User talk:Smith609

Please note: to avoid duplication, I'll reply to queries about DOI bot on this page.

Archives

1, 2, 3, 4, 5

/DOI bot

I've noticed that refs that used to have URLs seem to be losing them. URLs are one of the the few parts of citations that I'm somewhat fanatical about, as they can actually help the non-specialist reader (unlike accessdate or doi). I'd guess you've been using your citation tool. Can it be enhanced to capture the URL of at least an abstract, and complete accessdate automatically? Philcha (talk) 13:04, 28 June 2008 (UTC)[reply]

I agree that DOIs act as URLs, and usually lead to at least an abstract. The trouble is they don't look like URLs because they do not contain meaningful link text. As usual I'm concerned with the non-specialist reader who may think DOI = Department of Immigration, and / or may be reading his / her first scientific Wikipedia article after reading several about subjects in which DOIs are not used (e.g. chess, video games, ...). There also situations where DOI is not that helpful but a URL including the title is, e.g. the DOI leads to a "please log in / subscribe" page but the author has placed a copy at a publicly-accessible URL. I found a few of these for Cambrian explosion and also for Permian–Triassic extinction event. Philcha (talk) 14:02, 28 June 2008 (UTC)[reply]

Template_talk:Cite journal (which fizzled - I despair) makes the interesting point that Template: citation uses the DOI as the link destination for the title (which appears as a link), but does not display DOI. Another implication of the discussion is that Template: Cite journal does not use Template: citation under the covers (since Template: Cite journal does not make the title point to DOI if no URL) - which looks silly to me. My own preference would be for DOI to be the default URL for title (as in Template: citation) but also to show and link DOI in case the URL is wrong or "goes bad". Does this make sense? Philcha (talk) 14:50, 28 June 2008 (UTC)[reply]

I hadn't seen the term until earlier to-day, so I doubt if I can contribute content. Give me a call if you want comments on how it looks to an ignoramus. Philcha (talk) 14:52, 28 June 2008 (UTC)[reply]

That does appear to be a gastropod you see munching away in the background. Whatever was Seilacher thinking of? Seriously though, "The revolution put an end to the conditions which allowed exceptionally preserved fossil beds such as the Burgess shale to be formed" implies that the revolution was later than the Burgess Shale.

PS I think that takes it out of the Cambrian Explosion.

PPS "Agronomic revolution"? Whatever was Seilacher thinking of? It sounds like something that happened in the 18th century. Philcha (talk) 07:07, 29 June 2008 (UTC)[reply]

I've created Template:Cambrian substrate revolution as I think it will be useful for e.g. Microbial mat Philcha (talk) 07:25, 29 June 2008 (UTC)[reply]

Bottjer et al (2000) described the pre-revolution substrate as "a fairly stable, relatively low water content sediment surface with a sharp water-sediment interface". PS: should we start using the article's Talk page, so other can laugh at our antics? Philcha (talk) 09:17, 29 June 2008 (UTC)[reply]

I think we need to sort out Template:Annotated image before we go any further, as we're starting to make serious use of it - and so far we appear to be the only users. I think it needs to be revamped so that users have much more explicit control. I think a suitable box structure might be something like:

• Outer DIV, which also contains caption. Sized as next div plus 2px all round.
  • Inner border DIV. Sized explicitly by user.
    • Picture DIV. Absolute positioning within Inner border DIV, default to centered horizontally and vertically (vertical centering will need exprs).
      For flexibility and maintainability of position, it might be nice (not essential) to support right and bottom params. I'd leave this until "phase 2".
    • Annotation DIVs. Absolute positioning within Inner border DIV. Sized automatically. Maximum control over internal formatting (color, font-weight, etc.). Should if possible support any (X)HTML that's allowed in a DIV.

I suggest we develop it in a sandbox page. Once we're happy, I'll convert existing uses, which will also serve as test cases - only fair, since I'm the one suggesting this. Philcha (talk) 10:07, 29 June 2008 (UTC)[reply]

Did you have a plan for publicizing this template (for example at WT:TOL)? I'm sure that the regulars there will discover it before too long, but I didn't know whether you wanted to make some kind of announcement/proposal, or try and spend a little time working out issues such as the questions I asked at Template talk:Paraphyletic group. Kingdon (talk) 15:01, 28 June 2008 (UTC)[reply]

Starting now at User:Philcha/Sandbox/Template - Annotated image. I don't want to advertise this unless necessary, in case too many cooks spoil the broth. Can you please give me an hour to play around - unless I wail for help! Philcha (talk) 12:19, 29 June 2008 (UTC)[reply]

OK so far see - User:Philcha/Sandbox/Template - Annotated image/doc. Tested in Firefox, IE and Opera, all under Win XP; looks OK in all. AFAIK the outstanding item are:

• Getting control over formatting of annotation text.
• Wikilinked annotations.
• Test with extinction graph.
• See how much (X)HTML we can get away with in params.
• See if we can sensibly use positional params - they're all keyword at present. Right now I'm not confident, becuase the price of greater flexibility is more params. Philcha (talk) 14:28, 29 June 2008 (UTC)[reply]

I'm not sure exactly what you've changed. I've put some notes on the sandbox talk page. However, both of the test case images displayed fine with the old template; perhaps it would be sensible to create some "test cases" which attempt to use some of the fuller functionality you're trying to introduce. (The old version still did the job, and seems to do everything on your "outstanding" list perfectly well. As any earthworm will tell you, complexity isn't always necessary!)

Also note the "image-css" parameter I added, which might help keep parameters to a minimum. Martin (Smith609 – Talk) 14:40, 29 June 2008 (UTC)[reply]

I've done all I can think of at present - see User:Philcha/Sandbox/Template - Annotated image/Samples

My original problem with Template:Annotated image was the hard-coded padding, which was wrong for what I wanted to do with the Coelomate and Cambrian substrate revolution templates. It also had the annotations' font-weight hard-coded to bold.

My new versions give, as far as I can see, complete control over the size and position of the image, plus the ability to set default values for CSS formatting params, plus can float right or left. Philcha (talk) 21:11, 29 June 2008 (UTC)[reply]

Looks great! Nice work. Martin (Smith609 – Talk) 21:31, 29 June 2008 (UTC)[reply]

Would you like me to start converting existing stuff now? Philcha (talk) 21:34, 29 June 2008 (UTC)[reply]

If you don't mind - go for it! And I wonder where the best place to publicise the template would be. Martin (Smith609 – Talk) 21:38, 29 June 2008 (UTC)[reply]

I think it is good the DOI bot flags DOIs that are presently broken, but does it do anything else with the information? I'm sure publishers like to remain in CrossRef's good books and have DOIs that actually register properly. I notice that dx.doi.org has a special form that allows you to send annoyed emails to publishers. Will you be sending lists of broken DOIs to publishers for fixings? JFW | T@lk 13:30, 29 June 2008 (UTC)[reply]

I'm sending them an XML file - when I get round to coding up something to check that the DOIs really are broken and didn't just time out when accessed. (Especially important as Blackwell synergy are moving at the moment!!) Martin (Smith609 – Talk) 14:05, 29 June 2008 (UTC)[reply]

Thanks. I think the added scrutiny will eventually be appreciated by publishers. And yes, don't get me started on Blackwell... JFW | T@lk 08:28, 3 July 2008 (UTC)[reply]

Hi,

Thnaks for your speedy work on WP Palaeontology! The new infobox looks much better! Oh, cheers for sorting out all those redirects too! Best, Mark t young (talk) 16:13, 30 June 2008 (UTC)[reply]

Template:Italic title has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. —Remember the dot ^(talk) 02:52, 1 July 2008 (UTC)[reply]

I started an article Microbial mat to support various current aticles, then found Biofilm. To avoid duplication of content we need to agree a division of labour between articles. I've posted a comment about this at Talk:Biofilm#Microbial_mat, and suggest we keep an eye on it. Philcha (talk) 23:49, 1 July 2008 (UTC)[reply]

I suspect we'll wind up doing our own thing. The first response to my post was "do you have a WP:RS to confirm that they are the same?" The responder assumed I was proposing a merge, despite how carefully I'd phrased my post to indicate that I expected multiple articles. Philcha (talk) 09:18, 2 July 2008 (UTC)[reply]

I notice Wikipedia:WikiProject Geology/Cambrian explosion says, "aims to ... bring the article first to a "Good Article" standard, and ultimately to a "Featured Article" status." What happened to "By the advice of a very wise man, I try not to play the "featured article" game any more"?

I suggest the bit about "intelligble to the non-specialist reader ..." should be repeated in the section about GA and FA criteria, to make it absolutely plain that there is no room for the "academics envy"^(TM) that afflicts many WP:MOS mavens. Philcha (talk) 09:23, 2 July 2008 (UTC)[reply]

I'm not going to concern myself with box-ticking; it is simply intended to give participants an idea of what kind of level we're aiming for. (although it might be nice to see the article on the main page someday...) Not that it is likely to make much difference in any case; the project has had precious little interest so far! Do feel free to amend the page as you wish.

Martin (Smith609 – Talk) 09:47, 2 July 2008 (UTC)[reply]

It's removing accessdates from references (example). I've blocked it for now, once it's fixed let me or any other admin know, or drop a note on WP:AN. Neıl 龱 01:25, 3 July 2008 (UTC)[reply]

Never mind, I realised it's supposed to if no URL exists. My bad. Bot unblocked, hopefully I haven't broken anything. Neıl 龱 01:34, 3 July 2008 (UTC)[reply]

On 3 July, 2008, Did you know? was updated with a fact from the article Agronomic revolution, which you created or substantially expanded. If you know of another interesting fact from a recently created article, then please suggest it on the Did you know? talk page.

--Gatoclass (talk) 06:52, 3 July 2008 (UTC)[reply]

Are you done with Talk:Cambrian explosion yet? Philcha (talk) 22:21, 3 July 2008 (UTC)[reply]

A reader of Chiropractic recently complained "The source given is Cooper & McKee 2003. The doi link is broken". I was surprised to see this, since I thought the DOI bot would catch that. I verified that the DOI (doi:10.1111/1468-0009.00040) is indeed broken. I also verified in my sandbox that the DOI bot indeed does not catch the broken URL; here's the diff. I fixed Chiropractic by hand for now, by adding a doi_brokendate= parameter to that citation. Eubulides (talk) 00:45, 4 July 2008 (UTC)[reply]

Perhaps due to the current merger of Blackwell with Wiley? --Rifleman 82 (talk) 01:39, 4 July 2008 (UTC)[reply]

I guess that's why the DOI is broken. Despite their proclamation that the merger would not break any links! The bot only tests DOIs when in "slow mode"; I ran the bot in this mode and it found the broken DOI. (I've just updated my widget to give users the choice between slow and turbo mode.) Martin (Smith609 – Talk) 09:40, 4 July 2008 (UTC)[reply]

Explanation for recent outages

That announces a planned outage tomorrow; it doesn't explain recent outages. It looks like Wiley's reliability standards are somewhat lower than Wikipedia's. Eubulides (talk) 17:07, 4 July 2008 (UTC)[reply]

I can't wait until that's true for content, too!

They've changed the content of the page since I posted it; it listed a few times the servers had been down recently. But the 404s are still (annoyingly!) prolific! Martin (Smith609 – Talk) 17:12, 4 July 2008 (UTC)[reply]

That same diff illustrates another problem: the DOI bot inserted doi:10.1001/archinte.147.3.492 for Chou et al. 2007 (PMID 17909210), but this DOI actually points to a different article, namely Evanoff et al. 1987 (PMID 3827426). Somehow the DOI bot got the Archives of Internal Medicine confused with the Annals of Internal Medicine? Eubulides (talk) 01:19, 4 July 2008 (UTC)[reply]

A false positive from the database the bot queries. I'm afraid the only thing to do in this instance is to add a hidden comment in place of the doi - something like this:

| doi = <!-- explanatory note -->

The database allows a little flexibility when matching the names of journals so it can pick up abbreviations and mis-spellings; unfortunately these two journals are so similar it would appear that they are equated! Martin (Smith609 – Talk) 09:40, 4 July 2008 (UTC)[reply]

• Aaack. I'm afraid this solution is unsatisfactory. How many invalid DOIs will be inserted because of this mess, DOI bugs that will confuse the readers and are not likely to be caught by editors? The only reason I caught this bug is because the other problem caught my attention. Surely there's a better way.
• Can the DOI bot be modified so that it has less flexibility when matching journal names? It's better to be safe and not insert bad DOIs.
• Or at least can a special case be made for this particular pair of journals, so that they don't get confused?

Eubulides (talk) 09:51, 4 July 2008 (UTC)[reply]

The scale of the problem seems to be very small. As the matching of journal names is performed automatically by the CrossRef database, I don't have any control over it. I will try to implement a way for the bot to check how close a landing page's metadata is to that in the citation (which may take a while to get right) - this is probably the only way of catching "wrong" DOIs, but even this isn't 100% reliable. Martin (Smith609 – Talk) 10:00, 4 July 2008 (UTC)[reply]

Hello Smith609, may be you remember me. In December 2007 I had ask you to use your script on the de:WP. The script worked very well until recently a guy on the de:WP moved the parts of the script to the standard Vorlage:Grafische Zeitleiste/... (template:Grafische Zeitleiste/...) Now it does display anymore the time scale on the left margin. Several templates e.g. [1] or [2] are mutilated and partly useless. I asked the guy who moved the parts of the script to fix this mistake it, but he must confess he couldn´t find what´s wrong it. Do you have any ideas, why the time scale isn´t displayed anymore. May be you know what´s wrong. Here the link to the Category [3]. It would be nice if you could help us. Otherwise I have to revert all actions of this guy. Best wishes. -- Engeser (talk) 09:09, 7 July 2008 (UTC)[reply]

In response to a query at Wikipedia Talk:Citing sources#Reference Space, do you know of a DOI bot-like bot that handles ISBNs, to provide similar aid in citing books? Along the same lines, a {{cite isbn}} could be a very useful counterpart to {{cite doi}}. (Apologies if it seems like I'm volunteering you, but I don't have the expertise to work on such a project, and I imagine you might be able to provide some suggestions.) ASHill (talk | contribs) 16:47, 7 July 2008 (UTC)[reply]

You can try {{cite book}}. "isbn" is a parameter which can be set. Have you tried [diberri's tool? It pulls a fully formatted cite book template from just the ISBN. --Rifleman 82 (talk) 17:48, 7 July 2008 (UTC)[reply]

Or my Universal reference formatter now does the same job. The "Cite DOI" template has been discontinued (a drain on resources, apparently), and while an ISBN version would probably be even more useful still (given that the same book is probably cited a lot more often!) I suspect the same argument will be raised.

However I think that's the answer to a different question to the one you asked. It would be pretty simple to make the DOI bot work for ISBNs too - feel free to add a proposal on my behalf at the WP:BRFA ( I don't have the energy for bureaucracy at the moment, but would be happy to do the coding.)

I do first need to spend a little time at Template:citation, however, to ensure that it uses the same parameters and formatting as Cite Journal and Cite Book. Martin (Smith609 – Talk) 19:04, 7 July 2008 (UTC)[reply]

I've added a timeline to Microbial mat in preparation for a section about the role(s) of microbial mats in evolution. The trouble is that I can't get the note to use the full available width of the box - and I need that, otherwise notes overlap badly around the start of the Cambrian. Can you help? -- Philcha (talk) 21:55, 9 July 2008 (UTC)[reply]

To get the colours right. Besides being revolting, the standard ones are a terrible background for text. -- Philcha (talk) 18:40, 15 July 2008 (UTC)[reply]

Go ahead, but I think you'd save yourself a lot of work in the long run if you documented the template properly. -- Philcha (talk) 18:52, 15 July 2008 (UTC)[reply]

Is there a "min-width" for images? -- Philcha (talk) 09:05, 16 July 2008 (UTC)[reply]

Microbial mat just got into DYK - you're a bad influence, I never used to be such a publicity-seeker! Philcha (talk) 16:45, 13 July 2008 (UTC)[reply]

Greetings. I left you message at User talk:Smith609/Cite#Bot collaboration about a bot that I run, asking for advice or source code. Since then I've been researching what CrossRef offers, and I've read Wikipedia:Bots/Requests for approval/DOI bot and Wikipedia:Bots/Requests for approval/DOI bot 2 with great interest. I think I can narrow down my request to these questions:

1. Given a link to a journal article or abstract, such as this or this, what regular expressions do use to find the DOI on the page?
2. Given a DOI, how can one use Crossref to determine authors, title, publication, etc.?

Thanks again, – Quadell ^(talk) 23:36, 13 July 2008 (UTC)[reply]

No offence meant, but I've undone your last change to Kimberella. The GA reviewer requested, nay demanded, an explanation of protostome & deuterostome. I spent 2 hours searching and most explanations are uninformative - probably scared of being proved wrong if they said anything informative, as the subject is a can of worms (boom! boom!). Eventually I found Arendt et al, and let him have it. If you want a bit of fun, see the last few paras at Talk:Kimberella/GA1. PS the reviewer is a actually a very nice, helpful guy, but I don't think he's familiar with the minefield that is Precambrian palaeo.

PS What could a microbial mat have done to achieve infamy? My mind is boggling. -- Philcha (talk) 13:20, 15 July 2008 (UTC)[reply]

But I like your latest addition. you obviously know these sources better than I do. I just noticed that it was up for GA and thought the biggest missing pieces ought to be added asap. -- Philcha (talk) 18:56, 15 July 2008 (UTC)[reply]

The version at User:Philcha/Sandbox#Time_line_for_Kimberella now behaves nicely when I CTRL+ in K-meleon. Many thanks! I'll ask the GA reviewer to check it out too.

I'm not sure what your edit comment "long lines don't wrap" meant. Can you explain, please? -- Philcha (talk) 23:12, 15 July 2008 (UTC)[reply]

I had to add some BR tags to the legend entries, as the lines made the div fatter instead of wrapping to a new line. I guess I could get around it by making the enclosing div's width default to a certain number of em rather than px. Martin (Smith609 – Talk) 23:16, 15 July 2008 (UTC)[reply]

I'm sorry the problem, and thank you for alerting me. Much appreciated, and happy editing, Leonard^(Bloom) 15:08, 16 July 2008 (UTC)[reply]

There's a new version at User:Philcha/Sandbox#CSR_mark_2, including undermat miners. What do you think? -- Philcha (talk) 12:09, 17 July 2008 (UTC)[reply]

I just opened an account with them. The say "CrossRef provides an OpenURL interface for its key query services and this service is available at no charge for non-batch, non-commercial queries." Does this bot qualify as non-batch? VasileGaburici (talk) 13:52, 17 July 2008 (UTC)[reply]

Yes - I'm pretty sure I checked everything out at the time, and queries are run at a rate comfortably lower than their limit. Martin (Smith609 – Talk) 15:31, 17 July 2008 (UTC)[reply]

I've used all the sources I've found, and think Microbial mat now looks like a usable article. Would you like to give it a look over? -- Philcha (talk) 15:30, 17 July 2008 (UTC)[reply]

Hi! This is Dale (User: Dale S. Satre). Thankyou for your comment on My article "The 'Big Five.'" It was grately appreciated. However, since I am new, can you direct me on asking for opinions?--Dr. Dale S. Satre (talk) 18:01, 18 July 2008 (UTC)[reply]

Hi, you've already noticed that I've started on Small shelly fauna. The timeline's a pig, because there's so much happening 530-520 MYA. Any ideas on how to improve it? In particular is there any way to make Template:Timeline skip the least relevant part of the Ediacaran, 620-690 MYA? -- Philcha (talk) 19:28, 18 July 2008 (UTC)[reply]

Thanks! I opted for the earlier of your options as the representation of the episodes mentioned was more consistent. Is there any way to reduce the dead space on the right without afftecting the rest of the layout? -- Philcha (talk) 20:29, 18 July 2008 (UTC)[reply]

  :-)  :-)  :-)  -- Philcha (talk) 21:18, 18 July 2008 (UTC)[reply]