Hacker News Reader

Comment by throwanem | Hacker News Reader

> ...each Standard Ebook is lightly modernized to feature consistent and modern spelling and hyphenation, so old-fashioned ephemera doesn’t distract you from timeless content.

I like old-fashioned ephemera...

View on HN

Yeah this is something I find a bit awkward. Good typography is definitely a boon, and fixing obvious errors (often mistakes that may not even have been present in each edition) like 'tne → the' is fine, but this modernisation step oversteps the boundary of keeping the original intent and content as the author envisioned it.

If the text of an older work presents problems for a lot of modern readers, I would prefer an annotated version with a glossary or even full-blown explanations of antiquated customs, concepts, and words.

And I agree, it also takes away some of the flavour of an older book.

Not to sound too negative — the project itself looks interesting and of lasting value.

> like 'tne → the'

I know it's not what you meant, but one thing I'd genuinely love from a project of re-typesetting old public-domain works is to ensure that any "typewriter symbols" like "->" or "* * *" are turned into their proper dingbats ("→" and "⁂", respectively.) Also, ensuring the (double and single) quotation-mark characters are used, but that scare-quotes and actual measurements (6'1", etc.) don't use them. Just, general heavy-handed replacement of ASCIIisms with proper Unicode equivalents.

Though, once you start down that road, it brings you to interesting places: if a book says "ye olde shoppe", do you want to keep the "ye", or do you want to consider it a typographical limitation for what should have been properly typeset as "þe"?

The typographical equivalent for thorn is 'th'. You would therefore want to do "the" -> "þe".

> Also, ensuring the (double and single) quotation-mark characters are used, but that scare-quotes and actual measurements (6'1", etc.) don't use them

Scare quotes are a adaptation of quotation marks for use/mention distinctions, and shoukd use proper quotation marks the same as anything else; measurements should use the prime and double prime characters.

I've seen 'ye' for 'the' on tombstones from the 1600s. The stone carvers weren't picking type out of a box.

Agreed, maybe they could include some way of annotating the original content and display a modern version, to preserve the original work but also make it more accessible.

Here's the tasteful modernization script [0], it contains a ton of regexs (cause regexs never go wrong) with such tasteful changes as:

    # pease -> peas (but "pease pudding")
    # régime -> regime (but "ancien régime")

The last section of the script gets more interesting

    # manœuvre -> maneuver (for the Americans)
    # manœuvre -> manoeuvre (for the British)

    # cosey -> cozy (for the Americans)
    # cosey -> cosy (for the British)

I've never seen the word manœuvre and thought that I wanted it spelt any other way. Is it really just going to stop at cosy? What's so special about cosy that it needs converting but not other words.

Further they have script to [1]

> Try to convert British quote style to American quote style

So this is not 'standardized', this is 'Americanized'.

I was going to make a joke about how they'd probably try and get rid of the pesky 'e' at the end of Shakespeare, however it turns out, instead they actually have the opposite rule:

    Shakespear -> Shakespeare

[0]: https://github.com/standardebooks/tools/blob/master/moderniz...

[1]: https://github.com/standardebooks/tools#british2american

One can way overdo the modernization. From "The Message", a modernized Christian Bible:

"God, my shepherd! I don’t need a thing. You have bedded me down in lush meadows, you find me quiet pools to drink from. True to your word, you let me catch my breath and send me in the right direction. Even when the way goes through Death Valley, I’m not afraid when you walk at my side. Your trusty shepherd’s crook makes me feel secure."

Don't go there.

I dunno other than Death Valley... well bridal mysticism has a long tradition.

I thought I was reading an overtuned ELI5 machine translation for a minute. This is awful and significantly more difficult to read than even the KJV, which isn't exactly friendly for modern readers.

It's a tough translation. That section starts "The lord is my shepherd", and continues with a sheep metaphor, with "green pastures" and water, the things sheep need. Sheep draw confidence from the shepherd; they're herd animals and the shepherd is their alpha. Sheep are very herd-bound, far more than cattle or horses; a little bit of guidance and the whole flock follows. They've been bred for docility for millennia; the ones that were easy to herd were kept and bred. The people who wrote that were writing for an audience which knew that; today's audience has probably never seen sheep being herded. Modern versions face the question of how much explanatory material to attach, or whether to try to express that in the main text. Scholars prefer translator footnotes; preachers don't.

Still, if you're not translating between languages, it's probably best to stay with the original. Compare, say, Kipling's "007".[1] This assumes some knowledge of railroading in the steam era. "Modernizing" Kipling would be a terrible mistake. It's probably best to leave anything post-1800 entirely as original. And don't even try to "clean up" Shakespeare. It's been done.[2]

[1] https://ebooks.adelaide.edu.au/k/kipling/rudyard/days/chapte... [2] https://en.wikipedia.org/wiki/Thomas_Bowdler

"Death Valley". Wow.

Since they're keeping everything in git, maybe the modernized version could just be a branch---commits rebased on top of a more conservative base.

I've often thought it would be interesting to see an ebook format that would allow one to toggle between versions of the text or display text inline as some kind of diff.

I say this because there are some authors that I read who have revised their works many times, and sometimes I prefer an earlier version over a later revision.

I've been using this approach and it works amazingly well (for text-based formats like .tex and .md). I switched from mercurial to git because rebasing feels much more natural. I'm currently actively maintaining five different versions of a book in parallel as separate branches. The `master` branch sees the most fixes then I rebase the five branches.

Anyway, it's great to see books stored in git. Mark my words, this will be a big thing for education!

Most people find it hard to read books with long s, for example, or even know what long s is or where it occurs. "Why are all the s's replaced with f's?" people will ask. It's fine if it's "ash-tray" instead of "ashtray", but the further you go back and the more of these archaicisms you encounter, the more of a barrier they become to ordinary readers (instead of English majors). I don't necessarily like this kind of textual modernization, but in general it's necessary to keep the text accessible to moderns.

I'm not sure I can agree–seeing 'ſ', where I'd ordinarily expect to see 's', was briefly confusing at first, but I quickly derived from context what it must mean, and found my surmise confirmed in further perusal of older texts.

On the other hand–I'm not sure that someone, who has written an Emacs minor mode to perform automatic ſ-insertion where it is grammatical to do so, is necessarily best placed to speak to the general case here.

I don't think the parent poster was talking about being confused as to the meaning; rather, they're talking about the presence of these characters impeding fluent reading speed even after one knows what they are. Enough of these changes and you feel like you have dyslexia, a bit: you have to consciously analyze each word on the page to extract its "reading", rather than just letting your visual cortex pass the words straight into your brain's audio loop.

Seems like it'd be just a matter of developing fluency, the same way as with any other orthography. Adding one new glyph isn't that big a problem for someone who's already likely to be reading texts where long-s is found - so, at least, has been my experience, and while I always hesitate to generalize therefrom, I don't know that it's so unreasonable to consider doing so here.

> Most people find it hard to read books with long s, for example, or even know what long s is or where it occurs.

That's not old-fashioned spelling or hyphenation: that's old-fashioned typography.

> That's not old-fashioned spelling or hyphenation: that's old-fashioned typography.

A critical difference, because once you've corrected OCR errors so that you correctly have the spelling-as-in-original, but not done any “modernization”, there is no issue with long-s to addres.s

I personally would see this as a disaster.

I want the old language warts and all. I want to learn how words were expressed in a historical context. That teaches me more than just the words themselves, that transports you to their time.

I did a search in the Google Groups about 'modernization' and you get quotes like this coming up [0] (about William Wollaston's "The Religion of Nature Delineated"):

> I've got a first draft ready, after about a month's steady work. I could use some proofreading help, in three areas in particular:

> 1. general typos (the use of the long-s in particular I'm sure has led to several)

> 2. suggestions for improved use of commas. They did this weirdly in the 18th century, and I think a new edition could do well by bringing the practice up to date, but it isn't my strong suit

... and in reply ...

> then it sounds like we'll have to do some significant spelling modernization

So they've moved from just 'tasteful' to 'significant'.

I don't see how you can draw a line for this kind of thing. But I guess I don't know very much about the painful process up digitising old books.

[0]: https://groups.google.com/d/topic/standardebooks/etzVBgBo1VE...

I liked this note on Hodgson's The House on the Borderland:

The original print edition of this novel contains a pathological number of commas—so much so that a modern reader would find them distracting at best and plainly ungrammatical at worst. The editors of this Standard Ebooks edition have made an effort to remove the most egregious cases of these ungrammatical commas, so that modern readers can better enjoy this unique tale.

Yeah that was a hard editorial decision :)

I hope we did a good job, but if you disagree then all of the commas we removed were done in a single commit, that you can roll back: https://github.com/standardebooks/william-hope-hodgson_the-h...

You can also review the changes there to see if you agree with our judgment.

This book was a unique case. I think the only other time we've done a big editorial change like this was for Pride and Prejudice, also to remove some crazy commas, but for P&P there is a lot of precedent--other editions of Austen very often make those same kinds of changes.

If it's acceptable to edit a "classic" like "P&P" in that fashion, then surely it's even more so for a weird book like this one. b^)

Exactly. And today's "consistent and modern" form is tomorrow's old-fashioned ephemera. The form used at the time of publishing is relevant, like the content from that time.

Today's typography is not going to change in the manner it did in the past, for the same reason why dialects are less pronounced: the prevalence of mass media and the Internet makes things converge rather than diverge.

This little annoyance aside, I really appreciate the time and effort they put into this.

I agree, and I think it steps slightly and firmly over the line of changing the author's work. I really hope they "re-think" this.

What will they do with Jane Austen's inconsistent spelling, which was common at the time? https://duckduckgo.com/?q=jane+austen+spelling&t=lm&ia=web

Reply on news.ycombinator.com