User talk:Diberri/Archive 7

Wikify domains
I own wikify.biz and wikify.us. I know you are running a site with a wiki convertor. Are you interested in purchasing the domains. If yes, take a look and drop me a note. --Damis 05:11, 1 September 2007 (UTC)
 * Thanks, but no thanks. I'm on a student's budget, after all. :-) --David Iberri (talk) 02:06, 2 September 2007 (UTC)

New MCOTW
Hope you can help... --WS 17:53, 3 September 2007 (UTC)

Obesity
Hey Dave. A little while ago you added some useful neurophysiological information to obesity (LH/VMH and lesion studies). Could you add a source that most accurately reflects this content? JFW | T@lk  10:53, 4 September 2007 (UTC)
 * Yup, I remember that. I'll see if I can dig up my reference. --David Iberri (talk) 23:31, 4 September 2007 (UTC)
 * I fleshed out the section on neural circuits a bit more using the Flier paper as a reference. I think the section now gives the reader a better understanding of leptin's importance in some rarer forms of obesity. Your comments are welcome. I'm especially interested in your thoughts on whether we're making too big a deal of leptin's role in obesity. Will readers automatically think that obesity is due to an error in leptin signaling? --David Iberri (talk) 00:29, 10 September 2007 (UTC)

Gracias amigo. Leptin is only one of the many counterregulatory mechanisms, but I don't think the content focuses excessively on leptin. It was simply the first of a whole raft of mediators, and has been tried therapeutically. I'll be interested to hear what the fatso transcript behaves like. JFW | T@lk  12:31, 10 September 2007 (UTC)

Reference desk comment
Howdy, Nil Einne. I wanted to point out this RD/Science edit and ask you to refrain from making such comments in the future. Regardless of whether the OP is a medical student, the comment is just plain rude and doesn't help answer the question in the least. Thanks, David Iberri (talk) 20:12, 8 September 2007 (UTC)
 * When people show absolutely no effort on their part to answer a basic homework question I occasionally (not always or even often) tell they should expect to have to do so if they want to succeed. Frankly, I don't give a damn whether they or you are offended and I'm not going to change. I'm well aware of policies like don't bite and being civil and I do follow them. But I don't consider telling someone they need to put in an effort if they want to succeed as a student incivil or biting and nothing you're going to say is likely to change that. I've responded to your comment on the ref desk too. Feel free to respond to this and to my ref desk comment although I consider this matter over so won't be responding to either. Cheers... Nil Einne 08:22, 9 September 2007 (UTC)
 * The thing is, you didn't just simply say that the OP should "put in an effort if she wants to succeed as a student." You made what seemed to be an intentionally hurtful comment regarding the OP's presumed career goal, suggesting that she should give up her goal of becoming a physician. That's not helpful; that's a borderline WP:DICK move. --David Iberri (talk) 16:52, 10 September 2007 (UTC)

MCOTW
JFW | T@lk  11:23, 21 September 2007 (UTC)

Batch mode for citation tool
Two suggestions / requests:

1. Any chance of allowing batch creation of citations? I'm thinking of allowing >1 PMID # in the box, maybe separated by commas, which would then return all the needed citations. This may work with PMID, or ISBN, but maybe would get clumsy with URLs.

2. How about an option that allows user to NOT clear the box that the citation shows up in? Just append current result to previous result(s)?

Regards&mdash; G716  &lt;T·C&gt; 01:59, 1 October 2007 (UTC)
 * Thanks for the suggestions. I hadn't thought of supporting a batch mode in the past. I'm more in favor of the "do one thing and do it well" approach to things, and as it is the template builder is already getting pretty bloated what with all the options and everything. I'll keep the batch mode in mind, but I'm not sure I'm down with the idea just yet. Let me think about it some more. Cheers, David Iberri (talk) 12:22, 2 October 2007 (UTC)

Citation tool
Hi David, is there a way to share the code for your great tool? I ask you, because your tool is also very useful for the german community, with some minor configurations for our citations. Greetings, --Hoffmeier 02:50, 4 October 2007 (UTC)
 * Absolutely. The code for fetching data from various databases is readily available, but there's no clean code that actually makes the citations. Let me see what I can do. Will get back to you. --David Iberri (talk) 11:38, 4 October 2007 (UTC)
 * After a little thought, it might be easier to just add a "German citations" option to the existing code. Alternatively, I could give you the code, but it's a bit idiosyncratic (it uses Perl and HTML::Mason, for example). Up to you. --David Iberri (talk) 15:09, 13 October 2007 (UTC)
 * Hi David, I wrote an email to you. --Hoffmeier 12:31, 15 October 2007 (UTC)

HTML2Wiki Bugette
Hello David, I like the HTML2Wiki conversion tool. Found a small bug in it though... hello there david is converted to hello there david (i.e. hello there david) rather than hello there david (i.e. hello there david) as you might expect. Not a biggy, but thought you'd like to know ;)

Keep up the good work! —Preceding unsigned comment added by 69.251.22.175 (talk) 05:03, 13 October 2007 (UTC)
 * Ooh, that's an interesting one. Don't have the time to really tackle that one now, though, unfortunately. --David Iberri (talk) 15:09, 13 October 2007 (UTC)

WikiProject Cell Signaling
I see you listed yourself as a participant on WikiProject Cell Signaling. This project has been reactivated, if you are still interested. Biochemza 20:48, 3 November 2007 (UTC)
 * Glad to hear it. I'm definitely still interested. Thanks for dropping me the note. --David Iberri (talk) 16:58, 18 November 2007 (UTC)

Citation Tool down
Seems to occur from time to time ...  Skopp   15:04, 4 November 2007 (UTC)
 * Yup, blasted server. --David Iberri (talk) 16:57, 18 November 2007 (UTC)
 * Will the server be back up again? I miss being able to use this wonderful tool for ISBNs. Finell (Talk) 04:21, 10 December 2007 (UTC)
 * The server's been up for a while, but I just now realized that some of the HTML output for ISBNs was buggy, which prevented some of the page from loading. Should work all right now. --David Iberri (talk) 17:18, 10 December 2007 (UTC)
 * Yes, it is working again! Thanks very much for fixing this, and even more for providing this wonderful tool. Finell (Talk) 10:01, 15 December 2007 (UTC)

Additional Request
For the URL citation form, can you make it so a unique ref name is created from the webpage title. So you end up with <ref name="TITLE"...etc. -- Demantos (talk) 17:11, 16 November 2007 (UTC)
 * Done. --David Iberri (talk) 16:56, 18 November 2007 (UTC)
 * Any chance that you could make this new feature optional by use of a check box? Two reasons: some webpage titles are very loosely formed and can change fairly frequently, so do not make good unique names. Also, some webpage titles are very long, and including the entire title in the ref name makes for a messy inline ref citation, making subsequent editing a chore. Best regards,&mdash; G716  &lt;T·C&gt; 18:44, 26 December 2007 (UTC)

Citation tool not reading journal name
Hi PMID 17999832 fails to pick up on the journal: Gives i.e. Rather than i.e.

David Ruben Talk 23:30, 17 November 2007 (UTC)
 * Got it, thanks. It's fixed for that article and for any article that uses the Medline-indexed journal title only. --David Iberri (talk) 16:50, 18 November 2007 (UTC)

Citation tool enhancement suggestion
I'm not sure how your citation tool works, but would it be possible for it to detect, when given a PMID, whether a free version of the article is available, and include a link to it, under the "url=" attribute? --Seans Potato Business 12:13, 18 November 2007 (UTC)
 * It works by reading XML straight from PubMed. Right now, there's a 'doi' field that seems to only appear for freely available versions, and this doi field is output by the citation generator. For example, PMID 14688414 gives this output:
 * Seems to get the job done in my experience. Do you have a specific example where this isn't working correctly? Or is it just that you'd rather have this URL populate the 'url' field in addition to the 'doi' field? --David Iberri (talk) 16:55, 18 November 2007 (UTC)
 * Well I don't really know what the DOI field is supposed to include, but my guess right now would be only the pubmed central ones? Try PMID 17998692 for instance - that is available for free online, yet there's no filling of DOI (it's not available on pubmed central).
 * You can find loads of these by doing any search on pubmed, going to "limits" and selecting "links to full free articles" and then half/most of the resulting PMIDs wont get a DOI from the template filler machine. --Seans Potato Business 19:25, 24 November 2007 (UTC)
 * Ah, I see. I bet you're right that a DOI is only included for PMC articles. I'm not sure if there's a way around that, but it's something I could look into if there's enough of a need for it. What do you think about the new 'url' field filling? --David Iberri (talk) 19:37, 24 November 2007 (UTC)
 * Is the url field now filled instead of the doi field? --Seans Potato Business 20:01, 24 November 2007 (UTC)
 * No, it's filled in addition to the doi field. --David Iberri (talk) 20:09, 24 November 2007 (UTC)
 * In the example given below, DOI is blank but URL is filled. Why is that? --Seans Potato Business 20:19, 24 November 2007 (UTC)
 * Because the data from PubMed doesn't contain a DOI for that article. DOI and URL information are obtained separately (DOI from PubMed's XML, eg ; and URL from PubMedCentral). If a doi is available from PubMed, it'll be filled in. If a primary link is available from PubMedCentral, it'll be filled in for the 'url' field. If one is available but not the other, then only one will be filled. The doi field is not generated from the url; likewise, the url is not generated from the doi. --David Iberri (talk) 20:35, 24 November 2007 (UTC)
 * Any idea why pubmed doesn't give a DOI for that article? What dictates whether or not pubmed will provide a DOI? If there's an affiliation between pubmed and pubmed central, I would have expected pubmed to provide DOIs to pubmed central-held articles. --Seans Potato Business 21:11, 24 November 2007 (UTC)
 * Any idea why pubmed doesn't give a DOI for that article? What dictates whether or not pubmed will provide a DOI? If there's an affiliation between pubmed and pubmed central, I would have expected pubmed to provide DOIs to pubmed central-held articles. --Seans Potato Business 21:11, 24 November 2007 (UTC)

(Unindent) I'd expect the same thing. I don't have any theories to explain the lack of crosstalk between the them. --David Iberri (talk) 00:14, 25 November 2007 (UTC)

I'd certainly like to have the url field populated automatically. At the moment I'm doing this manually. Does the XML also return information on free availability on PubMed Central, and would integration with PMC be something worth considering? JFW | T@lk  00:01, 23 November 2007 (UTC)
 * PubMed does provide a 'primary link' option, but I'm unclear on the relationship between a primary link and a free full text link to journal articles. I've added an option to the template generator called 'add url if available' that uses the url specified in PubMed's primary link output for the given pmid. Feedback welcome. --David Iberri (talk) 05:52, 23 November 2007 (UTC)
 * Forgot to give an example: . --David Iberri (talk) 06:02, 23 November 2007 (UTC)

Couple of new problems with tool
Tool is great as always, but a couple of new problems:
 * 1) The concatenated tag name is not always picking up the title of the source url. Try http://www.nlm.nih.gov/cgi/mesh/2007/MB_cgi?term=low+back+pain . For webpages with long titles, the resulting tag name becomes quite long. Unfortunately, I cannot think of a good solution other than something tedious to code such as extracting the acronym of any title that has a space in it.
 * That's a specific problem with the HTML at that URL. It doesn't provide a &lt;title> in the &lt;head> section (in fact it has no head section), so there's nothing for the template builder to pick up. Interestingly, there is a &lt;title> elsewhere in that document, but since it's not in the &lt;head> section, it's not a legal way of specifying a window title. --David Iberri (talk) 16:51, 19 November 2007 (UTC)
 * 1) The slight change in the tag name from pmid####### to pmid-######## is causing problems in articles that already have references using the first format and now may have duplications using the second format.
 * Thanks for pointing that out. I made that change on a whim without considering the effects. I've reverted it. --David Iberri (talk) 16:51, 19 November 2007 (UTC)
 * 1) Here is an odd problem. This call to the tool creates a url attribute that has spaces in it, which invalidates the url. http://diberri.dyndns.org/wikipedia/templates/?type=url&add_ref_tag=1&id=http://www.nlm.nih.gov/cgi/mesh/2007/MB_cgi?term=low+back+pain
 * Got it. Fixed. --David Iberri (talk) 16:51, 19 November 2007 (UTC)

Thanks - Badgettrg (talk) 13:08, 19 November 2007 (UTC)

Hosting on toolserver
It seems the dyndns server crashes occasionally. Have you considered running your famous tool on http://tools.wikimedia.de ? JFW | T@lk  00:01, 23 November 2007 (UTC)
 * Great idea. Just posted a request over at . --David Iberri (talk) 06:03, 23 November 2007 (UTC)

Toolserveraccount
Hello Diberri, please send your real-name, your wikiname, your prefered login-name and the public part of your ssh-key to. We plan to create your account soon then. --DaB. 21:24, 23 November 2007 (UTC)

Suggestion: Google Books Integration?
Any chance that your application could check a Google Books link (e.g. http://www.google.co.uk/books?id=ivNZHslpQ5AC&pg=PA53&dq=il-12+th1&ei=sJ9MR7-qF5KuiQHyo9DlBg&sig=R8N7NBiBNGR1ZV6AWm28ulwsIoQ) and from that link retrieve the ISBN and page number and give a citation template in the same form as though an ISBN was given? --Seans Potato Business 22:59, 27 November 2007 (UTC)
 * If there's a way to extract the ISBN from the Google Books page, then yes. But I'm not familiar enough with GB pages to know whether that's the case. --David Iberri (talk) 23:59, 15 December 2007 (UTC)

Links to full text
Your tool continues to improve, thanks for your work. Suggestion: I think the urls to full text that publishers supply to the NLM are not as stable as DOIs. Consider not populating the url field is the doi field has a value. This might reduce the number of broken links in the future. - Badgettrg 16:04, 30 November 2007 (UTC)
 * It's unfortunate that PubMed's URLs aren't as stable. But is completely excluding them from the generated template code the best solution? Another idea would be to update the citation templates so that if they are given both the DOI and the URL field, they preferentially use the DOI field. Yet another idea would be to attack the source of the problem directly by contacting PubMed and seeing how the issue of unstable URLs might be improved. I'm open to suggestions, but I want to hear some extra input before I stop providing URLs in the generated template code. --David Iberri (talk) 15:58, 11 December 2007 (UTC)

Automatic filling of format field for URL references
If I enter the following URL:

www.promega.com/profiles/1001/ProfilesinDNA_1001_09.pdf I get this template in return:

cite web |url=www.promega.com/profiles/1001/ProfilesinDNA_1001_09.pdf |title= |accessdate=2007-12-15 |format= |work=

Is it possible that the template filler could read the final three letters on the URL and where appropriate (e.g. according to a list of predefined three character extensions) fill the format attribute?

Also, because I entered the URL as www., WP spat out an error when I tried to use the resulting template. Would it be possible to automatically append "http://" to URLs in the template when the given URL is in the format www.? --Seans Potato Business 22:03, 15 December 2007 (UTC)
 * There's no straightforward solution to the missing "http://" problem. If you don't give a proper scheme in the URL, then it's not a proper URL at all and can't be fetched. I should actually make the template filler complain if a scheme isn't provided, but haven't got around to that yet. Also, the last three letters don't universally provide the file type. Some URLs might end in ".pdf" and not actually be PDFs. The best solution would probably be to read the mime type of the file as it's loaded. I'll look into that. --David Iberri (talk) 23:58, 15 December 2007 (UTC)

Two requests
(1) For PMIDs, could you add an optional checkbox, adding the double-square brackets around the journal name? (2) For URLs, could you strip out quotation marks in the web page title used to generate the ref name? (The refs don't process properly when they have quotation marks in the name.) --Arcadian (talk) 18:44, 26 December 2007 (UTC)
 * Great ideas. All done. --David Iberri (talk) 12:17, 7 January 2008 (UTC)
 * Thank you! --Arcadian (talk) 16:22, 7 January 2008 (UTC)

Feature request: Cookies to remember preferences?
It would nice if some kind of cookie would remember my preferences for the different types of reference. For example, I always like reference tags to be added, and when using a PMID, I like to "add URL if available". Having to to re-check the boxes every time I switch reference type or go away and come back gets me down :( --Seans Potato Business 21:36, 1 January 2008 (UTC)

Titin
Any particular reason? User:Dorftrottel 01:22, January 13, 2008
 * Yes, see Titin's talk page. Sorry, I should've mentioned that in the edit summary. --David Iberri (talk) 15:51, 13 January 2008 (UTC)
 * All I see there is a majority wanting it there, and a minority being against. No one appears to be "arguing" as in "presenting arguments". I did provide two distinct arguments that I would like to see addressed. If you can manage (unlike Arcadian) to do so without resorting to ABF and blocking threats, based on a lack of comprehension of WP:VAND, I would appreciate your reply to those two arguments. User:Dorftrottel 15:58, January 13, 2008
 * I responded over at Talk:Titin before I got your latest message at 15:58. Let's keep the discussion there. --David Iberri (talk) 16:14, 13 January 2008 (UTC)

Template filling: special characters in DOIs and URLs
Hi Diberri. Thanks for producing a fantastic tool which has saved me a lot of work!

I ran into a small problem with a few number of non-functional URL links that the tool returns. For example, the following URL link does not work:

However if the special characters "<" and ">" in the URL are replaced by their hexidecimal equivalents "%3C" and "%3E" respectively, then the link works:



Would it be possible for the tool to scan the URL and make the appropriate hexidecimal substitutions? Cheers. Boghog2 (talk) 21:30, 20 January 2008 (UTC)
 * Thanks for the post. Should be fixed now. --David Iberri (talk) 22:10, 20 January 2008 (UTC)


 * And thank you. That was incredibly fast!  Boghog2 (talk) 22:51, 20 January 2008 (UTC)


 * And people wonder why I programmed the thing in Perl. :-) --David Iberri (talk) 14:14, 21 January 2008 (UTC)

Template filling: special characters in DOIs and URLs: Part 2
I have another request. The problem crops up not only with URLs but also with DOIs. For example the following DOI link is broken:



However as with URLs, if the special characters "<" and ">" in the DOI are replaced by their hexidecimal equivalents "%3C" and "%3E" respectively, then the link works:



I would appreciate if you would apply your Perl magic again. Cheers. Boghog2 (talk) 22:35, 10 February 2008 (UTC)
 * My gut feeling is that this should be fixed at the level of Template:Cite journal. I believe Wikipedia allows for url-escaping functions to be called within wiki markup, and this would solve the issue with the funky doi field. --David Iberri (talk) 19:44, 21 March 2008 (UTC)


 * Your right. This is really a bug in the template itself. I will redirect my request there. Thanks for the suggestion.  Boghog2 (talk) 20:22, 21 March 2008 (UTC)


 * The problem apparently is a Mediawiki bug (see Problem_DOI and URL_encoding). It has been reported, but as far as I can tell, there has been no commitment as yet to fix it.   It is not such a big issue if one is aware the percent-encoding solution, so I can fix this myself manually and hopefully the Mediawiki people will eventually get around to fixing the underlying bug.  Cheers  Boghog2 (talk) 21:12, 21 March 2008 (UTC)

CaMKII Image
Your CaMKII image is incredible! I'd really love to see more of that type of high-caliber image on Wikipedia. Miserlou (talk) 05:20, 30 January 2008 (UTC)

Action potential
Action potential has been nominated for a featured article review. Articles are typically reviewed for two weeks. Please leave your comments and help us to return the article to featured quality. If concerns are not addressed during the review period, articles are moved onto the Featured Article Removal Candidates list for a further period, where editors may declare "Keep" or "Remove" the article from featured status. The instructions for the review process are here. Reviewers' concerns are here.