Again, I don't know exactly how his current sheet looks to be certain,
but what he was asking for does not sound like a PDF printout. I am a huge fan of LaTeX myself. I wrote my 1,100 page textbook in it! And I even programmed a script to convert Microsoft Excel formulas -- of
any complexity -- into parsable LaTeX code with VBA for the purpose of generating Engineering reports. (
Here's the PDF documentation. If anyone is interested to actually view it, save to your machine and view with your PDF reader instead of in Google Docs to make the internal hyperlinks work.)
But the question is how is navigation for a specific case going to work in a PDF versus a database or a glorified (programmed) spreadsheet. For example, when I made
my F3L PDF for K4, on
page 2 is a diagram of the pieces that I numbered so that one can use the search in their PDF viewer to search for "8-7" (without quotes) to quickly find a case that has those two pieces need to be inserted (in their respective slots). But that is painful to do for a large quantity of cases! You can use internal hyperlinks and the table of contents in the LaTeX PDF, but it's still not efficient
to actually use.
When need a search engine,
which is possible with VBA (that's my search engine which mimics Google's, with autocomplete . . . in both directions . . . it can search a 2400+ page BIBLE in a manner of seconds . . . referring back to your example with spreadsheets failing to process large amounts of data as is in a book like LaTeX can).
Everything he's looking for (automatically create mirrors, inverses, etc.) can be easily done in a spreadsheet. And best of all, it can even be that the images can be dynamically created as the viewer is viewing the alg associated with the algorithms (rows) in the current view (in between the four corner cells visible in the current screen). So that absolutely no space is needed to store the images. (VBA can be used to do conditional formatting too.)
Any of the one to four versions (mirror, inverse, etc.) of a case need to be inputted in a given row. The other (three?) will be automatically calculated in the cells. And it can be that once there is some text in a cell, that Excel will not keep regenerating the move sequences. (It will only generate sequences in which all four cells in a row are not filled in . . . and it's easy to use Excel table filters to quickly find "incomplete rows" to fill. (Much much faster than a typical For loop.) So I'm sure he will run the code to generate the formulas. He can disable that code from running once he has it run once to generate all algorithms (so that it doesn't keep searching for "incomplete rows"). Then the user will use the search engine to pull up algorithms to cases that they describe some how with the search engine. Then as those rows come up, the images for those cases will be generated simultaneously as their corresponding algorithms come into view.
The main problem is people will have to have Microsoft Office installed to use this.
But if someone has a better idea, that's fine with me. But I don't like the idea of him having to depend on Lucas Garron (or anyone else's) software for the long-term. (alg.garron.us was taken down and replaced with alg.cubing.net, which completely wrecked years of posts which linked to it. So I don't think a big project like this should be "at the mercy" of anyone else's server. He's got to think
long-term, so that he doesn't have to keep "adapting" his work to changing technology as time progresses.)
1) Including alg.cubing like images in the database, which reads target stickers and generates an image. I pinged
@Lucas Garron about generating images with just target stickers as input. eg. UF-UB-RD-RU-LB as input should give me
View attachment 14853 as output.
I'm not sure if you ever thought about flat 3D images similar to visual cubes LL view
like these, for example, but that's also an option. (And are more feasible to generate with cell pixel representations with my idea.)
And
@qwr , I'm not "insisting" that this be the way to go (I saw that's what you wrote originally but reasonably changed it to "suggesting"), but I can't ignore the fact that he would have still been using Google Sheets if the data wasn't as large as it is now. This would be a way for him to continue to use his originally chosen form of medium.