 12:08, 1 February 2007 (UTC)

Hmmm... OK... Chris seems to have decided to get rid of it completely. So it's not my fault! – Qxz 14:58, 1 February 2007 (UTC)

[edit] You're going?

Oh no! Well, we'll cope somehow. Anyway, I agree with everything you just said on Talk:Wikipedia; I trimmed that section down a bit but it was Chris who removed it completely and I don't really agree. Good luck with whatever it is that's interrupting, anyway – Qxz 17:54, 1 February 2007 (UTC)

Well, now who's going to write such nice prose in our encyclopedia articles? Just kidding, we'll all do our best... hope everything works out for you in the real world, and my kitty sends a meow to yours ;) See?

(Not mine, but could be her mirror image, down to the white on the nose.) Opabinia regalis 01:46, 2 February 2007 (UTC)

Going? Say it isnt so !!--Filll 02:06, 2 February 2007 (UTC)

[edit] Prod (and some begging)

Physics is being rewritten and we are looking for contributors and/or moderators at Talk:Physics/wip. Is there any way I could prevail upon you? --Filll 16:46, 27 January 2007 (UTC)

I'd like to help out as well, but I can't; I'm really sorry. But please take some comfort from the fact that the article will still be there when I return to editing regularly, which I hope won't be too far off. Good luck; I know you'll do well Willow 14:44, 3 February 2007 (UTC)

[edit] Begging the question

Hi Willow,

you may or may not already know the following story - I'll tell it anyway.

In medieval times, there was a culture in formal debates. The contestants would take it upon themselves to argue a certain point. Part of the preliminaries of the actual debate was that contestant A would ask his opponent to grant certain things, a list of reasonable, general statements. In the course of the debate, contestant A would then try to show that the assumptions that were granted to him by his opponent imply the point that A wanted to argue.

In the formal style, each opponent would humbly beg the other to grant certain assumptions. If contestant A would try to slip in his point in a barely disguised form, his opponent would protest: "You are begging the question."

I noticed you using that expression on a Talk page. --Cleonis | Talk 18:58, 3 February 2007 (UTC)

[edit] Thanks for sayin hello!

Hi Willow

Thanks for coming along to my talk page and saying hi! Apologies for taking so long to get back to you. It's great to have a passion, isn't it?! I just love indulging my love for pageantry but spenting ridiculous amounts of time on here! Will take up your recommendation to see Little Miss Sunshine... I love Abigail Breslin anyway (first encountered her on an episode of NCIS... brilliant!) so it is a must-see :)

Thanks again, -- PageantUpdater • talk | contribs | esperanza 06:48, 5 February 2007 (UTC)

Hey Pageant, it is so good to hear from you. I totally know what you mean about spending too much time on WP; I have to be careful that it doesn't interfere with my job! But I confess, I'm so happy here; it's like walking into my own little Garden of Paradise, where nothing is scary or harmful; even the worst tigers here have rubber teeth. ;) And doesn't it feel wonderful to imagine that we're bringing something good into the world, showing the world how beautiful it all is? Even after ten months, I'm still hooked. ;) Willow 18:10, 5 February 2007 (UTC)

[edit] Textile arts

Tentatively yes, although first I would like to know what I'm committing to if I say I'm "interested in contributing".... /blahedo (t) 16:50, 5 February 2007 (UTC)

Hi Don,

I'm mainly concerned that we avoid seeming like an "inactive" WikiProject and then deleted. I don't know if there's a fixed criterion, though; maybe an article per month or something? I'm hoping that enough people will come out of the woodwork to make the project thrive; certainly, there's enough work to be done, if only in adding pictures to the existing articles! :) I was thinking of writing some software to visualize an arbitrary pattern of stitches; that way, we wouldn't have to reproduce Babara Walker's titanic achievements of making swatches for every conceivable pattern. Being a computer scientist and a knitter, do you know of such software already? I was just going to write a little GPL driver for Xfig and maybe a 3D visualization program. Willow 18:00, 5 February 2007 (UTC)

Heck, one a month, I practically do that already. Well, not so much in the last couple months. But yeah, I can do that much.

As for pattern vis, are you looking for "looks like a photo" or something more NPR? I can imagine a place for autogenerated diagrams even if you could get photos for everything. But no, I know of no such program yet. Seems like it'd be a pain to get all the fussy details right; a good student project, maybe. :) /blahedo (t) 03:36, 6 February 2007 (UTC)

Thanks, Don! :) If you could sign up at Wikipedia:WikiProject Council/Proposals#Textile_Arts, that'd be great! Hopefully, others will come out of the woodwork as well. I'll keep you posted, Willow 14:14, 7 February 2007 (UTC)

PS. I may try to tackle the software in my copious spare time, although it may well take six months for me, and would likely be "kludgy". Is there a better scriptable software than Xfig for 2D or (say) RasMol for 3D? Ideally, I'd like to only have to learn one new language.

[edit] We need you!

Dear Willow,

I don't know how you do it all! But ... I have one more thing for you. If you haven't heard of Distributed Proofreaders [1], go over there and take a look! We've turned over 10,000 public domain books into free ebooks and released them into the wild. The folks there are nice, dedicated, smart, and there are few, if any, difficult people. If you want to proofread Latin -- you can. For Greek, you'd have to go to our sister site, DP-EU. You can just do one page whenever you have a moment, and know that you're doing your mite towards the universal library.

I'd understand if you didn't, however. There are only 24 hours in a day, and they disappear quickly if you have intellectual ADD, like me. Zora 10:57, 6 February 2007 (UTC)

Hey Zora,

It seems right up my alley, being so nitpicky as I am. ;) I'd never heard of it (I'm pretty sheltered and clueless) but I'll be glad to help out, maybe a page a day? At first I thought you meant the WikiProject League of Copyeditors, which I'd heard of before, but I see that it's different.

It's great that you stopped by; I like getting visitors here! :D Willow 11:04, 6 February 2007 (UTC)

[edit] Sweaters and the curse

Um wow. I hardly know what to say. Unless it is something like: "By the way, my name has THREE Ls in it"...--Filll 13:22, 7 February 2007 (UTC)

Of course, TIPFIC¹--Filll 14:00, 7 February 2007 (UTC)

¹tongue is planted firmly in cheek

[edit] Bias

Although I have seen pictures similar to [2] many times, when I just saw this one of Phillip III, I realized that the fashion for men, especially men of a certain station in certain places, was essentially a sort of dress with a very short hemline. One gentleman even seems to be a bit more daring, with bare legs and a couple of slits! I am not sure what to make of it, but I thought it was strange that I have seen this my entire life and never quite made the connection.--Filll 13:39, 7 February 2007 (UTC)

Yes, doublet and hose were once very popular as men's clothing. I guess that the choice of whether to wear an overgown is a question of etiquette and warmth, but I haven't looked into it, really. Maybe next year! :) Willow 14:01, 7 February 2007 (UTC)

What looks like bare legs at first glance is tan-colored hose. Trust me on this. - PKM 20:47, 26 February 2007 (UTC)

[edit] Eclectic tastes!

I am really enjoying reading your stuff - you seem like the sort of person who would enjoy reading A Museum for Young Gentlemen and Ladies - ca. 1799. I spent a chunk of last year typing it in, and it is in process of being processed by Project Gutenberg. If my annotations annoy you, the PG version has no annotations - it is at [3]. Jpaulm 14:28, 13 February 2007 (UTC)

Hi Paul! :D

I'm glad that you enjoy my edits. :) That's one of the best pleasures of Wikipedia, don't you think — well-meaning, sensible people showing each other beautiful and interesting things that they've distilled into pithy (Pythic?) phrases. I'm especially honored and a little bashful, once I saw all of your accomplishments! Your little book is delightful as well, and has such nice wordings; the word "lest" is woefully underused in modern English. I didn't mind your annotations at all; on the contrary, I went looking for them. ;) The Etruscan one made me laugh out loud, which is bad because I'm writing from my local library. Of course, some of the text is rather benighted, even for its time; but I don't think you need to worry about the variant spellings and usages (e.g., "compleat" and "the mob were") since, as far as I know, English wasn't fully standardized back then. The title is a little misleading, though; the text seems targeted mainly at young gentlemen, and not at young ladies, wouldn't you say? Perhaps that reflects the editorial predilections of its author.

My own tastes aren't eclectic, at least in its original sense; rather, they're annoyingly catholic. I can't help but be hypnotized by practically everything and everyone I meet and I love learning new things and new people. An unintentional magic, too, causes me to get too much credit; people seem to see me as a reflection of themselves, and assume that I know what they know. Religious people see me as a devout believer of their faith; scientists credit me with knowing their fields; knitters see me as a friendly, patient Zen mistress of stitches; among farmers, I'm the hard-working girl-next-door who's good with animals; and so on. It's really strange to be given undeserved credit but protests and explanations don't help; and since I like being with people and being liked by people, I don't ask too many questions about how the magic works.

Thanks again very much for your kind note and for brightening my little out-of-the-way corner of Wikipedia, Willow 19:36, 14 February 2007 (UTC)

[edit] 3RR on Wikipedia:Wikipedia is failing

I've blocked you for 24 hours for your violation of the three-revert rule on Wikipedia:Wikipedia is failing:

1. [4]
2. [5]
3. [6]
4. [7]
5. [8]

During your absence, please reflect on the wisdom of edit-warring over a page so prominently linked to by outside sources. Kirill Lokshin 18:19, 15 February 2007 (UTC)

I accept the punishment, but I think this is somewhat unfair, since I was being nice to User:Worldtraveller by not reporting him for WP:3RR and trying to reach a reasonable compromise. None of his arguments were deleted, although they were softened somewhat by pertinent hard data and referenced arguments that he seemed unwilling to consider. I honestly did try to keep the spirit of his essay and even left my more positive assessment of Wikipedia (which was referenced and based on accepted standard criteria) in the Wikipedia:Wikipedia is not failing essay. Willow 18:24, 15 February 2007 (UTC)

I blocked him as well, actually. Kirill Lokshin 18:27, 15 February 2007 (UTC)

Willow, sorry to see that you have been blocked. I totally think this is unfair as you were truly being nice. MetsFan76 18:28, 15 February 2007 (UTC)

I support Kirill Lokshin's view that this was a valid three-revert rule block. You and Worldtraveller were engaged in an edit war and clearly violated policy. That said, I do feel that Worldtraveller started this edit war by deleting valid arguements and corrections which you placed in the essay. When you return from your block, I will support any attempt by you to edit the essay along the lines you proposed.--Alabamaboy 19:15, 15 February 2007 (UTC)

In fact, I've reverted the essay to your last version. You raised valid issues and points which shouldn't have been deleted.--Alabamaboy 19:18, 15 February 2007 (UTC)

Alabamaboy, there was a good debate going on and imposing a block just effectively ended it for the time being. I can't see how supporting a block is helping the editing of an article. MetsFan76 19:20, 15 February 2007 (UTC)

Please do not argue over my blocking; I accept my punishment and hope that no more of your valuble time will be spent discussing it. Instead, it would be better spent improving both articles, WP:WIF and WP:WINF; particularly the latter could use a good copy editing and overall integration. If both articles are to be kept (which I understand the reasons for), then it also might be useful to provide Template:See also links connecting each argument (i.e., section) of one essay with its rebuttal in the other essay. That should truly help to foster a good debate and clear exposition of the arguments, which will hopefully allow us to understand the issue in a larger and more objective light. Willow 19:31, 15 February 2007 (UTC)

PS. I think that an impartial reader will agree that the essay still reflects Worldtraveller's opinion that Wikipedia is failing. I do not think it is vandalism to point out the limitations of his data or of the conclusions that can be drawn from them. An essay in WP space is owned collectively, and all Wikipedians have a collective interest in reaching well-reasoned, NPOV conclusions; in particular, our interests are not furthered by treating poorly reasoned arguments of any editor, including me, as inviolable. Of course Worldtraveller has the right to defend his work, but I'm disappointed that he did not rebut my criticisms of his reasoning, or improve his reasoning, or engage with the subject matter at all; instead he chose the easy path of reverting. That is unworthy of a well-intentioned Wikipedian.

The edits were made in a good-faith effort to make the reasoning and arguments more rigorous and also to fix some grammatical errors (e.g., "that" vs. "which"). Calling it "vandalism" is uncalled for and silly. Willow 21:54, 15 February 2007 (UTC)

PPS. It's surprising to me that people imagine that I changed the sense of Worldtraveller's essay. His conclusion and all of his arguments/data were perfectly intact, although weakened somewhat for having their limitations pointed out. It seems to me, however, that any fair-minded person would prefer to base their argument on sound data and reasoning, and would be grateful for being alerted to any logical lacunae. It's not POV or NPOV, but just a reasonable request for honest reasoning. Instead, an angry Worldtraveller talks of me "emasculating" the article — which is ironic and funny, because I always wanted to attend Bryn Mawr. ;) Trying to stay cheery and hoping for a more tranquil future, Willow 02:08, 16 February 2007 (UTC)

Ah well, this is all a tempest in a teapot. I personally didn't care either way about the arguements--what bothered me is that your edits were deleted without cause, which isn't done here. I also agree with you that your edits didn't take away from Worldtraveller's essay; instead, they made the essay more reasoned and correct in why Wikipedia might be failing. Still, the issue is now resolved. Since the essay is now in Worldtraveller's user space, he can do with it as he pleases. Best, --Alabamaboy 02:37, 16 February 2007 (UTC)

I've unblocked you. Since the essay is now in userspace these blocks no longer serve much of a purpose and you are clearly an editor in good standing. JoshuaZ 04:06, 16 February 2007 (UTC)

[edit] Ideas for a new bot

I've never programmed a bot before, but the recent debate at WP:WIF and WP:WINF has suggested to me that we should make an automated bot that provides putative GA and FA assessments of articles that can be confirmed later by human intervention. Of course, a bot can't assess subjective qualities such as writing style, but it could assess quantitative properties to check whether an unassessed article matches the statistical profile of certified FA-class or GA-class articles. Please add your thoughts here!

Potential quantitative properties:

• Overall length of the article
• Presence of an info-box
• Presence of a navigational template
• Number of categories
• Number of sections
• Does it have a "history" section?
• Distribution of section lengths
• Distribution of paragraph lengths
• Distribution of sentence lengths
• Number of distinct references
• Number of distinct PMID or ISBN references
• Fraction of references with blue authorlinks
• Number of distinct images
• Length of captions
• Number of blue wiki-links within caption
• Dates of the references
• Number of wiki-links
• Fraction of wiki-links that are blue
• Maximum number of redundant wiki-links within a paragraph
• Stability of the article (say, 2-week running average of # bytes changed per day)
• Disqualifying features, such as the number of {{fact}} tags

One approach might be as follows:

1. We create a template that would go on the Talk page of every article, a template that stores the article's statistics (like those above), perhaps with a default "Hide" option so that it doesn't swamp the Talk page.

2. We write a separate bot for each type of statistic, a bot that would update the article-statistics template on the Talk page.

3. We generate statistical profiles from all the FA-class and GA-class articles and confirm that they are distinguishable from those of random articles. Ideally, we would do some sort of K-means clustering analysis on the statistical profiles.

4. Finally, we write a meta-bot to scan all of Wikipedia's articles and tag those that match the statistical profile of FA-class and GA-class articles as putatively FA-class or putatively GA-class, respectively. By placing such articles in a special category, we could expedite their evaluation by human Wikipedians and, thus, produce better articles faster.

5. In principle, the article-statistics bots could also show how a given article's statistics stack up relative to the FA-class, maybe with color coding (red=long way off, blue=at FA standard). Such a color coding might help Wikipedians decide what features of the article need the most immediate improvement.

Thanks in advance for your insights and suggestions! :) Willow 22:23, 15 February 2007 (UTC)

I like this. I do not have any brilliant ideas to add, but I think this is moving in the right direction.--Filll 23:50, 15 February 2007 (UTC)

My first thought here is that, while the idea is good, a tag for every talk page is unlikely to happen anytime soon. That only human-tagged articles could be assessed by the system may not be such a limitation, though, because articles that no one tags are disproportionately likely to be deleted or deletable. While it's clearly a 'never know till you try' situation, I'd predict that neither FAs nor GAs would cluster, though current FAs (promoted after citation requirement inflation) will, if the reference assessments are weighted. Clearly FAs will be distinguishable from stubs, but I don't know the utility of a system of bots and meta-bots to do that. Hm, some things to think about... Opabinia regalis 05:20, 16 February 2007 (UTC)

[edit] Note to O. regalis

Hi, I really hope you arrive soon; it was a rather depressing waste of my day off, both for me and methinks for User:Worldtraveller, whose essay I tried to improve. Would you be so good as to thank User:Tameeria for tagging all those MCB articles? It's amazing work. Willow 02:23, 16 February 2007 (UTC)

On a lighter note, this comic is always fun and sometimes uncannily accurate. ;) Willow 04:06, 16 February 2007 (UTC)

Sorry, I seem to be the one showing up at the party after everyone else has left. I dropped in to make a quick fix to RNAi earlier but have spent most of the night tracking down a weird bug in code I wrote months ago and never got around to documenting. (Note to self: don't do that.) It's really too bad this went as far as it did. Good to see you're unblocked though; I don't know why that was a good idea. On the bright side, I was somewhat surprised at the overall tone of the responses on slashdot; I expected them to be much more critical. (Or maybe I just filtered out all the 'Wikipedia sucks because they deleted my article on Vanishingly Small Linux!' posts ;)

As a suggestion - maybe move the Kister comparison material to its own essay, since it's not part of the direct response to Worldtraveller's? I'm not fully caught up yet, but the current arrangement of material seems very ad hoc. WP:WIF being userfied and WP:WINF not doesn't seem right either. Opabinia regalis 05:20, 16 February 2007 (UTC)

Hi, I'm working late, too; it's great to hear from you, though. :) I agree that WP:WINF seems inappropriately placed and oddly reactive. There are published standard criteria for evaluating encyclopedias; hopefully, we can measure Wikipedia fairly by those criteria and usefully assess its areas of strength and weakness. After my experience today, though, I'm really shy about becoming involved in the debate again; I'm going to have to overcome my fears. Basically, I'm afraid that people will edit irrationally from rancor, with no respect for cited references or logic, to say nothing of Talking, and I'll end up blocked again. :( I understand Worldtraveller's perspective, but it seems silly to feel "emasculated" because someone makes your publicly-posted arguments more referenced and more rational. Hopefully, he'll forgive my temerity and I'll forgive him for accusing me of "vandalism". So — do you think a new, neutral article is a good idea? Inquiring minds are inquiring. ;) Willow 06:26, 16 February 2007 (UTC)

I have to say that although both WP:FAIL and WP:WINF make some good points, I do not think either adequately describes the problems, strength and promise of WP. --Filll 14:43, 16 February 2007 (UTC)

Nice to hear from you, Filll! I'm inclined to agree, although I've started an essay with larger scope and perhaps more foolish ambitions. It'll be slow piecemeal work for me, since my various jobs are keeping me very busy, but hopefully other editors will warm to the challenge and give the essay wings. Gotta run soon, Willow 14:49, 16 February 2007 (UTC)

[edit] Hi, if you have a minute please read my essay.

Hi, Willow. I so loved your even-handed edits to WP:WIF that I would like to ask you to read and perhaps edit Wikipedia:A suggested improvement 0001 which I have created. Thanks if you are able to. Heathhunnicutt 03:12, 16 February 2007 (UTC)

Thanks Heath, your idea seems great! It'd be even nicer if all the PMID journal articles could be downloaded and rendered consistently as a bibliographic reference, without requiring any human typing, e.g., {{refmake|PMID=12033732}}. Is that possible? Sorry, I'm really sleepy and can't think straight now. Talk to you tomorrow, Willow 07:42, 16 February 2007 (UTC)

Well, if I understand your idea correctly, we can't do this locally, but this bookmarklet (more here)—which incorporates David Iberri's template filler—greatly reduces typing requirements ;) I don't know if you've heard of it before, but I've been using it on and off for a while now, and it works very well. Fvasconcellos 14:52, 16 February 2007 (UTC)

Thanks so much, F; your visit is a wonderful serendipity for my Talk page. :) I'll check it out when I have a free moment; right now, I have to dash off to work. :( I definitely need to learn more about the programming tools available to us here, although I'm also worried that it shouldn't distract me from writing articles. Willow 14:58, 16 February 2007 (UTC)

[edit] Essays and such

That's great that another project's getting off the ground! (Though I don't exactly have any insights about the topic :) FWIW, Worldtraveller knew what he was getting into when writing that essay; I'm surprised he didn't anticipate that the essay would be edited and seemed so frustrated by it. I suppose you feel 'in the limelight' when your essay's on slashdot, though.

Nice new essay, glad it has a home now. (I made a very minor addition to it, not very productive.) I haven't read Kister's study, but I keep wondering when I see it cited; are the criteria it lays out the 'standard' in encyclopediology (such as it is), or were the criteria defined for the purposes of that study? If the latter, I wonder whether it makes sense to describe criteria defined in one study from 1994 as 'the accepted/standard/common/etc' criteria. (For example, I can imagine a 2007 repeat of the study including multimedia availability as a criterion.) Opabinia regalis 04:18, 17 February 2007 (UTC)

No, I think they're standard; that's why I gave the other two references. I suppose that I ought to clarify that somewhere, though; perhaps at encyclopedia?

Thanks very much for your kind words and contribution to the essay; that's exactly the sort of thing that it needs! :) I'll try to brood some more over RNAi so that I can say something sensible before Supporting it. ;) Ta for now, Willow 13:21, 17 February 2007 (UTC)

I'm still a bit surprised that no one has updated those criteria more recently, but I suppose evaluating encyclopedias isn't something that happens all that often, either. What do you think of the current lead for RNAi? Oh, and also - thanks for your excellent co-nom; I imagine we'll have to give Tim his cabal robes before long ;) Opabinia regalis 23:31, 18 February 2007 (UTC)

It is going well, no? I'm very glad to see it, although it's not surprising, given his shining history. I had to laugh at my own metaphor, below; the idea of a watery worm and a weeping willow "singing" underwater amused me to the utter limits of sweet silliness. :D

I'm curious about the cabal initiation ritual, although I suppose you'd have to kill me if I ever found out. Does it involve tests of bravery such as drinking noxious potions or taking embarrassing photographs of D. Brandt in his underwear? I probably shouldn't make jokes about that. ;)

I had a few new ideas; can I try them out on you first, so that I don't embarrass myself elsewhere? ;) The first was to start a WikiProject Unassessed Articles to allow all articles to be rated for importance and quality before they get their own WikiProject. It was hard to get our Textile Arts WikiProject off the ground, and we couldn't assess our articles before then. I'm surprised that not even all FA's have a WikiProject home.

The second idea is even more crazy. What if we make an automated bot that would monitor only he FA's and GA's and automatically revert "obvious" vandalism such as page blanking, section blanking or obscenities and warn the malfeasers automatically? That might spare Tim a lot of work. We could even make an indvidual config file for every FA to look out for particularly common types of vandalism for that article. Some of those statistical profile measures might come in handy for this purpose. If one can describe and characterize vandalism statistically, one might be able to filter it effectively. Maybe a nice AI student project? Just a thought for help make "stable" versions without forfeiting progress, Willow 23:56, 18 February 2007 (UTC)

Mm, well, all I can tell you is that there are gummy worms involved. I make no claims as to the origins of any photos of notorious trolls in my possession.

I haven't given much thought to the assessment scheme after grousing about it when we got the MCB assessments going, so I'm not sure what the standard is - is the 1.0 project doing its own assessments or just trusting the wikiproject ones? (I frequently see articles tagged as both MCB and Chem where one project's rating is outdated, and I know that different projects use very different standards for A ratings, so it makes a difference to their quality distribution results.) I wonder if the 'unclaimed' articles are that way because they really don't have a project corresponding to their subject, or if they just haven't been stumbled across by someone who bothers to tag them? I rarely stop to tag my own articles (though I guess I should start). That would determine whether it's more useful to set up an 'assessment home' for loose articles, or a system for sorting loose articles into the projects they ought to belong to.

A very simple form of your bot idea was proposed at some point, as a separate instance of Antivandalbot to watch just the main-page FA. I don't remember the sequence of events, but I think it was obsoleted by the cascading protection feature. I wish the the {{maintained}} template, or some similar system, were more common and widely accepted (though I admit I've never bothered to use it, so I guess I shouldn't complain ;) Then the dedicated bot could leave a message on the maintainer's talk page if a 'questionable edit' occurred that wasn't obvious vandalism. I suspect that configuring it for individual articles is more trouble than it's worth, other than turning off obscenity detection for certain subjects (but I guess the existing bots don't have too many false positives on that front). I think 'edit creep' is a much bigger problem for long-term article degradation than vandalism, which is usually reverted quickly. Opabinia regalis 01:25, 19 February 2007 (UTC)

Hi O, I set up a tentative proposal for the new WikiProject on unassessed articles. We'll see if people are interested in helping with that. Oops, I discover that it's already been started by others! Willow 02:16, 19 February 2007 (UTC)

[edit] New WikiProject

Hey there, Willow!

I'm very excited by the new WikiProject you started, and am ready to buckle down and get to work on the Textile Arts articles (in fact, I've already started to do so on the Knitting article). However, not being an experienced Wikipedian, this is my first WikiProject, and I don't fully understand the whole process about it. I've read some of the pages on it in the Help namespace, but I'm still pretty confused. Could you explain just how WikiProjects work? Where can I find a list of all the articles that are currently under WikiProject Textile Arts, for example? And what's the deal with the importance and quality scales?

Much appreciated!

DroEsperanto 16:43, 17 February 2007 (UTC)

Hey Dro,

I'm writing from work, which I shouldn't do, so this will have to be brief. Basically, you should make articles as always, but now you can tag their Talk pages with something like {{WikiProject Textile Arts|class=Start|importance=Mid}} to make the article part of the WikiProject. Follow the links on WP:TA (e.g., the 2nd paragraph in the left panel) and maybe it'll be clear; the various classes and importance levels are explained at Category:Textile Arts articles by quality and its wiki-links. You can also ask some of our fellow members there; they're very nice and helpful! :) Willow 00:01, 18 February 2007 (UTC)

I'm also a bit new to Wikiprojects as well, but I'm excited to learn. Like I stated on the wikiproject page, I'd probably be only good to make crochet related articles. I really don't know that much about other textile arts. Thanks for starting the project!--Alizera 03:44, 19 February 2007 (UTC)

[edit] Thanks

I will try to stretch myself to fit the outline you drew on my admin nomination page. I hope I will succeed. TimVickers 05:59, 18 February 2007 (UTC)

Well, that confirms the "noble" part. ;) Although a measured self-knowledge does you credit, you should also have faith that others' assessments are sincere and apt, even when they're positive. Honestly, we didn't sing your praises to make you blush. ;) Willow 12:11, 18 February 2007 (UTC)

[edit] Help! TA template

Willow, I added a comment to History of quilting using the comment link on the Textile Arts assessment template, and it categorized the article as "MCB articles with comments" - I am not sure how to fix that, can you do it please? - PKM 17:53, 18 February 2007 (UTC)

Hey Paula Kate, I think it might be fixed; please let me know if it's not. Great work, by the way! :) Willow 20:55, 18 February 2007 (UTC)

Looks good now, thanks. - pk (PKM 02:22, 19 February 2007 (UTC))

[edit] WikiProject Unassessed Articles

Consistent with your ideas, I'm thinking of starting a WikiProject Unassessed Articles that would allow us to rank articles for importance and quality, even if they don't have a sponsoring WikiProject of their own at present. Does that seem like a good idea to the people here? Hoping for constructive feedback, Willow 00:34, 19 February 2007 (UTC)

Hi Willow. I do like your WikiProject idea, in principle. :) WP certainly doesn't have enough "data" on the overall quality of its articles, with respect to the "encyclopedic value" of each, to gauge its progress. However, I don't believe that the current article assessment approach is able to accurately capture "the state of the art-icles", even if it were more widely applied—such as by your proposed WikiProject. If Wikipedia doesn't have a down-and-dirty assessment system that does not require layers of administration for two/three out of five/six of its assessment grades, we'll never be able to apply it widely enough to arrive at the statistical picture that we may be seeking. Recognizing relatively good articles must not be left to a jury, if we're every going to "recognize" them all. I'm also concerned (and I wrote about much of this on the WP:FAIL talk page) that a single rating per article cannot capture the diverse qualities of articles on an open project like Wikipedia. So, I'd be the first to join your wikiproject if we proposed a new rating system along with it. Methinks that would be a little controversial. :) –Outriggr § 07:58, 19 February 2007 (UTC)

Hi Outriggr,

Thanks for coming to my page! :) I too have been thinking about the inadequacies and subjectiveness of the present evaluation scheme, and have been thinking of making a bot to assess articles according to quantifiable criteria; scroll up a bit and you'll see what I mean. I think that the qualitative assessments such as "B-class" could and should still be done by humans, but quantitative assessments might be useful, orthogonal and complementary. The quantitative assessments would be stored in a separate template and "hidden" by default. What do you think of this idea?

Unfortunately, I have effectively zero experience with Python and Pywikipedia, so don't expect too much too fast. :( However, perhaps we could recruit a good but bored bot writer to help out? Just brainstorming, Willow 13:32, 19 February 2007 (UTC)

Hi Willow, and thank you for the friendly reply. I read about your idea above, and it does pique my interest. My beliefs about article rating include the notion that the rating should not be "owned" by a wikiproject (wiki as a whole points to "WP:OWN", so I don't know why I should be in the minority with that opinion). I mention that because I would like to see this hypothetical bot review human-assessed articles as well--the point being that, whatever metric is chosen, the system has the entire article set to sample from. I would like to think about this proposal some more, now that I see where you're coming from! You are, of course, much more versed in statistics than I, but I love "data", and would find it an interesting experiment to see how quantitative measurements of articles predict their quality. (Being a perfectionist, I think it's too bad that this bot-stuff can't be implemented inside the software, especially for a computationally involved set of statistics such as you're proposing. Querying the data, once generated, would also ideally reside inside the software. Talk-page tags as metadata make me sad!) Given that you were pointed to an already-proposed project when you submitted your proposal, are you still considering this idea yourself? Regards, –Outriggr § 02:52, 21 February 2007 (UTC)

[edit] MCB worklist question

There's been some questions here about the MCB worklist updates that Clockwork had been doing for a while, because the list has gotten out of date after tameeria's tagging spree. I swear I remember someone else having the tool CS used to do it - but it's not me or Tim; do have it (or know how it was done), or am I imagining things? Opabinia regalis 17:56, 19 February 2007 (UTC)

I'm afraid that I don't know. All I did was set up the initial MathBot stuff. Is Clockwork on wikibreak at the moment? If need be, I could try to figure it out, but I'm probably not the best qualified person for that job. Willow 18:51, 19 February 2007 (UTC)

I guess I am imagining things then. It seems like it would probably be a pretty easy pywikipedia job to replace if necessary. Opabinia regalis 03:01, 20 February 2007 (UTC)

[edit] You've been busy

I see you managed to get yourself blocked, among other things... oh well. Just thought I'd drop by and mention that I noticed Wikipedia:Evaluating Wikipedia as an encyclopedia, and I think it will be useful. It didn't quite fit in Wikipedia but as a stand-alone essay it's fine. (There seems to have been a lot of drama regarding essays recently; I assume you've heard about Wikipedia:Wikipedia is failing, Wikipedia:Wikipedia is not failing and the associated edit warring) – Qxz 20:44, 20 February 2007 (UTC)

Hi Qxz, it's really nice to hear from you! :)

Yes, the essay episode was a big misunderstanding on my part. I disliked the idea of adversarial essays, each telling half the story, and wrongly assumed that no one would mind if I clarified the reasoning of Wikipedia:Wikipedia is failing. Being blocked was surprising and painful for me, but I think it was more so for Worldtraveller, since he had to spend the next four days in sometimes angry disputes. I tried to distract that energy into analyzing Wikipedia collaboratively and neutrally, but people seem to prefer the wrangling. :( Since the consensus was that there had to be twin adversarial essays, I went ahead and wrote Wikipedia:Wikipedia is not failing yesterday. On a happier note, peace has broken out between Worldtraveller and myself, which makes me very happy.

On another happier note, several of us succeeded in starting a new Wikiproject, which has long been a cherished wish of mine; I really wanted to assess all those knitting articles! I'm also adding a few extras in preparation for submitting Encyclopædia Britannica for FA; maybe you could take a preview? I'd very much appreciate your advice. (I'll fix the {{fact}} tags; please ignore those.) Talk to you soon, and thank you very much for visiting, Willow 21:04, 20 February 2007 (UTC)

I'll take a look at that article when I have some time. Yes, I noticed Wikipedia is not failing; I agree with many of those sentiments. Much as I'd love to add something constructive to the debate, sadly, I'm good at asking questions but useless at answering them. I'm also a little worried that the question of whether or not the project is failing is becoming a little polarized. I refuse to sit and do nothing, though, so I wrote Wikipedia:Wikipedia may or may not be failing instead – Qxz 21:55, 20 February 2007 (UTC)

Hilarious! I love it. Laughing at ourselves is the best way to shake free of the wrangling rancor — and, umm, poetic primness — and get back to writing articles. Simply brilliant. If you start a Clueless Wikipedians club, I'd like to join as "cleverly clueless Chloe". ;) Kudos for ludos, Willow 22:14, 20 February 2007 (UTC)

I fear membership numbers would quickly spiral out of control... also past events seem to suggest "clubs" don't go down too well :) On a project the size of Wikipedia, there's so much going on that everyone's inevitably clueless in all but a few small areas. For example, I've read the image copyright pages several times over and I still don't understand them. And as for the Arbitration Committee, I think the only course of action is to avoid doing anything wrong and getting sent there in the first place, because I have yet to figure out how it works, only that it takes everything very seriously and is somewhat scary.

One other thing: I notice you mentioned Wikipedia:WikiProject Missing encyclopedic articles in your essay; you may wish to know that I just did a mass-update of all the subpages of Wikipedia:List of encyclopedia topics, so the figures given in the table on that page are correct (ignoring any articles created in the last day or so). It was somewhat out-of-date before, and nearly 6,000 articles that were on that list have been created in the last 6-8 months; there are now 25,000 remaining out of the original 78,777 – Qxz 23:06, 20 February 2007 (UTC)

To honor your essay, and to make myself useful, I indigo'd four missing articles, three redirects and one stub on David Mermin. I see a few others to bluify, but I have to run. Ta and here's hoping that others take your questionable essay in the right spirit, Willow 23:34, 20 February 2007 (UTC)

Thanks! – Qxz 23:42, 20 February 2007 (UTC)

[edit] Return visit

Hi Geometry Girl ;) - I thought I would stop by your user page to return the compliment you paid by visiting mine. I have to say, I am very impressed by your dedication to wikipedia, not to mention the depth and breadth of your contributions - you certainly deserve those awards! The articles on whether wikipedia is failing or not were interesting, but I'm glad to see you have returned to the front line. With your energy, it seems that soon that all articles beginning with the letter 'D' will be completely sorted out! What letter do you plan to tackle next ;) ? Geometry guy 22:02, 21 February 2007 (UTC)

Hi Geometry guy, I'm so delighted that you came for a visit, and I'm also very glad that you sorted out that D-dimensional exposition thing. :) I know, today was a fun game! :) We'll see if my Wiki-brothers rise to the challenge; I hope they won't be able to resist their competitive urges. ;) I was almost tempted to propose a "guys vs. gals" battle to wake them up, but I was afraid of going a little too far. I still believe that Opabinia regalis and I could take on all of them, especially if we made Tim Vickers an honorary member of the gentler sex — and could convince him to play along with something so silly. ;)

I'm so glad that you're here; you're inspiring me to get back to my orthogonal coordinates articles and prettify them with some nice color pictures. Geometry is so hypnotizingly beautiful; I drink it in just as a cat laps milk. On the other hand, I'm easily distracted as you see, so don't be surprised if I wander off into knitting land or somewhere else tomorrow... Thanks for your friendly note and your kind forbearance with my silliness, Willow 22:26, 21 February 2007 (UTC)

If I can coax a cat back to colourful coordinates I can be completely content :) Geometry guy 22:47, 21 February 2007 (UTC)

Cleverly clear of cloying clarifications! Clipped closing, Clueless Clogging Chloe :D 22:56, 21 February 2007 (UTC)

[edit] Beverage production

I noticed in your essay you noted we don't have an article on beverage production, and I thought that needed to be fixed! I'm currently working on a draft in userspace until it's ready to go into main, if you have anything to add, please do. Seraphimblade ^{Talk to me Please review me!} 09:48, 22 February 2007 (UTC)

Hi Seraphimblade,

Thank you for doing that! I've added a few sundry notes; feel free to change/delete whatever you like! :) Willow 12:10, 22 February 2007 (UTC)

PS. Your name is wonderfully evocative; it makes me think of the blade of Manjushri, separating truth from not-truth. I'll admit, though, that I prefer to think of the seraphim more as singers than as warriors. <:O <-alto-wannabe

Thanks for the suggestions, and actually you nailed the name dead on! I used to enjoy playing an RPG called In Nomine (and occasionally still do), in which the seraphim are angels whose specialty is discerning truth from lies. They also are pretty capable warriors in that particular game, so I derived "Seraphimblade" from that. Seraphimblade ^{Talk to me Please review me!} 12:18, 22 February 2007 (UTC)

[edit] Well, well...

Your name was dropped in this week's Signpost! How's fame treating you? :) Fvasconcellos 13:37, 22 February 2007 (UTC)

Blush Yes, I was embarrassingly clueless on several fronts and ended up in stocks. :( But all's well that ends well; Worldtraveller and I are positively friendly these days, although I can't seem to break him out of his blue funk... :(

There've also been some happy developments unrelated to the Essay Wars. We started the WikiProject Textile Arts, I'm gradually filling in the gaps on the Encyclopædia Britannica articles and, on a lark, I took a little holiday yesterday starting sixteen missing encyclopedic articles that started with "D". Hoping all's well with you, too, Willow 17:53, 22 February 2007 (UTC)

I'm fine, thank you! I noticed WP:TA—I created its member category (guess you forgot ;) Ah yes, I wish I had the guts to create the odd article instead of just reverting childish vandalism and fixing those "yptos". Happy editing, Fvasconcellos 01:23, 23 February 2007 (UTC)

Well, let me convince you to be bold! I'm probably one of the most sheepish people you'll ever meet in real life (in many ways), but here nothing truly bad can happen; unfold your wings! :) It might help to start with something you truly love, no matter how ignoble or small; lavish its article with your enthusiasm and others will see its beauty through your eyes. Joy in the morning and joy in the evening, Willow 01:49, 23 February 2007 (UTC)

[edit] This may be helpful

I've just come across something which may or may not have already been thought of, but it may be helpful in the never-ending quest to turn red links into blue ones, specifically ones that have incorrect capitalization. User:Qxz/Redlinks/1 contains some redlinks from Wikipedia:Requested articles, sorted alphabetically. It ocurred to me that there is a quick way to check an entire page for unnecessary redlinks caused by incorrect capitalization; simply put {{lc: and }} around it. I did this and saved it to User:Qxz/Redlinks/1L, and lo and behold, a couple of blue links appear – compare the two. This might save a bit of time when confronted with a list of articles to create :) – Qxz 02:11, 23 February 2007 (UTC)

Thanks, Qxz, I'll remember that! :) How did you make your list; it was funny to see Bicep written as BiCEP. ;) Is it something automated? Curious and curiouser, Willow 02:16, 23 February 2007 (UTC)

BiCEP is probably an acronym for something and shouldn't be redirected. It's helpful for things like Quantum Biology instead of Quantum biology – trivial, but difficult to pick out from a sea of redness. I've written a script that collects up all the red links from a specified set of pages, weeds out duplicates and sorts them. I might try it out on some other lists when I'm finished with these ones – Qxz 02:20, 23 February 2007 (UTC)

Interestingly, many of the requests actually have the right capitalization, but an article already exists with the wrong capitalization – very annoying! Anyway, between this, removing blue links, and getting rid of duplicate requests (for some reason people seem to add requests without checking if they're already there), I'll have removed about 800 links today, all without writing a single article! I'll leave that bit for you to do :) – Qxz 02:51, 23 February 2007 (UTC)

You rock! 800 articles is simply amazing. How do you run scripts on Wikipedia? Is there a page where I can read up on it? Still more curious, Willow 03:33, 23 February 2007 (UTC)

I'm not running scripts *on* Wikipedia as such. Though you can do this; each user gets a page here where they can add JavaScript code, which can be used to add functionality to the interface, though you have to have a web browser with JavaScript enabled in order for this to work. (See for example Wikipedia:Tools/Navigation popups, which I highly recommend). What I'm actually doing in this case is just copying the wiki-text of each page into a text file and then parsing it. I should point out that most of those 800 articles were already there, it's just nobody had removed the links :) – Qxz 08:52, 23 February 2007 (UTC)

[edit] Ice cubes that fly

To answer your question - the easiest way is just to do rigid least-squares fitting of the system to itself. A better method uses the Eckart-Sayvetz conditions (we don't have an article, but there's an example for a vibrating molecule here), though I don't know how this is implemented numerically. (IIRC the current version of GROMACS does something related but simpler.) It's actually a very bad idea to remove rotation from most simulations (explicit solvent with periodic boundary conditions - hm, no article on that either?) - fortunately those are less vulnerable to ice-cubism ;) Rotational motion is really only removed during the simulation for gas-phase simulations, but it's almost always a good idea to remove translation. Of course you remove both after the simulation so that you can look at the trajectory without getting dizzy ;) Opabinia regalis 02:58, 23 February 2007 (UTC)

OK, be nice to me and help me understand. :) The Eckart-Sayvetz conditions were reasonably straightforward for a vibrating system but shouldn't they be modified for a system with no set equilibrium position? Also, the "rigid least-squares fitting" method for removing net angular momentum was dark and mysterious like a deep pool...hmmm, maybe it's hopeless for me. :P But instead of removing the spontaneously generated angular momentum, maybe you could take it as a measure of how accurate your simulation is? Befuddled Willow 03:31, 23 February 2007 (UTC)

The least squares fit is just optimizing a transformation matrix - if you write it as a separate rotation matrix and translation vector, it's minimizing

where i=1..N particles and a and b are the two structures. It's part of the class of Procrustean problems. So you just move the system at time t to the coordinates that are the output of a fit to the system at time t-τ, and scale the velocity vectors to match. (Done properly, this transformation preserves magnitude, so it doesn't induce more artifacts.) This is actually a rather bad method for solving the rotation problem mid-dynamics, but it was used in the original GROMACS rotational motion correction, and variants of it are pretty much universal in post-processing the data. I think what people do with the Eckart method is just to define an arbitrary orientation as 'equilibrium', since you shouldn't be trying to apply the correction to anything but a single molecule in vacuum. But I'm not really sure how this is implemented, or whether it's even used for reasonably complex molecules; it's very rare that you'd put anything the size of a protein in vacuum anyway, not least because it makes such a mess with charged sidechains. GROMACS is open source, so if you were really curious what a production removal method looks like, it's in there somewhere...

The flying ice cube is really a very weird pathological case; you essentially would never encounter it in an explicit-solvent simulation where you weren't trying to elicit it. (There are so many other possible sources of numerical error!) If you don't remove the rotational motion in a vacuum simulation at constant temperature, some artifacts will occur, though they won't necessarily be as weird-looking as the rigid-body flying ice cube. So accumulating angular momentum wouldn't really be a useful measure of accuracy because some 'extra' kinetic energy will accrue to those low- or zero-frequency modes in a vacuum simulation, and the whole trajectory (or most of it) is useless if that happens to an appreciable extent. In general, the common measure of simulation accuracy is energy drift (oh my, that's a redlink too?), which measures how well the integrator is conserving energy over time. If it's drifting a lot, that usually implies that the simulation parameters are poorly chosen; if your system itself is unstable, or the parameters really terrible, the energy will rapidly increase until the program crashes, hopefully with a cute error message like 'Your molecules are exploding' ;) Opabinia regalis 06:13, 23 February 2007 (UTC)

Wow... you actually understand all this stuff? I'm in awe – Qxz 08:54, 23 February 2007 (UTC)

Oh, and I added Energy drift and Periodic boundary conditions to requested articles, except I don't know where they should go, so I had to stick them at the bottom. It's a start, anyway – Qxz 08:57, 23 February 2007 (UTC)

Wait, I think I see my misunderstanding; please write back soon and tell me if if I'm on the right track! :) I thought that you were removing angular momentum spontaneously generated in the system as a whole (e.g., because the step size was too big), whereas I bet that you're talking about removing it just from a subsystem, e.g., from a protein in a bath of water, so that you can watch it on the screen without getting dizzy. ;) Speaking just for myself, I wouldn't dare to remove angular momentum from a subsystem; aside from violating the equipartition of energy, it would be an embarrassingly affected molecular mannerism, no? ;)

This makes me think that I've misunderstood flying ice cube as well. :( My understanding had been that, due to a overly large step size or somesuch, linear momentum was not being conserved in a closed system, and that the system's internal energy was gradually being converted into kinetic energy as the simulation progressed, cooling it off. But maybe you mean a subsystem is developing linear momentum, as again would be natural from equipartition, and that removing it by changing frames of reference would be bad somehow? Hmm, I need to read your article again; a flower blossomed right away for me when first I read it, but maybe it was the wrong flower?

Umm, really dumb question: are total linear and angular momentum conserved in molecular dynamics? I would suppose that they should be, and that deviations from their conservation might be useful as "canaries in the mine" probes of simulation accuracy, but that might be impractical for some reason that I don't understand.

Maybe I'll throw in what tidbits I've heard into the redlinks; please fill out and correct as needed! :) With many thanks and much affection for your patient explanations for the woolly-headed, Willow 12:01, 23 February 2007 (UTC)

I think I muddled things above. There's two cases where you'd want to remove rotational motion: one is after the simulation is over, so you can look at the results without the distraction of uninteresting motions. The other is during a vacuum simulation, with the same frequency with which translational motion is removed. They aren't really related except that the same method can be used to do both calculations. It's almost always a bad idea to remove rotational motion during the simulation from a subsystem of a molecule in water - almost because there's probably some reason out there, though I can't think of one - and almost always a good idea to remove it from a single molecule or isolated cluster of molecules in vacuum. The flying ice cube artifact isn't actually dependent on the timestep, except in the sense that rescaling happens at some multiple of the timestep and the rate of rescaling influences which motions, other than zero-frequency global motions, accrue excess energy. I was originally going to write out a derivation in the article, but decided not to because it would basically be republishing the meat of the paper. (On the other hand, it's not really so exciting anymore either, so maybe that's not such a problem.)

Linear momentum is (modulo numerical artifacts) conserved under periodic boundary conditions, but is usually zero; the center of mass velocity is set to zero in initializing the system velocities. (The temperature is then calculated over the remaining 3N-3 degrees of freedom.) Angular momentum is actually not conserved under PBC, because the system isn't rotationally invariant. If you want a copy of the original ice cube paper and can't get to it, I'll email it to you - but be warned, it's badly formed somehow and uses some sort of font that acrobat reader can never seem to find. Fortunately xpdf/gvv/etc. have no problem.

Oh hey, nice blue links! (And a pretty talk page.) Will expand upon them when I get a chance. Opabinia regalis 05:11, 24 February 2007 (UTC)

Ok, I think I understand that; thanks also for fixing up the little energy drift article! I definitely should've clued in about angular momentum not being conserved in PBC's, because my own example in periodic boundary conditions doesn't conserve it. ;) But I'm still curious about a situation that is likely too mundane for real scientists: constant-energy, constant-volume simulations of a droplet of molecules in empty space. Are total linear and angular momentum guaranteed to be conserved then, or do little numerical errors add up eventually? Maybe there's some correction system that people use? I can see how you'd correct for linear momentum (change the reference frame), but I'm mystified how you'd do it for rotation. if you could mail me the article, that's be fun; it's not at my library. Just curious, Willow 01:24, 26 February 2007 (UTC)

P.S. Tim's an admin, in case you haven't heard! Another successful chapter in MCB's bid for Wikipedian hegemony. ;D

OK, paper sent. Yes, plain NVE (contant particle number, volume, and energy) should conserve momentum; if you're not doing any velocity rescaling, the flying ice cube artifact shouldn't occur. However, without temperature coupling, energy drift in NVE is particularly visible because the temperature will drift along with it. I'm not sure if that 'extra' kinetic energy necessarily partitions 'correctly', but if not, the effect is definitely not very visible. The annoying thing with energy drift is that it can take quite a while to become obvious, but still be outside the acceptable range for your simulation, so you could be days into a calculation before it becomes clear that the parameters are non-optimal. This is, of course, especially true if you want to simulate over very long time scales. There isn't really a good correction that can be used for these artifacts due to the chaoticness of the trajectories - although it proceeds from the same idea that very minor numerical foibles may give you the 'wrong' trajectory for your starting conditions, but still a trajectory that's a reasonable approximation of the desired ensemble. The values of 'minor' and 'reasonable' depend on the context. Opabinia regalis 07:12, 26 February 2007 (UTC)

Thanks for the paper; now I really see just how patient you were with me! It's very interesting — I love mechanics :) — and it does seem like a pretty valid criticism of that instantaneous velocity rescaling approach, no? I'm a little surprised that such a basic problem is still unsolved in MD. A couple of alternative solutions suggested themselves; could I float them past you, like little toy ducks ready for their shooting?

• Couldn't you defeat energy drift by minimizing the energy in the 6N-dimensional phase space every so often back to its proper value? No energy drift = no problem, no?

• Maybe you could introduce a constant or very slowly changing non-physical damping to compensate for the non-physically increasing energy? Admittedly, this approach seems ad hoc, although you could maybe tune the damping constant on the fly?

• If you really don't want to change the atom positions but only their velocities, perhaps you could define other global system velocity variables to rescale? For an isolated system, maybe you could rescale each normal mode independently, similar to the Eckart-Sayvetz conditions you mentioned; each mode acts like its own independently oscillating molecule, right? That would decouple the zero-frequency and high-frequency modes. For more complicated systems like the PBC's, perhaps you could rescale the different Fourier components of the velocity field independently? I have this intuitive idea that that should be good, since the spatial averaging done in calculating the Fourier modes should be roughly akin to temporal averaging they want to do in the article. But these are really random thoughts before I've had any coffee; forgive me if they sound crazy — I'm probably still dreaming. :(

Even if I say silly things, I hope you'll be patient still; it's fun for me to chat with you and wander well-guided in a world that's so far from my mundane everyday life. Thanks muchly once again, Willow 12:36, 26 February 2007 (UTC)

Umm, I looked a little more online today, and I realized that I am totally out of my depth. Sorry for trying your patience. :( It seems that I'm conflating two problems: the energy drift problem and the problem of carrying out a constant-temperature MD simulation. I guess ideas #1 & #2 might pertain to the energy drift problem, whereas #3 might pertain to the thermostatting problem. But they seem pretty lame to me now, even though they made so much sense in my dreams... ;) I should definitely stick to knit one, purl one! :D Willow 21:36, 26 February 2007 (UTC)

Heh, no, I think you should definitely stick to more mechanics articles ;) The thing about energy drift is that it's more of a symptom than an inherent problem; it's rare to simulate so long that it really does become an issue. There's a few people who kind of make it their religion - it's a bit like the footnotes people at FAC; it's an easy way to sound like a hard-headed defender of quality without having to do much thinking - but realistically, it's very context-dependent. Even in NVE, for a given set of parameters, the difference in dynamics between the mean temperature of 300K at the beginning of the simulation and 302K at the end is essentially nonexistent. The problem is that (a significant amount of) energy drift implies that the parameter choices are bad in some way, and that may affect the dynamics. So just fixing the energy doesn't really fix the underlying problem.

The flying ice cube thing is actually not an unsolved problem - I think the article is a bit clearer about that now - there are other thermostat schemes that don't introduce that artifact. The Berendsen theromstat and barostat are not really well-regarded in terms of their thermodynamic properties, but they're faster than the alternatives, and in most practical cases the artifacts they introduce are minimal, so they're still commonly used. (Some current implementations might use a short-time average for scaling; I've never really checked, come to think of it.) Apparently we have articles for exactly zero of the common algorithms for these things; if I ever get an infusion of patience, I'll have to create them if I don't get beaten to it. For what it's worth, doing anything with normal modes mid-run would be, if nothing else, prohibitively slow; updating the covariance matrix and running the eigensolver is in the range of seconds to minutes for decent-sized systems.

On an unrelated note, I see Tim has been bringing down the hammer on the poopyheads of the world ;) Opabinia regalis 05:41, 27 February 2007 (UTC)

Tim is ongoing marvel for me, but I'm also constantly surprised and disappointed that people get their thrills from vandalizing Wikipedia. I'd like to believe that the attraction is not destroying other people's work or uglifying the world, but rather rejoicing in absurdities — I know the appeal of that! ;) Perhaps it's also powerless people relishing the power to change at least one thing in the world as they see fit and having those changes recorded for posterity in a database. Oh, yeay. :p

Hmm, you think I should work on mechanics articles? I was thinking of fleshing out virial theorem when I get a chance, but I've still a lot left to do: the EB, knitting, nature,... No rest for the wicked, and absolutely no rest for the absolutely wicked. ;)

Your explanations really helped, yet again — thanks! I never realized how complicated it could be to numerically integrate Newton's equations of motion, and how many "tricks of the trade" there were. No wonder people go to grad school!

A few blue links for you to discover, affectionate ta-ta, Willow 23:00, 27 February 2007 (UTC)

Eh, I think the garden-variety vandals are mostly bored kids who can't think of anything more intelligent to do than write 'penis' and 'gay' on every possible surface. If they weren't scribbling on Wikipedia articles, they'd be doing it on their desks, their classmates' books, and the door to the men's room. (Hurrrr, she said 'doing it'!) Three-year-olds throw toys on the floor, and thirteen-year-olds vandalize, but I suspect it's mostly the same impulse to make a mess and see if other people will clean it up.

Wow, virial theorem really needs some text! EB is looking nice; I'm looking forward to its FAC. Very nice blue links, by the way ;) When I get a little more time, I really have to get back to all the Alzheimer's-associated proteins - those have been red way too long. Oh, and I still have some new stuff for constraint algorithm that I keep getting distracted from, too.

I can't remember which paper it was in now, but there's an MD review out there somewhere that starts with (something like) 'MD shouldn't be a state-of-the-art method, and it really shoudn't work at all'. What kind of bright idea was it to use classical mechanics on atoms anyway? ;) Opabinia regalis 06:03, 28 February 2007 (UTC)

[edit] Minor thiols in little-known organisms

I took no offense! TimVickers 18:26, 23 February 2007 (UTC)

Yeay! nice words are their own kind of antioxidant. :) Eagerly anticipating the next FA, Willow 18:31, 23 February 2007 (UTC)

Wow, your talk page is now deeply funky. Thanks for your help and support in the RfA, it is much appreciated. If I can help you in any way, please don't hesitate to get in touch. TimVickers 16:59, 25 February 2007 (UTC)

[edit] Oops! Slight mistake!

Slight mistake on the Encyclopædia Britannica article, if you read the web source you would have realized that it was a legit viewpoint and not random vandalism. I think the problem was the use of the word "vandalism" in the article, which rightfully triggered some alarms. I'm going to put it back and then mention it in the article's discussion page for open discussion.

Happy editing!

128.61.36.21 03:38, 26 February 2007 (UTC)

Hi 128.61.36.21, I replied over there. Thanks for writing! :) Willow 13:22, 26 February 2007 (UTC)

[edit] Thanks

Just a note to say thanks for your optimistic essay about Wikipedia. I think it's amazing what we already have, after so few years. I've only been on Wikipedia since June, and so this lark is still great (unlike you, I don't know what crack is) fun. I can't believe the way people get all depressed and angry on here. Sheeesh—to use an American expression. Anyway, you seem to have the right idea, judging by your pagy bit. Middlemarch is my favourite novel, too: in some ways I wish it wasn't, because it led me to vainly read the rest of Eliot in search of something as good (if only I could have back the time I wasted labouring through Daniel flippin blasted Deronda). Thanks again for the sustained burst of analytical optimism. qp10qp 23:33, 27 February 2007 (UTC)

Dear qp10qp, thank you for your most excellent note, which cheered me up a lot. I know that my friends like me, but sometimes I get down on myself; that's when notes like yours come to the rescue! :) Your name is tantalizing; is there a story behind it?

I like what you wrote about "analytical optimism" very much. I've often had the experience of looking at something and finding it first beautiful, then ugly, then beautiful again as I come to appreciate it in the fuller light of all its qualities, both good and ill. I think it's important not to shut your eyes at the "ugly" stage, lest you miss out on the final "beautiful" stage. :) I do understand the vexation and disappointment of people like Worldtraveller, though; perhaps it's just that we have different expectations of the world and different ideas of what should be sacrificed for a good end.

I also know what you mean about George Eliot's other works; for me, they're respectable stories (a little moralizing, perhaps), but they don't transcend their era in the way that Middlemarch does. Of course, that's a rather high standard to hold authors to, especially since they have to make a living by appealing to their readers. ;) I've never tackled D.D. but the ones I've read all have little gems of painted personalities.

I haven't written about him on my user page, but I'm also a huge fan of Anton Chekhov's short stories; I love the way he sketches people, with a doctor's keen eye for their stranger-than-fiction mannerisms. I find his stories generally elegiac and beautiful, which often suits my mood perfectly. Willow 13:15, 28 February 2007 (UTC)

P.S. I've lived a rather sheltered life, so I've no experience with crack; but I wanted a little racy something that got across the rush of the feeling as though I'm making a delightful and surprising present for other people. It's one of the best pleasures of life, and something that's relatively easy to do here. :)

You are quite an extraordinary person WillowW. I'm not surprised you have so many warm friends here. Does your username have a story behind it? Geometry guy 21:14, 28 February 2007 (UTC)

Thank you – what a glowing day! I'm sure to have good dreams tonight. :)

Sadly, there's no cute story around my name. My joining was rather impulsive, as with many things I do, so it never occurred to me that I should invent some cool handle. In retrospect, though, that's probably a good thing; I probably would've chosen something really silly. Luckily, CatherineMunro was one of the first people I looked at here (thanks to her most encouraging essay and poem) and, like her, wanted to use my own name. Besides, I don't mind "Willow"; it's a little unusual, and has good connotations of gentle strength. But you can call me "Geometry gal"; I really like that name, too! :) Willow 21:43, 28 February 2007 (UTC)

"Gentle strength" seems to summarize you very well from what I have seen! So is "Willow" then your real name, with "W" your surname? I hope you don't mind if I stick to "Geometry girl" (old fashioned Englishman that I am), but from your contribs, a better name would surely be "Everything girl" (not to be confused with the wonderful music of EBTG of course). Geometry guy 22:34, 28 February 2007 (UTC)

That's almost correct; my last name isn't "W", and I can't be held responsible for any other "W"'s in the world. ;) But "Geometry girl" would be wonderful; thank you! It'd be a constant reminder of some of the most beautiful things. Maybe I could even emulate Prince and choose some nice geometrical construction as my signature here, such as Soddy's hexlet or a tetrakaidekahedron; then people could call me "the editor formerly known as Willow"? Umm, maybe not. ;D Cheery cheers, dashing off to another job, Willow 22:58, 28 February 2007 (UTC)

P.S. Why those two? I'm rather inverted and "of imagination all compact". ;)

Hello Geometry girl. I may be biassed, as a differential geometer, but for me "geometry" and "everything" are almost the same thing, as geometry is (for me) more a way of looking at the world, rather than a branch of science. A geometer asks "what is the most natural way to consider this question?". And it seems to me that you strive to look at the world in quite a geometrical way (in this sense). Anyway, I shall think of you as Willow (The Geometry Girl) henceforth and hope that others follow suit. (But what happened to the clueless Chloe and her cloying clarifications?) Geometry guy 23:19, 28 February 2007 (UTC)