Right now, at least for my tastes, the UI is a little crowded. (I suspect the crowding might also turn some potential users of the program away from it, since it makes it look pretty complex). I made a couple janky MSPaint illustrations of some changes which I think could help. (It might help to view them at full size.) This isn't a "Feature Request" in the system because these aren't really added features, they're just ways to redesign things if you're so inclined.
First, mostly for the search UI:
The suggestion is to take all the stuff at the bottom and move it to the top. Instead of a bunch of separate buttons for searching people, institutions, etc. just have one dropdown list that lets you select what you're searching. Another benefit is that this moves all the search UI next to the "Advanced Search" button which is more intuitive. I also put an X through the toggle buttons for launching works vs. launching previews, because if these were moved to the Settings menu that would give more space to the search UI, and I never use them. But maybe there's a good reason to have the buttons easy to access. (The two buttons could be one dropdown to save a slight amount of space.)
The chief drawback for this is that if people have a very small horizontal resolution it will fuck stuff up, especially the second row which has all the tabs plus the UI for moving through records. This could be mostly solved if that second row ballooned into two rows if it ran out of horizontal space, but I don't know how easy that is. Ballooning would be nice though because worst case scenario, that row would take up two rows, which just brings the row usage back to where it was before the UI change was made (well, technically you still gain a row since you lose the bottom row of buttons by turning them into a dropdown menu).
Another issue is that you lose the "Last Saved" text for the XML database if you just drop all the bottom UI. That could be moved to the program's title bar, I think - I've used programs where "Last Saved" shows up in the name of the thing so that you can read it even from your taskbar. I didn't put this on the mockup but it would just say on the top of the screen "Hypernomicon - C:\Data\Hypernomicon Database\database.hbb - Database last saved to XML files: 10:10:52 AM." I guess actually that text down there is maybe the status report so it gets used for other stuff, in which case maybe it could stay. I dunno. I'm spitballing.
Finally, I've kept it all in the mockup, but is there even any reason to display the record # or record ID to the user? Would anyone use that for anything? The record #s are not useful information. The record ID is I suppose in principle usable, since I can customize it, but I have trouble imagining what I'd do with that power. So, maybe just nix that stuff from the UI? Similarly, I've kept the arrows for moving through records, but since the ordering is basically arbitrary, I don't ever use those arrows, so maybe nix them too?
Second, mostly for the work UI:
I Xed out the WorldCat and Google Scholar buttons because I never use them but maybe they're useful so whatever. The main suggestion is to combine the two bottom menus into one menu. They're both just a series of tabs so why not grab the tabs from the bottom menu and consolidate them into the top menu? Maybe there are reasons to want both menus open at once but I basically always leave both minimized so I haven't encountered that use case yet. If there aren't reasons, then combining the tabs seems like the way to go.
Aso, this isn't on the image, but I think in the "Arguments" tab it would be nice to be able to reorder columns by drag and dropping (the program lets you drag but nothing happens when you drop). Or at least specifically I want the "Argument name" on the left since that's how I identify arguments, especially because some arguments don't have associated positions (since they're just counterarguments to other arguments, like the second argument in this screenshot). Probably I should put that in as a feature request.
Last edit: Danny Weltman 2020-07-27
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Do you not use the WorldCat and Scholar buttons because you look stuff up on different websites? Or do you never find yourself wanting to look up a work on the web while using Hypernomicon? The buttons are actually configurable from the Settings window.
I will giving these a close look soon because I think a lot of it is probably a good idea.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I don't use WorldCat or Scholar because I can't figure out what I'd be using them for. I can imagine using the Scholar button if I want to find stuff that cites the work, but the two seconds I save by not having to copy and paste the work name into Scholar is mitigated by the fact that this happens so rarely that I'd prefer the button just to be gone. WorldCat I don't even use ever, let alone from Hypernomicon. Maybe once Hypernomicon gets much more integrated into my research habits I'll end up doing all my Google Scholar searches from there, but that seems unlikely: the main time I search Scholar to see what is citing an article, it's when I first download the article, which means it's before the article is entered in Hypernomicon.
Also I thought of another thing I'd change if possible:
I always just hide the author and label panes/panels/whatever since they take up too much space. If they were just tabs down in that bottom thing that would be easier. Or perhaps the author list could be a dropdown menu or something over where URL goes, because I don't care about having the URL displayed - that information is never any help. (At most I'd want a "Web" button which just launches the URL, but I don't need to see the URL. And I don't even think I need a Web button.) (The reason I suggest maybe moving author up top rather than into a tab is because it would be nice to see the author more obviously than having them usually listed by last name via the search key. But I don't know how you'd easily display multiple authors.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I use google scholar all the time, not just when I first download an article. Sometimes to see who cites an article but also because it is sometimes the best way to find a downloadable version of a work. Or sometimes if a work has the wrong title google scholar is the best way to search it and figure that out. The button uses multiple fields to do the search and gives you options for how to search if the author has a first, middle, and last name (depending on the name one way of searching may yield the results you want while others don't). To me, WorldCat is the best overall database for books, and if you are logged into a university library (at least in the US and UK) links to other resources, like electronic copies of the book or a link to request the book through the library, will be shown. Sometimes WorldCat is the only place on the internet where I can actually see the table of contents for a book. So the buttons are doing a lot more than just a replacement for copy-pasting the title into a web browser.
Another thing WorldCat is very useful for: sometimes I'll find a book chapter I want but it's in an old book that's not accessible online, and it's not otherwise downloadable. If I search for that chapter on WorldCat, and that chapter was reprinted in a much newer book, it will show up in the results. Then that newer book will often be downloadable on libgen. (If you never read stuff that's older than about 20 years then none of this might be an issue for you.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Also, this is not intended as a criticism of any of the above ideas because many of them should perhaps be implemented, but I want to emphasize that an important aspect of the overall UI design from the start is that the application is constantly putting information in front of you that you aren't specifically seeking out. A lot of your ideas seem to reflect a desire to hide a lot of the information that is currently not hidden, to free up more space for the parts of the UI where you are actively editing information, or perhaps for aesthetic purposes of visual neatness. But this can defeat the purpose of jogging and reinforcing things in your memory, reminding you of things, increasing familiarity and cognitive incubation and internal mastery of the information, and getting you to see that things are related that you forgot or didn't realize. The application is supposed to play an active role in helping you think and have ideas. Think of it like the difference between having a bold and talented research assistant who hustles and takes initiative (and therefore can even be a little annoying sometimes, but in a way that's good for you), as opposed to merely having an efficient and dependable administrative assistant who always tries to stay out of your away and never distract you. The fact that a bunch of stuff is always showing, that you may not have already felt like you needed to see, is by design.
Similarly with the buttons to search things online: if I can search for something in scholar or google or philpapers with a single click, I am more likely to do it on a whim, than if I have to do a longer series of clicks and keypresses (copying and pasting, opening a webbrowser, etc.). If the button is staring me in the face already I don't even have to think about doing it on my own. The threshold level of effort and will-power needed to quickly glance at what is in philpapers, SEP, and IEP is reduced from a pretty small amount (so I might do it once every 3 weeks) to virtually nothing (so I might do it on average once every 12 hours of time spent working). That might not sound like a big difference but it can make the difference in whether or not you find that crucial idea you needed for a paper to really come together, like I demonstrated in the introductory tutorial.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I get the idea of jogging memory/reminding myself/etc. by displaying stuff, but as far as I can tell the main jogging is from the list of arguments, the records listing the work as a key work, and the labels. The list of arguments and records listing the work as a key work already fight with each other, so I can already only see one of those at a time, and whichever of those I pick can just be the main tab down at the bottom that I leave open. I don't need to see the list of files, the sub-works, the bib data, etc. to jog my memory - none of that helps jog anything except "here's how many PDFs of this work I have." So, that's why it would be nice to consolidate the tabs in the bottom thingy into the top thingy.
As for the labels, I don't use those yet (I can't really figure out how they work) but maybe it would be helpful to have those always on screen. The only other info in that side panel is author, which I do like to have on screen, but not really as a memory jogging thing so much as a convenience thing. So, maybe the side panel can stay.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
For the WorldCat/Scholar stuff, I typically only put stuff in Hypernomicon if I have a PDF (or some other copy) of it, because otherwise it's something I haven't read and I don't have anything useful to say about it. For the rare case in which I'd want something in there that I don't have a copy of, it would be something I'd already tried to find a copy of via the usual methods, so it wouldn't help to have (e.g.) the WorldCat button. None of this matters much since the buttons don't take up much space, but since at most they save about 10 seconds of copying and pasting, my personal tradeoff would be to lose the buttons (or make them optional via the settings menu, etc.).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have a lot of work records with no associated PDF, usually because I'll read where somebody says "Such and such concept was coined by so and so in..." and I might have already had a term record for it, so I create that work record and put the information that the term was supposedly coined in that work (especially if more than one paper has now said that the term was coined there). Or somebody will say that so-and-so argues against the view being put forth in so-and-so, adn I'll quickly make an argument record and work record. In those instances I might not currently feel a need to look at the actual paper and it will slow me down or interrupt my train of thought to fiddle with finding a good PDF of it.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Right now, at least for my tastes, the UI is a little crowded. (I suspect the crowding might also turn some potential users of the program away from it, since it makes it look pretty complex). I made a couple janky MSPaint illustrations of some changes which I think could help. (It might help to view them at full size.) This isn't a "Feature Request" in the system because these aren't really added features, they're just ways to redesign things if you're so inclined.
First, mostly for the search UI:
The suggestion is to take all the stuff at the bottom and move it to the top. Instead of a bunch of separate buttons for searching people, institutions, etc. just have one dropdown list that lets you select what you're searching. Another benefit is that this moves all the search UI next to the "Advanced Search" button which is more intuitive. I also put an X through the toggle buttons for launching works vs. launching previews, because if these were moved to the Settings menu that would give more space to the search UI, and I never use them. But maybe there's a good reason to have the buttons easy to access. (The two buttons could be one dropdown to save a slight amount of space.)
The chief drawback for this is that if people have a very small horizontal resolution it will fuck stuff up, especially the second row which has all the tabs plus the UI for moving through records. This could be mostly solved if that second row ballooned into two rows if it ran out of horizontal space, but I don't know how easy that is. Ballooning would be nice though because worst case scenario, that row would take up two rows, which just brings the row usage back to where it was before the UI change was made (well, technically you still gain a row since you lose the bottom row of buttons by turning them into a dropdown menu).
Another issue is that you lose the "Last Saved" text for the XML database if you just drop all the bottom UI. That could be moved to the program's title bar, I think - I've used programs where "Last Saved" shows up in the name of the thing so that you can read it even from your taskbar. I didn't put this on the mockup but it would just say on the top of the screen "Hypernomicon - C:\Data\Hypernomicon Database\database.hbb - Database last saved to XML files: 10:10:52 AM." I guess actually that text down there is maybe the status report so it gets used for other stuff, in which case maybe it could stay. I dunno. I'm spitballing.
Finally, I've kept it all in the mockup, but is there even any reason to display the record # or record ID to the user? Would anyone use that for anything? The record #s are not useful information. The record ID is I suppose in principle usable, since I can customize it, but I have trouble imagining what I'd do with that power. So, maybe just nix that stuff from the UI? Similarly, I've kept the arrows for moving through records, but since the ordering is basically arbitrary, I don't ever use those arrows, so maybe nix them too?
Second, mostly for the work UI:
I Xed out the WorldCat and Google Scholar buttons because I never use them but maybe they're useful so whatever. The main suggestion is to combine the two bottom menus into one menu. They're both just a series of tabs so why not grab the tabs from the bottom menu and consolidate them into the top menu? Maybe there are reasons to want both menus open at once but I basically always leave both minimized so I haven't encountered that use case yet. If there aren't reasons, then combining the tabs seems like the way to go.
Aso, this isn't on the image, but I think in the "Arguments" tab it would be nice to be able to reorder columns by drag and dropping (the program lets you drag but nothing happens when you drop). Or at least specifically I want the "Argument name" on the left since that's how I identify arguments, especially because some arguments don't have associated positions (since they're just counterarguments to other arguments, like the second argument in this screenshot). Probably I should put that in as a feature request.
Last edit: Danny Weltman 2020-07-27
Do you not use the WorldCat and Scholar buttons because you look stuff up on different websites? Or do you never find yourself wanting to look up a work on the web while using Hypernomicon? The buttons are actually configurable from the Settings window.
I will giving these a close look soon because I think a lot of it is probably a good idea.
I don't use WorldCat or Scholar because I can't figure out what I'd be using them for. I can imagine using the Scholar button if I want to find stuff that cites the work, but the two seconds I save by not having to copy and paste the work name into Scholar is mitigated by the fact that this happens so rarely that I'd prefer the button just to be gone. WorldCat I don't even use ever, let alone from Hypernomicon. Maybe once Hypernomicon gets much more integrated into my research habits I'll end up doing all my Google Scholar searches from there, but that seems unlikely: the main time I search Scholar to see what is citing an article, it's when I first download the article, which means it's before the article is entered in Hypernomicon.
Also I thought of another thing I'd change if possible:
I always just hide the author and label panes/panels/whatever since they take up too much space. If they were just tabs down in that bottom thing that would be easier. Or perhaps the author list could be a dropdown menu or something over where URL goes, because I don't care about having the URL displayed - that information is never any help. (At most I'd want a "Web" button which just launches the URL, but I don't need to see the URL. And I don't even think I need a Web button.) (The reason I suggest maybe moving author up top rather than into a tab is because it would be nice to see the author more obviously than having them usually listed by last name via the search key. But I don't know how you'd easily display multiple authors.)
I use google scholar all the time, not just when I first download an article. Sometimes to see who cites an article but also because it is sometimes the best way to find a downloadable version of a work. Or sometimes if a work has the wrong title google scholar is the best way to search it and figure that out. The button uses multiple fields to do the search and gives you options for how to search if the author has a first, middle, and last name (depending on the name one way of searching may yield the results you want while others don't). To me, WorldCat is the best overall database for books, and if you are logged into a university library (at least in the US and UK) links to other resources, like electronic copies of the book or a link to request the book through the library, will be shown. Sometimes WorldCat is the only place on the internet where I can actually see the table of contents for a book. So the buttons are doing a lot more than just a replacement for copy-pasting the title into a web browser.
Another thing WorldCat is very useful for: sometimes I'll find a book chapter I want but it's in an old book that's not accessible online, and it's not otherwise downloadable. If I search for that chapter on WorldCat, and that chapter was reprinted in a much newer book, it will show up in the results. Then that newer book will often be downloadable on libgen. (If you never read stuff that's older than about 20 years then none of this might be an issue for you.)
Also, this is not intended as a criticism of any of the above ideas because many of them should perhaps be implemented, but I want to emphasize that an important aspect of the overall UI design from the start is that the application is constantly putting information in front of you that you aren't specifically seeking out. A lot of your ideas seem to reflect a desire to hide a lot of the information that is currently not hidden, to free up more space for the parts of the UI where you are actively editing information, or perhaps for aesthetic purposes of visual neatness. But this can defeat the purpose of jogging and reinforcing things in your memory, reminding you of things, increasing familiarity and cognitive incubation and internal mastery of the information, and getting you to see that things are related that you forgot or didn't realize. The application is supposed to play an active role in helping you think and have ideas. Think of it like the difference between having a bold and talented research assistant who hustles and takes initiative (and therefore can even be a little annoying sometimes, but in a way that's good for you), as opposed to merely having an efficient and dependable administrative assistant who always tries to stay out of your away and never distract you. The fact that a bunch of stuff is always showing, that you may not have already felt like you needed to see, is by design.
Similarly with the buttons to search things online: if I can search for something in scholar or google or philpapers with a single click, I am more likely to do it on a whim, than if I have to do a longer series of clicks and keypresses (copying and pasting, opening a webbrowser, etc.). If the button is staring me in the face already I don't even have to think about doing it on my own. The threshold level of effort and will-power needed to quickly glance at what is in philpapers, SEP, and IEP is reduced from a pretty small amount (so I might do it once every 3 weeks) to virtually nothing (so I might do it on average once every 12 hours of time spent working). That might not sound like a big difference but it can make the difference in whether or not you find that crucial idea you needed for a paper to really come together, like I demonstrated in the introductory tutorial.
I get the idea of jogging memory/reminding myself/etc. by displaying stuff, but as far as I can tell the main jogging is from the list of arguments, the records listing the work as a key work, and the labels. The list of arguments and records listing the work as a key work already fight with each other, so I can already only see one of those at a time, and whichever of those I pick can just be the main tab down at the bottom that I leave open. I don't need to see the list of files, the sub-works, the bib data, etc. to jog my memory - none of that helps jog anything except "here's how many PDFs of this work I have." So, that's why it would be nice to consolidate the tabs in the bottom thingy into the top thingy.
As for the labels, I don't use those yet (I can't really figure out how they work) but maybe it would be helpful to have those always on screen. The only other info in that side panel is author, which I do like to have on screen, but not really as a memory jogging thing so much as a convenience thing. So, maybe the side panel can stay.
For the WorldCat/Scholar stuff, I typically only put stuff in Hypernomicon if I have a PDF (or some other copy) of it, because otherwise it's something I haven't read and I don't have anything useful to say about it. For the rare case in which I'd want something in there that I don't have a copy of, it would be something I'd already tried to find a copy of via the usual methods, so it wouldn't help to have (e.g.) the WorldCat button. None of this matters much since the buttons don't take up much space, but since at most they save about 10 seconds of copying and pasting, my personal tradeoff would be to lose the buttons (or make them optional via the settings menu, etc.).
I have a lot of work records with no associated PDF, usually because I'll read where somebody says "Such and such concept was coined by so and so in..." and I might have already had a term record for it, so I create that work record and put the information that the term was supposedly coined in that work (especially if more than one paper has now said that the term was coined there). Or somebody will say that so-and-so argues against the view being put forth in so-and-so, adn I'll quickly make an argument record and work record. In those instances I might not currently feel a need to look at the actual paper and it will slow me down or interrupt my train of thought to fiddle with finding a good PDF of it.