gemini://birds-are-nice.me/
In an act of pure egoism, I submit my own site! I just finished the rewriting in Gemtext. Here you may see all the software I wrote (most of it useless) and some of the circuits I design (most of which you can buy as modules anyway), and all the essays I write.
birds-are-nice.me/
Posted in: s/discoveries
🐦 JustASillyBird
May 04 · 3 weeks ago · 👍 norayr, Homer
🌆 skyjake [mod...] · May 04 at 15:28:
(No self-promotion, please.)
Boo hiss.
🚀 stack · May 04 at 17:58:
Thank you. I am looking forward to reading your site when I have a few minutes...
dupdelete is something I always want to write... There are alternatives but each one I've tried had some issue..
🐙 norayr · May 05 at 21:35:
so many cool programs!
🐦 JustASillyBird [OP] · May 07 at 22:25:
Shameless I may be, but you cannot deny the site is interesting to the type of UberNerd who you are likely to find in this community. See, I already have positive comments! It's a proper old-school personal programming collection, like we had in the days before every app was expected to come with a Discord community.
🦂 zzo38 · May 07 at 23:31:
I read the article about PDF.
I agree that PDF is bad for many reasons, although I do not agree with the suggested alternative of using JSON, XML, and UTF-8, which I think are also bad (also, some parts of PDF files do use XML, but most don't, so it is rather messy). I think that Unicode is not a good character set (and it is also not suitable to use one character set for everything, regardless of which one it is), and that DER is a superior format. (That JSON, XML, Unicode, PDF, HTML, etc are common does not inherently make them good.) I might use a archive format with a DER for each page, and another one for the table of contents.
Reflow is mentioned. Reflow is also possible in PostScript, although the PostScript code must be written to do this, and the reflow will be lost when converting to PDF (although your PostScript file might support an argument to specify the reflow width, and then the PDF will have that width). PostScript is good as a programming language, but it is not good as a replacement for PDF. However, an implementation of PostScript could be made which can support output to whatever formats that you might want as an alternative of PDF.
EPUB can be an alternative for books but I would not really call it an alternative of PDF since it is not made for positioning on a page (although this is possible), although EPUB is a ZIP archive containing HTML files so it involves the complexity of HTML as well (although this does mean that it can be opened in a web browser that supports the "jar:" scheme, such as Firefox, so a separate EPUB program is not strictly required). Gempub is similar but using Gemini text instead of HTML, which makes it simpler.
OpenXPS is mentioned as well. It says it lacks cryptographic signing, interactive forms, DRM, and encryption. I think that encryption and cryptographic signing should not be needed since that feature should be done using a separate format (I think password protection should also be done with a separate format if needed, and 3D graphics should not belong in the format intended for printed documents, and DRM should be avoided entirely). Also, it uses XML which I mentioned above as being not very good in my opinion.
Plain text is mentioned as an alternative, and for many uses it is; plain text, especially plain ASCII text, is very portable and is compatible with most computers. I mostly agree with that you should not need PDF for plain unformatted text, although there are sometimes uses of this, e.g. that you want to transfer such a file to another computer to print and want to ensure the width, fonts, page numbering, etc is how you want it to be formatted (I have done this before, using a PostScript program that I wrote in order to produce the PDF from the plain text file); however, there is not much point to publish such a PDF file (it would be better to use plain text file instead). (The Z-machine interpreter in PostScript can also produce such PDF files when you enable transcripting (normal output is displayed on the screen, so no pages of output are produced if you do not enable transcripting), although in that case it does handle the Z-machine font styles (fixpitch vs variable pitch) so it is not entirely plain.)
CBZ can be used if you only have the pictures, but this lacks useful features of PDF such as a table of contents menu, hidden text layers, etc. CBZ can be viewed by extracting the files and viewing them using any program that can display the pictures, and can also be viewed by browsers supporting the "jar:" scheme (like with EPUB).
Another feature not mentioned is spot colours, which your article does not seem to mention. For some applications, such as some types of printing, spot colours can be useful feature to have, and I would think that a format intended for use like PDF should support it.
A feature I would want to see in a suitable alternative format would be to be able to specify multiple colour variants, so that you do not need a separate file for monochrome vs colour printing and can use the same file for both.
(I have a directory with PDF files and I wanted to make a file with the combined table of contents from all of them. I tried to do this by writing a PostScript program which redefined the "pdfmark" operator to write the table of contents entries to another file (but not produce any pages of output), and then execute all of the PDF files in the directory, but this was extremely slow, and some of the table of contents entries seemed to be UTF-16 but not all of them. Probably there is a better way but I don't know what it is.)
🐙 norayr · May 07 at 23:54:
i think for the future all ubernerds may just add a link to their capsule in their profile.
when we engage in comments we may go to that person's profile and discover the link.
still thank you for sharing the link, it is amazing and i shared with many.
🐦 JustASillyBird [OP] · May 09 at 21:20:
zzo: You make good points about all those formats, and I should alter my article to clarify. I am not saying that these formats should be considered potential PDF replacements. I am saying that there are many different document formats, each with their own area of focus, but PDF has become such an easy answer that it is used even in situations where another format would be more appropriate. The real advantage of PDF isn't anything in the specification: It's just the one document format you know everyone can open without having to install more software, so it wins out on pure convenience despite all the many, many shortcomings.
🐦 JustASillyBird [OP] · May 09 at 21:26:
Norayr: Good idea. Revising!