#124626 - keldon - Sat Apr 07, 2007 10:26 am
Wait, do we not see a problem....
... with websites going down, no longer being maintained/manned by anyone, or not keeping legacy links! I see this as a big problem to the net. Web archive and mirrors are currently the best solutions but is there a better one?
And then there are domain/directory/sub domain changes when the company/project change name.
When work is published on the Internet it should be published forever. At least when your library moves there are signs and warnings about this. Our local library moved about 15 years ago when it expanded, and the signs are still there!
The Internet has been mainstream for only a short while, and already we are plagued with countless broken links (look at devrs.com). Give it another 10 years and the Internet as we know it could be only accessible through mirrors and the web archive! Anyone remember GBADev? Yes, webarchive.com/blahblah/http://gbadev.org ^_^
Hmmm, I wonder!
Discuss!
[Subject Line Fairy was here]
#124667 - tepples - Sat Apr 07, 2007 5:44 pm
| keldon wrote: |
Wait, do we not see a problem....
... with websites going down, no longer being maintained/manned by anyone, or not keeping legacy links! I see this as a big problem to the net. |
Yes, it is a problem. I try to follow the guidelines and techniques described in "Cool URIs don't change", but I recognize that a lot of people relying on advertising-supported web hosting or ISP-provided web hosting cannot financially afford to implement these techniques.
| Quote: |
| And then there are domain/directory/sub domain changes when the company/project change name. |
Sometimes, these subdomain changes are forced by trademark holders. I try to name my own projects so that trademark holders can't touch them, as "This Can't Be Yogurt" did when it became "The Country's Best Yogurt" after a suit from "I Can't Believe It's Yogurt". "Carbon" became "Tetanus" after Apple introduced Mac OS X including a refactored Mac OS classic API called Carbon. "Tetanus On Drugs" became "Lockjaw: The Overdose" when The Tetris Company cease-and-desisted a few other tetromino game projects, keeping the TOD acronym. (Quinn stayed Quinn, but Neave Tetris became N-Blox.) Luminesweeper might get attacked, but I'll find another name that fits /lu/.
| Quote: |
| When work is published on the Internet it should be published forever. At least when your library moves there are signs and warnings about this. |
Unless the project used to be on ISP-provided web hosting, and the developer doesn't want to pay $240 per year to keep an account with the ISP just to have the moved notice up. Some people lose their ISP when they graduate from university, and they don't want to stay for a master's degree and take a T.A. job just to keep the domain valid.
_________________
-- Where is he?
-- Who?
-- You know, the human.
-- I think he moved to Tilwick.
#124724 - keldon - Sun Apr 08, 2007 12:11 am
I never noticed before that all pages on the w3c site have no extension!!! Interesting article, and the Alertbox one too. I've read that Alertbox article before, but it wasn't so relevant then. Now I understand.
#124736 - HyperHacker - Sun Apr 08, 2007 2:23 am
I try to keep all links valid on my pages. If a page moves I add a redirect, and I don't delete files from my FTP unless they were only intended to be given to one or two people. Check all my posts from ever since I got the server a few years ago, and the images linked in them will still be there (unless the server's down). It is annoying when URLs don't work, but the Internet is just like that. It's very dynamic, constantly changing.
_________________
I'm a PSP hacker now, but I still <3 DS.
#124743 - sgeos - Sun Apr 08, 2007 4:19 am
Server died. My home page vanished. The data is somewhere, but I don't have time to fuss with that now. Such is life.
-Brendan
#124764 - keldon - Sun Apr 08, 2007 8:56 am
Not according to Tim Berners Lee (Godfather of the Internet). But I was talking about it being a problem with the internet in general; and TB supported it.
Are you saying that 'such is life' is an excuse for there not being solutions to likely problems? Are you suggesting that the world is perfectly fine and nothing should change/improve? No need for nothing new/improved/better because 'such is life'? ^_^
#124771 - sgeos - Sun Apr 08, 2007 2:21 pm
Sometimes people know what the "correct thing to do" is, but they can't manage it for some reason- usually time or money. (Retrofitting is expensive.) Other times things just don't work out according to plan.
Things will naturally change and better, but not everything will. 'Such is life' is not an excuse, it is reality; the difference between what should and can be done. There are times when nothing can be done or there are better things to do. I never expected the server to go down and I wish my site was still up. I wish I had time to put it back on line, but the specifics of my life right now don't permit it. It is one of the lowest priority things in my life right now.
If you sell dancing bottles, then actually selling dancing bottles is more important than state of the website. The website will be updated if it will increase projected bottle sales by an amount that justifies the cost of the update. If the update is effecively free, guess what the obvious choice is. =)
-Brendan
#124981 - ScottLininger - Tue Apr 10, 2007 5:34 pm
There has been some interesting talk that Google's massive mirrors of the internet could in a funny way end up replacing the internet. Their bots crawl through billions of pages on slow, backwater servers and store archives of them (for free) on their megaservers. Why go anywhere else?
As soon as search offers more ability for archiving and version tracking, your problem will largely go away, I predict.
-Scott
#124986 - keldon - Tue Apr 10, 2007 6:21 pm
It's funny because Google are always adding new functionality based on the way people use Google. For example if you google for "about programming" (or almost "about *anything*") you are treated to an immediate definition of it based on Google's interpretation of Wikipedia.
#129333 - keldon - Mon May 21, 2007 10:39 pm
Solution: "Internet library archive"
How: In the same way *some* websites allow search bots, websites can allow bots to store their site in an online library of some sort.
But: Sounds lame, sure someone can help boost the idea into something clever'er -_-