4 ms·

I don’t know why CI or a custom CMS would be the problem here, the output is still static HTML. IMO the problem compared to years ago is that the client side d

I don’t know why CI or a custom CMS would be the problem here, the output is still static HTML. IMO the problem compared to years ago is that the client side dependency trees are so much more complex, involving third party domains and such. But dependencies are still a problem even in Flash: it can load external JSON data for example, and it’s a lot more difficult to sniff all that out than it is with JS. Don’t get me wrong, the web is still miserable in this regard. But IMO it’s the mobile apps that are going to be a giant hole in history: we’ll just have some screenshots to look back on.

javajosh 5y ago
Surely someone is archiving apks, just for the fun of it? I can imagine that someday we'd have a webarchive that embeds a simulation of the device where you can run that apk in your browser. Of course, if it's just an application shell dependent on a running backend, then yes it's not really an application so much as an "application shell" at this point, and you ALSO need to be able to move around the data corpus to systematically capture all reachable application state. This is a hard problem in general, but pretty easy I think when applied to specific cases.
- pharmakom 5y ago
  This is why piracy is so important.
- maximedupre 5y ago
  Haha well that just blew my mind xD I personally don't really care about achieving anything. Why is this such a big thing? Why would we want to achieve all the APKs? There are too many of them lol History? Memories? Who cares, we're all going to die lol And with the qty of APKs, only 0.00000001% of them will ever be dug out of the archives.
  - selfhoster11 5y ago
    It's definitely feasible to store all APKs. After all, Google does just that. It probably wouldn't even be a significant chunk of what the Internet Archive is currently capable of storing.
  - sosborn 5y ago
    > Why would we want to archive all the APKs? It is hard to know what APKs (or anything really) will be interesting to people 30 years out. Archiving all of them gives our future selves a chance at finding what they want to see. Here is an example pulled from a discussion I once had. Someone was looking for a really obscure feature phone game from the late 90's. I had played the original game on original hardware and commented about how bad that game was and that they weren't missing anything by not finding it. Their reply was essentially "I don't care if the game is any good, I really just want to experience what was like," which made sense to me after I heard it. Anyway, the game was so bad at the time it was released that I doubt anyone felt like it would be worth revisiting in the future. And yet, 25-ish years later, they would be wrong.
    - genewitch 5y ago
      Was it E.T.?
  - anthk 5y ago
    Your statement is dumb. With FDroid and some Android 4.1 ARM chinese netbook, you can have a portable and usable machine for several years on. Backing up these APK's for future usage it's essential.
  - rchaud 5y ago
    I'm currently hunting down an old Firefox Android apk, why? Because that had a "Print" option integrated with Android's built in printer functionality, which was removed from future Firefox updates. Why did they do it? No idea, and there is no official response in the forums, despite users complaining about its removal for over a year. https://support.mozilla.org/en-US/questions/1301314 https://support.mozilla.org/en-US/questions/1301314
- dunnevens 5y ago
  I would imagine the Internet Archive is already storing them. APKMirror is fairly comprehensive. Wouldn't take much for the Archive to mirror them. Sometimes, I wonder what the Archive stores in their non-public repositories. Just waiting for the day when it's safe to show the world.
b112 5y ago
It isn't static html, it is an ever changing tapestry of js fetching this and that, on demand loaded content, and so on. To archive, one need to not only snag the page, and loaf time js, but also js triggered by user actions. This means you need to load the whole page, then hit js triggered by page load, etc, then save output to html. Because, some of those js libs won't work in future browsers, others won't work if they can't phone home, or 100 other things. Archive.org is about 100 years from now too, and js content is a PITA in that respect.

javajosh 5y ago

Surely someone is archiving apks, just for the fun of it? I can imagine that someday we'd have a webarchive that embeds a simulation of the device where you can run that apk in your browser. Of course, if it's just an application shell dependent on a running backend, then yes it's not really an application so much as an "application shell" at this point, and you ALSO need to be able to move around the data corpus to systematically capture all reachable application state. This is a hard problem in general, but pretty easy I think when applied to specific cases.

pharmakom 5y ago
This is why piracy is so important.
maximedupre 5y ago
Haha well that just blew my mind xD I personally don't really care about achieving anything. Why is this such a big thing? Why would we want to achieve all the APKs? There are too many of them lol History? Memories? Who cares, we're all going to die lol And with the qty of APKs, only 0.00000001% of them will ever be dug out of the archives.
- selfhoster11 5y ago
  It's definitely feasible to store all APKs. After all, Google does just that. It probably wouldn't even be a significant chunk of what the Internet Archive is currently capable of storing.
- sosborn 5y ago
  > Why would we want to archive all the APKs? It is hard to know what APKs (or anything really) will be interesting to people 30 years out. Archiving all of them gives our future selves a chance at finding what they want to see. Here is an example pulled from a discussion I once had. Someone was looking for a really obscure feature phone game from the late 90's. I had played the original game on original hardware and commented about how bad that game was and that they weren't missing anything by not finding it. Their reply was essentially "I don't care if the game is any good, I really just want to experience what was like," which made sense to me after I heard it. Anyway, the game was so bad at the time it was released that I doubt anyone felt like it would be worth revisiting in the future. And yet, 25-ish years later, they would be wrong.
  - genewitch 5y ago
    Was it E.T.?
- anthk 5y ago
  Your statement is dumb. With FDroid and some Android 4.1 ARM chinese netbook, you can have a portable and usable machine for several years on. Backing up these APK's for future usage it's essential.
- rchaud 5y ago
  I'm currently hunting down an old Firefox Android apk, why? Because that had a "Print" option integrated with Android's built in printer functionality, which was removed from future Firefox updates. Why did they do it? No idea, and there is no official response in the forums, despite users complaining about its removal for over a year. https://support.mozilla.org/en-US/questions/1301314 https://support.mozilla.org/en-US/questions/1301314
dunnevens 5y ago
I would imagine the Internet Archive is already storing them. APKMirror is fairly comprehensive. Wouldn't take much for the Archive to mirror them. Sometimes, I wonder what the Archive stores in their non-public repositories. Just waiting for the day when it's safe to show the world.

b112 5y ago

It isn't static html, it is an ever changing tapestry of js fetching this and that, on demand loaded content, and so on. To archive, one need to not only snag the page, and loaf time js, but also js triggered by user actions. This means you need to load the whole page, then hit js triggered by page load, etc, then save output to html. Because, some of those js libs won't work in future browsers, others won't work if they can't phone home, or 100 other things. Archive.org is about 100 years from now too, and js content is a PITA in that respect.