# User talk:MarkSweep/archive2

## Image:Spoon-lure.png

Hi. You tagged this for cleanup. I've replaced it in the article (Spoon lure) with Image:SpoonLureWithCrosssection.png, which I drew from scratch. It's still kinda crude, but at least it's symmetrical (and more accurate from what I remember--it's been a long time since I've gone fishing). Is that the level of improvement you had in mind, or should I mark it for further cleanup? Niteowlneils 22:09, 13 Feb 2005 (UTC)

Thanks. I was hoping it could be redrawn in a vector graphics program like Inkscape, Ipe, or good old XFig. With transparent background and antialiasing. I've put up a request over at Wikipedia:Image recreation requests. --MarkSweep 01:55, 14 Feb 2005 (UTC)
OK. At least it's a better (IMHO, of course) placeholder until a more sophisticated one is developed. Anyway, it was only the third image I'd editted with any graphics program, and the first I started from scratch, so it was good practice. First was Image:SanJoseMapWithLAFCOandCityLabelsandCA.png--you can see what I started with at the top of the Stick Map section; second was removing the background clutter of Image:MeerkatAtHappyHollow.jpg and uploading to commons Commons:Image:MeerkatAtHappyHollow.jpg. How about Image:Hammersmith and city line.png on commons to replace the one on en:? Most everything but Plaistow seems legible now. I think to get any better it would have to be changed to a vertical orientation, which would also have the advantage of fitting nicely in the right margin instead of leaving so much white space across the entire page. I could do it, but certainly not tonight--I'm going to turn into a pumpkin soon. Niteowlneils 04:45, 14 Feb 2005 (UTC)
I agree that your replacement is a great improvement over the original, I didn't mean to dismiss your contribution. The schematic drawing may be OK, because drawing a reflective surface by hand is extremely difficult. However, there are people here who are good with ray-tracers (see e.g. Image:Dodecahedron.jpg); a ray-traced image of a spoon lure would make a fantastic illustration. The other line drawings you mentioned are also improvements. They can be improved further, by the use of anti-aliasing. What I sometimes do in Gimp is to start with a design size that's an integer multiple of the intended maximal final size. So for the subway map, I would start with an image 4500 pixels wide, use a readable vector font for the labels, and save a private copy of the design in a lossless format. Then I would blur it very slightly and scale it down to 1500 pixels, using the best available scaling method (Gimp's default of linear interpolation should be changed to cubic interpolation), save it as PNG if it's a drawing, compress it further with pngcrush, and upload the resulting PNG file. For an example, compare the various revisions of Image:Logarithmic spiral.png. For function plots, I've automated this (see the script at Image:Gamma distribution.png). --MarkSweep 05:51, 14 Feb 2005 (UTC)

## Commons:Taxonomy

Hey there, I started a Taxonomy page on the commons to discuss the animal/plant gallery handling there. I thought you might be interested, so I left a note for you here. :-) --Conti| 17:24, Feb 14, 2005 (UTC)

## Spelled/spelt etc.

Hi. I just thought that I'd say (as a cross between an apology and an excuse) that it's often very difficult for a British-English speaker to tell the difference between genuine U.S. English and mistakes that are commonly made by U.S. writers (and, I've no doubt, the same applies the other way round). I mean, the busines of colour/color, centre/center, lift/elevator, etc. — that's easy enough. But (as I've just been discussing with another user) there's a huge grey area. For example, I'm sometimes told that the use of 'disinterested' to mean 'uninterested' or 'alternate' to mean 'alternative' is genuine U.S. English — but then other people, such as my U.S. colleagues, assure me that it isn't, it's just that a lot of Americans make those mistakes. Do So (doh!) when I come across what I'd normally think of as a spelling or grammar mistake, I have to try to work out what its status. And sometimes I get it wrong. I had no idea until someone made the same point on another article that 'spelled' was U.S. English (I corrected it automatically, as one would correct 'kneeled' to 'knelt'). I won't do it again... Yours was a neat solution, though. Mel Etitis (Μελ Ετητης) 22:12, 14 Feb 2005 (UTC)

Pardon me for butting in, but as a British-English speaker who's had a lot more exposure to written American than to written British English, and who is alternately amused and appalled by the concern over alleged American adulteration of "the language" that is expressed by otherwise sane British people, I have an interest. But I'm disinterested. Where was I? Right, the question was of "disinterested" and the like. Perhaps the simplest solution is to get a couple of good, recent dictionaries, one British, one American, making sure that these have not been influenced by soi-disant "language mavens". (If I remember right, some "American Heritage" dictionary has little comments by "experts". Avoid!) There's a splendid little book titled American Tongue and Cheek that can be picked up used for peanuts and that destroys a lot of the myths propagated by the "mavens". As for "disinterested", its use to mean something rather like "bored" is old in Britain as well as the US; I think this is rather a pity as "uninterested" does this well, and it's convenient to have to single word that unambiguously means "not gaining personally from any decision on". But there's no point making an issue out of this: I only use "disinterested" with the latter meaning, but I don't touch others' use of it with the former meaning. Sorry to go off at a tangent, but for the rare well-informed view on issues (or non-issues) of "good English", see Geoff Pullum: this (on singular "they") is a superb appetizer. -- Hoary 22:48, 2005 Feb 14 (UTC)
Hi Mel, I like your nick, very clever. Hey Hoary, thanks for stopping by! Yes, the two Geoffs, Pullum and Nunberg, are among the few voices of sanity in this whole mess.
I don't know of any good recent American dictionaries that would be comparable to the Collins Cobuild English Dictionary in terms of being thoroughly grounded in reality. My own view on matters of orthography is that in many cases it is possible to simply avoid using words and expressions that would be somewhat puzzling to a sizable group of people. Of course it's not always possible to find good workarounds: I wouldn't advocate the exclusive use of "hue" instead of colo(u)r, or "dissect" instead of "analy[zs]e", "unconcerned"/"unbiased" instead of "disinterested", etc. It would be interesting to see if an article could be completely rewritten to avoid all of these forms while still sounding natural. --MarkSweep 02:26, 15 Feb 2005 (UTC)
(I'm having this discussion in parallel, prompted by exactly the same issue in both cases — purely coincidentally.) Interestingly I'd thought that 'analyze' (like 'compromize' or 'surprize') was a misspelling on both sides of the Atlantic (it's a back-formation from 'analysis', after all), so it's another example of a correction I'd have made automatically, not thinking that it was a U.S./U.K. matter. (Is 'disect'–'dissect' another, or was that a typo?)
That was a genuine typo, now corrected.
I once received an e-mail (nothing to do with Wikipedia) from someone who was angrily protesting at my comment that philosophy should be done disinterestedly and dispassionately...
I see. "No, philosophy should be done passionately…"
The problem with work-arounds is that either they (as with the 'hue' example) use literary words to replace everyday words, or (as with 'dissect' for 'analyse', or 'unconcerned' and 'disintersted') they're not fully synonymous. Indeed, English is unusual among European languages in having very few genuine synonyms, which is part of the reason that (again, unusually for a European language) elegant variation is frowned upon.
I don't think that's true. I doubt that French has more true synonyms than English for example.
Finally, with regard to Hoary's point, the fact that a usage can be found in older English isn't enough to demonstrate correctness — otherwise we'd have to accept people using 'silly' to mean 'holy', and 'nunnery' to mean 'brothel'... Mel Etitis (Μελ Ετητης) 09:03, 15 Feb 2005 (UTC)
Regarding "analyze", you at least have to admit that the AE convention results in less ambiguity. In BE orthography, "analyses" is ambiguous, and can either correspond to AE "analyses" (plural of "analysis") or to AE "analyzes" (3rd person singular form of "to analyze").
I think Hoary's point was only that "disinterested" has a long history of its own and can traditionally mean two different things. It appears that one of those meanings is gradually falling out of use and is no longer understood by a growing number of people. --MarkSweep 09:26, 15 Feb 2005 (UTC)
But if it can be shown that older usages have continued (despite complaints to the contrary), this might be an additional argument for correctness. -- Hoary 09:11, 2005 Feb 15 (UTC)
Of course — but then you'd have evidence of recent accepted usage too, and wouldn't need to appeal to the older usage. Having said that, there are many other problems, not least being dialect. (I remember the first time I used the word 'mardy' after moving to London at the age of eighteen; I'd no idea that it wasn't standard English. I could have looked it up in a dictionary, of course, but children don't generally look up words that they know (unless they're rude words, of course; my image of the French was very peculiar for a long time after reading in an old French-English dictionary an illustrative use of the verb 'pisser': 'pisser au feu').
When alternative forms exist side by side, they're often not geographical, though those who use one form or the other sometimes think that they are. And, of course, people often simply don't know that there are alternative spelings (on one page I've had one anonymous user repeatedly changing my use of 'artefact' to 'artifact', insisting (though not in so many words) that I'm illiterate).
Oh, by the way, 'mardy' isn't easily translated into standard English; someone who's mardy is a sulky spoilsport — at least, in Lincolnshire. Mel Etitis (Μελ Ετητης) 09:22, 15 Feb 2005 (UTC)

## Template:TAL episode

(posted to User_talk:Leif) Hi Leif, I noticed that you reverted the TAL episode template to re-enable the external links. I don't disagree with that revert at all, I'd just like to give you a quick run-down of what happened and why they were disabled. Recently, list of This American Life episodes was listed on VfD. The decision was to keep it, but enough people felt that the article was no good with the external links enabled. I personally don't see how that comes into play (as I see it, if this episode list is deemed encyclopedic, adding information such as external links can only enrich the article, not render it unencyclopedic). In order to address some of the concerns expressed during the VfD, I refactored the episode list to use the newly created TAL episode template and disabled the external links. I don't know what's going to happen next, maybe someone will nominate the episode list again for deletion. Anyway, just a quick heads up, no further action required for now. Cheers, --MarkSweep 03:35, 16 Feb 2005 (UTC)

I was actually the person who nominated the article for VfD. I didn't see anyone who said the article would be worth keeping if only the links were disabled and instructions for linking were added to the article. Saying that the article should be deleted "because it is just a list of links" is not the same as saying the article should be kept but the links should be disabled. Now that there are a few details about some episodes, there is more reason to keep the list (and of course, the majority of keep votes is reason enough). Disabling the links was completely pointless and I don't think it made the article any more keepable in anybody's view. ~leifHELO 03:49, Feb 16, 2005 (UTC)

## PRC, ROC, mainland China, Taiwan, etc.

Hello MarkSweep. Thank you for joining the discussion over the titles of China-related topics articles. Would you mind help explain to the contributors who opposed renaming because they thought the new titles are confusing, that how these terms differ from each other, and how the choice of one of these terms as a title is important. Thanks in advance. — Instantnood 18:04, Feb 19 2005 (UTC)

I think that this whole controversy is indicative of a general problem of insufficient separation between policy and conventions on the one hand, and the implementation of these policies on the other. Based on the naming conventions, it is perfectly clear what each article should be called. However, some people oppose the moves because they are not familiar with the naming conventions, or because they disagree with them. This leads to pointless discussions when the main objective is to implement the policies. I'm not suggesting that the policies should be set in stone, clearly those should be up to debate. But we can't keep reopening the same discussion every time a change mandated by a policy is proposed. In the cases at hand, the question should simply be "according to the current naming conventions, what are the most appropriate names for these articles?". Personal opinion beyond the scope of these conventions should play no role, so if someone disagrees with the move, they should participate in the discussion of the naming conventions. In any case, a clear distinction between setting policy and implementing existing policies should be drawn, so that we don't have to revisit the same issues over and over. --MarkSweep 02:48, 20 Feb 2005 (UTC)

In fact some who opposed the moves have already proceeded to discuss the naming conventions itself. Curps has even edited the naming conventions, although being reverted.

It has become not only a matter of naming conventions, but a choice between familiarity and accuracy. Those who opposed at WP:RM probably also oppose the current naming conventions, and not follow it when they create new pages.

Thanks again for the helpful inputs. :-) — Instantnood 09:45, Feb 20 2005 (UTC)

It should be simple: every article should have an accurate title. If familiarity is an issue, it can be addressed by adding a redirect. (But in some cases this won't work: history of Taiwan is obviously different from history of the Republic of China, and neither can be simply redirected to the other.) In related news, there was a recent discussion on Talk:Elizabeth II of the United Kingdom#Suggested_page_titles of a similar nature, where someone suggested renaming that article into something "simpler", even though the result would have been less accurate and would have been inconsistent with the naming conventions for monarchs. I think all of this indicates that to the extent that naming convnetions and editorial policies are desirable, there should be a centralized discussion and separate implementation of those policies without having to go through the arguments for the conventions again each time they are implemented. --MarkSweep 09:56, 20 Feb 2005 (UTC)

What makes the situation more complicated is that everybody knows the current Queen of the UK (or "the Queen") is "Elizabeth II of the United Kingdom" with little objection, which is not the case for "Republic of China", a name many people have never heard about. The current discussion and vote can perhaps be the foundation for later moves, avoiding lengthy discussion again each time. — Instantnood 10:16, Feb 20 2005 (UTC)

And interestingly the results so far for X in the People's Republic of China → X in China and Politics of Taiwan → Politics of the Republic of China contradicts. — Instantnood 10:35, Feb 20 2005 (UTC)

Hello MarkSweep. The vote and discussion at Wikipedia:Requested moves#Economy of Taiwan → Economy of the Republic of China is getting messy. — Instantnood 12:30 Feb 28 2005 (UTC)

## Image deletion

Greetings! Over at the Commons, some images I've uploaded are being considered for deletion. I'd be interested in hearing your honest opinion on the matter there. Thanks! – Quadell (talk) (sleuth) 13:07, Feb 21, 2005 (UTC)

## RfC

Hello there. I am recently being listed on RfC. Feel free to comment as you wish to. I regard it as a way out and to have the matter settled. Thanks. — Instantnood 18:07 Mar 1 2005 (UTC)

Thank you very much MarkSweep. I am still a newbie to Wikipedia, and I did not expect some of the actions would have catched fire in this way. There are still a lot to do to build Wikipedia a better encyclopedia.
If you don't mind and are interested, please also take a look at Talk:Hong Kong, Talk:Victoria City, Talk:Cathay Pacific destinations, User talk:Huaiwei, Talk:List of city listings by country (and its archive), Talk:List of cities by country, Talk:List of countries that only border one other country, Talk:List of roads and highways and Talk:List of cities in China, and the edit histories of the articles. I am also discussing with Grutness and trying to resolve the stub sorting trouble at Wikipedia:WikiProject Stub sorting/Criteria#The China-Taiwan mess.
Once again thanks for your help. Seems like I am always coming to you and ask for comments. :-P — Instantnood 00:54 Mar 2 2005 (UTC)

The sharing at RfC seems to be over. I have made a response there. Please take a look. I do hope that with everyone's effort Wikipedia will soon be the best encyclopedia ever. :-D — Instantnood 21:23 Mar 5 2005 (UTC)

## Deleting Image:Cumulative normal distribution.png

I put Image:Cumulative normal distribution.png up for WP:IFD since I obsoleted it with Image:Normal distribtion cdf.png. Cburnett 03:52, 10 Mar 2005 (UTC)

In that case I'll have to recreate your image, because it's not antialiased. And BTW, you can use the "norm" function in gnuplot. --MarkSweep 03:56, 10 Mar 2005 (UTC)
OK, awesome. I'm trying to use your gnuplot to redo the expo distribution plots but I can't get it converted to png. I used the script on Image:Cumulative normal distribution.png to no avail (I get a segfault on pnmrotate!). Is that the same script you're using? Cburnett 05:17, 10 Mar 2005 (UTC)
Actually, what I've been doing lately is the following:
1. Create PS output with gnuplot (the main reason is that you can use the Postscript Symbol font and it has some support for subscripts and superscripts).
2. Open the PS output in Gimp at high resolution (3000 by 4000 pixels or larger). (Update: I've been using 600 dpi recently, which results in images of about 4500 by 6000 pixels.)
3. Blur the image slightly. (Gaussian blur, 5 pixel radius.)
4. Dilate.
5. Autocrop.
6. Scale down to 960 by 720. 1300 by 975
There must be is a way to automate that and to run Gimp in batch mode, but I haven't played with it yet. I don't use the old script anymore, because it leaves too much space around the figures. I could gradually (re-)create all of these plots in the same style (and perhaps upload them to the Commons instead), starting this weekend. --MarkSweep 06:58, 10 Mar 2005 (UTC)

## "Mainland China" in titles

Hello. I have proposed at Wikipedia talk:Naming conventions (Chinese) to change the title of some articles and categories. Would you be interested to join the discussion and say something? — Instantnood 20:26, Mar 11, 2005 (UTC)

Hello again MarkSweep. Can you come and say something? Thanks in advance. — Instantnood 15:01, Mar 14, 2005 (UTC)

## Template:deletebecauseoncommons

Agreed, there are some issues here. I only used it for images I uploaded myself, as well as a few with no history that I had no qualms about uploading, all {{trainweb}} images. --SPUI (talk) 22:12, 15 Mar 2005 (UTC)

## Gnuplot

I really like Image:Normal distribution cdf.png. Where can I learn to do that? This is the best I could figure out: Image:Dba plot.png - Omegatron 01:14, Mar 16, 2005 (UTC)

The trick is oversampling: use Gnuplot/Gimp to create an image several times larger than the size that you intend to upload, optionally blur it slightly, then scale it down to the desired size (which may have to be scaled down further for use in articles). I would strongly recommend using the PostScript output ("terminal") setting of gnuplot, since it has the most features (plus you can edit the generated PostScript code if necessary), then open the .ps file in Gimp at a high resolution (I use 500dpi or at least 3000 by 4000 pixels for the image dimensions). Some of the steps for the PDF/CDF plots are described a few entries further up on this talk page. I can plot the dba function for you, but I would like to do it directly in gnuplot, if possible. My understanding of maxima is limited, as I've only started using it recently: so H is (the real part?) of a (complex?) rational function. But what does db() do? I can't find it in the maxima documentation. Could you write down the definition of H in Mathematica notation? --MarkSweep 03:43, 16 Mar 2005 (UTC)
oops. db() is my own function... i'll have to fix that. - Omegatron 05:46, Mar 16, 2005 (UTC)
Is db(x) = 20 log10(x)? --MarkSweep 08:50, 16 Mar 2005 (UTC)

Does the following gnuplot code work?

kA = 7.397057228624818216e9
H(s) = kA*s**4 / ((s+129.4)**2 * (s+676.7) * (s+4636) * (s+76655)**2)
dba(f) = 20 * log10(abs(H(2 * {0.0,1.0} * pi * f)))

set grid
set logscale x 10
set xlabel "Frequency (Hz)"
set ylabel "Gain (dB)"

plot [10:39999] (x>2e4)? 0/0 : dba(x) lt 3 notitle


If it works, it's fairly easy to generate smooth PNG output via PostScript. --MarkSweep 09:19, 16 Mar 2005 (UTC)

Excellent fix! Thanks! - Omegatron 16:51, Mar 17, 2005 (UTC)

## Proposed reorganization of Taiwan

Someone proposed something at Talk:Taiwan#Proposal_for_New_Organization_.283.2F19.2F2005.29.2C_Please_Comment, if you would take a look and comment....--Jiang 01:10, 21 Mar 2005 (UTC)

## The proposed resolution to the RFC arising from the Sollog page

In essence I agree with the resolution - especially since my posts elsewhere are now being trashed on the RFC pages. But surely, just as people tracked down the Sollog sockpuppets, so Wikipedians will track down my IP and start all over again? The Number 01:57, 22 Mar 2005 (UTC)

I replied on your talk page. --MarkSweep 00:55, 24 Mar 2005 (UTC)

## Patent template

Template:US patent is a neat idea, thanks. — Matt Crypto 10:10, 22 Mar 2005 (UTC)

I'm glad you like it. --MarkSweep 00:55, 24 Mar 2005 (UTC)

## Cauchy Distribution graphic

Ha! Another picture person. Excellent job on the Cauchy distribution graphics. PAR 00:45, 24 Mar 2005 (UTC)

Cburnett and I have started to systematically (re-)generate all existing distribution plots and upload them to the Commons. I might eventually get to the Lévy distribution too, but have been putting it off because it requires writing C code. --MarkSweep 00:55, 24 Mar 2005 (UTC)
Can I ask you to keep putting it off? I would like to try to generate them, while trying to conform to the look that you have on the ones you generated. I'll play around with it, trying to get the look, and let you know when I upload something to Levy, then let me know what you think. Also, I didn't generate those point-by-point, I just took the FFT of the discretized characteristic function, which makes things simpler. PAR 01:33, 24 Mar 2005 (UTC)
Sure, nothing easier than putting things off. ;-) --MarkSweep 01:51, 24 Mar 2005 (UTC)

Hi - can you tell me what font you are using for your standardized plots? PAR 03:39, 26 Mar 2005 (UTC)

It's Times (standard PostScript font), because it matches the PostScript Symbol font. You can find the gnuplot source code for the plots e.g. at Commons:Image:Cauchy distribution pdf.png. It's best to use the enhanced PostScript backend of gnuplot, because it's the most flexible. Once I have the .ps file, I open it in Gimp. The remaining steps in Gimp are described above on this talk page. I'd be happy to help out, if necessary. Cheers, --MarkSweep 04:57, 26 Mar 2005 (UTC)

I finished the Levy distribution plot, but haven't had time to do the infobox. Could you check it for consistency with your plots before I start churning these out for a few more distributions? Thanks - PAR 07:21, 3 Apr 2005 (UTC)

Sorry, didn't see this in time. Your Levy plot is excellent: smooth, without appearing fuzzy or blurred. --MarkSweep 02:05, 11 Apr 2005 (UTC)

Hello MarkSweep. Despite the ongoing discussion at Wikipedia talk:Naming conventions (Chinese) over the use of the terms "mainland China" and "People's Republic of China", SchmuckyTheCat and Huaiwei have listed category:Cities in mainland China, category:Companies of mainland China and category:Laws of mainland China onto Wikipedia:Categories for deletion.

Your vote is now essential and vital for the survival of these categories, and to avoid attempts to bar the Wikipedia:Naming conventions (Chinese) from truly enforced. — Instantnood 20:44, Mar 24, 2005 (UTC)

Update: there's also a poll at Wikipedia talk:Naming conventions (Chinese). — Instantnood 01:02, Mar 25, 2005 (UTC)

Thanks MarkSweep. — Instantnood 02:27, Mar 25, 2005 (UTC)

## Arbitration

Hello MarkSweep. I've got listed onto requests for arbitration by SchmuckyTheCat. Could you have a look? Thanks. — Instantnood 04:29, Mar 29, 2005 (UTC)

## Re: Template:Isd

Hi MarkSweep,

Thanks for your message, Template:Isd has been restored. My apologies for deleting it in error. I just started out in the cleaning department. :)

- Mailer Diablo 22:50, 2 Apr 2005 (UTC)

No problem. Thanks for restoring it. --MarkSweep 20:50, 3 Apr 2005 (UTC)

## Thanks Mark for helping on the "Junkyard" Article!

Hey dude, you do a great job, please can you help on the Junkyard (disamiwhateveration) and Junkyard (band) formatting/configuration, so it looks acceptable, Thanks again, TJ (this stands for "Terrible Joe")

## Unverified orphans

This is regarding your comments at Wikipedia talk:Images and media for deletion/Unverified orphans#So what do we do about those UOs? Request for comments.. Please note that there are many UOs that satisfy one of more of the following:

1. They weren't orphaned to begin with and were removed from articles by people paranoid about copyrights.
2. They were uploaded before mid-2004 which is when people decided to include non-GFDL non-PD non-fair-use images. Long ago everything was GFDL or PD (we had to agree to GFDL-redistributability while uploading) and there was no requirement to tag, then fair-use came to be allowed and then came allowing other licenses and image-tagging.
3. Their uploaders have since left the project and hence can't be contacted for updating the image with licensing info.
4. They might be redundant images in an article, but may be useful as a commons resource on the subject.
5. Orphans don't load the image server since nearly nobody sees them.

Please see my messages here, here, here, here, and here and the meagre response I was able to get.

If you feel I am not a trusted person enough to be bothered about, you could see Jamesday's opinion on Wikipedia talk:Copyright violations on history pages and Anthere's opinion here, here and here and Sj's opinion here.

As you could see from my messages, I would like a proper place where many people discuss the current image deletion policy, but I don't know what to do. Thanks for any help. -- Paddu 08:53, 3 Apr 2005 (UTC)

Following the long discussion at Wikipedia talk:Naming conventions (Chinese) regarding proper titling of ROC-/Taiwan-related topics, polls for each single case has now been started here. Please come and join the discussion, and cast your vote. Thanks. — Instantnood 06:15, Apr 6, 2005 (UTC)

## binomial (disambiguation)

Hello.

I think links to binomial (disambiguation) should usually be in a notice at the top of the article, saying

For other topics using the name "binomial", see binomial (disambiguation).

The "see also" list at the bottom should be for related topics, not for unrelated topics that happen to use the same word. Also, people who find the long article titled "bishop" when looking for an article about the chess pieces called "bishops" are not likely to scroll all the way to the bottom and find the notice about "other uses". Michael Hardy 20:43, 7 Apr 2005 (UTC)

Good point. I see you already fixed the relevant cases. I was merely zapping double redirects without thinking. --MarkSweep 21:17, 7 Apr 2005 (UTC)

## Plea for help

Hi Mark,

sorry for asking you directly but I have again a color problem and I have to admit that I did not managed to fix this with Gimp. I tried to follow the advises you gave me, but it did not become better - I guess I need some more experience. If you have time and desire to fix this you might have a look at Featured_pictures_candidates/Image:EuropeanParliament.jpg. Tillea 21:14, 7 Apr 2005 (UTC)

## RfA thanks

Thank you for the vote at Wikipedia:Requests for adminship/Henrygb2. It has made my week. --Henrygb 01:57, 8 Apr 2005 (UTC)

You're welcome. Congratulations! --MarkSweep 23:43, 8 Apr 2005 (UTC)

## Help with inserting image from commons in wikipedia

Hi... I hope it isn't highly inappropriate to ask you personally. I uploaded an image to the Wikimedia commons, Commons:Image:Orphan-typesetting.png, to be used in article Orphan (typesetting). For the life of me, I cannot figure out how to make it display in the article!! Can you please help me? Thanks, Smooth Henry 23:20, Apr 8, 2005 (UTC).

It's cool. I've added the image to the article for you. Have a look at the source – it works just like adding a local image. I've also changed the PD-US tag on the Commons: if you created the image yourself (which I assume is the case here) and want to release it into the public domain, you can use the {{PD-user|Smoothhenry}} license tag. Cheers, --MarkSweep 23:40, 8 Apr 2005 (UTC)
Hah. It never even occured to me that it would be that easy... Thanks a lot! --Smooth Henry 18:30, Apr 10, 2005 (UTC)

## "China"/"PRC" vs. "mainland China" for page titles

Following the long discussion at Wikipedia talk:Naming conventions (Chinese) regarding proper titling of Mainland China-related topics, polls for each single case has now been started here. Please come and join the discussion, and cast your vote. Thank you. — Instantnood 12:49, Apr 9, 2005 (UTC)

Hello MarkSweep. A request for arbitration has been filed against me at WP:RFAr by Snowspinner as the AMA advocate for jguk. What do you think I can do? — Instantnood 20:44, Apr 10, 2005 (UTC)

Thanks for your advice MarkSweep. I am seeking assistance from AMA, to have the case better handled. I'll take your advice, and I won't be hestitant to seek further assistance from you, say, making a statement or hunting for evidence, when it is necessary. Thanks again for your generous help. :-D — Instantnood 21:55, Apr 10, 2005 (UTC)

## Instantnood RFAr

Thanks for your comments and info on the case, I appreciate it. If you ever have anything else you think of feel free to let me know about it! --Wgfinley 03:33, 11 Apr 2005 (UTC)

Thanks for your additional comments, I agree with your points and am going to be revising the response here shortly. --Wgfinley 23:14, 11 Apr 2005 (UTC)

## Roasted coffee

Hi Mark! I'm a bit annoyed you suppressed both the Featured Picture tag {{FeaturedPicture}} and the copyright license tag on this picture : http://commons.wikimedia.org/w/index.php?title=Image%3ARoasted_coffee_beans.jpg&diff=0&oldid=85750 Is there something wrong with the picture? Are you the copyright holder? I'm mainly embarrassed because the picture still shows on Commons Featured pictures ... villy 18:55, 11 Apr 2005 (UTC) (villy on the Cmmons)

I've replied at Commons:User talk:Aurevilly. --MarkSweep 22:06, 11 Apr 2005 (UTC)

Thanks for your answer. I did not notice the actual chronology and yes, it is weird. I juts "reverted" to the last version ... villy 06:51, 12 Apr 2005 (UTC)

Probably a software glitch. Perhaps we can ask a developer to look into this? --MarkSweep 06:53, 12 Apr 2005 (UTC)

## Re: Archiving

I intended to make the history of my talk page more easily located, by cutting the records according to archive pages. I have tried, say [1] or even [2] and it still works. Only urls using &diff=next or &diff=prev will stop working. I'm not sure if it is necessary to have things restored tho. — Instantnood 21:38, Apr 12, 2005 (UTC)

Brilliant. I didn't realize those diffs would still work. But just to be safe, check with an admin and/or developer (perhaps on the #mediawiki channel) if restoring the talk page is recommended. --MarkSweep 21:43, 12 Apr 2005 (UTC)

## Commons vs Wikipedia question

Hi - I have a question - I uploaded a file to the commons named PlackianLocus.png. The image I was expecting to update has the same name but is in Wikipedia. The commons image does not replace the Wikipedia image in articles. Do you know the best way to get articles to access the new image? Thanks for any help - PAR 06:11, 13 Apr 2005 (UTC)

You'll have to list the local picture for deletion (perhaps those cases qualify for speedy deletion now, but I'm not sure). Point out that an image of the same name exists on the Commons, which means that once the local image has been deleted, it will be transparently replaced by the version from the Commons. It's not possible to embed images from the Commons (or anywhere else) directly. The only way to embed a Commons image is to say [[Image:foo.jpg]] where foo.jpg exists one the Commons but doesn't exist locally. If it exists locally, it will preempt the inclusion of the Commons image. --MarkSweep 19:10, 13 Apr 2005 (UTC)

I was afraid of that. - Thanks for the help. PAR 22:25, 13 Apr 2005 (UTC)

## Wikipedia:Chinese naming controversy

Dunno if you have noticed there's another discussion at Wikipedia:Chinese naming controversy, started by Uncle Ed. Do take a look. :-D — Instantnood 14:42, Apr 14, 2005 (UTC)

If you can, please take a look at the accuracy dispute at Talk:Economy of China--Jiang 06:05, 20 Apr 2005 (UTC)

## Request for help

My User page keeps on getting vandalised. Under the 3-revert rule there is a limit as to how many times I can revert the vandalism. You are clearly buddy-buddy with one of the vandals, can you please exert some pressure....or does Wikipedia really consist of people who carry out such childish behaviour?

Sollogfan 11:19, 22 Apr 2005 (UTC)

There is no limit for vandalism....especially on your own talk page. Cburnett 14:32, Apr 22, 2005 (UTC)
It must be noted that 'Sollogfan' is a sockpuppet account; the vandalism he refers to involves re-instating the 'sockpuppet' image.-Ashley Pomeroy 23:00, 23 Apr 2005 (UTC)
It must be noted that .Ashley Pomeroyis telling (deliberate) untruths to justify his obsessive vandalism. He has not provided one iota of evidence that I am Sollog. Sollogfan 11:15, 13 May 2005 (UTC)

It should be noted that my account is NOT a {{sollog}} account. Do you even know what a {{sollog}} is? Sollogfan 11:58, 25 Apr 2005 (UTC)

You vandalise my page and ignore Sweep - so I descend to your childish level. Sollogfan 11:58, 25 Apr 2005 (UTC)

Without commenting on the substance of the alleged vandalism, whenever you're thinking that you need to stoop to someone else's "childish level", you are automatically wrong. Don't bite the bait, if you think that's what's going on. --MarkSweep 19:05, 25 Apr 2005 (UTC)
My page continues to be vandalised by Pomeroy. I have ignored it but it continues. I now expect you to reprimand me for simply doing to Pomeroy what he does to me - after all, unfairness is a by-word hereSollogfan 11:15, 13 May 2005 (UTC)

Interesting that when Sollogfan asked for help and complained about vandalism you waited and waited and then.......sent the bullying for arbitration because Sollogfan fought back! Incredible! The Number 13:02, 15 May 2005 (UTC)

## Gamma & inv-gamma plots

Looking at commons:Image:Gamma distribution cdf.png, your gnuplot code uses this for the cdf:

pgamma(x, shape, rate) = (x<0)? 0 : igamma(shape, x*rate)


except the parameterization is the k/theta not alpha/beta so the cdf should be x/rate not x*rate (x/rate was my original code and x*rate was your fix). I don't understand how the k/theta notation works for the pdf but doesn't for the cdf.

I'm trying to generate an inverse-gamma cdf but can't get a cdf generated. Maybe the cdf is wrong. Cburnett 21:47, Apr 22, 2005 (UTC)

You do realize that the function that actually gets plotted is f and not pgamma, right? Function f is defined as
    f(x, k, t) = pgamma(x, k, 1.0/t)

The only reason the function is called pgamma is for compatibility with R. The gnuplot code is taken from a bigger gnuplot file I use that defines many pdfs and cdfs in a way that is as close to R as possible, for my own sanity. Notice that the third paramter of pgamma and dgamma is called "rate" and not "scale", just the way R does it. I think it was for this distribution that there was a problem at an earlier point with the pdf and cdf plots not matching, so I uploaded one gnuplot script with matching definitions and either the cdf part of the pdf part commented out.
Or did I just completely misunderstand your question?
Now for the inverse-gamma distribution, something is wrong in the current version of that article. Let me work on that and get back to you (on this talk page). --MarkSweep 22:03, 22 Apr 2005 (UTC)
DOH! I missed the reciprocal in f(). I'm quite certain that the inverse-gamma pdf is correct and I'm sure I got the cdf from the same source as the pdf so that should be correct as well. Cburnett 22:54, Apr 22, 2005 (UTC)
Actually, the inverse-gamma cdf was wrong, at least according to the conventions for ${\displaystyle \gamma (a,z)}$ and ${\displaystyle \Gamma (a,z)}$ that are used in Wikipedia. The correct cdf (now fixed in the article) is as follows:
${\displaystyle {\frac {\Gamma (\alpha ,\beta /x)}{\Gamma (\alpha )}}=1-{\frac {\gamma (\alpha ,\beta /x)}{\Gamma (\alpha )}}\!}$
In other words, this is the upper incomplete gamma function (where the Gamma integral starts at ${\displaystyle \beta /x}$ instead of at zero) divided by the complete Gamma function. This fraction is also known as the regularized upper incomplete Gamma function (but there is widespread confusion about the terminlogy). gnuplot implements the regularized lower incomplete Gamma function as igamma:
${\displaystyle {\mbox{igamma}}(\alpha ,z)={\frac {\gamma (\alpha ,z)}{\Gamma (\alpha )}}\!}$
So the InvGamma pdf and cdf can be implemented in gnuplot as follows:
    dinvgamma(x,a,b) = exp(a*log(b) - lgamma(a) - (a+1)*log(x) - b*1.0/x)
pinvgamma(x,a,b) = 1 - igamma(a, b*1.0/x)

I've plotted these and they seem to pass integration-by-eye. --MarkSweep 23:05, 22 Apr 2005 (UTC)
Thanks a lot! Got them uploaded now. Cburnett 23:27, Apr 22, 2005 (UTC)

I was nominated for administrator and I'd like to hear your opinion at Wikipedia:Requests for adminship/Cburnett. Cburnett 07:23, Apr 24, 2005 (UTC)

I am really offended with your saying my revision to "blardy" made no sense. I just used a visual example which seemed far clearer to me than the first example. If I'm mistaken, that would be one thing, but I'm not. When blue is written in black print, it is antiblardy. When blue is written in blue print, it is blardy. So blardiness or antiblardiness are not inherent qualities of a word. What is the mistake in that?--The Dogandpony 01:11, 25 Apr 2005 (UTC)

Your example made no sense to me precisely because you tossed in another dimension. The previous examples were about words and their conventional orthographic representations, which are still inherent properties of the words themselves, not of specific occurrences. A second objection is that "blue" may not appear blue everywhere – for example, if I'm reading the article in a text-based browser, the link will not necessarily be blue. This is very hard to fix. I thought about "italicized" vs. "italicized", but that runs into similar problems. The best I could think of vs. "CAPITALIZED" vs. "capitalized". But even if this second problem could be resolved, there is still the issue that you're talking about occurrences of words, not about words as abstract units. --MarkSweep 01:35, 25 Apr 2005 (UTC)

## Amicus curiosus brief

Did you mean Amicus Curiae Brief?Xiongtalk 04:34, 2005 Apr 26 (UTC)

Well, there is a link to amicus curiæ, but I was going for the idea of the curious bystander. Though I guess curiosus does not quite mean "curious". ;-) --MarkSweep 04:42, 26 Apr 2005 (UTC)

Can you do something about 50Stars and the List of national flags article? I've exhaused my reverts for the day and posted everything that's I wanted to post on the talk page...and he's still reverting. --Jiang 04:29, 28 Apr 2005 (UTC)

## Rendering Math

Hi Mark - On both my machines, the \, works fine to force TeX to PNG. Could you do me a favor and check it again? Did you make sure it wasn't an old cached image or something? Thanks. PAR 05:25, 6 May 2005 (UTC)

It depends on your user preferences. I've set mine to "HTML if possible or else PNG" for editing, which results in \, being rendered as a space. --MarkSweep 06:30, 6 May 2005 (UTC)

Hi Mark - I think the \! character is too strong - it makes "HTML if possible, else PNG" give PNG, and for somebody who prefers HTML and has chosen preference 3, thats bad. If you choose "HTML if very simple, else PNG" then the \, character will give PNG. Please look at User talk:Cburnett#Rendering Math

Now I'm confused: I thought you wanted to force PNG rendering in displayed formulas? Here's my current understanding of the situation: If you want to force PNG rendering, then use \!, because \, will not work for all settings. If you don't want PNG, then it's simple: don't add anything. But I must be missing something here. --MarkSweep 16:58, 6 May 2005 (UTC)
My current understanding of what I want is:)
1. Preference = "Always render PNG"............................All inline=PNG, all displayed=PNG
2. Preference = "HTML if very simple, else PNG"........All inline=HTML, all displayed=PNG
3. Preference = "HTML if possible, else PNG"...............All inline=HTML, all displayed=HTML
Articles should be written to maximize the above. They probably cant always be written so that the above is completely true. When \, is added to a "simple" displayed formula, it goes from HTML to PNG under preference #2 but stays HTML under #3 which is good. When \! is added, it gets displayed as PNG even under #3 which is not good. PAR 18:54, 6 May 2005 (UTC)
I basically agree with you, but I don't think point 3 is realistic, simply because HTML rendering is too limited. Things like multiple subscripts or superscripts, continued fractions, under-braces, etc. cannot be rendered as HTML at all, or do not look intelligible when rendered as HTML. Also, most big formulas (big sum, product, integral signs, even fractions) look just plain ugly in HTML. Finally, there is the issue of consistency and control over fonts: if you have a kappa in a PNG formula and a kappa in an HTML formula, they look very different in my browser (in fact, I can't distinguish a kappa from a k at all in HTML, but that may just be my font choices). Anyway, for consistency and control over fonts (since you know PNG will use Computer Modern), PNG may be preferrable. Overall I don't see anything wrong with always forcing PNG rendering for certain displayed formulas. I thought that's what you wanted to do. --MarkSweep 19:16, 6 May 2005 (UTC)
Well, it was, but my thinking was clarified by reading the discussion between you and Cburnett. I just want us to get a semi-standard way of doing things so we don't waste time reverting each others edits, since we are working on some of the same stuff. How about:
1. Preference = "Always render PNG"............................All inline=PNG, all displayed=PNG
2. Preference = "HTML if very simple, else PNG"........All inline=HTML, all displayed=PNG
3. Preference = "HTML if possible, else PNG"...............All inline=HTML, all displayed=HTML
with the proviso that #3 can really look bad, and if it does, add a \! to force PNG under all circumstances.
If you agree with that, how about we run it by Cburnett and see what he thinks and then we can go back and edit the chi-square article? If we can agree on that article, I think we are good to go. PAR 20:18, 6 May 2005 (UTC)
That sounds good. Let's work on getting the chi-square article into a format we are all happy with. I'm not going to edit it for the time being. If you think I went overboard with my latest changes a couple of days ago, feel free to revert. --MarkSweep 02:06, 7 May 2005 (UTC)