This is really impressive. It's exactly what I imagined the original Microsoft Network in Windows 95 would have been like.
And so The Microsoft Network wasn't a program you loaded like CompuServe. It was part of the OS, with folder icons that looked just like real folders. It was a kind of version of the Web where you could browse online data the same way you browsed your file system. This is what made it cool.
It was as if the data was suddenly free of the shackles of being displayed in a program. Data wasn't just a web page, or a program showing its own internal databases. The Microsoft Network made it look like the data was right there, and you could click it and drag it around! For a brief time, back in 1995, it felt like we were on the verge of the true object-oriented web, a world filled with open data and free from the tyranny of the walled gardens.[1]
It also reminded me what an excellent job Wikipedia does with their hierarchical classification which you don't see when you're often only searching by article name.
Talk about data being separate from programs always reminds me what a good job Microsoft did with the spacial filesystem (that means one folder is one window, and they remember their location), single-document interface (a UI paradigm) and COM (a cross-process communication system). As a novice user not understanding a whole lot about the system, your documents were in the operating system and not in a specific program (this still wasn't perfect and a lot of new users did think their documents were in programs, which might be why we gave that up) and those programs could talk to each other and embed each other's documents.
This stuff probably seemed moderately innovative if you didn't grow up with it, seemed blindingly obvious if you did grow up with it, and somehow, like idiots, we've managed to lose it again!
> this still wasn't perfect and a lot of new users did think their documents were in programs, which might be why we gave that up
To this day there exist office workers—ones old enough that no, it's not because the were introduced to computers via smartphones—who use a computer for hours every single weekday day but get totally turned around in a file manager, and don't know even the extreme basics like how to copy and move files.
There are offices full of such folks, in non-tech offices, where the person who knows how to sort-of use a GUI file manager is the "computer whiz" they go to with questions.
The "files and folders" hierarchical tree model for a file system is one where I wonder about the limits or effectiveness of the skeuomorphism approach to convey such a concept. If you're coming from a place where information was generally held and organized on paper, it _should_ be natural that you can group files within a container like a folder, and the kind of folder the iconography showed should be able to contain sub-folders.
While many did pick up on the idea, where were the shortcomings? Were the early graphics not enough to build the mental link. Was it the common grid view of icons. Was it the icon being an abstract thing you needed to open to see the contents instead of looking at it directly (as previews on the icon which came later), was it things opening in separate windows. It's not as though other more visually 'rich' methods to show a file system such as 3D or animated took off.
There's also the modern version that gets brought up occasionally where people who are using devices with mobile instead of desktop OSes apparently don't know how to work with file systems to manage data, and presumably they'd have even less exposure to the physical paper concept that inspired it.
OLE objects are just FAT like filesystems; nothing too disimilar to Unix if it mounted disk images with text files and different images with it in order to create a document format.
> It also reminded me what an excellent job Wikipedia does with their hierarchical classification
As someone who once tried to use that supposedly hierarchical classification for data organization, it is unfortunately not excellent at all.
It is rather arbitrary, inconsistent, extremely incomplete, and not infrequently circular. Think of it more like a bunch of haphazardly applied tags that make perfect sense in the context of a single page, but quite frequently make very little sense when you look at the actual pages and sub categories that belong to a category. Category membership is just not something visible enough for it to wind up being organized and curated in any kind of systematically accurate way.
On the other hand, the presence of an infobox of a certain type is extremely reliable for categorizing many types of articles.
> They had this project called Cairo that was supposed to throw out that scruffy old file-based filesystem and bring in a shiny new Object Based File System instead. It never happened, so we'll never know exactly how it might have turned out.
Nowadays we call those APIs. They are REST based rather than file-based to make them distributed, the main difference is that you don't get a common user interface that all providers adjust to; you need to choose your own client to read them and write into them.
And because they're created by programmers for programmers, they're not what you'd call user-friendly. Usually the only efficient way to use them is programmatically, so that you need to create a specific user interface for each API. Somehow, I doubt that Cairo would have come to be anything much different from that in the end.
Somehow this reminds me of https://news.ycombinator.com/item?id=19249373 released by CERN on 30th anniversary. Pretty sure Berners-Lee in recent years was contributing to decentralized web/Internet concept that does also reminds a little bit of early WWW.
Are you mabe thinking of Solid[1]? This is an interesting idea, and I have looked into it on several occasions recently, but despite it being around for a really long time it appears to be going exactly nowhere, unfortunately.
It seems a bit of a missed opportunity really. If they had more agressively pursued alliances, it could potentially have been a solid (pun only semi intended) foundation for Mastadon and Bluesky.
The name is unfortunate as well, it is really difficult to search for.
> This is really impressive. It's exactly what I imagined the original Microsoft Network in Windows 95 would have been like.
That's actually not far off. It was an old-fashioned BBS like Compuserve in a Windows Explorer-like window. The topic-specific icons you see in this mockup are actually very on-point, though on the Microsoft Network they would be for general BBS sections not encyclopedic articles or media.
Incredibly beautiful, possibly because it maps so well to the mental model we typically use to organize knowledge in our heads. I don't know how we lost the folder/container vs. document/content iconography, and other things (like layout of items, sorting) during the shift to web applications.
Knowledge doesn’t neatly align to a nested hierarchy. Especially written knowledge.
Language is an imperfect means to convey knowledge, and people store that knowledge in subjective and highly personal ways.
You may mentally recall balloons within “entertainment” or “party”, whereas I might store that knowledge under “horror”.
Add onto that the massive focus on using graph theory to scale social networking technologically, and you effectively lose any motivation for rigid hierarchy.
A folder system doesn't have to be strictly rigid, you can still have "symlinks" so the same article appearing in different folders (aka labels if you can easily duplicate content inside folders, but you retain the nested, drill-down approach)
Wikimedia Commons has this feature. Editors can manually bless certain combinations of traits as "subcategories".
For example, https://commons.wikimedia.org/wiki/Category:Paintings_of_cas... contains the subcategories "Paintings of castles by country" (nested hierarchy), "Frescos of castles" (a medium), "Paintings of Château de Chillon" (a subject), and "Young Knight in a Landscape by Carpaccio" (multiple views onto a specific item). Each item may appear in multiple subcategories. As far as I can tell, the UI won't let you search for frescos of Italian castles (unless somebody's made a subcategory for that), or view all paintings of castles regardless of their subcategory.
I'm not very fond of this approach. I'd prefer for each item to have an unstructured set of tags ("fresco", "depiction of a castle", "depiction of Italy"), with automatic derivation of parent tags ("fresco" implies "painting") and the option to search by multiple tags. It should be possible to automatically discover tags which best refine a search, so that the UI can still suggest them to the user, as it does today.
> I'd prefer for each item to have an unstructured set of tags ("fresco", "depiction of a castle", "depiction of Italy"), with automatic derivation of parent tags ("fresco" implies "painting") and the option to search by multiple tags.
It's definitely possible to do this. IMSLP (a large repository of freely available sheet music, which differs by cross-cutting features such as genre, historical period, contributors (composers and others), instrumentation etc.) is MediaWiki based and has a plugin that does exactly that. These days the would probably want to host all the tags on Wikidata so that they can be multilingual and queryable out of the box, though.
Which is actually done on commons, it just isn't very popular (on images, click the structured data tab and then look at depicts) [admittedly i think a big part of the problem is is implementation choices and UI decisions].
That's only "depicts" claims and is nowhere near comprehensive. It doesn't even come close to matching what's currently stated using categories. Running searches on that data is also hard compared to what IMSLP gives you for their own system.
The Library of Congress uses both approaches, to an extent.
The cataloguing system uses a hierarchical classification, based on one originally developed by Thomas Jefferson, on whose initial donation the Library of Congress is based. This is known as the Library of Congress Classification, and is used to specifically locate a given title or work within the stacks, that is, each item has one and only one location.
There are also subject headings which are more tag-based, though also on a controlled vocabulary. A given work is given a (relatively small number) of subjects to which it's associated. These are not hierarchical, though of course the listing of subject headings itself follows a sequence. Unlike the classification, which assigns a single location to each work, the headings are a search aid to patrons searching for a set of related works within a subject heading, or facilitate branching of a search to possibly related subjects.
Tagging systems, especially ad hoc tags supplied by untrained users, are popular but tend to produce numerous issues over time. Not that formal systems (as with the LoC systems mentioned here) are immune to same. One feature of the LoC systems is that they've evolved processes for managing change over time. Examples would be terminology or classifications which are now deprecated, or of regions and polities which have changed or no longer exist (e.g., the Austro-Hungarian empire, the USSR), or of changes in underlying classifications (e.g., of chemical elements or of biological classifications, both of which have evolved significantly over the life of the Library of Congress).
The history of hierarchical information classifications is long and IMO fascinating, dating at least to Aristotle and his Categories, as well as numerous variants used in classifications of knowledge (such as Francis Bacon's) or encyclopedias, including Diderot's and Britannica.
Which isn't a hierarchy, it's a tagging system. The tags have some hierarchy but that's not uncommon. The distinguishing characteristic of "nested hierarchy" is that a particular thing should only appear exactly once in the hierarchy.
Since this is so terribly impossible most systems almost immediately make it possible for things to show up in more than one place, which means it's actually hierarchial tagging, whether or not the organizer(s) realize it.
You could also make a distinction based on how many tags things end up with; if it's almost always one, you could call it a nested hierachy with some exceptions, but if it's almost always more than one, and often much, much more than one, it's a tagging system. Even by that criterion that creates a spectrum rather than a binary distinction, Wikipedia is very much organized by tags and not hierarchies. I don't know what the average is but every Wikipedia page I've ever looked at the tags for has quite a few.
The ICD-11 has an interesting "nested hierarchy with some exceptions": each entry has a primary location, and secondary locations are implemented a bit like symlinks.
I agree, for some reason I have always alternated between wanting not just the universal search box but a browsable hierarchy to mentally run my fingers over and discover in a structured way.
We let go of the the manual index somewhere along the way since it doesn’t scale like search, obviously, but for the same reason I keep a library and enjoy traversing others’ private ones and visiting public ones, I keep coming back to browse.
Did you prefer the Yahoo/internet frontpage approach to google search though? I didn't, but I remember a time when it was a live debate. It has been interesting to see some sites like youtube or wikipedia evolve a quasi-hierarchical frontpage though.
I did. I made my own custom start page back in the day with frequently used searches, along with headlines, weather, stocks, etc. Instead of menus, it just divided topics using horizontal rules, much like an actual newspaper.
Large scrollbars! Windows with borders! What a relief!
This has become a forgotten art: we focus so much on CONTENT these days that we forget that people want to use the mouse to scroll, and use the mouse to resize windows.
I agree. We focus on content where usually it’s so poorly structured and displayed it lacks final touches. I prefer wider scrolling bar than extra white space in the browser. Same goes for finder. I find old systems to be much more usable for me than modern ones
This is genuinely a really fun way to browse Wikipedia. Only drawback is that folder names that contain ellipsis don't show the full name when clicked.
It is nice. I randomly click on something interest just appear in my mind and lead to this: life -> death -> last_words -> More milk. But I can't find it on Wiki. I search More milk. and the first result is this page: https://en.wikipedia.org/wiki/Death_of_Michael_Jackson. Hmm, why is the name different?
> After several hours and several drug injections, Jackson was still unable to fall asleep, and, according to Murray, was repeatedly asking him for "milk", a nickname for the powerful surgical general anesthetic propofol, which Jackson had used in the past as a sleep aid. At 10:40 a.m., with Jackson still not asleep, Murray relented to his requests and injected him with 25 milligrams of propofol diluted with lidocaine. With Jackson finally asleep, Murray testified that he left his bedside to go to the bathroom, and after returning two minutes later, discovered that Jackson was not breathing and had a weak pulse.
Not sure why they downvoted you because you have a point - icons are not the same as Windows XP's, wallpaper flat color reminds me more of Win 95/98 and the taskbar design has some details that do not match precisely with Windows XP's. I'd also bet it's due to copyright concerns
I’m surprised the search function in the start menu doesn’t do anything. Seems like that would be super useful. But I did enjoy this, nice job with the polish.
This is cute, but the UI is uncannily not there (I think there were multiple attempts of designing the XP for web already which looked more authentic).
But my biggest gripe is, why represent it as a file system with WordPad displaying HTML? I get the idea for media, but not for the articles.
It's pretty obvious that Wikipedia should be a single CHM file. That would be nice and much more immersive.
idk, it appears this thing is, since after clicking around a few times it becomes very obvious all the content is english. Guess that is just how the author of this project decided to make it.
Generally you can delete your comment if its relatively soon after you posted it and nobody replied.
Honestly, a testament to current Wikipedia design, because while it's fun to click around it's literally impossible to find what you need in this kind of GUI. (Geofile explorer is simply baffling.)
What does Wikipedia do differently? I don't think I have ever used Wikipedia's internal navigation - if I want to find something I usually type 'XYZ wikipedia' into Google and click the first link.
I often use category links on the bottom or in the sidebar when researching. Also I use search, which doesn't work in this alternative GUI. Sometimes I check the article in another language.
Ok this is a genuinely perfect way to research an entire field by article instead of having to jump recursively link to link and forgetting what you were doing 5 minutes ago.
I've never seen wikipedia from this categorized vantage point. If we're being real their UX is kinda crap outside the usual search->article->link flow and could use a complete rework.
3: the tree style tabs plugin in combination with middle-click is criminally underrated for navigating hierarchical data. (middleclick open-in-new-tab is only mildly handy, tree style tabs seems tepid by itself without it)
TIL, I think I've landed on these pages a decent number of times but never from wikipedia's internal nav. I assumed they were more of an adhoc occasional thing, not a standard for sorting all pages.
The little arrows next to the subcategories can be clicked to open up trees, so you have hierarchical data in there as well. Try click open eg Classes of Computers (With 41 direct subcategories, and 91 pages directly in the top level category, that's a big tree!)
Somehow the format makes me feel like its easier to learn here than the intimidating encyclopedia theme of wikipedia. It's interesting to consider the effect that presentation of information might have on learning. We know that physical books are said to be better for learning (I have heard people go up by an entire grade if they use them), but maybe there is something to be said for themes, too.
And so The Microsoft Network wasn't a program you loaded like CompuServe. It was part of the OS, with folder icons that looked just like real folders. It was a kind of version of the Web where you could browse online data the same way you browsed your file system. This is what made it cool.
It was as if the data was suddenly free of the shackles of being displayed in a program. Data wasn't just a web page, or a program showing its own internal databases. The Microsoft Network made it look like the data was right there, and you could click it and drag it around! For a brief time, back in 1995, it felt like we were on the verge of the true object-oriented web, a world filled with open data and free from the tyranny of the walled gardens.[1]
It also reminded me what an excellent job Wikipedia does with their hierarchical classification which you don't see when you're often only searching by article name.
[1]: https://web.archive.org/web/20260129143542/https://www.coder...
This stuff probably seemed moderately innovative if you didn't grow up with it, seemed blindingly obvious if you did grow up with it, and somehow, like idiots, we've managed to lose it again!
To this day there exist office workers—ones old enough that no, it's not because the were introduced to computers via smartphones—who use a computer for hours every single weekday day but get totally turned around in a file manager, and don't know even the extreme basics like how to copy and move files.
There are offices full of such folks, in non-tech offices, where the person who knows how to sort-of use a GUI file manager is the "computer whiz" they go to with questions.
While many did pick up on the idea, where were the shortcomings? Were the early graphics not enough to build the mental link. Was it the common grid view of icons. Was it the icon being an abstract thing you needed to open to see the contents instead of looking at it directly (as previews on the icon which came later), was it things opening in separate windows. It's not as though other more visually 'rich' methods to show a file system such as 3D or animated took off.
There's also the modern version that gets brought up occasionally where people who are using devices with mobile instead of desktop OSes apparently don't know how to work with file systems to manage data, and presumably they'd have even less exposure to the physical paper concept that inspired it.
As someone who once tried to use that supposedly hierarchical classification for data organization, it is unfortunately not excellent at all.
It is rather arbitrary, inconsistent, extremely incomplete, and not infrequently circular. Think of it more like a bunch of haphazardly applied tags that make perfect sense in the context of a single page, but quite frequently make very little sense when you look at the actual pages and sub categories that belong to a category. Category membership is just not something visible enough for it to wind up being organized and curated in any kind of systematically accurate way.
On the other hand, the presence of an infobox of a certain type is extremely reliable for categorizing many types of articles.
Nowadays we call those APIs. They are REST based rather than file-based to make them distributed, the main difference is that you don't get a common user interface that all providers adjust to; you need to choose your own client to read them and write into them.
And because they're created by programmers for programmers, they're not what you'd call user-friendly. Usually the only efficient way to use them is programmatically, so that you need to create a specific user interface for each API. Somehow, I doubt that Cairo would have come to be anything much different from that in the end.
There was also this submission from 9 years ago: https://news.ycombinator.com/item?id=13361523 - and probably not the only one of such ideas
It seems a bit of a missed opportunity really. If they had more agressively pursued alliances, it could potentially have been a solid (pun only semi intended) foundation for Mastadon and Bluesky.
The name is unfortunate as well, it is really difficult to search for.
[1] https://en.wikipedia.org/wiki/Solid_(web_decentralization_pr...
That's actually not far off. It was an old-fashioned BBS like Compuserve in a Windows Explorer-like window. The topic-specific icons you see in this mockup are actually very on-point, though on the Microsoft Network they would be for general BBS sections not encyclopedic articles or media.
Language is an imperfect means to convey knowledge, and people store that knowledge in subjective and highly personal ways.
You may mentally recall balloons within “entertainment” or “party”, whereas I might store that knowledge under “horror”.
Add onto that the massive focus on using graph theory to scale social networking technologically, and you effectively lose any motivation for rigid hierarchy.
For example, https://commons.wikimedia.org/wiki/Category:Paintings_of_cas... contains the subcategories "Paintings of castles by country" (nested hierarchy), "Frescos of castles" (a medium), "Paintings of Château de Chillon" (a subject), and "Young Knight in a Landscape by Carpaccio" (multiple views onto a specific item). Each item may appear in multiple subcategories. As far as I can tell, the UI won't let you search for frescos of Italian castles (unless somebody's made a subcategory for that), or view all paintings of castles regardless of their subcategory.
I'm not very fond of this approach. I'd prefer for each item to have an unstructured set of tags ("fresco", "depiction of a castle", "depiction of Italy"), with automatic derivation of parent tags ("fresco" implies "painting") and the option to search by multiple tags. It should be possible to automatically discover tags which best refine a search, so that the UI can still suggest them to the user, as it does today.
It's definitely possible to do this. IMSLP (a large repository of freely available sheet music, which differs by cross-cutting features such as genre, historical period, contributors (composers and others), instrumentation etc.) is MediaWiki based and has a plugin that does exactly that. These days the would probably want to host all the tags on Wikidata so that they can be multilingual and queryable out of the box, though.
The cataloguing system uses a hierarchical classification, based on one originally developed by Thomas Jefferson, on whose initial donation the Library of Congress is based. This is known as the Library of Congress Classification, and is used to specifically locate a given title or work within the stacks, that is, each item has one and only one location.
There are also subject headings which are more tag-based, though also on a controlled vocabulary. A given work is given a (relatively small number) of subjects to which it's associated. These are not hierarchical, though of course the listing of subject headings itself follows a sequence. Unlike the classification, which assigns a single location to each work, the headings are a search aid to patrons searching for a set of related works within a subject heading, or facilitate branching of a search to possibly related subjects.
Tagging systems, especially ad hoc tags supplied by untrained users, are popular but tend to produce numerous issues over time. Not that formal systems (as with the LoC systems mentioned here) are immune to same. One feature of the LoC systems is that they've evolved processes for managing change over time. Examples would be terminology or classifications which are now deprecated, or of regions and polities which have changed or no longer exist (e.g., the Austro-Hungarian empire, the USSR), or of changes in underlying classifications (e.g., of chemical elements or of biological classifications, both of which have evolved significantly over the life of the Library of Congress).
The history of hierarchical information classifications is long and IMO fascinating, dating at least to Aristotle and his Categories, as well as numerous variants used in classifications of knowledge (such as Francis Bacon's) or encyclopedias, including Diderot's and Britannica.
The category tree being displayed comes directly fron wikipedia. E.g. Wikipedia has pages like https://en.wikipedia.org/wiki/Category:Art
Since this is so terribly impossible most systems almost immediately make it possible for things to show up in more than one place, which means it's actually hierarchial tagging, whether or not the organizer(s) realize it.
You could also make a distinction based on how many tags things end up with; if it's almost always one, you could call it a nested hierachy with some exceptions, but if it's almost always more than one, and often much, much more than one, it's a tagging system. Even by that criterion that creates a spectrum rather than a binary distinction, Wikipedia is very much organized by tags and not hierarchies. I don't know what the average is but every Wikipedia page I've ever looked at the tags for has quite a few.
We let go of the the manual index somewhere along the way since it doesn’t scale like search, obviously, but for the same reason I keep a library and enjoy traversing others’ private ones and visiting public ones, I keep coming back to browse.
This has become a forgotten art: we focus so much on CONTENT these days that we forget that people want to use the mouse to scroll, and use the mouse to resize windows.
https://brynnbateman.com/
https://en.wikipedia.org/wiki/Category:Art
https://en.wikipedia.org/wiki/Category:The_arts
Unlike Wikipedia these days.
https://en.wikipedia.org/w/index.php?title=More_milk.&redire...
The "Windows XP" website displays the same article when you click on "More milk" there
> After several hours and several drug injections, Jackson was still unable to fall asleep, and, according to Murray, was repeatedly asking him for "milk", a nickname for the powerful surgical general anesthetic propofol, which Jackson had used in the past as a sleep aid. At 10:40 a.m., with Jackson still not asleep, Murray relented to his requests and injected him with 25 milligrams of propofol diluted with lidocaine. With Jackson finally asleep, Murray testified that he left his bedside to go to the bathroom, and after returning two minutes later, discovered that Jackson was not breathing and had a weak pulse.
He calls taking propofol for sleep as Having chemotherapy because you're tired of shaving your head
The main CSS comes from XP.css [0], but the AI additions [1] have definitely messed it up in some way.
The whole thing is pure JS which is nice but the comments give a good impression isn't not hand written IMO [2]
[0]: https://github.com/botoxparty/XP.css/
[1]: https://explorer.samismith.com/css/base.css
[2]: https://explorer.samismith.com/js/explorer.js
But my biggest gripe is, why represent it as a file system with WordPad displaying HTML? I get the idea for media, but not for the articles.
It's pretty obvious that Wikipedia should be a single CHM file. That would be nice and much more immersive.
https://en.wikipedia.org/wiki/Microsoft_Compiled_HTML_Help
Should put a shortcut to it on the desktop as well, so that users who experience significant lag can defrag at will.
Generally you can delete your comment if its relatively soon after you posted it and nobody replied.
I've never seen wikipedia from this categorized vantage point. If we're being real their UX is kinda crap outside the usual search->article->link flow and could use a complete rework.
Three tricks if you didn't already know:
1: you'll find categories at the bottom of regular mediawiki pages
2: if you click one, you'll end up on a page like eg. this: https://en.wikipedia.org/wiki/Category:Computers
3: the tree style tabs plugin in combination with middle-click is criminally underrated for navigating hierarchical data. (middleclick open-in-new-tab is only mildly handy, tree style tabs seems tepid by itself without it)
The little arrows next to the subcategories can be clicked to open up trees, so you have hierarchical data in there as well. Try click open eg Classes of Computers (With 41 direct subcategories, and 91 pages directly in the top level category, that's a big tree!)
Categories are criminally under-used.