Yes, we are sorry for this, incoming links will be mostly broken.
The problem is that in the old site we only supported a single translation in one language, so the URL just has the language and the text ID. This was short-sighted!
Now we support multiple translations of the same text in the same language, so we had to add the author name to the URL.
We did discuss ways of automatically redirecting old URLs, but found no solution, I’m afraid. We will look into it again and hope to find some solution.
Meanwhile, if you remove the language code from the URL it will give you the “suttaplex” card with links to translations, etc. Alternatively, if you prefix all the old URLs with “legacy” they should work just fine, but go to the old site.
Safari should work, so there may be a last minute bug happening here, we will look into it. But if you could update to Safari 11, web developers around the world will send you a special blessing!
The problem I have is that there are thousands of these links on dhammawheel.com, and probably other forums, embedded in user posts. Editing all of those is not a simple matter.
So if there is a software solution on SC to fall back to a valid URL, that would be very much appreciated.
Sorry, @dheerayupa, Edge is not supported as of now. Over 12,000 developers have voted for the relevant issue in Edge to be fixed, but it is still not done:
It could in principle be made to work with polyfills (i.e. lots of ugly code trying to do what a small amount of elegant code should do) but as you see this doesn’t seem to work at the moment either.
Yes, this is probably the simplest idea. Of course nothing in tech is really simple!
The problem comes from the complexity of the data, and from the fact that the old URLs simply have less information than the new ones. But the good news is that a fix doesn’t have to be 100%, as many of the more obscure texts probably have few links. So long as we can capture the majority of the main ones, we should be okay.
Our techs have discussed this, and will do more in future, but meanwhile if there’s any programmers out there who’d like a crack at this problem, let us know!
I’m not quite ready to crack on this issue, however something propped up whilst I let that question be.
Basically, it depends on how the site handles http 404 errors, and how we could use that to our benefit.
So, not knowing how it works under the wood, I wonder if it would be possible to customise the content shown on http 404 error, and possibly to do url manipulation to show (valid) url(s) that could be derived form the “wrong” url?
This way we don’t make the decision for the end user, but we can offer a panel of options on potential places he may want to go to?