Hacker News Reader

Comment by robin_reala | Hacker News Reader

As with every thread like this (seriously, tell me if you want me to stop) I’d recommend the project I contribute to: Standard eBooks: https://standardebooks.org/ . We take the Gutenberg transcriptions and prep high quality electronic editions, then proofread against the original scans to make sure all the formatting is properly represented, and finally release them as public domain again. Being limited to pre-1923 means that we’ve focused on the ‘classics’, but we’re always doing more and should break through the 200 mark shortly. Currently I’m producing a Woolf and Pepys’ Diaries (although that last one is definitely a long term project). Let me know if there’s a book you specifically want to see produced and I can add it to the list or help you produce it yourself.

View on HN

A warning: these books are not the original editions, but edited versions with modernized spelling and grammar. The editorial changes are kept in version control so they could theoretically be reverted, but they are mixed with the technical changes so in practice this is not easy.

Some people might prefer the changes, but when I read classics I want to read something as close as possible to what the author originally wrote. Despite the name, "Standard Ebooks" are actually non-standard.

To clarify, SE does not modernize grammar. 99% of the (very light) modernization we do is one-to-one spelling changes for wildly archaic spellings like "shew -> show" or "swop -> swap" or "to-night -> tonight". Grammar is left alone.

Whether or not you prefer slightly modernized spelling is a matter of taste, and something I've discussed at length in previous HN comments. (So I won't repeat myself here.) At SE we take the same stance as editors going back generations, who have all been quietly modernizing spelling right under everyone's noses as a matter of course since Shakespeare's days.

From Wikipedia's "grammar" article:

"Outside linguistics, the term grammar is often used in a rather different sense. In some respects, it may be used more broadly, including rules of spelling and punctuation, which linguists would not typically consider to form part of grammar, but rather as a part of orthography, the set of conventions used for writing a language."

In common usage, punctuation is often considered part of grammar. This matches well with the computer science meaning of grammar (rules for parsing), because punctuation can change the meaning of a sentence.

Of course. We do not change punctuation, with a few exceptions for some short-lived archaic constructs like comma-emdash.

We do change from single-quoted British style to double-quoted American style, but I think most everyone would agree that that is merely a presentational change that does not change the actual grammar of a sentence. (And that in and of itself was a very common thing for editors to do going back hundreds of years too!)

That seems an odd thing to want to change. Do you also move proximal punctuation inside the quotes, AE-style?

Doing something like this has the advantage of reducing the distance between reader and the translator or the author, but at the same time removes context. If it’s a high Victorian translation of a Greek text, what you’re reading is from ancient Greece, but given a Victorian cast, with all the ideological implications that brings with it. I’d rather you didn’t remove that context, but I can see why you’d do it.

Fantastic project, anyway.

‘Standard’ as in a standard imprint. The typical argument is that any print edition beyond the first is altered, especially with punctuation, according to the whims of the day. Even the first edition is rarely what the author originally wrote, having had a multitude of spelling corrections and changes from the original MS.

But yes, we do do this (sympathetically to my mind), and yes, it’s all marked as editorial in the commit history. It’s not trivial to untangle, but it’s not impossible by any means. We also list the sources used to build the books which include the transcriptions and scans if you want to find the unaltered versions.

(for what it’s worth, by far the most common adjustments are to-day to today, to-morrow to tomorrow, and good-night to good night)

I really enjoy SE, and I just found out about these corrections. Not really sure what to think about them but for trivial things like to-day I don't really see the point of correcting it since it's obvious to understand and gives some free authenticity. Just my 2¢; thanks for the project as a whole.

Also see LibriVox[1], where volunteers convert public domain books (many of them from Project Gutenberg) to freely-available audiobooks.

[1] - https://librivox.org/

That's wonderful and I will be recommending it to my friends from now on, but I've found in my experience of reading classic literature, it's at least sometimes much more useful to read such books with modern annotations, which is why I just check out annotated versions from my local libraries.

The Pepys I’m working on at the moment is stuffed full of footnotes (check the Gutenberg source); I’m currently nearing the 700th. These are being converted to a standard ePub endnote format, so compatible readers, i.e. most of them, will get popup modals with the annotation when you click on the note number.

Thanks for posting, awesome resource! I was surprised to see recent science fiction works and authors, like Phillip K. Dick and Algis Budrys, which I would expect had been published well after the 1923 copyright deadline.

They are, but there was a point in the 50s I believe when copyright in the US had to be renewed, and that didn’t happen for a bunch of sci-fi and pulp.

This is a great project! Sadly SE has a focus on ePub and similar formats. These are impossible from a typographical standpoint. Text in ePub reader apps always comes out with a ragged right or impossible word breaks.

Latex with the microtype package is the only solution that does text justification on the level of a human professional. Can’t SE convert to that and provide PDF for common screen sizes?

It would be so easy if only Project Gutenberg had a standardized master source format. So conversion tools can be built around that. But all efforts in that direction were killed of by trolls in the community.

All the source files of standardebooks.org books are up on Github, it might be possible to write a script to automatically convert the files to LaTeX.

Pandoc [0] could probably do most of the heavy lifting, starting from the ePub files (after all, an ePub is just an archive containing HTML, CSS, images, and metadata)

[0]: http://pandoc.org/

I spent a lot of time with Gutenberg epubs in a former life, and I found that their quality was far better than most other digitizers with comparable scale. However, the tables of contents were extremely inconsistent and I had to essentially rewrite them to get anything like a usable outline.

SE looks interesting. I don't have an epub reader handy but I will definitely be back to understand the motivation (knowing it's a lot of work to do this sort of thing).

SE books all have rebuilt ToCs, yes. Gutenberg’s ones follow the presentation of the original scans, which basically isn’t useful for an ereading experience.

This is amazing. Thank you.

In your experience what’s the most convenient e-ink device for free ebooks? Ie Kindle is optimized for amazon so probably not the most convenient for this.

It’s not the easiest to get SE books onto Kindle unfortunately, though we obviously supply them. I personally read on a Kobo, and I think they’re starting to be available in the US now? You can just drag and drop our kepubs into the mass storage and it’ll import them. Apple Books works well too if you’ve got an iDevice (although there’s a lingering bug with forced page breaks that I really need to fix).

Thank you! We homeschool and use a “classical” program focused primarily on grammar, rhetoric, and logic. Resources like yours increase our children’s access to classical literature while helping us reduce book costs (which can be surprisingly high).

Are the books supposed to work with Kindle Paperwhite? I am trying to email it to my account and Amazon is saying that it cannot be converted. Super cool project!

Have you read https://standardebooks.org/help/how-to-use-our-ebooks and tried it yet? Your device is listed!

I have yes. Emailing the file to an address associated with the Kindle is also a standard way to upload a book. That's why I was thinking that this must be some sort of bug. BTW I tried to upload Pride and Prejudice.

Might be late / irrelevant but a while back amazon was failing to convert anything i emailed to my kindle and it turned out to be that i wasn't specifying a subject line in the mail. once i did that the very same files started working.

Reply on news.ycombinator.com