User talk:Mjodonnell

From Wikisource
Jump to: navigation, search


Hello, and welcome to Wikisource! Thank you for joining the project. I hope you like the place and decide to stay. Here are a few good links for newcomers:

Please consider putting a brief description of yourself on your user page. If you are already a contributor to another Wikimedia project, such as Wikipedia, please mention this on your user page so we know how to contact you. Also, mention which languages you understand if English is not your first language.

In any case, I hope you enjoy donating your time to grow the Wikisource library that is free for everyone to use! In discussions, please "sign" your comments using four tildes (~~~~); this will automatically produce your IP address (or username if you're logged in) and the date. If you need help, ask me on my talk page, or ask your question here (click edit) and place {{helpme}} before your question.

Again, welcome! John Vandenberg (chat) 13:04, 11 February 2008 (UTC)

We aim to be very anal retentive, but we have an large percentage of texts that are "dumped" here by Wikipedians. The Germans have a policy of refusing new texts which are not accompanied by page scans. oldwikisource:Wikisource:ProofreadPage_Stats shows that the English project is a long way behind the Germans and French projects. There is some help at Help:Side by side image view for proofreading. John Vandenberg (chat) 05:25, 12 February 2008 (UTC)

Hey! this looks good. Sorry I haven't been responding promptly to your various ideas and suggestions. I tapped out a long message last night, but then closed the window, blissfully ignoring Firefox's lovely "are you sure you want to close 50 tabs" warning. It looks like you're moving ahead quickly without assistance; I'll try to catch up and add whatever whatever comments are still relevant to Talk:The Ballad of Reading Gaol. John Vandenberg (chat) 02:05, 14 February 2008 (UTC)
Regarding uploading, please read through Help:Adding images and Wikisource:Image guidelines when you have time.
In short: we prefer to upload media onto Wikimedia Commons.
That said, please dont stress about your existing uploads onto Wikisource - you havent done anything wrong, and you may continue to upload as you have been uploading if you find it easier to do that - I can have those images transferred to "Commons" in a jiffy.
Commons is a "sister project" that collects and organises media that may be used on any project in the Wikimedia Foundation fold. If you are enjoying working on Wikisource, you'll probably want to register an account on the Commons project as well.
The "Commons" project only accepts media that is "free-libre" both in the US and the country of origin, so there are still occasions when images need to be uploaded to the local project.
John Vandenberg (chat) 02:51, 14 February 2008 (UTC)

Continuing GIMPing work, no stress[edit]

Well, I'm a tenured full professor ("full of what?" one should wonder) in a ridiculously run CS department (some redundancy there), so I don't stress about breaking rules. I get anal about the content, but not about my methods along the way.

For, now, I'm just posting images in the discussion here as drafts, and I'll try to absorb the WikiMedia Commons method as I go along.

--Mike O'D 04:12, 14 February 2008 (UTC)


fyi, if you want to set up experiments and trial ideas, you can do them under "User:Mjodonnell", such as on a new page "User:Mjodonnell/sandbox". John Vandenberg (chat) 10:13, 20 February 2008 (UTC)

technical discussion[edit]

Hi, The Ballad of Reading Gaol is now using the new text, with a few minor technical issues. The path you were taking to fix those problems was horribly complex, which makes it difficult to teach the next newbie. We need better software. :-)

Yes, I'm glad I only fooled around with one page. --Mike O'D 23:22, 20 February 2008 (UTC)

To that end, I have started a discussion here: Wikisource:Scriptorium#Poems_that_flow_across_multiple_pages (that page is where any discussion that affect the entire project occur). John Vandenberg (chat) 18:01, 20 February 2008 (UTC)

I may poke into it. But, I think that the whole software/data-structure base just lacks a clear model of transclusion (which I think is several Ph.D. dissertations away from being understood, anyway), and the specific case of the poem may not be all that important. --Mike O'D 23:22, 20 February 2008 (UTC)

interleaving comments[edit]

Just an fyi, it is bad practise to modify another users comment by interleaving your comments into their comment, such as this. It makes it hard for 3rd parties to comprehend how the discussion flowed. Sometimes it is useful, but in general it is not done. Also, if you want a real-time discussion, I am nearly always available on IRC channel #wikisource. John Vandenberg (chat) 03:42, 21 February 2008 (UTC)

got it[edit]

Thanks for the tip. I actually yearn for an even more disciplined style of discussion, such as the old dead HyperNews. Thought that the Wiki game was trying to be looser. I'll try to unravel things as I notice them---won't be offended if you do it yourself. Generally resort to chat as a last resort, because it doesn't leave a record--Mike O'D 03:47, 21 February 2008 (UTC)

There is no need to change it now. I'm just letting you know the "norms" so you can settle in here. John Vandenberg (chat) 03:54, 21 February 2008 (UTC)

moving on[edit]

The Ballad of Reading Gaol is 90% done, and it is a 500% improvement on the previous version of the page. The law of diminishing returns kicks in now. I strongly recommend that you tackle another few projects and come back to this one later. With each project you will face a different set of problems; each will give you a better understanding of our community practises and technical wiki features and issues. This experience will prepare you to come back and apply the polish to The Ballad of Reading Gaol. John Vandenberg (chat) 06:13, 22 February 2008 (UTC)

excellent advice—quick audit of dangling problems[edit]

Sounds right. I'm familiar with small project groups (typically size 1) with more rigid goals, so the last 10% problem tends to dominate. In this collaborative regime, it makes sense to drive the time economy by return on investment.

I'm making a final mental scan for those things that are likely to be harder to fix later than now, while we have some things in cache memory. They are all in the category of very easy now, and modestly harder later, rather than big time bombs.

  1. I'd like to make sure that everything about the provenance that might be helpful to a future scholar/user is easily traceable. That may already be the case, but you pointed me to the Dejavu, and I'm not sure of all of its background. I gave a quick look at what seemed to be the source in Internet Archive, and didn't see a very thorough treatment of provenance there. Even if it is there, a few really careful lines on provenance are worth copying over, in case the URLs change.
  2. All of the relatively raw materials (nothing in information processing is quite raw) should be easy to survey. To this end, I prepared an appendix with an index of the images that I extracted. The appendix disappeared, apparently due to {{simultaneous editing by you and me. I will probably restore it. BTW, I use Wiki in classes that I teach, and I always keep a "Miscellaneous" page to throw pointers and uploads while pondering how to integrate them.
  3. A large part of the end of the discussion disappeared. I can't see any explanation in the edit history at the list level (didn't look into all of the differences). Your comment on the links to individual pages was among the lost ones. I was able to read it through the saved differences, but I don't see it on the discussion page itself. I noticed that the discussion page was flagged as "too long", so perhaps a splitting operation ganged aglay.

I should probably just lay off Wikisource for a week, hit other things entirely. Then, I would be willing to take the collaborative equivalent of assignments, which I guess we'd call "hints how to be most useful." I think that I was pretty efficient at the transcribing of poetry. I will probably have difficulty with very long works, since I have a weird neurological problem that sabotages my concentration from time to time. I find transcription an interesting way to stimulate a novel style of reading.

Personal anecdote: I stuck my nose/foot/whatever into Reading Gaol because I started out wondering whether I could sing the thing through in the style of a ballad. I found a version of "Lord Derwentwater's lament" on an old vinyl disc, with a tune that seemed adaptable. Was working on a printed lyric sheet. You know the rest. I may sing the thing yet.--Mike O'D 17:33, 22 February 2008 (UTC)

candidate project[edit]

I found the first page of A.A. Michelson's manuscript notes on measuring the speed of light: [1]

Copyright is not a problem: Michelson died more than 100 years ago.

I'm well qualified to transcribe this work:

  • I know the science,
  • I can read the handwriting,
  • I know how to cook up the interleaving of formulae/diagrams,
  • I have my own personal high-quality photographic copy at home.

The main trouble: I can't find the remainder of the work online, although it has clearly been photographed at least once.

Worth further thought. --Mike O'D 00:10, 23 February 2008 (UTC)

That is a good "next project", as we dont have any works by that author, so you will need to do some research into what material already exists. The first step is to create "Author:Albert Abraham Michelson" using the template {{Author}}.
Note that an edition already exist at Project Gutenberg. Does 1882 sound right for the original publishing date? OCLC:998616all editions.
More tips to come... --John Vandenberg (chat) 01:45, 23 February 2008 (UTC) has two copies of "Light waves and their uses" (1903), one from University of Toronto (contributed Dec. 07) and the other from University of California (contributed Feb. 07). I have downloaded them both, quickly checked for differences and found very little (except the Toronto copy appears to have clearer colours in the plates at the end), so I've uploaded the Toronto copy to Commons (Image:Light waves and their uses.djvu), and set up Index:Light waves and their uses.djvu. Note that you do not need to type this up; has the text which is roughly proofread, and the DJVU file also has a text layer which you can copy using most DjVu software, such as this. John Vandenberg (chat) 02:34, 23 February 2008 (UTC)
fyi, my bot has populated Index:Light waves and their uses.djvu, using the OCR text in the DJVU. John Vandenberg (chat) 04:19, 24 May 2008 (UTC)

I saw the Project Gutenberg text, which looks very good (better than other texts I've used from them). But it's a different item. I have a copy of the manuscript notes, which were revised into the published version that someone transcribed for Gutenberg. Substantial sections are common to both, but each has material missing in the other. The notes were never really published openly (that I know of).

I have a facsimile edition that preserves the grease stains on the cover, which appears to have been fashioned from a grocer's bag. It was produced in 1964 by Honeywell in Minneapolis, apparently to show off their capabilities. I found a copy in a used book store in Baltimore in about 1984. Once, years ago, I tried to find those responsible for creating the copy, in hopes that they had done other similar projects, but I couldn't trace them at the time.

The manuscript notes have errata that are especially interesting to those who care about the logistics/economics of science, because they show how careless he could afford to be about some things (easily correctible) while he was obsessive about others (impossible to correct after the fact).

The other book that you found is cool, but quite a different item, covering the basic science rather than the specific experiment.

Anwyay, I think that the manuscript notes are particularly wonderful for scholars, but I don't know where to find a complete set of photos for Dejavu. I could do a transcription from my copy, at the cost of it being less proofreadable. I don't have the resources/skills to scan my copy. I'd be happy to have it done once I figure out how. BUT, that might violate the copyright on the 1964 publication of that particular copy? I don't understand the law on that sort of thing. I'm confident that I can use the underlying public domain text, even though I read it in a copy. But I don't know whether I can copy the copies themselves. Hoped you might have some insight into discovering whether there is online facsimile somewhere.

Cheers, --Mike O'D 03:13, 23 February 2008 (UTC)

I realise that those are not the same edition as the copy you have. The page "Author:Albert Abraham Michelson" should contain a list of as many works by or about Michelson as you have the dedication to find and list.
I'll have to think about the copyright issue. John Vandenberg (chat) 03:28, 23 February 2008 (UTC)

I found another copy of the first page, photographically superior, at [2]. I wrote to the supervisor of the Naval Academy web site asking whether he/she knows of sharable images.

A Google search for "velocity light michelson osa facsimile" turned up the response:

To obtain the complete angular resolution vs light account with ... journal a splendid facsimile of Michelson's experimental note-. book on his 1879 velocity of light measurements. This had ... - Similar pages

But the link took me to a [3] explaining that I needed an account to proceed.

Examining the materials that came with my facsimile, I found that Honeywell contracted the printing to Lund Press, Inc., Minneapolis. But I doubt anyone will be frisky enough to call them and chase the unlikely chance that they'll have a useful lead.

--Mike O'D 03:47, 23 February 2008 (UTC)

Ideally, we want page scans, which can be normal images obtained using a typical scanner. The images do not need to be bound into a DjVu file. For example, Index:History of West Australia is using standard JPG files that were uploaded onto Commons. How many pages is this Honeywell edition? Perhaps you can ask for assistance at the library.
If you cant obtain page scans, or it turns out that redistributing them is illegal, the English project does still accept normal transcriptions without images.
The 1964 Honeywell edition is by default protected by copyright; had it been printed in 1963 it would probably be free, because they are unlikely to have renewed their copyright on it. See {{PD-US-no-renewal}} for details. However, if it is primarily "sweat of the brow", i.e. substantially free of any creative elements, then they have no right to assert copyright, as that is called "copyfraud". Bridgeman Art Library, Ltd. v. Corel Corp. touches on this. John Vandenberg (chat) 04:14, 23 February 2008 (UTC)

Well, I've rather spectacularly failed to ignore Wikisource for a week. At least I haven't transcribed today.

The notebook has 113 pages.

I'm not confident of my judgement regarding copyright. The facsimile is intended to capture the original (so it's not novel in that sense), but they were creative (in the intuitive, perhaps not the legal sense) in doing that. They used separate techniques to produce the different inks: mainly Michelson's pen vs. the blue lines from the notebook paper. They did a remarkable job on the crappy cover, which seems to be cut from a grocery bag.

But the original document seems to be reasonably accessible, probably at the Naval Academy, maybe at Clarkson, possibly at my own U of Chicago (but I checked an inventory of Michelson stuff here, and it had a much later repeat of the experiment). It would be much better to get 1st-level copies than copies of a somewhat whimsical copy. I'll at least wait for a reply from the Naval Academy.

Personal anecdote: my office is in Michelson's old U Chicago building—conceivably he may have used the room (U Chicago sensibly discourages sentimentality about office identities). A wooden duct used in a repetition of the ether-drift experiment runs through my office, and holds ethernet cables now.

--Mike O'D 05:33, 23 February 2008 (UTC)

It sounds like there efforts resulted in a lot of sweat, but their technology and effort was not to create something new - it was to faithfully reproduce.
The note that copper now runs through the ducts is quite amusing. John Vandenberg (chat) 06:45, 23 February 2008 (UTC)

Good to see that this next project is moving ahead!

You need to create an "Index" page to link all of your "page"'s; e.g. Index:Experimental Determination of the Velocity of Light. Also, as each page is prefixed with "Page", the use of "page" again at the end is superfluous. i.e. I think Page:Experimental Determination of the Velocity of Light/page 7 should be moved to Page:Experimental Determination of the Velocity of Light/7.

p.s. on [4], it links to 404, instead of [5] John Vandenberg (chat) 05:14, 13 March 2008 (UTC)

Thanks for the tips
  • I'll create the index page this evening.
  • Before renaming the Page pages, I have one thing to ponder. In addition to the pages of handwritten text, which Michelson marked "Page 1", "Page 2", etc., and I just copied his mark, there is a cover with a hand-printed title, and some front and rear matter, which I think is all blank, but I haven't looked closely yet. When you created the "Reading Gaol" transcription with DejaVu, the page numbers were just sequential from the photos, including cover and blank pages. They disagreed with the page numbers in the book. I was pondering whether to use labels, such as "Front Cover", "Front Matter 1", or some shorter abbreviations, or whether to number sequentially through all page and cover images as in the "Gaol" project. Not a biggie, and I'll settle it one way or t'other before there are too many pages to rename.
  • While we're thinking about names/titles, I think that "Experimental Determination of the Velocity of Light" will collide with the title of the published article, which surely belongs on Wikisource to compare with the manuscript. I started out thinking of the manuscript under the title "Velocity of Light", which is printed on the cover, but I decided it's better to view that as an abbreviation of the full title, which appears at the head of the first page of text. I could add "(manuscript)" or the date "1878" or something. The date is a bit fuzzy: the booklet included with my copy gives 1878 as the date of the experiment, but doesn't mention a date for the manuscript, which is neat enough that it may be rewritten from rougher notes. I think I even saw the experiment attributed to 1879 somewhere else.
I expect to have good images eventually. Having received no reply to my email to the Naval Academy, I've pinged a colleague in physics and another in optics for leads. I'll probably call Naval Academy Library on the phone next week and see what I can get from them. I found that the American Institute of Physics has a specific program to promote preservation of scientific source materials ([6]). I'll probably contact them at some point, and perhaps the link should be incorporated into Wikisource documentation somewhere.
As a last resort, I'll find a way to scan my own copy, but I'm pretty sure that much better quality is available somewhere in a first-generation picture.
Thanks for the tip on my Habeas Corpus link. Another case of incorrect copying of a relative link.

--Mike O'D 15:45, 13 March 2008 (UTC)

Besides the need to create an "Index:" page, the other page names are best decided by yourself, as you have the physical edition in your hand.

You may also find that your uni library may be interested in helping; they may have the facilities to scan it using equipment such as this. John Vandenberg (chat) 22:09, 13 March 2008 (UTC)

I moved pages to eliminate the "page" redundancy, and to distinguish the "1878 manuscript" from the published article of the same title. I created the index page Index:Experimental Determination of the Velocity of Light (1878 manuscript), and linked it into a "textinfo" template at the top of the main document discussion page. I modelled the Index page on the one for Reading Gaol, using <pagelist/>, but it didn't appear to work. Left it alone for now, since it's not a good idea to list the pages manually in too many places.

--Mike O'D 01:05, 14 March 2008 (UTC)

<pagelist/> only works for djvu files, so I have added a list of pages to the index.
John Vandenberg (chat) 02:12, 14 March 2008 (UTC)

a complex case[edit]

So you an see how detailed we can get, I have revised our main edition of Oh My Darling, Clementine using the page scans in Index:College Songs (Waite, 1887).djvu, pages 22 and 23.

As you can see, the text can be divided into in a million little pieces by creating "sections", and then those sections can be used in a variety of different layouts. John Vandenberg (chat) 19:46, 27 February 2008 (UTC)

gimp work[edit]

Index:A selection of war lyrics with illustrations on wood.djvu is a nice little collection with images that we can use on our pages. e.g. All Quiet along the Potomac. If you want to, it would be nice to trim those images and upload them onto commons, and then we can use the cropped image instead of the image of the full page. John Vandenberg (chat) 19:46, 27 February 2008 (UTC)

proposed deletion[edit]

Please look at Wikisource:Proposed_deletions#Index:Experimental_Determination_of_the_Velocity_of_Light_(1878_manuscript).

The way you have used the page namespace ignores the basic purpose of that system: to provide side-by-side comparison with the original. Can you get the document scanned an uploaded! Chris55 (talk) 13:49, 26 August 2012 (UTC)