Wikipedia:Reference desk/Archives/Computing/2021 June 22

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 22

Modern shape of digit 1

Look at this:

https://medium.com/@PostHasteCo/why-old-typewriters-lack-a-1-key-83d777f1e9d0

It says that until as late as 1948, all 1's were simply written as I's in all available fonts (this means that no 1's had an upper left serif with no upper right serif.)

But look at this:

https://www.google.com/search?q=%22World+almanac%22+1934&rlz=1C1CHBF_enUS782US782&sxsrf=ALeKk00MJ46riFF4--mVA8b7giHwczDxiA:1624326042695&source=lnms&tbm=isch&sa=X&ved=2ahUKEwissJiGjqrxAhXVSzABHbtLAN8Q_AUoAnoECAEQBA&biw=1366&bih=625#imgrc=i6ZaSaASIXtRhM

This is the front cover of a 1934 book, and it reveals a 1 that meets the criterion that the first link says didn't exist until 1948. Any corrections to what I'm saying?? Georgia guy (talk) 01:42, 22 June 2021 (UTC)

For clarification, I'm putting this on this reference desk because it's primarily about the reason early typewriters had no 1 key. The first of the 2 above URL's explains the reason, but the second appears to disprove it. Georgia guy (talk) 01:51, 22 June 2021 (UTC)

My 1944 copy of The Aftermath by Winston Churchill, which happens to be the nearest pre-1948 book to hand, has both forms - I in the text, 1 on maps. DuncanHill (talk) 02:04, 22 June 2021 (UTC)

And the Lloyd George Liberal Magazine from the early '20s of the last century has 1 in the page numbers, but I in the text. Clearly the article you linked to is wrong. DuncanHill (talk) 02:07, 22 June 2021 (UTC)

And my Universal Encyclopedia by J. A. Hammerton, again early '20s, has 1 in both text and page numbers. DuncanHill (talk) 02:11, 22 June 2021 (UTC)

And the 1914 advert reproduced in the article, supposedly supporting their hypothesis, includes at two occurrences of a 1! DuncanHill (talk) 02:15, 22 June 2021 (UTC)

Randomly checking Newspapers.com around the year 1900, the 1 with a serif is all over the place. ←Baseball Bugs ^{What's up, Doc?} carrots→ 02:50, 22 June 2021 (UTC)

The "Post Haste" writer may have been confused by old style numerals, which, as that name suggests, were used more commonly in the past than they are now (although they certainly haven't disappeared). In old style numerals a 1 typically does look like a small capital I.

As to typewriters, it was common right up to the time typewriters started disappearing that there was no 1 key on cheaper ones: having fewer keys saved money. My computer keyboard has "!" on the 1 key, but typewriters generally didn't have that character either: you could make one out of a period and an apostrophe. But as to 1, what people typically used was not a capital I, it was a small L. Here are a few examples I found by a Google Images search on "typewritten page": [1] [2] [3]. --184.145.50.201 (talk) 04:24, 22 June 2021 (UTC)

You can look at http://www.mrmartinweb.com/type.htm and see when typewriters started having 1 keys. --Guy Macon (talk) 06:38, 22 June 2021 (UTC)

• From that source you can see that typewriters started having 1 keys, but since it's one person's collection, you can't tell when it first happened. --184.145.50.201 (talk) 04:24, 23 June 2021 (UTC)

As well as ones and exclamation marks the zero was also omitted. You simply used capital "O" for "0". See Drummond, Archie; Coles-Mogford, Anne (1988). Typing, first course (5th ed.). London ; New York: McGraw-Hill. p. 33,35. ISBN 978-0-07-707036-6. (NOTE: If your typewriter does not have a figure 1 key, use small 'L'.) and (If your machine does not have a 0 key, use capital 'O' and 'L' finger.) for example. It was so common to use these substitutions that punch sheets in the 1970s provided space where you could put your manucript versions of "1" (one), "I" (uppercase i), "l" (lowercase L), "0" (zero) and "O" (letter) to make it clear to the punch operators what you meant. (See here for an example of a punch machine.) Martin of Sheffield (talk) 07:54, 22 June 2021 (UTC)

This may be right for typewriters, and I don't know much about handwriting conventions. For printing Arabic numerals I can assure you it is straightforwardly wrong and in several ways.
Up until the late eighteenth century, Arabic numerals were written as text figures, with the 1 at the height of an 'i'. So it's a different height to a capital letter. However, it was not simply a dotless 'i' character, as the top and bottom were symmetrical. Is it the same as a small-cap 'I'? Well, yes, I suppose it looks similar, but typefaces (e.g. some of Robert Granjon's italics) existed that had a '1' and didn't have small caps. So it may have been similar in form but clearly had different semantic meaning. Towards the end of the eighteenth century, people began to use typefaces with figures at the height of the caps. At this point a new form for the '1' emerges, which is asymmetrical and has a top serif only on the left. I'm not sure where this came from, presumably handwriting or lettering conventions. The Bell typeface is an example of this. Is this an 'I' being used as a '1'? No, it's asymmetrical. Is it then an 'L' in lower case? Not that either! It's shorter. The first "lining figures" were at a slightly different height to cap height.
This is a quick summary; I could give more information and sources if needed. I have occasionally seen lettering with the Arts and Crafts movement, influenced by Roman capitals and Roman numerals in which a '1' is treated as the same as an I and at cap height, e.g. Eric Gill's Perpetua Titling. But this would have no connection to typewriter manufacturers. I suspect someone simply wanted to save some parts. Blythwood (talk) 10:07, 22 June 2021 (UTC)

"simply wanted to save some parts" – exactly. Mechanical typewriters were heavy and bulky, so anything that saved both weight and complexity was welcome. My portable (c. 1974) had neither 1 nor 0 whereas an office machine of similar date had both. The office machine also had provision for different spacing whereas the portable was just a made. The big difference was that the portable could be carried in one hand whereas the office machine needed both hands! Martin of Sheffield (talk) 10:19, 22 June 2021 (UTC)

I've just had a look at Pétiet, Jules; Flatchat, E (1849). The student's guide to the locomotive engine : being a description of the different modes of constructing locomotives, details of their component parts, and the nature of their employment : with observations on the management of locomotives. London: John Williams & Co.. The numbers are of uniform height with upper case. The ones have an asymmetric top and the zeros are less bold and slightly slimmer than "O"s. A Church of Ireland prayer book of 1794 has the symmetrical ones of lowercase height and the descenders of 3, 5, 7 and 9 come below the line. Martin of Sheffield (talk) 10:55, 22 June 2021 (UTC)

DuncanHill, noticed your comment on the Medium post! Looking more closely at the Medium channel, I think it's some kind of prank. It's advertising a joke "app" that lets you send telegrams, and most of the posts seem to be in character, e.g. lamenting that nobody wants to telegrams any more. The article may be written in sincere belief that what it describes is the case, but I think it's more likely to be a prank. (The reason I fell for it is I previously have seen an article claiming that the modern '1' used in fonts with a left-hand serif was copied over from typewriters, which was wrong wrong wrong, but having seen it meant I was willing to believe this was a belief going round.) Blythwood (talk) 13:00, 22 June 2021 (UTC)

Yeah... the post is dated "April I"... Elizium23 (talk) 21:48, 22 June 2021 (UTC)

For what it's worth, the Blickensderfer No 5 typewriter introduced in 1893 is generally regarded as the first portable typewriter. I happen to possess one that was made in 1896, and it has all ten numerals (1–0) as separate entities from the letters, as can be seen in the first photo in the article.

What may not be obvious at first is that this model has two levels of shift, the first level ("CAP") for capital letters and the second ("FIG") for numerals and other symbols. (Mine has both shift keys labelled "SHIFT".)

Unlike the typewriters most people will be familiar with, it doesn't have the striking characters on the ends of type bars (and thus only one available typeface): instead it has a type wheel or drum, similar to the "golfball" of the much later IBM Selectric typewriter.

Usually Blickensderfer supplied at least two different and easily swapped typeface drums with each machine, and dozens of others were available, including with other alphabets for various foreign languages. The mechanism and key layouts also reportedly had distinct advantages over what later became standard. (I haven't yet dared to try mine out!)

By contrast, my Brother portable typewriter (similar to that article's first photo, but not a 'De Lux') dating from about 1970 indeed has no 1 or 0, requiring substition by l and O. It's also heavier and with a much less robust case – so much for progress!

(No, I don't collect typewriters; these are the only two I have :-) .) {The poster formerly known as 87.81.230.195} 2.122.59.177 (talk) 01:26, 24 June 2021 (UTC)

Jimmy Wales and Larry Sanger dispute over co-founder

Why Jimmy Wales credited as founder of Wikipedia, why not Larry Sanger in the first place and what reason to removed and eventually resigned to Wikimedia Foundation? The Supermind (talk) 09:56, 22 June 2021 (UTC)

the article you link to does say that both Jimmy and Larry created it. What are you referring to? RudolfRed (talk) 17:25, 22 June 2021 (UTC)

I believe that he is referring to "Wikipedia was launched on January 15, 2001, by Jimmy Wales and Larry Sanger" and asking why it isn't "Wikipedia was launched on January 15, 2001, by Larry Sanger and Jimmy Wales".

In my opinion the answer is in one of the sources cited:[4]

" Internet entrepreneur Jimmy Wales launched an online encyclopedia project, Nupedia, in the beige-and-gray Lamont Street building. It was Wales' vision to have an encyclopedia written by volunteers. You could count that as the beginning, since Wikipedia grew out of Nupedia.

Wales hired graduate student Larry Sanger as editor in chief of the Nupedia. During the project's excruciatingly slow start, Sanger had dinner at the Balboa Avenue restaurant with a friend. It was there that he was introduced to wiki technology, a type of software that makes possible communal editing of wiki Web pages.

Sanger says it struck him on the spot as a way to open up the writing and editing of the encyclopedia project and get it off the ground. So the restaurant could also be seen as the spawning ground.

Sanger later came up with the name Wikipedia, a second undisputed point in the history of the site."

Also see Sanger's resignation,[5] where he says

"I was more or less offered the job of editing Nupedia when I was, as an ABD philosophy graduate student, soliciting Jimbo's (and other friends') advice on a website I was thinking of starting. It was the first I had heard of Jimbo's idea of an open content encyclopedia, and I was delighted to take the job."

So Wikipedia was launched on January 15, 2001, by Jimmy Wales and Larry Sanger. Jimmy Wales came up with the idea of an online encyclopedia, and Larry Sanger, who was hired before the encyclopedia launched, came up with the name "Wikipedia". --Guy Macon (talk) 20:35, 22 June 2021 (UTC)

BTW, do we know who the friend was who introduced Sanger to wiki technology? --Guy Macon (talk) 20:38, 22 June 2021 (UTC)

@Guy Macon: Ben Kovitz [6] The Supermind (talk) 21:12, 22 June 2021 (UTC)

If you are going to include Ben Kovitz in this discussion, you should also include Ward Cunningham. 97.82.165.112 (talk) 14:59, 24 June 2021 (UTC)

Splitting a PDF file with hyperlinks

I'm looking for a way to split a PDF file into several parts, but keep the hyperlinks intact. What I want is this: If the hyperlink and its target are in the same part, the link should be kept as it was before, but if the target is in another part, it should be changed into an external link.

For example, I have a PDF file of four pages, with two links on the first page, one to the second page, one to the fourth. I now want to split it into two files part1.pdf and part2.pdf of two pages each. The first page of part1.pdf should still have two links, one internal link to the second page, and one external link to the second page of part2.pdf. All tools I tried so far failed in doing so. The best was PDFsam which at least kept the internal links intact (and also the document outline etc.), but even this failed to convert the links to other parts as I want it. --132.230.195.144 (talk) 17:26, 22 June 2021 (UTC)

I just did an experiment. I created a 3-page PDF. Each page has a link to itself and a link to each of the other two pages. I use pdftk to split the multi-page PDF into three separate files (pdftk threepage.pdf burst). Each of the single-page files has three links, one to itself and one to each of the other two files. So, it maintained the internal link and updated the external link. 97.82.165.112 (talk) 19:00, 24 June 2021 (UTC)

I tried to compile pdftk from source but failed (which isn't that surprising for a program that seems unmaintained for 7 years). Any alternatives? --132.230.195.147 (talk) 17:20, 29 June 2021 (UTC)

How to edit html files inside .rar file?

I am tired of extracting html, edit it, putting it back into .rar file. Is there any way to edit html files inside .rar file?

Here html file is offline word document file saved as web page file. Edit means opening html inside Microsoft office, add text. Rizosome (talk) 18:41, 22 June 2021 (UTC)

Rizosome, no; by definition, a RAR is an archive format. You don't say what OS you are using. In general, archive formats are compressed and sometimes encrypted. They tend to be sensitive to changes of length of the archive components, i.e. files. It is usually more trouble than it is worth to support in-place editing of archive files. Because archive files are meant, after all, to be archival, i.e. backups, or for distribution. Elizium23 (talk) 20:57, 22 June 2021 (UTC)

This can be done with .zip files, but AFAIK, there's no software out there that allows native browsing and file streaming from inside a .rar archive. If anyone with more experience with Win10 integration wants to make one, you can count me in, however. It sounds like a fun challenge, but I haven't done any windows integration since 3.0. ᛗᛁᛟᛚᚾᛁᚱPants Tell me all about it. 21:03, 22 June 2021 (UTC)

@Elizium23: I am using windows 10. Rizosome (talk) 15:21, 23 June 2021 (UTC)

Your business is your business, but this just seems like an odd thing to be doing. RAR files are archives; HTML files are meant to be read by browsers to display content. If you're editing the HTML to update the page, why then re-archive it? Nobody can see it then. This feels like an example of the XY problem; we might be able to help better if you provide more context. Matt Deres (talk) 18:48, 23 June 2021 (UTC)

Matt Deres, this is a good point. Also, it reminded me of the {{Xyp}} template I made a while back, specifically for this page. ᛗᛁᛟᛚᚾᛁᚱPants Tell me all about it. 19:21, 23 June 2021 (UTC)